PowerTap G3 : Complete Power Meter and Training Wheel Analysis

Esta página está disponible en español

Español

Español

Power measurement is becoming more and more important in the world of cycling, not only at the professional level where power measurement is essential for any rider, but increasingly among amateurs, especially in triathlon. Both when preparing your training and when choosing a running tactic, working with a focus on the watts you are able to produce is undoubtedly the most effective way to compete.

One of the best known manufacturers of potentiometers is PowerTap. They have been offering power measurement on the rear hub of the bike for a few years, specifically since 1998. This PowerTap G3 you will see today at the test is their latest hub, which has also recently seen its price reduced.

Are you new to power measurement? You can take a look at the article I published a few days agoThis is a guide to learn more about the history and the different points where it is possible to obtain data from the cyclist.

Tamalpais, PowerTap distributor in Spain, has provided me with a training wheel to perform the different data analysis during these last weeks. And once I have finished with the different tests I have submitted it to, I will send it back, there is nothing left with me. Of course this time I have been using other power meters with the objective of comparing the data obtained, without which this type of test would not make any sense.

I won't keep you any longer, let's go through the full analysis.

Unpacking

For those of you who are not completely new to the site, you know that normally I receive the complete product in its packaging, the same one you would receive when you buy it. In this case it is a little different, because what I have received is a training wheel that, before it came to me, had been in different training camps. That is to say, it is not a new unit, but I suppose that it is conveniently adjusted by Tamalpais.

But to continue with something that has already become a tradition, let's go with the unpacking, but something different.

As I say, in this case the PowerTap G3 came along with a training wheel, so the box is quite a bit bigger than usual.

Inside that box, in its own case, the PowerTap training wheel. It's a normal wheel, made of aluminium and standard weight. No profile, hard and resistant. Perfect for making miles and miles on any kind of asphalt without worries, but it wouldn't be what you want to use for your racing day.

And with the wheel out of the casing, all that's left is to put it back in place. I finished the unpacking today, didn't I?

Installation

You can purchase the PowerTap G3 in two different ways. First, as a separate hub, in which case you will usually purchase it with a wheel that you order from your trusted fitter or replace the hub of your current wheel, in which case you will usually take it to your trusted mechanic for wheel spoking and he will deliver it fully assembled, with your brand new potentiometer.

There is a second possibility, which is to purchase the wheel directly from PowerTap from the different offers on offer, from a basic training wheel with aluminium rings (like the one in this review) to different profile wheels made of carbon, even with the possibility of Enve, Zipp or HED rings.

In any of these cases, the only thing you will have to do is to mount some sprockets on the hub (available for Shimano/SRAM or Campagnolo). There is no mystery, you just have to notice that the body of the wheel has several notches where to place the sprockets, one of them is thinner than the others, it will be the one that will serve as a guide

The only thing you have to be careful with is not to lose the order of the sprockets. And more than the order of the sprockets, be clear about where the spacers are placed. So place them in an orderly fashion and slowly insert them into the body to avoid confusion.

This is what they will look like when they are all inserted into the hub, with the lid missing. Screw it on, starting to tighten with your fingers.

Finally, you will need to use the sprocket wrench and a good lever. To fix the screw you will not need anything else, but when you want to replace them you will have to help yourself with a sprocket blocking key that will allow you to hold the body and prevent it from turning freely.

Once the wheel is mounted with the hub you will not need more than 5 minutes to put your tire or tubular (well, the tubular will take a little more time).

And if you want to move your PowerTap to a different bike, it takes as long as it takes to remove one wheel and put the other one on.

Calibration

The best thing about the PowerTap hub is its simplicity and ease of use. It's something that's there, that just works, and that you'll only remember when it's time to change the battery. You don't even need to perform a calibration before each run or workout.

Of course you can force the calibration from the cycling computer you use, but you won't actually be calibrating anything. You will simply be telling the unit what the status is when there is no load. The procedure is very similar to that of any other power meter

• Activate the power meter by pedaling backwards or forwards.
• While standing and without applying any pressure, start the procedure (from the power meter options on your unit).

