My sister has lived in France for as long as I’ve lived in the US, which is around 23 years. In 2021, my Mum moved to France and bought a house with my Sister. I hadn’t seen either of them since my Father’s funeral in 2019, and with COVID restrictions lifting, I started making plans to visit them in their new home in the summer of 2022.

My Mum and Sister live in a tiny hamlet called Verthemex, which is close to Chambéry. This is in the Auvergne-Rhône-Alpes region of France, located between the Jura mountains and the Alps. It’s great cycling country with the local roads frequently used by the Tour de France and the Critérium du Dauphiné.

I started to plan the trip before the Tour de France route had been announced and mapped out all the famous climbs of the Tour de France and their proximity to Verthemex (red dot on the map below).

I’m not a climber, but it’s the mountain stages that are famous, and from photos and TV coverage I’ve seen, the views are spectacular, so I figured I should plan to ride at least one of the famous climbs. I noted that the start of the climb up Mont du Chat, which was last used by the Tour de France in 2017, and has a reputation for being one of the toughest climbs in France, is only 5 km from Mum and Sister’s house.

In late October, the 2022 Tour de France route was announced. The second week was starting with a Monday rest day in Morzine, leaving Morzine on the Tuesday for stage 10. Stage 11 was starting in Albertville – which is only an hour away from Mum/Sister’s – then stage 12 was finishing on Alpe d’Huez on the Thursday, which also happened to be Bastille day. I made travel arrangements to be in France this week, made tentative plans to watch the start of the stage in Albertville and then ride Alpe d’Huez on the Thursday to watch the stage. (I ended up in Morzine for the rest day and to watch stage 10 leave, but that’s a story for another post)

I found a number of useful websites and blog posts with information about riding Alpe d’Huez and watching the Tour de France race up it, but this blog by DC Rainmaker was by far the most useful. Ray had walked up the mountain, but recommended cycling up there to watch the race.

I arrived on the Friday following a long journey – four hour drive from Detroit to Toronto, 7 hour flight from Toronto to London Heathrow, 3 hour layover then 1h 45m flight to Geneva, then a 1.5 hour car ride to Verthemex. I’d chosen to take my Colnago V1-R with me, as it was cheaper and more convenient than renting a bike, I’d have the bike for the duration of my stay, and it’s always nice to ride your own bike rather than a rental. I decided to take my Bora WTO carbon wheels that I’d been using on my C64 rather than the aluminum Zondas I had one the V1-R. This proved to be a mistake, and more on that later.

I’ve owned a Trico Iron Case bike box since I moved to the US and it’s been very dependable for my multiple bike trips out West. I’m aways amazed that everything fits in it, and I’m always a little anxious as I compress everything together to close it up – but I haven’t damaged anything yet.

The day after I arrived, my Mum and Sister had organized a get together to meet some of their friends. This included a friend of my Mother’s called Roger, an ex pat who’s lived in France for twenty years. He was a cyclist with an e-mountain bike, and I invited Roger to join me on my day out on the Alpe.

Roger arrived on the Thursday morning, we loaded up the bikes on his hitch mounted bike rack and we left around 7 am for the 1h45m journey to Le Bourg d’Oisans, the village at the bottom of the famous climb to Alpe d’Huez.

The drive was easy and uneventful. We drove South on the A43/A41 to Grenoble for an hour – surrounded by mountains, then around 45 minutes on the D1091, which was quite scenic.

One of my concerns when planning my day out on the Alpe, was that I was uncertain how easy it would be to park, and how close to the start of the climb I’d be able to park.

Almost from the moment we turned off the A41 onto the D1091 to Bourg d’Oisans, we started to see groups of cyclists, large and small, presumably heading to Alpe d’Huez – like cycling pilgrims making the journey to cycling Mecca, which I suppose is exactly what it was. There were many more parking at the side of the road, from around 20 miles out getting ready to ride. This was reaffirming my concern about parking in Bourg d’Oisans.

We decided that we were early enough that if we couldn’t find parking in Bourg d’Oisans, we could always backtrack the way we came and park at the side of the road as others had.

Realizing we were approaching the town center, we turned down one of the streets off the main road to see if we could find a somewhere to park. A short distance down, just at the side of the road we found a collection of camper vans and cars parked next to a stream. It appeared they were all there for the same reason we were. We parked next to a young French couple who were getting changed into cycling clothes to ride up the Alpe, and they confirmed it was ok to park there. We unloaded the bikes and got changed.

We were parked less than half a mile from town, and less than a mile from the start of the climb.

Once changed, we headed into town for a quick coffee.

…then set off for the climb, by which time it was around 9:45.

The road out of town was busy with cyclists and walkers making their way to climb up the Alpe. Barriers and sponsor banners served as a reminder that this was the road the race would take to the climb.

Following other cyclists and foot traffic, we continued across a bridge over the Romanche River and took the second exit from the roundabout onto the Route de l’Alpe d’Huez. I’d been told that the roundabout marked the start of the climb, but it didn’t register that this was that roundabout – because after the roundabout the road was still flat. After about a mile and a half the road started heading upwards and road painting had started – with flags and messages to the riders – and I realized, this is it, I’m climbing Alpe d’Huez.

It was around 10 am and the riders weren’t expected on the climb until sometime after 5 pm. However as the photos show, the road was already teaming with walkers and cyclists making their way up the climb.

Tour de France stages start in the early afternoon, usually between 1:00 or 1:30 pm, presumably to maximize live TV viewing around the world. The organizers estimate the pace of the stage with slow, medium and fast pace estimates, and they select a start time so that the stage will most likely finish around 6:00 pm. This particular stage started at 1:05 pm.

Each year the organizers publish a ‘Roadbook’ for the race, which is 200+ pages long providing details on the teams and detailed information on each of the stages including timing. They also publish route and timing information on the official Tour de France website (letour.com) and the le Tour app.

Here is the route and profile for stage 12 that was finishing on Alpe d’Huez:

The timings for the stage cover two pages, here is the second page showing expected times to reach Alpe d’Huez:

The Roadbook is available to buy in paper form but sells out quickly. There is also a smart phone app available to download, and I found a pdf on the Internet. Here’s a QR code to download the app:

Continuing up the start of the climb, I passed the Beefeater Bend walking up the climb, pushing their gear in a shopping cart. They’re a group of British cycling fans who organize a party on one of the bends of Alpe d’Huez (or on another major climb in years that the race doesn’t climb the Alpe). Their slogan is ‘Probably the best party at the Tour de France’. On this day they set up their party on bend 20.

Past the first of 21 hairpin bends, the road continued up and was very busy.

Although the road closed to cars at 10 am, hundreds of team and official vehicles were still making their way to the finish at the top of the climb.

Each of the bends are marked with a sign counting down from 21 to 0, and each sign carries the name of a previous winner of a stage ending on the Alpe (the bottom 10 bends having two winners names as Alpe d’Huez has featured in the Tour de France 31 times).

The climb to Alpe d’Huez is 13.9 km (8.6 miles) long with a total elevation gain of 1,118m (3,668 ft). The average gradient is around 8% with a short section up to 13%. The bottom of the climb between bends 21 and 16, is the steepest at around 10%.

From the outset I never intended to try to ride the climb within any set time, or to achieve any other goal. I was there to enjoy the experience, soak up the atmosphere, admire the scenery and take plenty of photographs.

My Colnago V1-R also had very high gearing for a big guy on a climb such as this, with a standard 53/39 crankset and 11-29 11 speed cassette, so I knew I was in for a hard ride.

I’d stop at most of the bends to take it all in, catch my breath and snap a couple of photos.

Once we got to the top, we found that the road to the finish line was barricaded with the Gendarmerie stopping people from passing.

There were hundreds of cyclists gathered by a roundabout, supposedly trying to figure out what to do next, and with the Gendarmerie asking them to move along.

It was around noon, two and a half hours since we started out, of which I’d been riding for 1 hour 40 minutes. The weather was hot, the sun was blazing, and there was barely a cloud in the sky.

We decided refreshments were in order.

Although the race wasn’t due to come up the Alpe until some time after 5, I wasn’t sure how difficult it would be to find a good spot to watch the race from, and we needed to find somewhere with good shade. No way I was going to sit in the sun for over four hours. So after an ice cold beer we headed back down the mountain – in no particular rush – to find a good vantage point where we could hang out for a few hours and watch the race.

The descent was fun, and I felt like a kid with a new toy.

At ‘Dutch Corner’ which is bend 7, the party was in full swing:

A great video here of the madness later in the day:

Deciding that Dutch Corner would be too busy and too crazy, we continued down.

Between bends 8 and 9 there seemed to be plenty of shade and space, so we decided this would be a good spot and ‘camped-out’ here:

Roger had everything planned out. Deciding the trip was going to be ‘first class’ rather than ‘economy’ – he packed and carried a lunch for each of on his e-bike, including a small bottle of Bordeaux and Champagne each.

With plenty of time to kill, I rode down to bend 9 to see if there was anything interesting happening, and found everyone in good spirits

Back up the hill, cyclists, team and officials cars continued to make their way up the Alpe, and the crowds continued to grow.

Sometime shortly after 4:00 pm, the Caravan passed by.

My phone reception was not good on the Alpe, and my battery was running low, so it wasn’t possible to follow the race, however the was a rumor circulating through the crowd that Chris Froome was in a breakaway with Tom Pidcock. Apparently Tom Pidcock had put on a display of his descending prowess on the Galibier

https://youtu.be/3f4Pp4oYh28

We realized the race was close when we could hear helicopters overhead.

At around 5:55 motorcycles started to come through, leading the car of the race officials followed by the cyclists.

I’d been undecided what to do about taking photos of the race. I’m a keen photographer and had taken my DSLR with me to France, and in the car to Bourg d’Oisans, but I decided I didn’t want to have to carry that up the climb and that I’d make do with my iPhone. I didn’t want to get in the way of the riders, so decided to hold my iPhone waist high and point and hope, as Roger captured me doing here:

Overall, apart from a few misses, and glare from the motorcycle’s headlights, I got some decent photos.

