Powermeter choices

13

Comments

  • neeb
    neeb Posts: 4,473
    If you only zeroed the torque 5 times in 12 months on your power tap, then almost certainly the data isn't data. it's junk.
    Is that necessarily the case? I calibrate mine most rides, but the value I get is always between 497 and 500. That suggests that if I didn't calibrate it would only be out by 0.6%. I've only had it 2 months though, and I guess the temperature has rarely varied by more than 5 degrees.
  • bigpikle
    bigpikle Posts: 1,690
    rozzer32 wrote:
    But is there any evidence to say the Power2max are inaccurate? I thought it was the drift that was the issue.

    I think this goes right back to the question of variation between units....

    Some report that drift settles down after 10 mins, and some say as long as 40 mins, even with no significant change in temperature. After 40 mins I'm well stuck into intervals and the last thing I'd want to think is that the unit might still be drifting and reading inaccurate numbers. As we also dont know what/how/when the units are zero'ing themselves, and it may be doing it while you are stomping on the pedals at the time, it just leads to all sorts of uncertainty. Thats what concerns a lot of people.
    Your Past is Not Your Potential...
  • rozzer32
    rozzer32 Posts: 3,923
    Bigpikle wrote:
    rozzer32 wrote:
    But is there any evidence to say the Power2max are inaccurate? I thought it was the drift that was the issue.

    I think this goes right back to the question of variation between units....

    Some report that drift settles down after 10 mins, and some say as long as 40 mins, even with no significant change in temperature. After 40 mins I'm well stuck into intervals and the last thing I'd want to think is that the unit might still be drifting and reading inaccurate numbers. As we also dont know what/how/when the units are zero'ing themselves, and it may be doing it while you are stomping on the pedals at the time, it just leads to all sorts of uncertainty. Thats what concerns a lot of people.

    I agree with you but if the drift issue is sorted as they say then surely everything is sweet. It auto zeros when you stop pedalling for 2 seconds.
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • Doobz
    Doobz Posts: 2,800
    If you only zeroed the torque 5 times in 12 months on your power tap, then almost certainly the data isn't data. it's junk.

    if you start at say 300 W, and increase to 350 W, and then change your power meter (when your previous one wasn't correct) then your new one could say 320 or 380 or whatever. I can't stress enough how important it is to have consistent, accurate, data.

    I've seen incorrect data - some power meters have been way out. but even when the error is small it makes a huge difference.

    if you're not interested in accurate, calibrated data, then i'd suggest not getting a power meter. having a power meter is a bit like having a pet. you have to look after it.

    Those times I recalibrated was only when I noticed things were not quite right. Most other times my HR and power zones were matching up spot on so I never really questioned anything. If it aint broke then why fix it?

    I was always under the impression that with "auto zero" enabled on the Edge 500 that the Powertap's would Zero themselves when not under load / freewheeling?

    I wonder if I had been more vigilant and made sure things were always zero'd things would have been any different this season? I doubt it..

    What is cool though is since buying the P2M I have been doing a lot of reading and posts like yours and others make me aware of how important it is to make sure you are getting the most accurate data you can from your training.
    cartoon.jpg
  • Doobz
    Doobz Posts: 2,800
    rozzer32 wrote:
    Bigpikle wrote:
    it just leads to all sorts of uncertainty. Thats what concerns a lot of people.

    I agree with you but if the drift issue is sorted as they say then surely everything is sweet. It auto zeros when you stop pedalling for 2 seconds.

    Check this from one of the dudes from Power2Max

    Question
    I was wondering if you could clear something up that has been bugging me for a few days.

    Hopefully this question makes sense!

    I understand that the unit updates the calibration when no pedalling action is happening. What is happening if there is force being applied during this recalibration moment? Wont the new zero value be adjusted as the cranks have force applied resulting in different power values?

    thanks..

    Answer
    The chances are very small of getting a wrong number while it gets a new zero. We measure 50 times a second and only if all those numbers are in a certain area the algorithm works. It is developed that it is nearly impossible to get a wrong number. If you really try and know how it calculates, it is possible to trick it, but this will not happen during the ride. We tried to make it bullet proof.