There, you can start pedaling like crazy. But as I say, this procedure is not even necessary, because PowerTap has the auto-zero function that is activated by default (and I recommend that it stays that way). This feature allows that when you advance without pedaling, a calibration is carried out automatically. The hub detects that there is no pedaling when there is no torque and, therefore, the power at that moment is zero. This way the power measurement is always accurate, regardless of possible temperature changes. As you see, all simplicity.

Connectivity

When you buy the PowerTap G3 it only supports ANT+ protocol. Although PowerTap has its own cycling unit (the Joule), thanks to this universal protocol you can use it with any cycling unit or watch that has ANT+ connectivity and supports the use of cycling power meters (such as Garmin Edge or triathlon watches).

But keep in mind that there are cycling units that even though they support ANT+ they do not have the power measurement support (like the Garmin Edge 25), and units or watches with Bluetooth sensor support that will not be compatible with the PowerTap G3, at least not directly.

For the latter there is a solution, as the hub "cap" can be replaced by another that transmits data via Bluetooth and additionally, there is also another cap with dual technology, transmitting simultaneously via ANT+ and Bluetooth. But in both cases they are added as an option, since the cover included in the standard version of PowerTap G3 only has ANT+ connectivity.

From here, simply search for devices from your compatible watch or cycling computer and add them to the list of external sensors.

Battery Replacement and Upgrade

And speaking of caps, that's where all the electronics reside, so that's what we have to take apart to replace the battery when it runs out (CR2032 button cell, easy to find in any supermarket). You'll also find a micro USB port that, by connecting it by cable to the computer, will allow you to update the firmware as soon as PowerTap provides new versions.

The change is very simple. You just have to remove the hub cover with the tool that is included with the hub. The location is, of course, at the opposite end to where you install the sprockets. I didn't have that specific key, but it is very easy to turn it with a simple flat screwdriver. But let's assume that you do have the right tool.

Once the plastic screw that blocks the cover has been removed, you will have to use a flat screwdriver or a plastic spike to remove the module.

If you turn the lid over you will see the battery socket. Simply remove it and replace it with a new one. If you need to update the connection module, you can do so with a standard micro USB cable.

And just in case you're curious, this is what the hub looks like with the cover removed. This is where the most important part of the power meter is, what I've removed simply takes care of the connectivity part.

Do you prefer a video? Don't worry, PowerTap has a complete guide detailing the procedure.

I'm sure you've seen it more clearly on video.

Details to consider

Because of the location of the potentiometer on the bicycle there are some details that you should take into account. Firstly, it is obviously at the end of the transmission chain, which means that the measurement you have is not necessarily your power as a cyclist, but the power that effectively reaches the roadSo you may be developing 280W but on screen you see 270W. And where is that 10W difference? Well, in the transmission.

It's just an example and there doesn't always have to be that difference. There can be more, or there can be less. It depends on many factors, such as the rigidity of the frame, the state of the transmission, the bottom bracket bearings, etc.

To make it clear, imagine pedaling uphill with a full suspension bike without blocking the rear shock. Obviously with every pedal stroke the shock will compress, losing some power. So the power you are applying is not the one that reaches the ground, because you are suffering from losses on the road. Or as it happened to me once, a too tight pedal set that did not turn properly. My power on the pedals was quite high, but all my effort was spent turning the pedals and the hub barely reached the power.

Is this bad? No, it is neither good nor bad, simply as it is. In the end the figure you will see and with which you will prepare your training will be constant, and it is this figure that you will work with. Although generally in road cycling there will be hardly any difference between the power applied to the pedals and that which reaches the ground, provided that you carry out proper maintenance of the bike.

Not only power, but cadence and speed

By design, a potentiometer must also calculate the cadence, because to measure the total power it needs not only the applied force, but also the angular velocity (i.e. the cadence). Therefore this will also be sent to your cycling computer, so you can dispense with any other cadence sensor you have mounted on your bike.

Depending on the location of the potentiometer, it can be obtained in one way or another (using magnets or accelerometers). In the case of the PowerTap G3, it will be through an internal accelerometer and its own calculation algorithm, because unlike a power meter installed on cranksets or connecting rods, it cannot measure the revolutions per minute constantly because of the position of the crank or the number of times it detects the rotation.