As soon as the last riders and team cars had passed, it was time to ride back down the mountain to make the journey home.

Unfortunately, all that braking overheated the carbon on the brake track of my front wheel, causing it to start to delaminate. By the time I got to the bottom of the descent, my front wheel was out of true and rubbing against the inside of the fork, andI could feel vibrations under braking.

One thing that sets Garmin cycling computers apart from other manufacturer’s units, is the ability to personalize a compatible Garmin head unit through the Garmin IQ store. This offers thousands of apps, widgets and data fields with more being added all the time. Most are free, some require a small payment, others offer free versions with limited capability with the option to upgrade to a more capable version for a small fee – normally in the form of an annual subscription.

The image below is a screenshot of a small selection of some of the apps, widgets and data fields available for the Garmin Edge 1030 from the Connect IQ store.

Amongst there shown here are:

• Komoot. An app that integrates a 1030 unit with a user’s Komoot account (Komoot being a popular route planning and navigation application)
• Trailforks. An app that provides access to Trailforks off road maps and navigation, particularly useful for mountain biking, hiking etc.
• AccuWeather. A widget that provides current weather information and forecast information
• Starbucks. An app that allows a user to add their Starbucks card to their Garmin so they can pay for purchases from their Garmin unit
• Strava Relative Effort. A data field that displays relative effort as calculated by Strava so this can be monitored during a workout

An app is an application that is run from the IQ Connect menu on the head unit (Komoot for example) or a data screen that is displayed continuously during a workout. A widget is one of the secondary data screens in the background that can be accessed momentarily e.g. AccuWeather weather forecast information. A data field is one of the items displayed on a data screen during a workout – speed, heart rate, distance are all examples of data fields. Below is a typical data screen showing 6 data fields on my Garmin 1030.

A popular data field is SPH Graph, which is shown below. This is a data field that takes up an entire data screen.

The SPH Graph data field attempts to display a maximum amount of data on a single screen and displays select high priority information (in this case speed and heart rate) as an analogue gauge. I’ve tried similar layouts, but found them too cluttered and difficult to read.

There are two IQ data fields I’ve used for over a year that I’ve found useful and wanted to bring to your attention.

My Bike Radar Traffic

I bought a Garmin Varia RTL515 at the same time I bought my 1030 in 2019, and I’ve found it to be indispensable for road riding.

The Varia RTL515 is a rear light, but it also uses radar to track cars approaching from behind and relays the information to compatible Garmin (and Wahoo) cycling computers & watches. The relative positions of approaching cars is displayed on head unit which beeps to warn of an approaching car and then beeps again when the road is clear.

The photo below shows a typical display on a Garmin 830 computer connected to a Varia device. The red bands provide a warning that a car is approaching from behind, and the white dot on the left tracks the location of the approaching car relative to the cyclist’s position.

Data from the Varia is displayed in real time and once cars have passed, the data is gone and not accessible.

Adding the free My Bike Radar Traffic data field to an active data screen on a compatible Garmin device, stores Varia data in the head unit. This tracks vehicle location, speed and vehicle count. After the ride is complete, the data is available on Garmin connect, on the Garmin smart phone app or on the mybiketraffic.com website.

Once the ride is uploaded to the mybiketraffic.com website, the location and speed of each passing vehicle is shown.

I find this data useful to identify how busy certain roads are that I ride and how fast the traffic is moving. If I’m ever unfortunate to be hit by a car approaching from behind, this will give me the evidence I need to show how fast that car was traveling when it hit me. I discovered it will also help to determine the location of a Varia that becomes detached from its mount and lost during a ride. Vehicle count will stop incrementing when the Garmin head unit is out of range of the Varia and the route is color coded on the website to show where the Varia was enabled or disabled (i.e. connected or disconnected).

WindField

Another data field I have loaded into my Garmin 1030 and use routinely is WindField.

WindField is a data field that shows wind direction and wind speed (in addition to other weather information, depending on the subscription level and width of the field used to display the data field on the Garmin). While there are other weather widgets available for Garmin, what makes WindField useful is that the wind arrow will point in the direction that the wind is blowing, and the arrow also takes into account the direction that you are traveling.  So if you are riding East with wind out of the North, the arrow will point to your right. I find this to be useful on a ride, to know whether I’m riding into a headwind or tailwind, and what to expect later in the ride. It’s also useful to know whether it’s worth attempting to beat my time on any Strava segments based on wind direction as I approach the segment.

There is a free version of WindField that will update weather information once per hour. A $7 annual ‘Standard’ subscription will update every 15 minutes. The PRO version has a ton of extra features and will update every 5 minutes for $12 per year. If you want the absolute most accuracy possible or want to provide extra support for the app, subscribe to the $30/year Ultimate version that averages several sources together.

In order for WindField to work, the Garmin head unit needs to be paired and connected to a smart phone running the Garmin Connect app.

My Windsock

While not a Garmin app, if wind conditions are important, especially if you’re trying to take a KOM/QOM or PR on Strava, then MyWindsock.com is worth knowing about:

My Windsock is a web based application. It links with Strava, Komoot, Ride with GPS, or rides can be manually uploaded. Following a ride, the weather conditions, and in particular, wind speed and direction, can be reviewed for the entire ride or particular sections, with tons of data analysis including interesting features such as ‘Feels Like Elevation’ whereby it estimates how high you would have climbed if the effort you put into riding into a headwind was a hill. As an option, it’ll add details of the weather and wind conditions to the description of the ride in Strava.

In addition to reviewing previous rides, My Windsock can be used to analyze and estimate times over Strava segments and help plan for future rides. For example, below is one of my favorite Strava segments ‘Levan Knoll Climb WB’ on Hines Drive. My Windsock loads in all the pertinent information for the segment such as distance, direction, elevation profile etc., I pick a time and a date from the drop down menu, then My Windsock will estimate the elapsed time for the segment based on rider & bike weight, average power (allowing for different power levels for flat roads, climbing and descending) and aerodynamic drag (CdA) based on riding position. Power and CdA values can either be entered manually, or My Windsock will calculate values based on past rides.

This allows the average power requirement to be determined to take a KOM/QOM or beat a PR on a particular day and time based on the weather forecast.

As if this wasn’t enough, My Windsock will track up to five user selected Strava segments (windsocks) and send e-mail notifications like the one below to alert to upcoming wind conditions that are ideal for a fast time on that segment:

As I excitedly awaited delivery of my new espresso machine and coffee grinder, I immersed myself in YouTube videos to learn how to make espresso and its many nuances. I was surprised how it was more involved than I expected.

Undoubtedly, the best source of good information and well made videos was James Hoffmann.

My delivery arrived. I unpacked and set up the machine and grinder on my kitchen counter. The machine was very heavy weighing a reported 55 lbs. The coffee grinder was also very heavy (12 lbs) due to its diecast construction, and it was certainly built to last.

Water quality can be an issue with espresso machines. Hard water containing high levels of soluble minerals can cause excessive scale buildup inside the machine, but pure water without any mineral content can impact the taste of the coffee. The Specialty Coffee Brewers Association of America recommends total dissolved solids (TDS) in the range 75 – 250 mg/l. A TDS meter was included in the starter kit I received with my machine, and my tap water measured 115 mg/l. I figured this was good to use out of the tap and I didn’t need to use a water filter or bottled water. Incidentally, I also measured the TDS content of water from my refrigerator after fitting a new water filter, and the TDS levels measured about the same as the tap water.

I filled the Izzo’s 3 liter water tank, switched it on, heard the pump run to fill the boiler with water and waited for it to heat up.

The boiler temperature was preset at 252°F, as was displayed on the control panel on the front of the machine.

With a heat exchanger espresso machine such as this, the temperature of the boiler is set to boil the water and produce steam, which is then used to steam milk or provide boiling water to make tea. However, if this boiling water were used to make espresso, the coffee would be over extracted which would lead to bitter and burnt tasting coffee. Heat exchanger machines have a second circuit for the brew water. This circuit passes through the boiler and the water in the boiler heats the brew water through a heat exchanger. The brew water circuit feeds the brewing group (an E61 group in the case of my Izzo Vivi – which is the most common brewing group found in prosumer semi-automatic espresso machines).

As the water cools at the brew head, its density increases and this cooler water falls downwards in the circuit returning to the inlet of the boiler. This sets up a thermal syphon whereby water circulates through the brew head keeping it warm.

Due to heat losses in the heat exchanger circuit and at the brew head, the temperature of the water that reaches the coffee for extraction (brew temperature), is much lower than the boiler temperature – typically around 50°F lower than the boiler temperature.

It’s difficult to measure brew temperature directly. A device called a scace allows brew temperature to be measured, but these devices are expensive and are used to set up an espresso machine, not to make coffee.

My machine came with a brew head thermometer installed. This provides a means to measure water temperature close to where the water meets the coffee grinds, and while the temperature measured here can still be around 7°F higher than at the coffee, it provides a better estimate of brew temperature than merely assuming the temperature will be 50°F lower than the boiler temperature.

Boiler temperature can be adjusted, and providing the temperature exceeds the boiling point of water at the pressure in the boiler, steam will be produced. The higher the temperature, the higher the steam pressure. Changing the boiler temperature also changes the brew water temperature, this allows brew temperature to be optimized to achieve the required coffee extraction while still providing steam in the boiler. The PID controls the boiler temperature to maintain the desired temperature.

Switching my machine on from cold, the boiler reaches its setpoint temperature (currently set at 252°F) in around 10-15 minutes, but it takes up to 40 minutes for the brew head to reach a stabilized temperature (typically around 210°F measured by the brew head thermometer on my machine).

Initially, I saw this as a drawback of using an espresso machine. I was used to switching my Keurig machine on in the morning and having a coffee 2 minutes later. Also I could program the Keurig to switch on and be ready to brew in the morning, there are no such timers on these traditional semi-automatic espresso machines. I wasn’t prepared to wait 40 minutes for my morning coffee. I got round this by using a Wyze smart plug that I have set to switch the machine on at 5:00 am and off again at 6:30 pm. All I have to do then is to make sure there’s sufficient water in the water tank.