    Michael
    cartoon.jpg
  • bigpikle
    bigpikle Posts: 1,690
    Doobz wrote:
    The chances are very small of getting a wrong number while it gets a new zero. We measure 50 times a second and only if all those numbers are in a certain area the algorithm works. It is developed that it is nearly impossible to get a wrong number. If you really try and know how it calculates, it is possible to trick it, but this will not happen during the ride. We tried to make it bullet proof.

    what is small? or nearly impossible? 1 in 10, 1 in 20, 1 in a million?

    50 times a second for a 3 hour ride = 540,000 times
    if the chance is 1 in 100,000 then already its likely to have got it wrong 5 times...even if its 1 ina million you have a 50% chance of the zero being wrong.

    Dont get me wrong, as I'd like to own a P2M, but it justt seems a little too vague right now. It wasnt long ago P2M said their product wasnt significantly affected by temperature changes....
    Your Past is Not Your Potential...
  • rozzer32 wrote:
    That makes sense. But say the Power2max is 20 watts off "real values" does that really matter? As it will always be 20 watts off? So is it consistency that matters or absolute values?
    With instrumentation, precision and accuracy tend to go hand in hand.

    Nevertheless, whether it matters depends on its intended use. If you never intend to use another power meter, ever, then it probably doesn't matter as much as long as you are confident that the data is consistent, within a ride and from ride to ride.
    rozzer32 wrote:
    As I can see the other side of the coin that says if you're power is reading 20 watts out then you could think you're training is say zone 5 but in reality it's zone 4.
    Not if the slope of the meter is linear (I've no idea if this has been tested with P2Ms). If the slope is non-linear, then this will be a problem.


    I wouldn't use a meter I couldn't calibrate for tasks such as aerodynamic field testing, or other physical modelling. And any assessment of performance involving more than one power meter becomes problematic (be it contemporaneously, or over a longer time frame).

    Keep in mind that the impact of drift in torque zero, and potentially of a power meter's slope, on power numbers will not be a fixed number, nor a fixed percentage of watts.
  • doyler78
    doyler78 Posts: 1,951
    Bigpikle wrote:
    Doobz wrote:
    The chances are very small of getting a wrong number while it gets a new zero. We measure 50 times a second and only if all those numbers are in a certain area the algorithm works. It is developed that it is nearly impossible to get a wrong number. If you really try and know how it calculates, it is possible to trick it, but this will not happen during the ride. We tried to make it bullet proof.

    what is small? or nearly impossible? 1 in 10, 1 in 20, 1 in a million?

    50 times a second for a 3 hour ride = 540,000 times
    if the chance is 1 in 100,000 then already its likely to have got it wrong 5 times...even if its 1 ina million you have a 50% chance of the zero being wrong.

    Dont get me wrong, as I'd like to own a P2M, but it justt seems a little too vague right now. It wasnt long ago P2M said their product wasnt significantly affected by temperature changes....
    Bigpikle wrote:
    Doobz wrote:
    The chances are very small of getting a wrong number while it gets a new zero. We measure 50 times a second and only if all those numbers are in a certain area the algorithm works. It is developed that it is nearly impossible to get a wrong number. If you really try and know how it calculates, it is possible to trick it, but this will not happen during the ride. We tried to make it bullet proof.

    what is small? or nearly impossible? 1 in 10, 1 in 20, 1 in a million?

    50 times a second for a 3 hour ride = 540,000 times
    if the chance is 1 in 100,000 then already its likely to have got it wrong 5 times...even if its 1 ina million you have a 50% chance of the zero being wrong.

    Dont get me wrong, as I'd like to own a P2M, but it justt seems a little too vague right now. It wasnt long ago P2M said their product wasnt significantly affected by temperature changes....

    That assumes that you are trying to zero every second for 3 hours which of course you are not. The vast majority of the seconds will be spent turning the pedals I'd hope and even then what is happening is that 50 records are compared to see if they fall within a given spec to effect an auto zero so you aren't multiplying it by 50 but instead by 1 as there is only one calculation that is deterministic and that's the final result from all those samples.

