Tried Garmin Vector 2's vs Quarq Dzero on the same ride

[gurk700]

This might be kind of a "cool story bro" but I confirmed my suspicions. Quarq Dzero measures far less power than Vector 2's. Obviously we can't know which one's more accurate and ultimately it doesn't matter as long as you keep using the same power meter it just needs to be accurate relevant to itself only. But anyway...

My first PM was Garmin Vector 2 then I switched to Stages Duraace which felt like it was measuring way less in the same rides. I've never stopped training and within few months of switching to Stages I still wasn't able to produce some watt averages of the Vector 2's. Then I switched to Quarq recently and it felt pretty much on part with Stages.

Here are the two readings I got on the same ride. GPS and barometer of the units seem to be a little different hence the different results in distance.

Bottom line, I'm still a weak noob and I'll stick to the same power meter from now on so I can have a better understanding if I'm improving or not

Garmin 510 + Quarq DZero


Garmin 520 + Garmin Vector 2

[jlok]

10w difference on avg. think about the location of the meters. it's because the loss of power from the crank (and maybe the pedals also).

[moonoi]

Exactly, it's to do with the location of the power meter, the further away you move from the pedal, the lower the power will display due to loss of efficiency as you move down the drivetrain. If you stick a powertap hub on as well, likely you'll see a similar difference as then you're measuring power at the point furthest from where it is produced.



Sent from my iPhone using Tapatalk

[gurk700]

I think I've mistakenly assumed they would compensate in their calculations to show a more "accurate" number. Guess not.
In any case, yes, they are close. I wouldn't call 10 watts average difference on a 1.5 hr ride "small" but it's an understandable difference if they aren't factoring in the difference in location.

[deek]

You're not losing 10W in the crank.

[jlok]

Definitely not losing 10w (it's one ride data only) but the point is the further down the "power line" the more the loss.

Pedals, crank arms, spiders, chainring and hub (where power meter sensors are located)... check DCR site.

[moonoi]

Taking the average is useless anyway, what you should do is overlay the power profile from both and that will tell you if they track well, 10w difference is within the margin of error for most power meters anyway


Sent from my iPhone using Tapatalk

[kwakekeham]

So the good. You're using energy correct head units -- awesome (Edge 500, 800, FR 910xt can lead to about 5% error recording the same PM).

I believe what they are calling Coggan power is Normalized power. Which is a pretty good quick and dirty comparison. So they are off a little. In depth comparison is usually better as mentioned. If you see them tracking well, then a sprint or a hill climb and it diverges then you can usually trace it to the Vectors. Re-zero and it'll get back to normal most of the time and then will track better after future sprints / high wattage.

I've used Vectors a lot for comparison to other meters. Based on my Vector versus Quarq history (several Quarqs, but NOT Dzero) and about 6 - 8 pairs of vectors, I'd put the nod towards your vectors are a wee bit off. These are finicky. The Quarq should be a little more stable, especially the new bolt system they use which is awesome. They say it's not needed, but first install, I'd zero the Dzero. However, I'm sure other people would trust vectors. Just my history with a bunch of units. I have had vectors that when zero'd showed a constant negative torque offset -- if yours do, RMA.

Some things that can help dial in accuracy on Vector side:
- Re-torque them with crowfoot adapter. This is a significant source of error if they back out even a tiny bit.
- Reset the zero install angle (this impacts as the percentage of each measurement used to figure out torque -- cosine, sine relationship stuff)
- Few sprints to get them "settled" especially if they are brand new. Zero them after
- check crank length is set correctly. This might need to ensure your vectors on a head unit with newer firmware. Some early bugs with older firmware on setting these. I find 510/520/820/1000 are best for this.

As for your Baro problems -- I feel your pain. I've ridden with 4 head units a lot and none of them line up. Most of the time the same model. Let them all stabilize to temperature first to help. Your's aren't too far off.

[Eclat]

Wow. Impressive experiment.
I think the nominal power comparison between power meters is meaningless.
No matter what power meter you use, it is more important to check my progress with the power values ​​extracted from the same power meter.

[deek]

Definitely not losing 10w (it's one ride data only) but the point is the further down the "power line" the more the loss.

Pedals, crank arms, spiders, chainring and hub (where power meter sensors are located)... check DCR site.

The majority of drivetrain losses are going to be due to friction and not flex. There is nothing moving between the pedal axle (where the Vector strain gages are) and the spider (where the Quarq strain gages are). There should be nearly zero difference between the two. There will definitely be difference between the hub and spider/pedals though, I'm not disagreeing with that.

[Stueys]

Both power meters rated at +\- 2%, if one is at the upper end of that scale and one at the bottom then you would get a bigger variation. Added to the fact that they are measuring power at different places, also creating a slight delta....

For what it's worth my p2m, vectors and neo all track pretty closely. Not bang on but close enough to train zones without making adjustments.

[niklasp]

Vector pedals differ depending on torque...