There are two pressure gauges on the front of my Izzo Vivi.

The gauge on the left shows the boiler pressure, which is a function of boiler temperature. The gauge on the right displays the brewing pressure during espresso extraction. This should be around 9 bar, and it was set correctly as received. I’d seen reviews of other machines where the brewing pressure needed to be adjusted on new machines, which in some cases involved removal of the casing to gain access to an adjustable pressure relief valve. I was pleased I was saved this trouble.

Onto making coffee.

A portafilter houses a basket which is filled with ground coffee. Ground coffee is evenly distributed across the basket and compressed (tamped) with a tamper forming a ‘puck’ of densely packed coffee. The portafilter is then fitted to the machine, a lever actuates valves in the brewing group which directs the flow of hot water through the coffee. The water flows under pressure created by the machine’s pump and resistance created by the small holes in the basket and the densely packed coffee. Features in the brewing group promote an even flow of water across the entire area of the coffee puck. The goal is to achieve even flow across all the coffee to evenly extract the flavor from the coffee.

Water will take the path of least resistance through the puck, so it will form channels through less dense regions of the puck leading to under extraction. To achieve even extraction the grinder should produce consistently sized grounds which should then be evenly distributed throughout the basket and tamped evenly.

My Izzo machine was supplied with two 58 mm (standard size) portafilters, one for a single shot with a single spout and small basket and one for double shots with twin spouts and a larger basket.

Espresso is commonly made with a double shot portafilter, the single shot portafilters are rarely used and single shot espresso is more difficult to make well consistently. The starting point is to grind, distribute and tamp 18 g of coffee in the portafilter. From there, the brew ratio (weight of the liquid coffee produced : weight of the ground coffee) can be altered depending on the desired strength and mouth feel (a lower ratio will produce a smaller quantity of coffee that tastes stronger and is more viscous). A brew ratio of 2:1 is often recommended as a starting point so 18 g of ground coffee would yield 36 g of liquid coffee.

Extraction time, which is measured as the time period from initial actuation of the brewing group and pump by throwing the lever, to the time when the desired mass of coffee has been extracted, should typically be in the 25 – 35 seconds range. This extraction time is achieved by adjusting the grind. If it takes less than 25 seconds to extract 36 g from 18 g of ground coffee then the coffee is ground too coarse which reduces the resistance to the water flow. This results in under extracted coffee, which can taste sour, salty and will lack sweetness. On the other hand, if it takes longer than 35 seconds then the grind is too fine and this results in over extracted coffee which can taste bitter, astringent and hollow/empty. Somewhere between the two the coffee should exhibit sweetness and taste fresh with complex acidity and have a long finish. Between the guideline 25 – 35 seconds, brew time and grind setting may be down to personal preference and taste for the coffee being used. Brew temperature can also impact the flavor profile but providing brew temperature is between 195 – 205°F it will have less of an impact on the flavor profile than the grind, puck preparation and brew time, and is normally the last variable to be adjusted, if it needs adjustment at all.

Different coffees will require different grind size depending largely on the density of the beans, and density will change based on whether it’s a light or a dark roast, the altitude the beans are grown at, and the process that’s used to roast the beans. So when switching to an entirely new coffee, it will take time a trial and error to dial-in the grind to achieve an ideal brew time. This is much easier with a fixed mass (18 g) of ground coffee and brew ratio (2:1) to reduce the number of variables.

So armed with the theory, it was time to put this into practice with my own machine.

It is accepted that the coffee grinder can have more of an impact on the quality of the coffee than the espresso machine. A grinder needs to be able to deliver coffee ground to the right grind size and the size of the particles should be consistent. The grinder I bought was an Macap M2D, which was recommended over other grinders I had been considering.

This is a burr grinder that uses 50mm flat burrs. With this kind of grinder, the coffee beans are crushed and cut between two rotating parallel burrs. Grind adjustment is achieved by moving the burrs closer together or further apart to alter the gap between them.

The particular model I have is an on demand type, meaning that it will grind as long as switch is depressed while the portafilter or dosing cup is held under the outlet. There’s an adjustable timer that can be set to achieve the required dose, but I haven’t found this to be useful as I weigh just the quantity of beans I need for each espresso I make and grind until all the beans have been ground and there is no more ground coffee coming out of the grinder. One notable feature is that this grinder has a worm drive microadjustment, which provides stepless fine adjustment of the grind size.

I’ve mentioned an 18 g coffee dose and 2:1 ratio several times, to achieve this and to be consistent, accurate measurement of weight is important when making espresso. As such, there are numerous scales available specifically designed for coffee making. The scale I bought was a TIMEMORE Black Mirror Basic Plus coffee scale. It measures to a resolution of 0.1 g and is low profile enough to fit under the brew head with an espresso or latte cup. It charges with a USB-C cable, but a charge seems to last several months.

To make a coffee, I start by weighing 18 g of beans which I then load into the hopper of the grinder. I grind until all the beans have been ground and load into the portafilter.

Initially I was grinding straight into the portafilter, but I found that ground coffee would spill out, both reducing the coffee dose and creating a mess on the counter. To counter this, I bought a portafilter dosing funnel.

This sits on the portafilter as shown, and helps prevent ground coffee from spilling out when coffee is ground directly into the portafilter. This reduced the mess, but I found I would still spill ground coffee when I removed the funnel and went to distribute the mound of coffee.

I’m currently using a dosing cup. I grind into the cup and then empty the cup into the portafilter. This seems to work well.

After distributing the ground coffee in the portafilter, I tamp with the portafilter on a tamping mat. The amount of pressure isn’t as important as ensuring consistant tamping pressure from shot to shot. The more pressure used to tamp the coffee the denser the coffee puck and the longer the brew time, therefore varying the pressure will introduce inconsistency in the quality or flavor of the coffee shot to shot.

When a heat exchanger espresso machine is left idle for some time, the water that circulates around the heat exchanger circuit with the thermal syphon becomes superheated, approaching the same temperature as the boiler. This water is too hot to make good coffee. To deal with this, it’s necessary to perform a ‘cooling flush’ to flush out all of the superheated water and replace it with water at the correct temperature. I do this without the portafilter fitted. When the valve is first opened and water begins to flow, the water will sputter accompanied by a hissing sound. This indicates that the water is flash boiling. Eventually the sputtering and hissing will stop and there will be a uniform water flow. At this point the water should be allowed to run for a further 5-6 seconds to reduce the temperature to what should then be the ideal brew temperature. This should be done every time the machine is left to idle for longer than 5 minutes, and even if left for less than 5 minutes a shorter cooling flush should be performed.

At this point we’re ready to ‘pull a shot’ of espresso, as it’s termed. To do this, I install the loaded portafilter in the machine, set the coffee scale on the grid of the drip tray and place a cup on the scale underneath the portafilter spouts.

After switching on and taring the scale, I pull the lever to start the extraction. One nice feature of my Izzo Vivi is that it has a built in automatic shot timer that starts as soon as the lever is pulled. It typically takes 6-8 seconds of the pump running before coffee starts to flow out of the portafilter, and I watch the scale and stop the extraction when I reach 36 g. At this point I look at the timer to see how long the extraction has taken. If the extraction took less than 25 seconds or longer than 35 seconds, I’d pour away the coffee and remake it with the grind setting adjusted to give me a brew time of between 25 – 35 seconds. If it took too long I’d change to a coarser grind and if the brew time was too short, I’d change to a finer setting.

When the lever is returned to the off position at the end of an extraction, then water is released from the bottom of the brewing group and into the drip tray. This is to release the water pressure. Some of this water typically puddles on the coffee scale, so it isn’t possible to stop the extraction and observe the final weight of the coffee yield. The weighing process at the end extraction is very dynamic, the weight is increasing steadily and quickly, and then increases sharply when the lever is returned to the off position and the water is dumped out of the groupset. The shot timer on the other hand, displays the total time for a few seconds before reverting back to displaying boiler temperature.

The result should be a cup of espresso covered in a 1 to 2 mm thick golden to reddish-brown layer of foam on top. This is referred to as crema, which is the result of a combination of coffee bean oils, carbon dioxide, and tiny coffee grounds that form as the espresso is brewed under high pressure.

Espresso naturally separates into different layers during brewing, with the crema on top, the body in the middle, and the heavier components settling at the bottom. Stirring helps to integrate these layers, ensuring that each sip is uniformly flavored.

A well-made espresso should offer a complex and robust flavor profile that is both intense and balanced.

A good espresso should have a rich, deep coffee flavor. The taste should be potent and full-bodied, often described as creamy or thick compared to regular coffee due to the concentrated nature of espresso.

While espresso is characteristically bitter due to the concentration of flavors and the extraction process, a well-made espresso will also have a noticeable sweetness. This sweetness comes from the natural sugars in the coffee beans that are caramelized during roasting.

A good espresso should have a pleasant acidity that brightens the flavor, rather than overwhelming it. This acidity is often described as crisp and can have a slight fruity or winey characteristic, depending on the coffee beans used.

Depending on the coffee blend and roast, espresso may exhibit notes of chocolate, nuts, caramel, fruits, or spices. These flavors should be discernible but well integrated, creating a layered tasting experience that evolve with each sip.

The aftertaste, or finish, of an espresso should be pleasant and lingering. It should continue the flavors of the drink without turning overly bitter or sour, encouraging another sip.

The crema should add a slightly sweet and slightly savory flavor that enhances the overall cup. It should be aromatic, adding to the complexity of the espresso’s taste.

Enjoy…

I’ve always enjoyed good coffee, and as coffee and cycling have long enjoyed a healthy relationship I thought I’d write about my experience buying and my first three months using an espresso machine at home.

I enjoy a post ride latte in Tuscan Cafe in my home town of Northville, MI, where I’ve been a regular customer since 2000. I would visit there with the family at the weekend, or grab a coffee on the way to work at least once a week.