    As to vagueness well he just answered the question put to him. If anyone wants more clarity then they should just ask him :wink:
  • rozzer32
    rozzer32 Posts: 3,923
    rozzer32 wrote:
    That makes sense. But say the Power2max is 20 watts off "real values" does that really matter? As it will always be 20 watts off? So is it consistency that matters or absolute values?
    With instrumentation, precision and accuracy tend to go hand in hand.

    Nevertheless, whether it matters depends on its intended use. If you never intend to use another power meter, ever, then it probably doesn't matter as much as long as you are confident that the data is consistent, within a ride and from ride to ride.
    rozzer32 wrote:
    As I can see the other side of the coin that says if you're power is reading 20 watts out then you could think you're training is say zone 5 but in reality it's zone 4.
    Not if the slope of the meter is linear (I've no idea if this has been tested with P2Ms). If the slope is non-linear, then this will be a problem.


    I wouldn't use a meter I couldn't calibrate for tasks such as aerodynamic field testing, or other physical modelling. And any assessment of performance involving more than one power meter becomes problematic (be it contemporaneously, or over a longer time frame).

    Keep in mind that the impact of drift in torque zero, and potentially of a power meter's slope, on power numbers will not be a fixed number, nor a fixed percentage of watts.

    But surely if you did change power meter, say from P2M to Quarq you would do another FTP test with the new Quarq and work of that value?

    I'm not wanting the power meter for aero testing etc, just for "standard" road bike racing training.

    Can you explain what you mean in the bit of highlighted in bold? (again sorry for the power newbie questions but I do appreciate your help).
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • Doobz wrote:
    If the new power meter is out by even 50w then I would be worried.
    50W!
    5W and I start to wonder.
    Doobz wrote:
    Accuracy has never been the issue with the P2M units. Its always been about the data drift due to temp issues which they have now sorted.
    1. drift is an accuracy problem
    2. P2M said they didn't have a problem before, now they've fixed the problem they apparently didn't have.
  • neeb wrote:
    I calibrate mine most rides, but the value I get is always between 497 and 500. That suggests that if I didn't calibrate it would only be out by 0.6%.
    No, you are mistakenly thinking that the "calibration" value is a measure of torque, when it's a marker for the zero point.

    The error would be calculated against the difference between the torque reading with force applied to crank and the zero point.

    e.g. in a 53/15 gear, riding at ~ 300W, then the equivalent average torque applied to the crank is ~18kg @ 93rpm.

    On that gear, then the Powertap would report a difference in the average effective pedal torque values of ~ 75 foot-lbs. i.e. the calibration number would move from a zero point of 500 to a loaded value of 575.

    If your PT has a zero that is out by 3 units, then the error in torque using that gear would be 3/75 = 4% = 12W

    When fit, I can spend many months to lift my power output by that much.
  • Doobz wrote:
    I was always under the impression that with "auto zero" enabled on the Edge 500 that the Powertap's would Zero themselves when not under load / freewheeling?
    The Powertap autozero only works if the zero point has shifted within a narrow range. If the zero point is beyond that range, then the autozero no longer operates.

    This is why with a Powertap it is especially important to set the torque zero before you start any ride. It's important for all meters, but that is a particular feature of the Powertap.
  • rozzer32 wrote:
    But surely if you did change power meter, say from P2M to Quarq you would do another FTP test with the new Quarq and work of that value?
    Of course. Year on year improvements might be of the same order of magnitude as the difference though.
    rozzer32 wrote:
    I'm not wanting the power meter for aero testing etc, just for "standard" road bike racing training.
    Until you want to find other ways to get faster than just training.
    rozzer32 wrote:
    Keep in mind that the impact of drift in torque zero, and potentially of a power meter's slope, on power numbers will not be a fixed number, nor a fixed percentage of watts.
    Can you explain what you mean in the bit of highlighted in bold? (again sorry for the power newbie questions but I do appreciate your help).
    For that, you need to understand how a power meter calculates power.