The PowerTap hub also sends information about the speed. This is very easy to calculate, as the hub can perfectly know the speed by the rotation of the hub. Therefore, in addition to the cadence as any other power meter, you will also have speed (and consequently distance) data much more reliably than if you only use GPS data.

Precision as a power meter

Information about the tests

Before going into the cold data, I would like to talk a little about the process of testing power meters. First of all, testing a potentiometer correctly is very complicated (and tremendously expensive). Analyzing a single device is useless, because we don't know if the data obtained is valid or not. It's like testing an optical pulse sensor and not comparing it with anything else. Yes, it reads pulse data, but... is it correct?

There has to be another device that can confirm that these data obtained are correct, in order to give validity. But what happens if the data of both differ? It is then that we need, at least, a third power meter, which is the one that will give the right perspective. Of course, each one sending data to its corresponding unit. And everything adds up to several thousand euros in material.

In addition to this, there is the problem of having a clear procedure for action, and any failure in this process can lead to a complete session of testing and data analysis.

• Calibrate each of the power meters individually, and all at the same time, so that the temperature settings made by each of them can be taken into account later.
• Match each one with its corresponding unit, with which we will take the data from that particular potentiometer.
• Confirm that all units are recording data in a similar way (every second, including zeros, etc.)
• Start recording data simultaneously so that you can then compare the results effectively without having to make numerous adjustments. This is much easier said than done.
• Make sure that each and every device used for testing has enough battery power both at the beginning of the test and to last until the end of the test.

There are many small details to consider before you can start each training session, and as I say, a slight glitch in any of these steps can make a scheduled test session absolutely useless.

Power data

I'll start with the first test after the initial installation. I don't take this test into account, as things are still in the process of settling down (especially for the bePRO pedals). The graph is smoothed out for 30 seconds to make it easier to see the data. More or less all three are in the same ranges, but with punctual differences.

As I say, I don't want to start drawing any kind of conclusion yet, but it helps me so that you can see what the initial settlement is like in the first test.

In this second test you will see that things are much more aligned, especially in the first part of the route. There are a couple of points where Stages is totally out of the graph, but it's not a defect of power detection but a punctual disconnection with the device with which it was recording data through ANT+ (I will talk about the signal power of Stages in its corresponding analysis).

If you enlarge the image you will see that sometimes there is some appreciable difference, especially in the peaks. These differences can have their origin in the following:

• Firstly, the recording frequency of the devices. Power meters normally record data at a rate of 8 samples per second. Cycling units and clocks record one sample per second, so the computer takes one of these measurements and discards the other 7. It may seem that one second is a short time, but with rapid acceleration, within that second there may be significant differences in pressure. That is why the power peaks differ, but in general the graph is perfectly aligned.
• Secondly, it is clear that the location in the transmission area will also affect the measurement. And looking at these peaks you can see exactly what the difference is between the three. bePRO, being placed on the pedals, is the one that registers a little more power than the others.

The differences are not important. Reviewing the graph at the highest power point we are talking about a 15W difference in 500W. It is a 3%, but I repeat that it is a point peak and there are more details in the equation.

Before moving on to the next chart, I want to expand on an area that is quite strange, and I think it deserves an explanation.

There are strange things there, right? In the 1697 time code (it's seconds, so 28 minutes of activity) I stop when completing a climb. I take the opportunity to take pictures of another device I was testing at that time and write down some things. Here for some reason the unit that records the data of Stages does not interpret it correctly and has a line in ascent very unreal, but that's not what I want to highlight, because it is a mere difference in data capture that being smoothed has that strange behavior. I want to focus on what comes next.

In that area I was making a descent in an urban area, that is to say, pedestrian crossings, speed bumps, etc. Why those strange and differentiated peaks? Well, they can be explained by the way the cadence is registered. In this case, as I was descending a quite steep slope in which it is almost not necessary to pedal (because in addition with the speed bumps, it is necessary to brake constantly) the cadence is very low.