At home I’ve used a Keurig machine for the past ten years, which I always thought made decent coffee and was very convenient. I’ve also used a cafetière (French press for my American friends) and a typical American drip coffee machine in the past, these are certainly less convenient when you just want a single cup.

In December 2001, my wife mentioned she’d been considering buying me an espresso machine for Christmas. The one she’d been looking at was expensive ($1,600) and she didn’t want to spend that much if it wasn’t what I’d want. As I’m better at researching such things, she suggested I should do the research.

An initial Google search came up with this article, which was interesting in that it suggested that espresso was best left to the experts in cafes and shouldn’t be made at home, but then went on to give some good advice.

An initial search for espresso machines in a similar price range to the machine my wife had been considering (c. $1500) came up with the Rocket Appartamento. Regular retail price was $1,850-$1,900 but there were deals available for around $1,750.

This seemed ideal. It looked the part – I wanted a traditional looking Italian espresso machine – and with my limited knowledge at the time it fit the bill and would do the job.

I discovered that this kind of machine is referred to as a semi-automatic, prosumer machine, meaning that it doesn’t grind the beans and prepare the coffee for extraction as a fully automatic machine would, and that it shares similarities with the commercial machines you’d see in cafes, but is intended for home use.

Further research taught me about the different kinds of espresso machine, which essentially fit into three categories: single boiler, double boiler and heat exchanger.

Espresso machines essentially have three functions 1) brew coffee by passing ‘brew water’ of between 195-205 degF through a prepared ‘puck’ of coffee under pressure (nominally 9 bar) 2) provide steam to heat and texture milk to prepare latte and cappuccino based drinks 3) provide boiling water to make drinks such as hot tea.

The Rocket Appartamento is a single boiler machine. As they are the simplest of the semi-automatic machines, with fewest parts, they are the cheapest typically costing around $1,800 – $2,000.

With a single boiler, water is heated in the boiler and the same water is used as the source for steam, boiling water for tea, and to brew coffee. To produce steam, the boiler must heat the water above boiling, but if boiling water is used to make espresso, then the coffee will end up over extracted which would make it taste bitter. Therefore with a single boiler machine, the boiler temperature must be changed between brewing coffee and steaming milk.

This introduces inconvenience and increases the time required to make a milk based espresso drink, and certainly a nuisance when making multiple milk based drinks back to back.

At the other end of the scale (and budget, costing $2,500 and up) is a double boiler machine. Here two boilers are regulated to different temperatures, one boiler provides steam and boiling water, and the second boiler provides brew water that can be set to a lower, optimal temperature for coffee extraction. Commercial espresso machines have two and sometimes three boilers to provide the correct temperatures to steam milk and brew coffee and to keep up with the demand of continual use.

In between single and double boiler espresso machines are heat exchanger machines. These provide a cost effective compromise costing in the range $1,900 – $2,300. A heat exchanger machine has a single boiler that provides steam and hot water, but a brew water circuit passes through a heat exchanger located inside the boiler. This heats the brew water to a temperature lower than the water in the boiler. This allows the water in the boiler to be set to produce steam, while providing brew water at a lower temperature optimum for coffee brewing. It also allows milk to be steamed and coffee to be extracted simultaneously.

Now that I understood the differences between the types of machine, I decided that I didn’t want to put up with the inconvenience of a single boiler machine, but didn’t need to extra expense of a double boiler machine so I was looking for a heat exchanger machine.

I found there was plenty of useful information on the internet. The two top online retailers in the US seemed to be Whole Latte Love and Clive Coffee both of which had extensive stock and provided lots of useful information both on their websites and and their own YouTube channels. There were also plenty of other YouTube channels providing lots of information about espresso making and espresso machines, one of the best being James Hoffman, who I think has one of the best YouTube channels I’ve come across.

Based on the reviews, the Profitec 500 seemed to be one of the best heat exchanger machines, and it was readily available (many weren’t, I assume due to COVID related parts shortages) and could be had for around $2,000.

To control the water temperature in the boiler of an espresso machine, there are two options. The basic option is a thermostat that switches the boiler’s heating element off when a preset temperature is reached, the water then naturally cools and the thermostat then turns the heater back on. This cycling of the heating element causes fluctuations in water temperature. The other option is to use a PID controller, which maintains the water at a more consistent temperature. Most of the heat exchanger machines I was looking at either had a PID controller as standard or were available as a slightly more expensive PID equipped model. As water temperature is one of the variables that affects the taste of the coffee, it seemed I should buy a machine with a PID for temperature control. The Profitec 500 was offered as a PID model:

The only thing I didn’t like about the Profitec, was that it was made in Germany. While that could be a good thing as I’ve always respected German engineering, the sentimentalist in me wanted an Italian espresso machine.

After more searching, I came across the Bezzera Magica which is a heat exchanger machine that sells for around $1,850 with a PID. Bezzera is one of the oldest espresso machine manufacturers dating back to 1901. Their founder, Luigi Bezzera patented a number of improvements to the espresso machine as is credited with inventing the modern espresso machine. Bezzera hand make their machines in a factory in Milan. Nothing seemed more Italian than this, the specifications seemed to fit my needs and the price was good.

However, it seemed there were none in stock in the US (again assuming this was due to COVID parts availability issues).

During my search, I came across an online retailer called Italian Bean Delight, they had Bezzera Magica machines listed on their website so I contacted them to ask about availability. They were very responsive, and I spoke to the owner Andreas later that evening. He explained that in addition to his retail company he also owned a company that serviced and repaired espresso machines. He advised me against Profitec machines as he’d seen too many at his service company with corroded boilers. He explained that while the boilers are made from stainless steel, they suffer from galvanic corrosion at the fittings. In his opinion, when paying $2,000 for a machine I should expect it to last 20 years and none of the Profitec machines would last that long due to the galvanic corrosion issue. He also advised against the Bezzera machines, stating their quality wasn’t what it once was. Instead he recommended an Izzo Vivi machine, one I hadn’t come across during my search.

This was a PID machine and also hand made in Italy that retails for $1,890. I ended up negotiating what seemed to be a very good price on the machine and a Macap M2 wormdrive coffee grinder (that Andreas highly recommended – with a normal retail price of $769). Andreas also included everything I’d need to get started (two latte cups & saucers, a milk thermometer, milk frothing jug, water hardness tester and 1 kilo of Italian espresso coffee beans). Also, due to a delay in shipping, Andreas included a Geisinger brew head thermometer worth $145. I got all this for $2,500 including delivery and highly recommend Andreas and his company, Italian Bean Delight.

So my C64 frameset arrived the week after Thanksgiving 2021, which was earlier than I was expecting and after only a four month wait. I took everything out of the box. The frame was nicely packed with a styrofoam piece at the top that held the fork and saddle clamp and kept the frame in place during transit. The seat post was in a box of its own and there was an accessories box and two water bottles and bottle cages I’d ordered with the frame. These were loose in the box but unlikely to move due to all the packaging.

There were a couple of things I noticed. The most concerning was that the brake bosses seemed to be missing from the fork, as though the inserts hadn’t been installed.

Also, the accessories box that contained all the fittings I needed to build the bike was for a disc brake frameset.

This mattered, because it meant I didn’t have the fittings needed to fit the rear brake cable through the top tube. Also it didn’t have the compression fitting for the fork/headset (it’s moulded into the fork for disc brake framesets) and the stem cap and bottom spacer seemed to be for a Colnago R41 stem. With the C64 disc frame, the brake hoses can be hidden inside the bars and stem, but it appears that a R41 stem is required for this.

One nice touch is the Colnago bottle cage bolts that are provided with the frame:

I contacted Maris from Brava Cycles where I had bought the frameset, and he explained that Colnago had started sending out forks with the brake bosses painted over. I just had to dig out the paint covering the holes with a small screwdriver. It came off easily and cleanly in one piece. He also apologized that I’d been sent the wrong accessories box and he sent out the rim brake accessory set I needed.

I’d bought a set of direct mount brakes and a Prologo M5 saddle off eBay, Deda Superleggera bars and a Superleggero stem from Chain Reaction and Merlin Cycles respectively while I was waiting for the frame. I started the build with the components I had, along with the wheels from my V1-R. The plan was to use these wheels for the C64 and buy another set, probably Campagnolo, but model TBD, for the V1-R.

One thing that’s unique with the C64 frame is that the bottle cage on the downtube is fitted into a recess for some reason – referred to by some as a ‘swimming pool’. This limits the bottle cages that can be mounted without the use of spacers, but the Colnago water bottle cages I bought with the frame fitted without a problem and without spacers. The tag with the serial number is riveted here also.

The seat post clamp is a wedge inside the seat tube, which seems to work very well, and it’s tightened by an M4 bolt underneath the top tube, which is very neat.

Next I had to decide what components to buy, my short list being Campagnolo 12 speed Record or Super Record. Campagnolo’s electronic shifting (EPS) is only offered with Super Record, it would have been significantly more expensive than a mechanical groupset. The reviews I’d read were praising the 12 speed mechanical groupset while reviews of the EPS groupset were a little lackluster. After some deliberation I bought a set of Super Record 12 speed shifters from Merlin Cycles (at a price similar to Record at other online retailers). I found really good pricing at Pro Bike Kit and bought Super Record front and rear derailleurs and a Super Record chain. A new cable set was bought from Excel Sports, although I later realized that the shifters included cables.

All these parts and the rim brake accessory box arrived in time for the weekend, which happened to be a weekend where my wife and daughter were away, so I had plenty of uninterrupted time to continued the build, and could do so in the warmth and comfort of the family room.

On both my CX-1 and V1-R the shifting has never been perfect, there have always been a couple of sprockets where I could never get the upshifts and downshifts dialed in. It would hunt in either direction – add more cable tension and it would shift to a larger sprocket but hesitate when shifting to a smaller sprocket, and vice versa. This could be due to cable friction, so I was going to ensure that this wouldn’t be an issue with my C64.