    The torque measurement is a linear equation (well it is on a good power meter) with a slope and an intercept. The torque zero point is the intercept. Power is then a function of the accuracy of both the slope and the intercept, and the rotational speed of the component involved (crank/hub).
  • neeb
    neeb Posts: 4,473
    neeb wrote:
    I calibrate mine most rides, but the value I get is always between 497 and 500. That suggests that if I didn't calibrate it would only be out by 0.6%.
    No, you are mistakenly thinking that the "calibration" value is a measure of torque, when it's a marker for the zero point.

    The error would be calculated against the difference between the torque reading with force applied to crank and the zero point.

    e.g. in a 53/15 gear, riding at ~ 300W, then the equivalent average torque applied to the crank is ~18kg @ 93rpm.

    On that gear, then the Powertap would report a difference in the average effective pedal torque values of ~ 75 foot-lbs. i.e. the calibration number would move from a zero point of 500 to a loaded value of 575.

    If your PT has a zero that is out by 3 units, then the error in torque using that gear would be 3/75 = 4% = 12W

    When fit, I can spend many months to lift my power output by that much.
    ...with a Powertap it is especially important to set the torque zero before you start any ride. It's important for all meters, but that is a particular feature of the Powertap.
    That's extremely useful to know, thanks for the lucid explanation.

    If I am doing a 10min warmup before starting a session, would it be best to calibrate as soon as the bike is out the door, or after the warmup? I suppose the unit will be nearer ambient temp. after 10mins, but is it designed to be calibrated during/after it has been running? I've seen advice about leaving the bike outside for 10mins before calibrating & riding, but as I'm in a 5th floor apartment, practically that's never going to happen...
  • neeb wrote:
    If I am doing a 10min warmup before starting a session, would it be best to calibrate as soon as the bike is out the door, or after the warmup?
    Both. Then you can observe the typical behaviour of your meter.

    Considering time for the components to acclimate to changes in operating temperature is a good idea.
  • As we also dont know what/how/when the units are zero'ing themselves, and it may be doing it while you are stomping on the pedals at the time, it just leads to all sorts of uncertainty. Thats what concerns a lot of people.
    A guy from P2Max posted on the Wattage list that the re-zero occurred if (paraphrasing somewhat),

    It's more than 30 seconds since the last zero,
    and the crank stops turning for two seconds,
    and the torque during that time doesn't vary more than plus or minus 1 of their units.

    The torque is sampled 50 times a second.

    You can force a false zero by coming to rest on a hill and keeping steady pressure on one pedal.

    It will be interesting to see what the reviews make of their temperature compensation, and I'd like an explanation of what appears to be non-temperature related turbo drift.

    Paul
  • rozzer32
    rozzer32 Posts: 3,923
    So does that mean you could fool it if;

    You did a trackstand?
    Roll up to a set of lights at slow speed when you're out the saddle?
    Roll along out the saddle?

    Obviously not pedalling in the above.


    Paul what issue is the non-temperature related turbo drift? Can you explain please?
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • bigpikle
    bigpikle Posts: 1,690
    he's referring to examples of people seeing drift in values when using turbos in environments where temperature doesnt change. Lots of people are banging on about the new fix for temperature related issues but there appear to be units doing weird things when the temperature is stable.
    Your Past is Not Your Potential...
  • So does that mean you could fool it if;

    You did a trackstand?
    Roll up to a set of lights at slow speed when you're out the saddle?
    Roll along out the saddle?
    A track stand is unlikely to fool it unless you can hold a constant pressure against the brakes. It probably wouldn't happen by accident. The other scenarios don't matter since if the force is constant and the crank isn't turning then the zero is as good as it can be.

    Paul
  • rozzer32
    rozzer32 Posts: 3,923
    Bigpikle wrote:
    he's referring to examples of people seeing drift in values when using turbos in environments where temperature doesnt change. Lots of people are banging on about the new fix for temperature related issues but there appear to be units doing weird things when the temperature is stable.

