Login required to started new threads

Login required to post replies

Prev Next
Zero offset calibration on new Quarq ELSA 10R
Quote | Reply
Forum members,

Please see the below post. I sent this inquiry to Quarq after purchasing a new ELSA 10R power meter from them. I had previous experience with a Quarq S975 Cinqo power meter but that unit had to be returned because I began to experience a lot of data drops. While I am waiting for a reply from Quarq I figured I would poll you guys and gals to see what you think. Any help and suggestions would be appreciated!

Sent to Quarq:
I just purchased an ELSA 10R. I recently returned the S975 model due to data drops. My question has to do with the zero offset calibration on the ELSA model. I have done about 10 rides on the ELSA to date (mostly indoor rides on a Kurt Kinetic road machine trainer using a Garmin 500). My procedure is to conduct a manual zero offset calibration prior to starting my training session (Drive side crank in the 6 o-clock position, unclipped). My reading is consistently around -195. I then start my warm up (15 minutes of light pedaling with two 1 minute higher intensity efforts). This takes about 20 minutes total. I then back pedal to conduct an auto zero calibration while on the bike prior to starting my intervals. When I do this it appears I have to work harder to generate the same wattage numbers as earlier in the workout. Kind of like the power meter is "warmed up" and giving different numbers. After noticing this I got off the bike and did another manual zero offset calibration (unclipped, drive side in the 6 o'clock position). My reading was about 20 units different -175. This training session was being done in a basement with a constant temperature of 65 degrees. This 20 unit deviation resulted in about a 10-15 watt difference for an identical effort (Same gear, same cadence, same indoor trainer, etc). According to your literature as long as the manual calibration numbers don't deviate more than 50 units then you are okay. I just want some direction from you guys as to what number is inherently more accurate. If I am doing an indoor training session and conduct a pre-ride manual calibration should I even do an auto calibration or another manual calibration during the session or was the first manual calibration the most accurate it will get? Or should I definitely do an auto zero calibration (by backpedaling 5 times) 10 or 15 minutes into my session to get the most accurate power numbers? It seems to me that the power numbers I was getting in the beginning of the session (after the pre-ride manual calibration but before the during session auto calibration) were inflated and that the numbers I was getting after conducting the auto calibration during the session were more accurate. Also I have checked to make sure my crankset is not loose and my chainring bolts are properly torqued (10 nm) since they are aluminum and my magnet is within 2-8 millimeters of the cadence sensor.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
A 20 unit shift in zero offset on a Quarq should only result in an ~6W difference in power reading at 90 rpm. If your rpms were lower, it would be even less.

Each unit of offset is "worth" 1/32 N-m.

http://bikeblather.blogspot.com/
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Tom A.] [ In reply to ]
Quote | Reply
Thanks for the info. What is your suggestion regarding "when" to conduct a zero offset calibration during indoor trainer sessions? Before only, before and during, and which one would be inherently more accurate?
Last edited by: mcmtbr: Feb 7, 13 22:16
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
See my post towards the bottom of this current thread, admittedly not the ELSA but a S975, am having similar issues. I am working through them with the UK Quarq guy, I need to re-torque my chainring bolts and do some more checks. Thought it might be interesting to you.

Linkydoodle
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [knighty76] [ In reply to ]
Quote | Reply
You do not need to re-zero the Quarq, just zero it before you start the trainer session. On the road in cold weather, I will zero at the start and zero again 20 minutes later, but not by backpedaling, by stopping and using the same procedure through the Garmin. The back pedal routine can cause inaccuracies.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
have you done a few hard jumps on it to seat the chain ring?
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
mcmtbr wrote:
Thanks for the info. What is your suggestion regarding "when" to conduct a zero offset calibration during indoor trainer sessions? Before only, before and during, and which one would be inherently more accurate?


Is the bike stored at that same temp? If so, that seems like a lot of drift. If not, then zero after you know everything is at ambient temps. Also, ditch the aluminum bolts and get steel ones. The extra micrograms for the steel are well worth the ability to torque them tighter.

http://bikeblather.blogspot.com/
Last edited by: Tom A.: Feb 8, 13 7:37
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Tom A.] [ In reply to ]
Quote | Reply
stainless, not rusty-o-rama, i hope :)
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [jeffp] [ In reply to ]
Quote | Reply
jeffp wrote:
stainless, not rusty-o-rama, i hope :)

Naturally.

http://bikeblather.blogspot.com/
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Tom A.] [ In reply to ]
Quote | Reply
This is the reply I received from Quarq. I have to admit I am surprised the reply came so quick. What still seems weird to me is that I have done off the bike manual calibrations pre-ride as well as manual calibrations after about 15 minutes on the trainer and am still getting these 20 units shifts. But I guess what Quarq is saying is that the one pre-ride manual calibration is the most accurate given the consistent temperature conditions of an indoor trainer workout.