A member (who went by the name ‘Calnago’ – now sadly passed away) had written a great and very extensive thread on the C64 frame and build on the Weight Weenies Forum which I found very informative and helpful in researching my build.

One thing he covered in length was the cable routing around the head tube. With the C64 (and other current Colnagos – V3R, G3X), the derailleur cables are routed internally through the downtube entering on top of the downtube towards the front. According to Colnago, the cable for the rear derailleur enters from the right (right shifter to right hand side) and the front derailleur cable enters on the left (left shifter to left hand side). On the forum, Calnago had an issue with this because 1) the cable hosing would rub on the head tube and damage the paint (and he didn’t see how the rubber frame protector on the cable would stop this) 2) this routing would lead to a tighter bend in the cable that could increase cable friction.

His preferred approach was to route the cable from the right shifter to the left entry to the downtube (and vice versa) and let the cables cross in the downtube. The benefits he claimed were that this would avoid the cables rubbing on the headtube and the wider bend would reduce cable friction.

Incidentally, that is what’s done in this build (see 5m:25s in):

I routed the cables this way at first, but I didn’t like it. Despite keeping cable length to a minimum while ensuring the bars could turn without pulling on the cables, there was too much loose cable at the front of the bike. It didn’t look good and there was too much opportunity for the cables to get caught by something, or for my knees to hit them, which would be annoying.

I reverted back to the standard routing using Jagwire Mini Tube Top protectors, which I thought was a much more elegant solution (I would have liked to have used the Colnago cable protectors as shown on the white and black C64 above, but couldn’t find any).

One benefit of rerouting the cables, is that it gave me opportunity to check that the cables weren’t dragging on anything and cable friction was at a minimum. The cables ran freely. I paid particular attention to where the inner cable entered and exited housing that had been cut. I’d made the ends of the cable housing round again after cutting by inserting a bradawl, but sometimes the inner cable can still feel tight. They were nice and smooth now passing freely into and out of the housing. Also I found it better to insert the housing into the shifters before feeding the cable through the internal routing of the Deda handlebars. This ensured that the cable housings were seated and installed in the shifters properly – any movement between the housing and shifter could also affect shifting consistency

The bottom bracket cable guide is moulded into the carbon on the C64, which is a nice solution, but requires nylon tubing to prevent the cables from digging into the carbon and it also helps keep dirt away from the cable. There was no explanation in the Colnago assembly manual about it, although some tubing was included in the accessories box for this purpose. I followed the recommendation from the Weight Weenies thread, and purchased some tubing from Amazon which worked very nicely.

The last note on the subject of cable routing is that Colnago included a Jagwire cable end with an extension in the accessory box. It wasn’t obvious to me where this would be used. It’s used where the rear derailleur cable exits the chain stay with the extension inside the chain stay, and is a nice way of preventing the ingress of dirt into the cable.

Brake cables were easily installed, the only thing remarkable was that the fittings that act as end stops for the cable ends in the top tube, clip in to the frame. I was worried about either breaking them or damaging the frame. There were no issues but there seemed to be more movement with the rear fitting than I would have expected.

One of the more difficult decisions I had was around the chainset. A Super Record chainset is ridiculously expensive and twice the cost of a Record chainset. The technical differences are that the Super Record chainset uses a titanium axle compare to steel on the Record, the Super Record has a carbon ‘brace’ that adds some stiffness to the large chainring when on the smaller sprockets of the rear cassette, and the Super Record chainset uses CULT ceramic bearings, which apparently are very high quality. Also, the Super Record chainset uses a bolt with a left hand thread to join the two halves. I’ve read that this is done to avoid a steel bolt being used which could lead to galvanic corrosion between the steel and titanium.

90 g of the 150 g weight difference between the Super Record and Record groupset is in the chainset. For me, what was more important was that, to my taste, the Super Record chainset looked much nicer than the Record. Also it seemed appropriate that the top frameset got the top components – it’s supposed to be a superbike after all.

After much deliberating and searching for the best price, I bought a Super Record chainset from Chain Reaction Cycles. Chain Reaction Cycles seems to change their prices periodically without it being obvious. When I first looked, there were other retailers with better pricing, but when it came time to pull the trigger, Chain Reaction suddenly had the best price. I’ve seen the same in the past on other components I was looking for.

One thing I found was that most places had 172.5 mm chainsets, but some didn’t have any 175 mm. Merlin Cycles had the best pricing, but only had 172.5 mm. I’ve never had a bike fit so I don’t know that 172.5 mm cranks won’t work for me, but I’ve never had a problem with 175 mm cranks and all my other bikes have 175 mm cranks. It wasn’t worth the risk of buying 172.5 mm cranks to save a few dollars and finding a problem in the way they felt or worse still, start suffering from some kind of pain while cycling. It’s worth noting that 175 mm is considered the standard crank length for mountain bikes and 172.5 mm is considered the standard length for road bikes. I’m 6’4″ and I used to use 180 mm cranks on my mountain bike.

The other thing I found was that not many places offered 53/39T cranksets, which is what I’d always run on my road bikes, however I didn’t think it to be an issue going with 52/36T which were much more available. I could see where the lower gearing of the smaller chainring could be a benefit, and a 52T large chainring wasn’t going to limit me in any way.

The next decisions were around the cassette. There were two decisions to be made, gear ratios and Chorus vs Super Record (there is no Record 12 speed cassette). The gear ratio options were either 11-29 or 11-32 (Chorus also has an 11-34, which might need the Chorus rear mech, I wasn’t interested in that so didn’t look into it). My 11 speed setup on the V1-R has an 11-29 cassette. The 12 speed 11-29 cassette would add a 16 tooth sprocket that is missing between the 15 and 17 teeth sprockets on the 11 speed cassette. The 11-32 would also add the 16 tooth sprocket but extend the last four sprockets out to 32T going 22, 25, 28, 32 compared to 21, 23, 26, 29 on my 11 speed bike. This seemed a better option, and I’ve had no problem with the wider gear spacing for the lower gears on my Ekar 9-42 cassette. The difference between Super Record and Chorus 12 speed cassettes is that with Super Record the largest 6 sprockets are titanium whereas Chorus uses steel. Super Record is almost twice the price and titanium sprockets are less durable. The weight savings is over 50 g, but I wasn’t out to build the lightest bike -just the nicest – so I went for Chorus and saved the money.

Parts arrived, I pressed in the BB86 bottom bracket cups I’d bought earlier, and installed the crankset and cassette.

I shortened the chain (according to the Campagnolo maximum chain length rule) joined it with a KMC Missing Link, installed my favorite SuperCaz bar tape, adjusted the gears and I was done.

In an earlier draft of this blog, I wrote at length about using quick links on Campagnolo chains rather than the recommended pin. I’ll just say that if you chose to use the pin as Campagnolo recommends, you should use the appropriate Campagnolo tool, not a third party tool. I personally prefer to use a quick link and I think this article does a good job of explaining why and why it should still be ok to use a quick link rather than a pin to join the chain. This whole topic sparked some lively debate when I posted on the Campagnolo Tech Talk group on Facebook.

Here are a few photos of the finished bike.

Although I started cycling as a mountain biker, I took an interest in the Tour de France, and became an avid follower during the Lance Armstrong years, which made me at least appreciate road bikes and road cycling.

Colnago was a name I knew through following The Tour, but back in the UK when I was considering buying a road bike I had my heart set on a Bianchi. I loved the look of the celeste and yellow Bianchi Marco Pantani was riding, and I could get an aluminum Bianchi painted in similar colors for around £1500.

A couple of years later I was living in the US and I’d bought a Yeti road bike from my friend Joe who worked for Yeti Cycles.

One Saturday afternoon I was in the bike store ‘Two Wheel Tango’ in Ann Arbor. They were a Colnago dealer and had a Colnago Carbonissimo frame hanging on the wall. It was both the most expensive and the most beautiful bike frame I’d ever seen. I think I was mesmerized by the carbon fiber weave and how perfect the frame looked. That must have been either 2000 or 2001 and I remember the frame being on sale for around $7,000, which is a lot of money now in 2022, and ridiculously expensive back then. That put Colnago on a very high pedestal for me – very desirable but out of reach – but that didn’t bother me because I would have been dreaming about buying the latest full suspension Yeti mountain bike back then rather than a different road bike.

Between 2002 and 2004 I was working as an engineer on the Alfa Romeo variant of the General Motors High Feature V6 engine. This entailed monthly trips from the US to Alfa Romeo in Italy who at that time were located just North West of Milan in a small town called Arese. On one of the trips in 2002 I had a couple of hours to kill before I needed to head to the airport for the flight home, so I decided to visit the Colnago factory in Cambiago which is North East of Milan. I figured it wasn’t too far from Alfa Romeo and even though it was taking me the wrong direction, I could still get to the airport in time. This was before the days of satellite navigation in cars and iPhones with GPS apps, but I found Cambiago on a map and fought my way through Friday afternoon traffic to get there.

I wasn’t quite sure what to expect. The factory/headquarters building was relatively small, on what I remember being an otherwise residential street. I arrived unannounced, hoping that they might let me take a look around, or at a minimum let me have a look in what appeared to be a showroom. I’d visited the Yeti factory numerous times in Golden, Colorado, and in 1998 I visited the GT factory in Santa Anna, California. The front of the building where the workshops were, had open doors so I could see in, which is how I got the photos below:

After standing gazing in for a while, one of the Colnago staff members came to see what I wanted. I asked if I could have a look around. It became obvious they weren’t used to entertaining visitors – especially those who show up unannounced – he gave me a very nice hardbound book/catalogue, and politely asked me to leave.

So much for that, but oh well – nothing ventured, nothing gained – and it was still nice to have seen what I had seen.

