Currently have a 4iiii power meter on my TT bike and a stages power meter on my road bike. What are your thoughts between a single vs a dual sided power meter.
I had a single sided stages for years and it worked well enough. I recently put a Quarq on my gravel bike and it reads a bit lower and seems to put out a better signal. Apparently my left/right balance is pretty even so the left only didn’t make a huge difference for me over the years. I’m a big fan of the Quarq as the “just works†factor is high and that’s important to me. I haven’t had many problems with my Stages until recently but have several friends who have had problems since day one. Small sample size, but just my two cents.
not to turn this back on you, but have you noticed a difference yourself? And, is one of them single sided and the other one dual? Or are they both single sided and your’e thinking of replacing one with a dual?
Given the issues in getting two independent power meters both reading correctly, zero’ed, linked, etc, I’m warming to the idea that single sided power meters aren’t all that bad. In fact, with a number of L/R power meters I’ve ridden lately, disconnecting the right side and riding left only gave me better numbers. There’s a whole rabbit hole of info as to why… but… my opinion on left only has changed. (only a little… +/-5%) 
My personal favourite PM has been my Quarq S-Works spiders. Which I’d call ‘dual-ish’… They’re years old now though, ANT+ only, but never missed a beat. And they could be calibrated/verified with a static weight. If I was picking parts off the shelf for a new bike, I’d be leaning towards a Quarq or spider based PM - based on my previous experience.
The concern I have about true L/R power meters is their calibration and accuracy. Is a claimed +/-1.5% accuracy on each side, or both combined? I’ll go with each side as I can’t recall seeing anyone claiming better than +/-1% on an outdoor power meter. So in theory 3% difference is possible. That’d throw L/R balance out of whack, while being within spec. Hmmm.
I was looking to pick something up along these lines for black friday coming up. Was leaning toward a single sided unless someone can convince me otherwise.
The concern I have about true L/R power meters is their calibration and accuracy. Is a claimed +/-1.5% accuracy on each side, or both combined? I’ll go with each side as I can’t recall seeing anyone claiming better than +/-1% on an outdoor power meter. So in theory 3% difference is possible. That’d throw L/R balance out of whack, while being within spec. Hmmm.
If a true L/R power meter is rated at 1.5% margin for accuracy for each side, then it’s also 1.5% margin for accuracy total, not 3%.
For example, if your left side is supposed to be 100 watts, but it’s off by 1.5%, then it’s either reading at 98.5 or 101.5. Let’s assume the same for the right.
So while you are supposed to be getting a total of 200 watts, the lowest it will possibly read is 197 (98.5 + 98.5), or the highest it will read is 203 (101.5 + 101.5). 197 or 203 is still within 1.5% of 200.
Arh, yes. Corrected in relation to absolute accuracy. But with one at -1.5% and another at +1.5%, they’re 3% different to each other. Which is what the L/R balance will show, they’re showing relative watts to each other (which was my wording, but not explained well enough). I have a few things here I’m scratching my head with… so I could still be well off the mark. I have two L/R power meters on the same bike showing different L/R balance values (crank vs pedal).
I have a stages one sided (3 years) and L/R (<1year). They have both been great, no issues; the singe side stages has been bullet proof for the most part for me. However, I have noticed there is a slight watt difference (the L/R is lower) between the two, somewhere in the ballpark of 10-15 watts. The ant+ pedal metrics are fun with the L/R, but the two sided leaves a bit more room for possible tech problems. Either are great, the single would be a little more budget friendly and easy use if you just looking for power numbers.
You are overthinking this way too much.
If a true L/R power meter is reading 1.5% high on one side and 1.5% low on the other side, then it pretty much means the total power is going to be spot on. That’s a good thing…right? And at the VERY WORST…it will be 1.5% too high or too low. Given that there is no 100% accurate power meter available, this pretty much as good as you’re going to get. There is nothing to scratch your head about.