Paul,

The manual calibration will be more accurate than auto-calibration. If you are planning to use auto calibrate during the ride, for consistency it may be best to also auto-calibrate at the beginning of the ride as well. As the trainer's rolling mechanism and tire warm up the resistance will also change so after 20 minutes there should be a difference in perceived effort.

For indoor rides at consistent temperatures, a single manual calibration pre-ride should be sufficient and most accurate.

Cheers,
John Stock
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [knighty76] [ In reply to ]
Quote | Reply
I have checked the crank set to ensure there is no play and the torque on the chainring bolts has been checked (loosened the bolts then re-torqued them to 10 nm).
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [jeffp] [ In reply to ]
Quote | Reply
jeffp wrote:
have you done a few hard jumps on it to seat the chain ring?

Yeah. The first couple rides on the ELSA were outdoor rides in excess of 3 hours with numerous "hard jumps" in both the big an small rings to "set" the chainrings and bolts.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Tom A.] [ In reply to ]
Quote | Reply
Tom A. wrote:
mcmtbr wrote:
Thanks for the info. What is your suggestion regarding "when" to conduct a zero offset calibration during indoor trainer sessions? Before only, before and during, and which one would be inherently more accurate?


Is the bike stored at that same temp? If so, that seems like a lot of drift. If not, then zero after you know everything is at ambient temps. Also, ditch the aluminum bolts and get steel ones. The extra micrograms for the steel are well worth the ability to torque them tighter.

The bike was left in the basement between these workout sessions (Constant temperature of roughly 65 degrees). Thanks for the suggestion about the chainring bolts.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
The new spiders "should" be less sensitive to chainring torque than the S975. It may still be affected by crank torque however (as per my understanding).

And, it's always best to do manual calibrations instead of auto. Manually calibrate before you ride, calibrate 15 minutes after (or before you start your intervals when things have had a chance to warm up), and then check your post-ride calibration value. Keep a log of your values and verify that that you're not seeing more than +- 50 before and after the ride.

Another point to consider is that your trainer is also "warming up" and its resistance may also change with temperature/use.

The more you fiddle with stuff, the more issues you're going to have and the more variability you introduce into the system. Give everything a chance to settle in and ride.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [riotgear] [ In reply to ]
Quote | Reply
For training and a lot of racing thats possible, but for a tri where you must leave yor bikein a cold temp and get on it an hour or so later its a concern, particularly when I want the most help with pacing at the begining of the bike leg.

Styrrell
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
mcmtbr wrote:
This is the reply I received from Quarq. I have to admit I am surprised the reply came so quick. What still seems weird to me is that I have done off the bike manual calibrations pre-ride as well as manual calibrations after about 15 minutes on the trainer and am still getting these 20 units shifts. But I guess what Quarq is saying is that the one pre-ride manual calibration is the most accurate given the consistent temperature conditions of an indoor trainer workout.

Paul,

The manual calibration will be more accurate than auto-calibration. If you are planning to use auto calibrate during the ride, for consistency it may be best to also auto-calibrate at the beginning of the ride as well. As the trainer's rolling mechanism and tire warm up the resistance will also change so after 20 minutes there should be a difference in perceived effort.

For indoor rides at consistent temperatures, a single manual calibration pre-ride should be sufficient and most accurate.

Cheers,
John Stock

This is what I sent back to Quarq after getting this reply:

John,

Okay, so based on what you saying the manual calibration is a little more accurate than doing the auto calibration. I imagine this has to do with the dynamic(auto) vs static(manual) nature of how these calibrations are done. But during these indoor trainer sessions I have done manual calibrations pre-ride and manual calibrations after about 20 minutes and am getting these 20 unit shifts. Given the the fact the bike has been sitting in in the basement on the trainer in 65 degree temperatures for the last week, temperature changes shouldn't be causing these 20 unit shifts. Shouldn't these calibration checks be closer to one another?
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
If you're not a member, you should really join the wattage group on google as this has been discussed with great detail.

The issue is that the quarq assumes a constant freehub drag during auto-calibration and this is not going to happen on a trainer. You can get close on a slight downhill road but manual will almost always be better.

For example, here's a good link:
https://groups.google.com/forum/?fromgroups=#!searchin/wattage/offset$20drift$20quarq/wattage/9GUTjK84Ed8/_3NCkATQDR4J




Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [styrrell] [ In reply to ]
Quote | Reply
styrrell wrote:
For training and a lot of racing thats possible, but for a tri where you must leave yor bikein a cold temp and get on it an hour or so later its a concern, particularly when I want the most help with pacing at the begining of the bike leg.

That's what the backwards pedaling autozero is for ;-)

Although, as has been pointed out in this thread, the manual zero is more accurate, that's only because an assumption is made about the freehub and drivetrain drag, so naturally manually zeroing is going to be the most accurate since that assumption is taken out of the equation. That said, the assumed drag in the "autozero" function is a fairly good guess. Something would have to be REALLY wrong with your drivetrain for it to not do an acceptable job for you. The fact is that most well-maintained drivetrains don't have that much drag in them to begin with...