In 2017, I started getting back into cycling after a hiatus due to family life and golf. In the spring I started riding my road bike while the trails were drying out, and I realized how much I was enjoying it. My Yeti Road Project was ok, but at that stage it was 16 years old and I started thinking about a change. That got me thinking about Colnagos and wondering if one could be had for a reasonable price used. I found a Colnago C50 frame for sale on eBay which seemed reasonable, but I was concerned it might be just that bit too small, compared to the geometry charts for my Yeti. I kept searching and after missing out on a Colnago M10 frameset, that looked great, I scored an incredible deal on a Colnago CX-1. The CX-1 has a monocoque front triangle with a lug and tube rear. It was Colnago’s race bike when it was introduced, and after an EVO model was replaced by the M10 which in turn was replaced by the V1-R.

I built up the CX-1 with the 9 speed Ultegra components and Mavic Ksyrium wheels from my Yeti Road Project, which meant I only needed to buy bars, stem, cables, bar tape and a braze on front mech (the mech front mech on the Yeti used a clamp). I also picked up a used Colnago Selle Italia saddle on eBay for $25. The bike as pictured below cost me less than $1,000 (excluding the value of the carry-over Ultegra components):

It was a big step up from the Yeti, it was lighter and more responsive, but importantly not as harsh as the Yeti was on the rough roads I ride. The carbon frame was more compliant than the aluminum frame of the Yeti where it needed to be, but stiff where it needed to be also.

I looked out for bargains on eBay and the online bike stores, and over time I collected all the components I needed for a Campagnolo 11 speed Record build. I had zero knowledge or experience with Campagnolo, but I loved the aesthetics and it only seemed right to have Campagnolo on a Colnago. Then came a set of Zipp 404s, which were great wheels.

The Ultegra components and Mavic wheels went back on to the Yeti which became a spare bike and has since been mounted to my Wahoo trainer apart from the occasional outing.

I thought the CX-1 was a superb bike and I had it spec’d just as I wanted it (I remember deliberating for hours deciding which water bottle cages would best suit the bike – I ended up with FSA carbon cages that were black/white/red). I rode it for around 5,000 miles, I did my first century and other notable rides on it.

I was now more of a ‘roadie’ than a mountain biker, realizing that the time I would have previously spent in the car driving to a mountain bike trail to ride, could be better used to ride my road bike from home – and time was limited due to my commitments as a family man. I was also enjoying group rides with the R3 cycling club I’d joined in the Spring of 2018.

The CX-1 was great, but there’s always something better – the next bike. For me that was a Colnago Concept, which was their aero bike. I thought the Concept was a cool looking bike – in certain colorways – in others looked a but ‘dull’. My favorite was the black and gold art decor colorway, which was my dream bike at the time, and anything but dull.

I looked for months, but couldn’t find a Concept frame with rim brakes (I didn’t want the expense of new carbon wheels, and Campagnolo disk brake components seemed very expensive) in my size, at the right price and with a colorway I wanted. It then became apparent that the Concept was being discontinued and there’d be slim pickings, so I started looking at different options.

I found a V2-R frameset for a good price, but it was in matt black (TNBK) which didn’t appeal, and by the time I’d convinced myself it’d be ok, it had sold. Disappointed I’d missed out, and now decided I was getting a new bike (or frameset at least), I kept looking and found a great deal on a new V1-R frameset on bikeexchange.com. I’d been put off the V1-R due to the bottom bracket mounted rear brake, but this was an incredible deal and my favorite of the V1-R colorways. Being a car guy, I must admit that having a bike that was developed in collaboration with Ferrari and having a Ferrari logo on the top tube also appealed.

In time honored fashion, the wheels and components from the CX-1 were used to build the V1-R.

Then, again over time, I searched and waited for the best deals – this time on 11 speed Super Record components. When I had what I needed I rebuilt the V1-R and swapped the Record components back over to the CX-1, so I could use the CX-1 as a spare bike (or second spare road bike, because I still had the Yeti Road Project – but that was my trainer bike)

I’d read of people complaining about the bottom bracket mounted rear brake on the V1-R. The complaints were around servicing and adjusting brake pads, and some complaints about the rear brake picking up more dirt in that location. I’d decided these were issues I could live with, but I found was that I was getting significantly more brake rub when out of the saddle sprinting, or climbing than I was used to with CX-1. I attributed this mainly to the bottom bracket mounted rear brake, but realized it was due to differences in stiffness between frame and the rear wheel, and the V1-R must have been different to the CX-1 in some way. This was bothering me to the extent that I was beginning to regret the purchase. I later found a review on Velonews describing this issue, which made me wish I’d done more research.

I was expecting the V1-R to be an upgrade relative to the CX-1 but how could it be if I was experiencing brake rub, which was zapping power. A fix was to open up the clearance between the brake pads and the rim, but at the expense of braking capability.

After quite some research, it seemed that the Zipp wheels weren’t particularly good in regards to lateral stiffness due to the hubs, which could be contributing to the brake rub issue. I picked up a set of Campagnolo Bora WTO 60s and the brake rub was very much reduced. The Zipps were sold on eBay for more or less what I paid for them two years earlier.

Upgrades also included a Deda Superzero stem and carbon bars, which looked great fitting in well with the squared off tubes of the bike and were very comfortable.

Other than the brake rub issue, which was mitigated by the Bora wheels to the extent that it was no longer a concern, the V1-R has been an excellent bike and has performed very well over the 5,000 miles I’ve ridden it.

Anther passion besides cycling has always been cars. That’s what got my into engineering which made it possible for me to live and work in the US. I always loved AMG Mercedes and in 2013 I was fortunate enough to buy an E55 AMG Mercedes, which had the 5.5l supercharged V8 engine. I owned that for three years and replaced it with an E63. I’d owned that for a couple of years and realized that it wasn’t the most practical car – especially during Michigan winters – and it wasn’t very good at transporting bikes. I concluded that bikes were more of a priority for me than fast cars. The E63 went and was replaced by a two year old Jeep Grand Cherokee. The logic was also that that the money I was saving by having a less expensive car could go towards a dream bike – a C64 maybe!

So by now, I was a bonafide ‘Colnago Fanboy’. As well as riding on the road, I was now riding gravel a lot. I’d been looking for months for either a Colnago Prestige or G3-X to replace the Kona Super Jake I was using to ride gravel. That didn’t pan out and it became apparent that I’d be waiting over a year for a G3-X frame. I ended up with a Look gravel frame and wrote about it here.

I’d been e-mailing a lot with Maris, the owner of Brava Cycles – a Colnago dealer in Latvia – about the availability of a G3-X frame in my size. He seemed very good and his prices were the best I’d seen. I’d also noticed his prices for C64 frames, which didn’t seem that far out of reach. I loved my V1-R, but the C64 was the ultimate superbike in my opinion, and I’d been watching YouTube videos such as this:

I’d seen a C60 frameset in a store before, and I could imagine how incredible a C64 must look, especially in one of the new ‘Frozen’ colorways such as this bike in Frozen red, code RCRD. The Frozen paint schemes are offered in a number of colors, but I really liked the red.

In the spring of 2021, I sold my CX-1 realizing I wasn’t riding it as I had my V1-R, and I didn’t really need two spare road bikes and a gravel bike that could do double duty as a road bike spare. By the summer, I was in a position to order a C64 frameset. I placed the order with Brava Cycles in July and was told it would be a 4-5 month wait, meaning I’d receive the frame in November or December. If I received the frame in December, then I wouldn’t be riding it until March due to the weather, but this would give me time to assemble the components I needed to build it.

After being told that frames were taking longer to arrive from Colnago, and it might be January/February before I see the frame, out of the blue I received an e-mail from Maris saying my frame had arrived and was ready to ship. He also sent me a photo of my frame next to two other Frozen C64s he was about to ship. I couldn’t wait to get my hands on it.

The frame arrived the Monday after Thanksgiving and it didn’t disappoint. Then I built the bike up over the next month or so, here it is finished. I wrote about the build here.

Michigan winters can be long and cold, and despite all my best intentions I’ve always struggled to ride outdoors in the bitter cold. Besides, even if there’s a brief break in the weather, the roads are often dangerous due to the patches of snow and ice or covered in a layer of corrosive salt.

Given that, I needed a means to keep fit in the winter months. Initially I joined a gym to go spinning, but realized it would be more convenient to be able to workout at home on my own schedule and avoid the travel time to and from the gym – time is valuable with a young family. So, for the past three winters I’ve been riding an indoor trainer at home, mainly on Zwift.

US houses often have basements, and I’m fortunate enough to have nearly 1,000 square feet of finished basement in my home. I remodeled the basement a few years ago, converting half into a carpeted ‘living room’ and the other half into a gym. The gym side of the basement is now my ‘pain cave’ that I’ve set up to make indoor cycle training as enjoyable as possible. I understood some time ago, that the more enjoyable I could make indoor training, the more I would be motivated to ride.

This post provides an overview of my setup, what I use and why, and was written to pass on what I’ve learned over the past three years of indoor training with the hope that some will find it useful.

I’ve included a breakdown of the complete setup with links for each item at the end of the post.

Trainers

I bought my first smart trainer – a Wahoo KICKR SNAP – back in November 2017. Several friends owned Wahoo trainers, rated them highly, and I’d been very happy with my Wahoo ELEMNT BOLT, TICKR heart rate monitor and Cadence sensors, so I looked no further. The two trainer options from Wahoo at the time were the direct drive KICKR ($1,100) or the wheel-on SNAP ($600).

I’m a member at REI, they have sales 2-3 times a year and conveniently I got my KICKR SNAP in the November sale, right when I needed it, with 20% off.

There are plenty of good reviews already and I’d direct you to DC Rainmaker for everything you need to know: https://www.dcrainmaker.com/2017/07/wahoo-kickr-snap-2017-v2-trainer.html

The KICKR SNAP has worked flawlessly. The only problem I can report are that I had a number of flat tires, which seemed very odd on an indoor trainer, and it’s still a mystery to me what caused them. Initially I used a regular road tire (Conti GP4000), then switched to a Vittoria trainer tire, then switched to a Conti trainer tire and have had no issues since. Regular tires are too soft for an indoor trainer and are not designed to withstand the heat that’s generated between the tire and the roller.