I’m with you on the total watts. My discussion is on the L/R balance. Which can only be relative to each side. If they’re both ‘out’ by the maximum amount (1.5%) then the pedal balance information will be wonky when in fact the rider might be 50/50… Leading to an unnecessary overcompensation on one side if they’re looking to ‘balance out’ their pedal stroke. Thus making pedal dynamics useless… ish.
Happy to be corrected. In fact, this whole Internet/technology thing is getting the better of me lately. 
First of all…power balance is relatively useless. If you’re computer is saying 49/51, right or wrong…are you going to start pedaling 1% harder on your left side because of it?
Secondly, not only is total power accurate to 1.5%, but so is power balance. It’s not a potential 3% off. Just do the math.
If you’re pedaling at 100 watts on each leg for 200 total watts. One side is reading 1.5% high, the other side is 1.5% low. That’s 98.5 and 101.5. Calculate the L/R balance…
It’s 98.5 divided by 200 and 101.5 divided by 200. Which results in a 49.25/50.75 balance. Each side is reporting a balance that is off by 0.75% in the worst case scenario.
LOL dude…you’re overthinking this.
First of all…power balance is relatively useless. If you’re computer is saying 49/51, right or wrong…are you going to start pedaling 1% harder on your left side because of it?
Secondly, not only is total power accurate to 1.5%, but so is power balance. It’s not a potential 3% off. Just do the math.
If you’re pedaling at 100 watts on each leg for 200 total watts. One side is reading 1.5% high, the other side is 1.5% low. That’s 98.5 and 101.5. Calculate the L/R balance…
It’s 98.5 divided by 200 and 101.5 divided by 200. Which results in a 49.25/50.75 balance. Each side is reporting a balance that is off by 0.75% in the worst case scenario.
LOL dude…you’re overthinking this.
I agree. This is doing my head in. But… while I’m punching on with YouTube in the background… I’ll reply again (sorry!). I’m totally ok with being schooled here. And I appreciate it, I need to understand / correct my thinking.
Relative. That’s what I can’t get my head around. There is no 200W to refer to in the calculation of balance. L thinks -1.5% relative to 100W (true accuracy) is correct, R thinks +1.5% relative to 100W is correct. They’re both incorrect by 1.5% to the theoretically correct figure (which neither know, so they can’t calculate anything based on it. 98.5/101.5=.97. 101.5/98.5=1.03. A 3% difference relative to each other. Making L/R balance pretty much useless… but the overall wattage accurate. It gets interesting if the power meter spec is ±3%.
Flagging someone else to come in and help hijack this thread even more than we have!
Assuming all power numbers are accurate, if your computer told you that you were pedaling 49/51, how far off is your pedal balance?
Is it 1% because each side is off by 1%?
Is it 2% because 51 minus 49 is 2?
Or is it 4% because 51 divided by 49 is 1.04?
Not that any of it really matters anyway because pedal balance is an irrelevant data element to train by, but do you see the mental gymnastics you’re putting yourself through. You’re just throwing around numbers and multiplying or dividing them till they become scary to you. LMAO.
I have a stages one sided (3 years) and L/R (<1year). They have both been great, no issues; the singe side stages has been bullet proof for the most part for me. However,** I have noticed there is a slight watt difference (the L/R is lower) between the two, somewhere in the ballpark of 10-15 watts**. The ant+ pedal metrics are fun with the L/R, but the two sided leaves a bit more room for possible tech problems. Either are great, the single would be a little more budget friendly and easy use if you just looking for power numbers.
You mention you are seeing your L/R powermeter lower by 10-15 watts. Do you happen to have the new stages L/R Ultegra or Dura Ace model or are you using a different crank brand? This thread comments on why the Shimano crank based meters or showing lower.
My personal favourite PM has been my Quarq S-Works spiders. Which I’d call ‘dual-ish’… They’re years old now though, ANT+ only, but never missed a beat. And they could be calibrated/verified with a static weight. If I was picking parts off the shelf for a new bike, I’d be leaning towards a Quarq or spider based PM - based on my previous experience.