In short, when riding outside the backwards pedaling zero is "good enough". The manual zero is "better", just by definition, but doing the backwards pedal zero is not going to make things wildly different.

Here's a tip for folks with a Quarq who may be curious as to how to know when to zero your unit while you are riding around:

When I'm riding outside and either the temp has changed, or if I'm just curious if the offset has changed (this is probably more an issue for me since I tend to be riding around on protoypes, which can have all sorts of odd behaviors...but, I digress), when the terrain allows "soft pedal" (i.e. allow the freehub to disengage and coast, but keep rotating your pedals) for a bit. You should see your display go to zero. You only need to rotate the pedals at 50-60rpm or so to keep things from being forced to zero by the head unit (assuming a Garmin, and due to repeated ANT+ messages if the cadence value isn't updated). If you see a non-zero power reading while doing that, then do the backwards pedaling zero. If it goes to zero...pedal on :-)

Most of the time, it reads out zero for me...

http://bikeblather.blogspot.com/
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Tom A.] [ In reply to ]
Quote | Reply
Thanks for everyone's reply. I guess the best thing to do is always do a pre-ride manual zero and if possible another manual zero during a ride where the temps fluctuate greatly. If conducting a manual zero is not feasible (race, etc.) then stick with the pre-ride manual zero and only do an auto zero (backpedaling) if you absolutely have to.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
mcmtbr wrote:
Thanks for everyone's reply. I guess the best thing to do is always do a pre-ride manual zero and if possible another manual zero during a ride where the temps fluctuate greatly. If conducting a manual zero is not feasible (race, etc.) then stick with the pre-ride manual zero and only do an auto zero (backpedaling) if you absolutely have to.

I would take out the word "only" and also change the rest to "if soft-pedaling shows a non-zero number". There's little harm in doing backpedal zeros even if you're not sure if it needs it.

http://bikeblather.blogspot.com/
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [mcmtbr] [ In reply to ]
Quote | Reply
I have experienced similar issues with both my quarq 2012 models. I too use the kinetic kurt Indoors. There's a small amount of drift. No biggie.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [125mph] [ In reply to ]
Quote | Reply
I'm a Quarq 2012 and KK trainer owner and I can confirm the -20 ZO drift from the start to the end of the indoor ride.
Every ride the same value at the start (-100), and 20 less at the end of 1 hour ride.
It seems something due to the Quarq warming up in indoor training, usually I don't see this drift in outdoor rides.
I've my theory, outside the quarq is cooled by air you get riding, but indoor the pm has no "cooling".

When i stop the indoor ride and i check the zero offset a couple of minutes after the stop, I see that the value slowly go back to the start value.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [simosl3] [ In reply to ]
Quote | Reply
simosl3 wrote:
I'm a Quarq 2012 and KK trainer owner and I can confirm the -20 ZO drift from the start to the end of the indoor ride.
Every ride the same value at the start (-100), and 20 less at the end of 1 hour ride.
It seems something due to the Quarq warming up in indoor training, usually I don't see this drift in outdoor rides.
I've my theory, outside the quarq is cooled by air you get riding, but indoor the pm has no "cooling".

When i stop the indoor ride and i check the zero offset a couple of minutes after the stop, I see that the value slowly go back to the start value.

This makes sense. I have an Elsa and a KK trainer that I ride in my garage with a large fan blowing on me and the bike. Thus far I've seen very little change in the calibration offset.
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [simosl3] [ In reply to ]
Quote | Reply
I'm a bit late to this thread, but it's super interesting to me given something that I have noticed and largely just ignored during my trainer [indoor] rides with my Quarq (trainer is a Cyclops Fluid). Basically, I'll manually calibrate at the start of the ride, and then again at the end, and the values are actually pretty similar...that said, I notice that without fail, for the first 15-20 minutes of my ride the power numbers are lower than they are for the rest of the ride at the same resistance/gearing/rpms. Like I said, I've largely ignored this because the first 15-20 minutes are always a warmup anyways, so it just means that if I see 125 watts, it's probably closer to 150/160. Then, very consistently, at around the 20 minute mark of my ride (after some spin-ups to a higher power) the PM seems to get "warmed up" and starts reading higher values.

I've manual zeroed at this point to see if the offset value has changed, and generally if it has, it's no more than by 3-8 (e.g., from -290 to -298). Anyone else experience something like this or have any thoughts?

*********************************************
Brad Stulberg
Author, Peak Performance
http://www.BradStulberg.com
Quote Reply
Re: Zero offset calibration on new Quarq ELSA 10R [Bstulberg] [ In reply to ]
Quote | Reply
It could also be that the fluid inside your trainer is still warming up as well.
Quote Reply

Prev Next