I bought my wife a gravel bike in the spring of this year and set her up with the bike on the trainer so she could get used to riding it. She took to it straight away and began riding daily on Zwift. To avoid competition for the trainer, and to avoid having to keep swapping bikes, I decided to buy a second trainer and at the same time upgrade the pain cave so we could ride together.

Happy with the KICKR SNAP, the second trainer was also going to be a Wahoo. I decided I wanted to upgrade to a direct drive trainer, so to avoid the flat tires, and to make it easier to take the bike off the trainer and ride outside if I ever wanted to. The choices were the KICKR ($1,200) or the KICKR CORE ($900). While the reviews noted the KICKR had a few advantages over the KICKR CORE, most commented that there was very little noticeable difference between the two when riding: https://www.dcrainmaker.com/2020/08/wahoo-kickr-v5-2020-smart-trainer-in-depth-review.html

So I was leaning towards the KICKR CORE. https://www.dcrainmaker.com/2018/09/wahoo-kickr-core-trainer-in-depth-review-2.html

I went to REI’s website the morning the November sale started, they had the KICKR CORE listed but not the regular KICKR. That made the decision easy, and I got the 20% reduced sale price despite the terms and conditions stating that indoor trainers were excluded from the sale.

The KICKR CORE worked great. I didn’t find it too different to the KICKR SNAP. It was quieter, essentially silent apart from drivetrain noise. It broadcasts cadence, which means I don’t need a separate cadence sensor.

Sometimes when I sprint out of the saddle on the trainer, the trainer can feel a little unstable and I get concerned that I could topple the trainer over if I’m not careful. This seemed slightly worse on the KICKR CORE than I remembered it on the KICKR SNAP. I measured the distance between the front feet on the two, the KICKR CORE has them 23″ apart whereas the KICKR SNAP has the front feet 29″ apart, which is why the KICKR CORE feels less stable. This is just something for me to bear in mind and be careful of, although I found myself holding back on sprints more than I would have on the KICKR SNAP – at least while I got used to the KICKR CORE.

Smart Trainer Workout Applications

The decisions about what hardware to use to control the smart trainer can be very dependent upon the intended applications, how important the graphics are and how much interaction will be required. The two applications I’ve used most extensively are Zwift and TrainerRoad, which are very different.

TrainerRoad https://www.trainerroad.com/ is a cycling workout application that can be used on smart trainers and also outdoors with Garmin and Wahoo head units. Like other applications it connects to the smart trainer to receive power measured and broadcast by the trainer, and where required, it will adjust the resistance to maintain a specified power level. It’ll also receive heart rate and cadence broadcast by their respective sensors. TrainerRoad has apps for either iOS or Android (smart phones or tablets), or can be run from trainerroad.com on Windows or Mac computers.

The TrainerRoad workout screen is relatively basic, displaying the workout profile and key metrics (actual and target power, heart rate, cadence and workout time etc.). Therefore computational and graphics requirements of the device running trainer road are basic. It runs well on a smart phone or tablet and there’s little to no benefit of using 4K high definition TV or Monitor. I ran TrainerRoad on my iPhone with my phone on a handlebar mount, but my 50 year old eyes found the information hard to read without reading glasses. I got around this by mirroring the display to a Gen 3 Apple TV I had connected to a 39″ TV in the gym (that pre-dated the smart trainer setup). Once a workout has been started, the only interaction required is to adjust the intensity level or to end the workout, which is easily done with a smart phone or tablet.

If you intend to run TrainerRoad on a computer, then you have the option to overlay the workout screen on another application, so you could watch a streamed TV show or movie while performing your workout. Indoor training can be monotonous and boring at times, so anything that makes it more bearable or enjoyable can only be good.

Zwift https://www.zwift.com/ describes their platform as ‘Serious Training Made Fun’. The platform allows cyclists (and runners) to interact, train and compete in a number of virtual worlds. The creators market Zwift as a game, and riders have the option of riding freely around the virtual worlds, joining organized group rides, competing in races, or performing workouts similar to the ones found in TrainerRoad.

Zwift has become very popular and has global participation. During April 2020 with many countries in lockdown due to the COVID pandemic, Zwift recorded a new record of 34,940 riders in the game at the same time.

The first ‘world’ that Zwift made available is known as ‘Watopia’ which is a fictional island in the South Pacific. Watopia is being continually expanded and new worlds are being added, some fictional and some that are replicas of real locations. The worlds are incredibly detailed with lots to attract and keep a rider’s attention and keep it interesting. The intent is to make Zwift an immersive experience. Although it’ll run happily on a smart phone, it lends itself to a large high resolution monitor or TV to draw the rider into the game.

While riding, Zwift displays a leaderboard showing riders ahead and behind with details including time gap, nationality, distance ridden and current power output (in W/kg). Riders can communicate with each other via messaging and riders can give each other ‘Ride-Ons’ which is a way of showing appreciation, similar to a ‘like’ in Facebook or ‘kudos’ in Strava. A ride-on is shown as a thumbs up symbol which is displayed above the rider and it is then put into the riders jersey pocket as they ‘collect’ ride-ons during the ride. Zwift can therefore become very socially interactive. Additionally, a number of game controls are available from the game screen. Again, while a smart phone could be used on its own to run Zwift, it’s not ideal, due to the small screen for both viewing and interaction.

When I started with Zwift, I used an iPhone 8 and mirrored the display to my 39″ TV via a 3rd Gen Apple TV. This worked satisfactorily. I also sent audio from the iPhone to the Apple TV via Airplay which played through an old AV Receiver I have connected to a pair of cheap speakers. I find music makes workouts much more enjoyable. I also tried using my iPad instead of the iPhone. While a better standalone option compared to the iPhone due to the larger display, it doesn’t work as well mirroring to a TV as the aspect ratio of an iPad causes the TV to display black bands to the left and right.

Connectivity

Smart trainers and sensors typically broadcast using Bluetooth or ANT+ (or both). These days most devices are capable of connecting directly via Bluetooth, but typically only computers can connect to devices directly via ANT+. ANT+ usually also requires a dongle for connectivity to a Windows or Mac PC, preferably on a 3′ USB extension cable to place it close to the trainer.

Bluetooth is more universal, but there are some limitations. With an iPhone, if an application is using a Bluetooth device then it can’t be used by a second application on that iPhone. I’ve found this to be problematic when setting up to start a workout and having difficulty pairing the trainer or sensors with Zwift. This then requires me to close all apps apart from Zwift, then toggle Bluetooth off and then back on again. Sometimes I find I need to restart the iPhone to resolve the issue.

ANT+ doesn’t have this issue, but can be prone to signal dropouts if not set up carefully. Bluetooth connectivity for Windows PCs running Zwift is only possible with Windows 10, before that ANT+ was the only option to connect a Windows PC to Zwift.

Internet connectivity will more than likely be required for a smart trainer workout session. However the bandwidth requirements and data usage for Zwift and TrainerRoad are low, so in a pinch cellular data could be used. With both TrainerRoad and Zwift, an internet connection is only required to start a workout, and to upload the workout at the end (although with Zwift, if the internet connection is lost during the workout, other riders will disappear from the game).

Computer and Display

Smart phones, tablets and computers can all be used to run the various training applications that are used to control smart trainers. In addition to workout application specific considerations discussed above, there are other factors that may influence decisions:

1. What do you currently own that can best do the job?
2. Will the trainer setup be temporary or permanent?
3. If the setup is permanent, how much space is available and is there space for a monitor or TV?
4. What can you afford or want to spend?

I’m fortunate that I have a space I can dedicate to indoor training and set up as my pain cave as I’m describing here. If I needed a more basic or a temporary setup, I’d probably use either my iPad or laptop PC, depending on which I could best place on a stand in front of the bike. The laptop would probably be the best choice as it allows access to the keyboard for interactions with Zwift, while providing a large display. With a PC, I’d make sure notifications from other apps or the internet are disabled so they don’t interrupt the workout.

As mentioned above, when I first remodeled the basement, I installed a 39″ TV for entertainment during workouts, mainly thinking about my wife who was running on a treadmill regularly at the time. When I bought the KICKR SNAP I ran TrainerRoad and Zwift on my iPhone 8 mirroring to the TV via a Gen 3 Apple TV.

4K Apple TV. I later learned that the newer (Gen 4 and 4K) Apple TVs would run Zwift through an iOS Zwift app for Apple TV that could be downloaded from the Apple App store. The 4K Apple TV was recommended over the Gen 4 as it had a faster processor. It cost $200 at the time, but I managed to get one free with a three month trial of Direct TV.

One limitation with the Apple TV is that it can only connect two Bluetooth devices at once, and having the KICKR SNAP, a heart rate monitor and a cadence sensor to connect, I needed to connect three devices. However Zwift has an app called the Companion App, which serves a number of useful functions during and away from the game. One function is to act as a bridge between the iPhone running the Companion App and an Apple TV, this allows more than two Bluetooth connections to the iPhone, and the iPhone then sends the data from those devices to Zwift running on the Apple TV. The Companion App also allows the iPhone to operate the game controls during the game and to message and give ride-ons to other players. The iPhone and Apple TV have to be connected to the same network (via ethernet, 5kHz or 2.4kHz WiFi, it doesn’t matter as long as it all goes back to the same router) and once everything is setup and the devices are paired, I’ve found it works flawlessly.

Once the device running Zwift meets some minimum requirements, the level of detail shown in Zwift is dependent upon the capability of the graphical processing unit (GPU).

There are two separate definitions related to the graphics in Zwift, resolution and detail.

Display resolution is user selectable from the settings menu in Zwift, with the options offered dependent on the capability of the device running Zwift. The most capable devices will allow resolutions of 576p, 720p, 1080p, 1440p and 2160p (which is 4k).

Detail is not user selectable, but is determined by Zwift based mainly on the capability of the GPU. Here detail refers to things such as depth of field, lighting/shadow effects, textures, and environmental elements such as water effects, flower, trees and wildlife seen. The levels are basic, medium, high and ultra.