    Most people use a turbo in a closed environment, garage, spare room etc. Even with fans going and windows open the temperature still increases. Well at least that's what I find when I turbo train.
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • bigpikle
    bigpikle Posts: 1,690
    A degree or 2 maybe, but what if your see a 20-40w drift in output at the same time? People also have temp data that shows only very small temp changes that shouldnt affect data like it does.

    Its also the fact that some do and some dont that worries many. One reviewer on Wattage has 2 units that exhibit wildly different behaviour in the same circumstances.

    You seem to have convinced yourself these units are all totally reliable and there arent any problems despite users tests and reports.
    Your Past is Not Your Potential...
  • amaferanga
    amaferanga Posts: 6,789
    How many of these people actually exist though? I think I only know of one, but he bleats on so much about it on the Wattage forum that you could be mistaken for thinking that almost everyone has the same issue.

    There are issues with all power meters (broken torque tube anyone?) and I think it's likely that there is some kind of problem with some P2M, but I just don't think non-temperature related drift is a problem with P2M in general.
    More problems but still living....
  • rozzer32
    rozzer32 Posts: 3,923
    Bigpikle wrote:
    A degree or 2 maybe, but what if your see a 20-40w drift in output at the same time? People also have temp data that shows only very small temp changes that shouldnt affect data like it does.

    Its also the fact that some do and some dont that worries many. One reviewer on Wattage has 2 units that exhibit wildly different behaviour in the same circumstances.

    You seem to have convinced yourself these units are all totally reliable and there arent any problems despite users tests and reports.

    I haven't convinced myself at all. And I was only looking at getting a power meter around Christmas anyway so it will give time to see if these issues have been solved. But as it is £600 cheaper than a Quarq that is a fair chunk of money, but you're right in that if it doesn't work then it doesn't matter.

    I was asking as I haven't heard of the turbo issue, so I just wanted some more info on it.
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • bigpikle
    bigpikle Posts: 1,690
    I'm in the same boat and would really like a crank system i can swap between bikes, instead of my PT wheel which limits choice but has been very reliable the last couple of years. I'm hoping the P2M issue is a thing of the past with the fix, as the cost saving is potentially huge.

    I think one issue is that not many people really understand the power meter technology or pay close attention to whats going on. I think a lot of people just fit them and ride and assume everything is going to be fine. As a result you end up with a very small number of people focused on any issues etc. I bet 9/10 of users never really check or do anything...
    Your Past is Not Your Potential...
  • jerome3
    jerome3 Posts: 209
    Anybody got any experience of quarq and any probs encountered thinking of swapping over my powertap for one so as I can switch cranks between bikes
  • Like i said, you probably need to think of a power meter like a pet (sort of). It needs to be looked after, checked, and calibrated. newer units are subject to less testing (i.e., consumers using them) than older units. Often it takes a while for people to come out with various tales, and it takes some times for experts to draw an opinion. I remember when Ergomo came out and everyone was very excited and this was followed by a loss of excitement.

    If i was being given a power meter, then i'd be happy to try any one. if i have to purchase a power meter then i want one that has been reliably tested, that i know is accurate, and one that has a good back up service (because all power meters need to go back at some point). this leaves me with 3 power meters i would choose at present.

    it may be that the P2max system has been sorted out now (i have no idea). but to be honest, i'd rather let some other fool test that out. When you start collecting power data, if something goes wrong it can become pretty frustrating. same with Polar or any other system. If companies want to pay me to test their system, then that's fine. being frustrated with incorrect data i can take if i'm being paid.

    Note that i don't sell any power meters, or receive anything from the power meter companies (although in the past i have done some testing for power tap, and did some non power related work for polar).
    Coach to Michael Freiberg - Track World Champion (Omnium) 2011
    Coach to James Hayden - Transcontinental Race winner 2017, and 2018
    Coach to Jeff Jones - 2011 BBAR winner and 12-hour record
    Check out our new website https://www.cyclecoach.com
  • rozzer32
    rozzer32 Posts: 3,923
    Well hopefully by December/January enough people will have the new units to tell us if they're any issues. If people are still complaining then it would probably be worth spending the extra for a Quarq.