To record the power, the potentiometer needs pressure and angular velocity (i.e. cadence) data. Note that the cadence data is different, for example Stages records cadence from 30rpm, while bePRO records 20rpm as I have achieved records around that figure. PowerTap has an estimated cadence (because unlike the other two, because of its position it cannot record a real cadence). So there you have an explanation for those strange peaks that have occurred in an anomalous situation, something you won't see in the working part of your workouts or in a competition. The low cadence speed, and the lack of high torque makes the measurement not very accurate. Which of the three provides the most accurate data? Impossible to say.

Let's go to one of the training sessions that I liked the most in order to extract important data. A session of short intervals with a peculiarity: I was very tired after a trail competition two days before. I'll tell you right away why I highlight that fatigue.

First of all, the graph is smoothed to 30 seconds, but there are too many peaks to see anything clearly. We only see that the power goes up and down and that the Stages graph is usually quite low, but I want you to have it available before removing all the small peaks (which in the end are the most important ones when determining the accuracy of the power meter).

We're gonna soften up to 60 seconds.

But more of the same, there's not much clarity to be able to see data clearly and show you what I want, so I'm going to go much further, and smooth out the graph to 5 minutes.

Much better now, right? Well, what you'll see here is that there are two graphics that line up almost perfectly, the PowerTap one and the bePRO pedal one. At the beginning there's a difference of about 7W between them, but the one that stands out is the Stages one, well below the other two.

This is where you begin to see the need to test with three (or more) power meters. Because imagine you were facing the PowerTap hub with the Stages crank. How do you know which one is measuring power correctly? And even more, why the difference in power? This is where the third unit comes into play, which is the one that allows you to give the whole perspective to the tests.

As I said before, in this workout I was quite fatigued, with tired legs (a trail of 2,400m of accumulated unevenness was to blame). Why this huge difference between the Stages meter and the other two? Easy, because of the way the three measure. The PowerTap measures the total power of both legs, the bePRO measures each leg separately and the Stages measures only the left leg and multiplies by two to give the total power.

Generally my power distribution is quite even, around 48-52, with the right leg predominating a little more, so after seeing the data I quickly went to see what the distribution was that day.

Indeed, due to accumulated fatigue the power distribution was quite unbalanced, so Stages is measuring 7% less power, and that less power is doubling, giving much lower values than those collected by the bePRO pedals and the PowerTap G3.

Let's move on to a different day. First of all I want to show you the graphics without smoothing, so you can see why I like to upload the smooth images (which is actually the same thing you do with your cycling computer when you select power at 30s, for example).

Impossible to appreciate anything, but that's how you'll see it when you sync your device with the platform you're using. However, I use it to see the peaks of each of the meters and to be able to judge their accuracy. But let's go with smoothed graphics.

At the beginning of the activity everything appears quite correct and aligned, at least until the stop that can be seen in the graph. Yes, the punctual power peaks are separated in some points, but it is totally normal because of what has been explained before. The trend followed by the three graphs is the same, with the reference and inflection points coinciding. There are minimal deviations at any given time by any of them, but it does not present any problem in its use.

It is from the stop when the deviations by the three potentiometers begin. And you can clearly see how they are deviations that increase over time. None of the three measures in a totally wrong way to rule it out directly, because the trends are coincidental as well as the turning points. The graphs are parallel and when one goes up, the others do too. How do you know which of the three is giving the most accurate data? Impossible to know.

But the variation and difference in power is increasing, in my opinion due to temperature changes and how they have affected each unit and its speed of adjustment to these changes.

So in the next training the goal was to see how they react to the expansion of the materials produced by the temperature, because it's an output with a constant rise in which there is a permanent variation in temperature, dropping as I go up the port. The rise in core temperature is not real, because at that time I was standing making different checks, but being in the sun the Edge 520 is simply warming up for that reason. As soon as I start walking again you'll see that the temperature drops again.

The temperature variation is up to 10 degrees. In the initial part of the climb I had to do 2 minute intense intervals and 1 minute calm. Here things are pretty even or at least the same way we have been seeing in all previous tests, with bePRO and PowerTap correctly aligned and Stages slightly below in terms of power (because of my personal power distribution, with predominance in the right leg).