Even though the 4K Apple TV is capable of streaming movies in 4K ultra high definition, it is only capable of running Zwift in the game at 1080p with basic detail.

Dedicated Zwift PC. As I was adding a second trainer to the pain cave, I needed a second device to run Zwift. There have been rumors about the imminent release of a new Apple TV, that may or may not have better graphics capabilities to allow more detailed graphics in Zwift, the other option was a dedicated PC that would be capable of running 4K resolution with Ultra detail.

For some time, I’ve been following a Facebook group called Zwift PC Master Race https://www.facebook.com/groups/zpcmr which is a group for Zwifters who want to build inexpensive PC’s specifically to run Zwift in high resolution with High or Ultra graphics detail. This is an excellent resource for those interested in buying or building such a device and there’s a lot of well documented information in downloadable PDF files that are regularly updated.

I was initially reluctant to venture down this avenue as I was concerned it would be too much effort and take up too much of my time. However, after reading through the documents, the requirements were clear to understand. I then discovered that reconditioned office desktops were available on eBay that would be a great basis for a dedicated Zwift PC – and that they were inexpensive.

Minimum requirements for a Zwift PC with high or ultra graphics capabilities are as follows:

• 4th Generation (Haswell) Intel processor, such as an i3-4170
• 8 GB of RAM (more won’t make a noticeable difference in Zwift)
• SSD hard drive on which to install Windows and Zwift, doesn’t need to be big for Zwift
• Tower case to give more options for inexpensive graphics cards
• Windows 10, can be installed free and run without activation if necessary
• GTX 1650 Super graphics card if you want to run 4K at 60 fps (for best graphics)
• 400W Power supply with a PCIe connector for graphics cards that require external power
• GTX 1650 graphics card, if you don’t have a PCIe connector, but this won’t run 4K at 60 fps

After some searching on eBay, and following recommendations in ‘The Frankenstein Guide.pdf’ on ZPCMR, I found what looked to be a great match: Dell Optiplex 7020 MT, i5-4590@3.3GHz, 8GB RAM, 256GB SSD.

This ticked all the boxes apart from the power supply, which was rated at 290W and didn’t have PCEi connectors. However I discovered that the power supply can be easily replaced with an aftermarket power supply with increased power and a PCEi connector. There are several videos on YouTube that show how to upgrade the power supply and graphics card in a Dell Optiplex PC:

After confirming that the PC I’d found on eBay would work for me, I bought it for $150 (including $30 shipping). I then bought a used power supply off eBay similar to the one shown in the YouTube video, and bought a GTX1650 Super graphics card and a 24 – 8 pin connector as shown in the video.

The computer arrived, and worked without an issue out of the box. It had been listed as not having an operating system, but it had a new, activated installation of Windows 10 Pro, only requiring me to finish off the installation. I installed WiFi and ANT+ dongles I’d bought years ago that were gathering dust in a drawer, downloaded Zwift, connected the PC to the TV and went for my 1st Zwift ride. It was that easy.

It took a couple of weeks for the power supply to arrive (due to delays with USPS), but when it arrived, the upgrade process was quick and easy. I installed the power supply and graphics card, made the cable connections, downloaded the drivers for the new card, adjusted the settings and was up and running Zwift in 4K Ultra.

When Zwift is running, it keeps a log of graphics performance and ANT+ or Bluetooth connections. https://zwiftalizer.com/ is a useful website for plotting and reviewing the data stored in the log files. They also provide good insight into the results and recommendations for improvements.

Here’s a comparison from zwiftalizer.com of the graphics performance as received and after upgrading:

These plots from zwiftalizer.com show that by upgrading the graphics card to a 1650 Super, the graphics went from 1080 with basic detail running at 23 frames per second, and varying between 17-30 fps, to 4K with Ultra detail and holding steady at 60 fps (the dip just after the beginning was when I joined a group ride).

TVs. Adding the second trainer also required a second TV. I thought it wouldn’t look right for the two TVs not to match. The 39″ TV I had was old and it would be hard to find a second the same, so I needed two new TVs. Fortunately TVs are cheap these days. I figured I could go bigger than the 39″ TV I had, and that bigger would mean more immersive, but I needed to make sure I could fit two next to each other and in line with the bikes. I bought two 4K Hisense 50″ Roku TVs from Costco for $220 each – which I though was incredible value – and mounted them with mounts I bought in a half price sale at Target for $25 each. Setup was easy and I’m very impressed with the TVs, how easy they were to setup and the picture quality.

Fans are essential in a pain cave. Anyone who’s ever tried riding a trainer indoors without a fan can attest to that. The gold standard for fans for indoor training is the Wahoo KICKR Headwind. This is a powerful fan that will pair with either a heart rate monitor or the speed signal from a trainer and vary the fan speed accordingly. Great idea, but I’m not going to spend $250 on a fan, even though it’s supposed to be a very good one.

Instead the fans I use cost around $20-30. They’re simple and straightforward and do everything I need, meaning they keep me cool enough.

One thing I did though is buy a couple of Wyze WiFi Smart Plugs into which each of the fans is plugged. These only cost $20 for two from Home Depot, they were easy to set up with the app and they work well. This allows me to start a workout with the fans switched off, then when I’m warmed up, I can switch the fans on from my phone while I ride. A good cost effective compromise.

Lighting. When I remodeled the basement, I drywalled the ceiling and installed recessed lighting. The lights are on a dimmer, but I thought it would be good if I could dim the lights remotely after I’d started a workout. I went to my local Home Depot to look for some WiFi connected bulbs and came home with five of these for the gym:

These are dimmable, can change color and are controllable via an app or remote control (which was free when you buy two bulbs). Now, the lights are on full and their normal color when I’m setting up for a workout, then once the workout has begun, I switch them to a setting via the remote where they each invidually change color slowly. This adds a nice mood and ambience to the room.

The final touch was to add hue lights to the rear of the TVs. I bought two strips off Amazon for $13 each.

They attach to the back of the TV using self adhesive strips and connect to the USB port on the PC for power. They’re controllable via an app and both light strips showed up in the app 1st time I opened it after downloading.

Stands. Before my pain cave upgrades I needed a stand to allow the TV and Apple TV remotes to be within reach along with a water bottle and tissues (I always seem to get a runny nose when I work out), and also somewhere for my reading glasses. This laptop/projector stand was $30 off Amazon and fit the bill perfectly.

Now I’ve found I need a second stand and room for a keyboard, another remote (for the lights). Wahoo’s KICKR Indoor Cycling Desk has always seemed the best solution, but at $250 I put this in the same ‘poor value for money’ category as the Headwind.

These days however, some copies are available for much cheaper and have received very positive reviews, including this review from DC Rainmaker https://www.dcrainmaker.com/2020/09/rad-lifeline-conquer-cycling-trainer-desk-cheaper-wahoo-desk.html

I managed to find a RAD Cycling Products copy of the Wahoo desk for $130 including shipping.

Summary. Hopefully you found some of the information here interesting and useful. As you will have noticed throughout, I’ve done my best to minimize the costs of the setup by looking for deals and waiting for items to go on sale. This is because ultimately, indoor cycling and the pain cave is a means to an end, and I’d prefer to spend my money on bikes. I’ve limited my costs to around $2,500 for everything I’ve discussed inclusive of all shipping costs and taxes. I think this is pretty good considering the trainers alone would normally cost $1,590 at full retail price including 6% Michigan sales tax.

Bill of Materials

Here’s a breakdown of everything I used for my setup, with links to help find each item (better prices may be available elsewhere):

• 1 x Wahoo KICKR SNAP smart trainer Wahoo Fitness
• 1 x Wahoo KICKR CORE smart trainer Wahoo Fitness
• 1 x Wahoo KICKR Trainer Floormat Wahoo Fitness
• 2 x Wahoo TICKR Heart Rate Monitor Wahoo Fitness
• 1 x Wahoo RPM Cycling Cadence Sensor Wahoo Fitness
• 1 x 2001 Yeti Road Project
• 1 x 2020 Giant Revolt Advanced 2 Giant Bicycles
• 1 x Continental Hometrainer Folding Tire Amazon
• 2 x Hisense 50″ R6 Series UHD TVs Costco
• 2 x Core Innovations Tilting TV Mount 30 – 79″ Target
• 1 x Dell Optiplex 7020 MT, i5-4590@3.3GHz, 8GB RAM, 256GB SSD PC eBay
• 1 x PNY – NVIDIA GeForce GTX 1650 SUPER graphics card Best Buy
• 1 x EVGA 500w gold 80plus power supply Amazon
• 1 x 11 Inch 24-Pin to 8-Pin 18AWG ATX PSU Power Supply Adapter Cable Amazon
• 1 x Mini USB + ANT Stick Amazon
• 1 x Powered USB Hub Amazon
• 1 x USB 3.0 Extension Cable 2 Pack[3.3ft + 6.6ft] Amazon
• 1 x USB WiFi Adapter for PC Amazon
• 2 x Nexillumi 9.8 Feet TV Backlights Amazon
• 5 x Color and Tunable White BR30 LED Smart Wi-Fi Wiz Light Bulb Home Depot
• 1 x Smart Remote Control for Philips Smart Wi-Fi Wiz Light Bulbs Home Depot
• 1 x RAD Cycle Products Adjustable Bike Trainer Fitness Desk Amazon
• 1 x Universal Laptop Projector Tripod Stand Amazon
• 2 x Lasko 16″ Oscillating Pedestal Stand 3-Speed Fan Walmart
• 1 x Apple TV 4K 32GB Best Buy
• 1 x Apple TV (3rd Generation) Black Market
• 450 sq ft TrafficMaster Rubber Gym/Weight Room Flooring Tiles Home Depot
• Sound system: Marantz SR4000 AV Receiver & 2 x JBL TLX151 Speakers (bought for $100 used) connected to Gen 3 ATV with optical digital cable
• Music source: Tidal/Qobuz/Apple Music