    Although there will always be stories of people having problems with their PM if it's a Quarq, SRM, Powertap, P2M. Things do go wrong. But obviously the big problems need to be sorted.
    ***** Pro Tour Pundit Champion 2020, 2018, 2017 & 2011 *****
  • Doobz
    Doobz Posts: 2,800
    Lets make no bones about it. P2M has had some issues and yea they are fairly new compared to SRM.

    P2M have addressed the major (pretty much only) issue that the units displayed and things are moving forward.

    Every company has its problems with their units. Take a Look at SRM for example. A while back they had a real issue with Data Drift. According to SRM the problem was due to a bad batch of epoxy and it took them a long time to resolve. People still went out and purchased SRMs though :) -

    This below I found interesting..
    Where does this drift issue stem from? Cranks and power meters and strain gauges are all made from metal. Metal is susceptible to changes in temperature; it expands and contracts. Have you ever noticed when you go over a bridge there are metal gaps in the pavement? That’s to allow the bridge to flex and handle the temperature change. How does a crank-based power meter handle the change? They offset the change by zeroing the readings from the strain gauges when there is no load. Each manufacturer does it differently. I’m told with Quarq you pedal backwards twice. Power2Max auto-zeroes every time you stop pedaling for two seconds. According to “Training and Racing with a Power Meter, 2nd Edition,” with the SRM “(the rider) must create a ‘zero-offset,’ or ‘zero-point’ to ensure that the wattage will be zero when there is no load on the power meter. This is a FIVE-SECOND procedure and can be done throughout the ride with no detrimental effects to your data.” Here’s a pic of that paragraph if you don’t believe me:

    bookpage.jpg?w=604&h=453

    Taken from here http://hiprouleur.wordpress.com/2012/05 ... xperience/

    One thing I would like to say is Alex and Rick, there is no doubt you have VAST amounts of knowledge in Training, Coaching, Power Meters etc and I respect and appreciate that.

    What you don't have though is first hand experience with running a Power2Max unit. You are using second hand old out dated information off the internet to make assumptions on a new "to be released" product.. It comes across as a bit unprofessional.
    cartoon.jpg
  • doyler78
    doyler78 Posts: 1,951
    Interesting how you can criticise others for their lack of first hand knowledge of a product when you have no knowledge of it either yet you feel you can use phrases like "has had some issues" and "P2M have addressed the major (pretty much only) issue that the units displayed and things are moving forward". Lets wait and see.

    As for requiring first hand knowledge before anyone can offer an opinion as to whether they would spend their money on something or not is patently stupid. Anyone that spends a large sum of money based only on manufacturer claims is foolish with their money IMHO. Of course someone has to be first however they take a very big risk, though you would hope a calculated one but that depends on what reassurances that a manufacturer can give and that of course is highly subjective but then again it is your money so your are entitled to set the bounds.

    As for P2M well all I can say is that they lost a lot of credibility in my eyes when they denied a problem existed when they were actually working on a fix. That is simply deceitful and that makes me wary of them. I too was looking at these however I'll be waiting for some real world feedback on these and not the assurances of P2M. I've no desire to replace one lump of sh1t with another.
  • neeb
    neeb Posts: 4,473
    neeb wrote:
    If I am doing a 10min warmup before starting a session, would it be best to calibrate as soon as the bike is out the door, or after the warmup?
    Both. Then you can observe the typical behaviour of your meter.

    Considering time for the components to acclimate to changes in operating temperature is a good idea.
    I tried calibrating twice today. Got 496 before the ride (bike outside for less than a minute) then 498 after a 15 min warmup. That surprised me a little, as I'd got the impression that I tended to get higher calibration numbers with higher ambient temperatures (or am I just imagining that?), and as it was about 12C I'd expected to get a lower number after 15mins outside. But then I was moving for 15mins, so does the hub spinning maybe cause the unit to warm up? Does it matter? Will the power readings always be more accurate if the unit has been calibrated recently, irrespective of what causes the calibration numbers to vary?

    I'm also using the autocalibration function on the Garmin, so perhaps I only need to calibrate at the start of the ride, unless there are extreme temperature variations?