The S-Works spiders have also been my favorite for years… I have been through quite a few of them. They do now make the D-zero in a Specialized spider version and that is what I am now running on my road and tt bikes… adds BT and is slightly more accurate (and a little cheaper since PM prices have come down.) It isn’t well advertised as the old version was now that Specialized has their own power meters.
First of all…power balance is relatively useless. If you’re computer is saying 49/51, right or wrong…are you going to start pedaling 1% harder on your left side because of it?
If you’re off by 40/60 once over 1000W, like I was - then yes I worked on correcting that imbalance and pedaling harder on the weak side. And successfully improved my track standing starts (measured in standing 200m time). 49/51, I wouldn’t worry about.
If you’re pedaling at 100 watts on each leg for 200 total watts. One side is reading 1.5% high, the other side is 1.5% low. That’s 98.5 and 101.5. Calculate the L/R balance…
That’s nice, but I have zero confidence that reported accuracy values mean much at all. Or that they’re nice normally distributed values like your math implies. It’d be nice if there was some sort of industry standard for how to measure PM accuracy. (and, no, hanging static weights off a crank arm isn’t going to cut it). Otherwise they’re just numbers that manufacturers report, and the best we can do is record performance vs. other power meters and see how well they correlate with each other. Which one is right? Who knows…just call the SRM the gold standard (despite it falling for the same 4-arm crank issue as everyone else, apparently).
, and the best we can do is record performance vs. other power meters and see how well they correlate with each other. Which one is right? Who knows…
Well, there are ways but they’re a PITA. Some ways require another device (for example, one researcher used a driven bicycle treadmill). If you don’t have another independent device, then you can still do it but it’s an even bigger PITA. (I’m not suggesting that anyone do things that are a PITA, I’m just saying that ways do exist rather than throwing up one’s hands and saying no one could possibly know).
I’m just saying that ways do exist rather than throwing up one’s hands and saying no one could possibly know).
Oh I’m sure there’s ways, but my point is there’s no standard. They all seem to be proprietary. There’s no way that you can do it, and then I can do it, and we get the same #'s. Use these instruments with these precision specs. Particularly profiles for testing across different temperatures, acceleration ramps, etc.
I’m just saying that ways do exist rather than throwing up one’s hands and saying no one could possibly know).
Oh I’m sure there’s ways, but my point is there’s no standard. They all seem to be proprietary. There’s no way that you can do it, and then I can do it, and we get the same #'s. Use these instruments with these precision specs. Particularly profiles for testing across different temperatures, acceleration ramps, etc.
Here’s my non-proprietary protocol: in order to test the dynamic accuracy of the PM across a wide range of powers, you could do VE tests where you know the Crr and CdA and derive the profile at different levels of power (and, thus, speed). You knew VE was going to show up somehow, right? I’m a one-trick pony.
How do you validate the Crr and CdA? You estimate them by coasting down a shallow hill at different speeds but at zero power (if your PM doesn’t read zero at zero power then you have other problems). Then you use those values for the powered tests up a hill, and on the flat, and on varying terrain. If you get the same VE profile while varying your power then the power meter must be reporting accurately. This works even if the road surface isn’t uniform, but you really really have to have it be calm.
I actually did this, one time, just to check that I could. It was reassuring but a huge PITA.
My personal favourite PM has been my Quarq S-Works spiders. Which I’d call ‘dual-ish’… They’re years old now though, ANT+ only, but never missed a beat. And they could be calibrated/verified with a static weight. If I was picking parts off the shelf for a new bike, I’d be leaning towards a Quarq or spider based PM - based on my previous experience.
The S-Works spiders have also been my favorite for years… I have been through quite a few of them. They do now make the D-zero in a Specialized spider version and that is what I am now running on my road and tt bikes… adds BT and is slightly more accurate (and a little cheaper since PM prices have come down.) It isn’t well advertised as the old version was now that Specialized has their own power meters.
I have a Specialized spider version and like it. I also had a Quarq on a SRAM Force 22 crank and it was also removable. In some ways, I liked that crank better since the BB on the SRAM was more “set it and forget itâ€