After stopping in the village at the top of the harbour and waiting for the temperature change we can see more acute differences. I could have done a manual calibration of the potentiometers, but I remember that the idea is to test them under real conditions of use. Besides, I am sure that the most normal thing is not to stop to do manual calibrations, but that you as a user of a power meter will simply calibrate at the beginning of the route (at most).

The PowerTap G3, as indicated above, performs constant calibration when you are not pedaling, so it should continuously adjust as the temperature changes.

But if during the ascent everything matched quite correctly, it's on the way down that there are discrepancies. In this second part I want to forget about the Stages reading, because those higher peaks when the power goes down are not caused by the power meter, but by the unit that is recording data.

The two down peaks are times when I remain stationary, but when smoothing the power does not go all the way to zero (and in the case of Stages, the unit takes a little longer to "note" that there is no power).

But I want you to look at some other details, not just the obvious one. If you look at the clip of the climb, bePRO and PowerTap are aligned almost perfectly, with Stages slightly below (for measuring left leg power and multiplying the two).

After the stop we can see that now the curves that coincide are the Stages and PowerTap ones, being the bePRO graphic above, although little by little they are aligned again perfectly as I approach the end of the training.

There is no other explanation for this behavior than the speed each unit has in adjusting to temperature changes. Or, how the temperature affects each unit.

Again, I highlight one of the best aspects of the PowerTap G3: its constant self-calibration. So while the other two power meters need a few minutes to adjust (both have automatic adjustment for temperature change), the PowerTap G3 is much faster to do so, because every time we stop pedaling, it recalibrates fully automatically.

If I had stopped at the end of the descent to make a manual calibration from the device, the power adjustment would have been faster in the other two units.

And what about the cadence? Well, there's nothing to report. At no time have I been able to see any strange behaviour. In the graphs you can see peaks above and below, but they are simply moments when I stopped pedalling and are the data received by the unit. For some reason PowerTap sends a peak to zero and bePRO a very high peak, but that doesn't affect the final average or, mainly, the cadence reading we have during training.

As you can see, all three devices detect the same cadence data, even though the PowerTap G3 can only estimate the number of pedals per minute.

My opinion

The PowerTap G3 has become one of the reference power meters in recent years. It is easy to understand why: very simple to use, no incompatibilities in its installation and the data obtained is totally reliable.

There's only one question that can come up. Is it the perfect potentiometer for you?

What you have to decide is if you are interested in having the power meter on one wheel. It will depend on whether you use one wheel for everything (racing, training, climbing, etc), or on the contrary you have your training wheel set and on competition days you use something special depending on the type of event you participate in. It also depends a lot on the type of competitions you attend, since short races where you ride a wheel and respond to attacks the power data will not have much weight when deciding whether to respond to that attack, but for longer events it can become a must.

If you only use one set of wheels you won't find any simpler, more reliable and cheaper option. Put a PowerTap hub in your life and forget about everything else. But if you use several sets of wheels then you must decide whether to buy several units for the different games or decide which one you want to have the power information on.

It is one of the most economical options when it comes to measuring power, and it does so in a precise way. There are only two little problems with it: being forced to have it only on one wheel and not seeing the distribution of the pedalling. In both cases it is due to the design of the device itself and its location. As we do not have power for each leg we may be losing data that other meters do provide, but the truth is that neither of these two points would prevent me from training regularly.

Did you like the test?

I hope that this review will be to your liking. There are many hours needed to do each of them. If you like the work I do remember that your support is essential.

Don't forget to share the test in your social networks and with friends, so that they can also be informed. And don't hesitate to comment and subscribe to the comments, many times you will find answers to questions that have not been dealt with in the text of the test. You can use the test comments as a forum and share not only your doubts, but also your opinions of the watch with the rest of the readers.

Buy PowerTap G3

TamalpaisThe PowerTap distributor in Spain, has reduced the price of the PowerTap G3 recently. You can buy the hub from only 599 euros. The complete wheels start at 699 euros, with the training model you have seen in this test.

I don't have any kind of relationship with Tamalpais, so if you buy directly from their store there are no commissions of any kind. So if you liked the test and want to help the page continue, in your Amazon purchases make them through this link, because this will help defray the costs. Thank you!