Direto XR problems with ERG mode

[Elite Admin S]

Buongiorno

Posso comprendere la sua frustrazione, ma temo ci sia stato qualche fraintendimento.
Non esiste nessun problema di compatibilità tra Zwift, Assioma e Direto XR.

Riguardo alle oscillazioni di potenza, fluttuazioni di potenza intorno ai 20-35W sono considerate nella norma.
Anche ciclisti professionisti hanno variazioni di 10-20W e sono dovute alla naturale differenza di potenza che si ha tra una pedalata e l'altra.
Se invece sono presenti fluttuazioni di maggiore entità, è necessario approfondire la situazione (potrebbe ad esempio esserci un problema di tensionamento della cinghia). Dal grafico indicato non riesco a capire l'entità delle fluttuazioni.

Passiamo ora alla differenza di potenza indicata tra Assioma e Direto XR.
Che un ergometro segni un po' di potenza in più è un fenomeno generalmente dovuto alla diversa posizione in cui viene misurata la potenza: nel Direto XR viene misurata sul mozzo mentre in un ergometro su pedale la misura sulla pedivella, dove ci sono maggiori attriti.
Se la differenza è troppa, potrebbe esserci un problema di accuratezza nell'ergometro o nel trainer.

Essendo lei già in contatto con uno dei nostri tecnici, le consiglio pertanto di continuare con l'assistenza con lui in modo da risolvere il problema.

Un'ultima nota riguardo alla calibrazione su Zwift.
Siamo a conoscenza del fatto che Zwift ha un problema sull'algoritmo di calibrazione, dovrebbero risolverlo a breve.
Nel frattempo consigliamo di fare la calibrazione utilizzando la nostra applicazione. La calibrazione viene salvata sul trainer e sarà quindi valida anche quando si utilizzano app terze come Zwift.

[Chris_4711]

I tested my new Direto XR today more extensively with Zwift in ERG mode - this is the main app I’m using and will be using in the future.

With the short gears it worked okayish - however, the “road feeling” was a bit unrealistic, as discussed, e.g. in this thread: https://www.trainerroad.com/forum/t/big ... ng/6119/55

With a long gears the power metering is all over the place and Zwift is not usable in ERG mode.

You can see the results in this photo (in the beginning I used short gears, then later long gears.

I used the same long gears I used on my Tacx Vortex Smart. There the adjustment it ERG mode works pretty stable.

This are the results you can find other people talking about the Direto XR on the Internet and in reviews.

Power Measures for Elite Direto XR (first short gears - first two 300W segments, later for the other 300W segments, long gears)

diretoXRZwift.png (189.48 KiB) Viewed 15007 times

Power Measures for a TACX Vortex Smart (only long gears)

tacxVortexSwift.png (152.76 KiB) Viewed 15007 times

Maybe a roll trainer has a different behaviour from a direct drive trainer or my configuration is somehow wrong? Or do the charts show normal behaviour for the Direto XR? This would mean I can only use ERG mode with short gears?

What about other direct drive trainers from Elite or competitors. Do the behave the same way?

[Elite Admin S]

Hi Chris

I see that you are already in contact with one of our technicians, so please go on with him.

But I want to make a little explication for the other forum readers.
The power oscillation you see when you use the Direto XR are normal.
They are due to the natural difference between a pedal stroke and the other. Even pro riders have 10-20W fluctuations when pedaling.

As the direto XR has a high precision power sensor inside, it tracks down those oscillations. If you use an external power sensor during the trainer, you can check it by yourself. The Vortex doesn't have a power sensor inside, so it simply doesn't show them.

That said, if you prefer to have a smoother power graph, you can increase the power smoothing option.
To do it, run my e-training and open the trainer setting. As the value will be saved on the trainer, you'll have the same effect also when you use Zwift.

[Chris_4711]

I totally understand. But the range in the 300W segment on a large front-ring is extremely high (250W to 350W). It would be okay if it was in the +/- 20W. So with this spread, Zwift thinks I am not in the right power zone and will get crazy (with the large front ring)

So that means I can't use the Direto with the big front ring in ERG mode with Zwift to simulate rather plane roads (from the road feeling aspect). I can only use the small / mid-front ring in ERG mode with Zwift to simulate an ascent (from the road feeling aspect).

[Elite Admin S]

Did you try to increase your power smoothing value?
I think that this will help a lot.

Also remember that the best way to train in erg mode is to keep a speed between 15km/h and 25km/h.
In this way you'll be able to use all your gears and have the best stability of the power.
Look at this video of GPLama which explains very well how to train in ERG mode:
https://www.youtube.com/watch?v=LbeliW8uObk

Regards

[audenaertdidier]

With default settings there definately is a large power variation in ERG bwith sensible changes in resistance. As suggested, I also tried increasing the power smoothing value in order to try to improve the ERG mode performance. This resulted in slower varying power but still resistance and power significantly varied. To my feeling this is caused by the power control loop which likely has a fairly large gain in order to achieve fast response to power setpoint changes. This led me to try working the opposite way reducing the power smoothing factor to zero. For me this results in the performance I would expect. Yes, there still is some power variation ( +/- 10 Watts ) but far less variation in resistance once power target is achieved.

[nenadjev]

Am I the only one who finds cadence sensor readings more than unreliable? I've just took GOAT session in The Sufferfest and to my surprise power targets were much overshoot at the very beginning (like 380W instead of 330W) or after few seconds it was impossible to reach power targets (close to given cadence max power was sth like 270-280 instead of 330). During recovery intervals app was showing cadence like 44-49, when I've felt like it was 80-90 and the target was 90 — power target of course were impossible to meet. Afer stopping and pedalling again sensor seemed to be adding additional 40 rpm in a magical way. Throughout whole workout neither power nor cadence targets were even approachable. Frankly speaking my old other trainer — tacx flow delivers better experience when data reliability are considered. Also in my view problem was more vivid during bluetooth connection than ant+ (I've switched after first interval). Next time I'll try with independent cadence sensor and see it it helps. Using just normal (not ERG mode) everything seems to be working just fine.

I've just bought Direto XR and I have similar problem with power when I step into interval, targets are much overshoot at the very beginning of interval maybe 20% or more than target. Did you solve it?

[nenadjev]

Am I the only one who finds cadence sensor readings more than unreliable? I've just took GOAT session in The Sufferfest and to my surprise power targets were much overshoot at the very beginning (like 380W instead of 330W) or after few seconds it was impossible to reach power targets (close to given cadence max power was sth like 270-280 instead of 330). During recovery intervals app was showing cadence like 44-49, when I've felt like it was 80-90 and the target was 90 — power target of course were impossible to meet. Afer stopping and pedalling again sensor seemed to be adding additional 40 rpm in a magical way. Throughout whole workout neither power nor cadence targets were even approachable. Frankly speaking my old other trainer — tacx flow delivers better experience when data reliability are considered. Also in my view problem was more vivid during bluetooth connection than ant+ (I've switched after first interval). Next time I'll try with independent cadence sensor and see it it helps. Using just normal (not ERG mode) everything seems to be working just fine.

I've just bought Direto XR and I have similar problem with power when I step into interval, targets are much overshoot at the very beginning of interval maybe 20% or more than target. Did you solve it?

@Elite Admin S
Definitely I have problem with power when I step into interval with large power difference. During the first effort of 90 sec or more the Direto XR deviated from the target by 20-30%!
On the other hand long intervals and intervals with a small difference in strength are very stable and very smooth.
I am using Direto with PerfPro Studio.
Please see images below. First two images are training with Direto XT and third is training with Flux S, both with PerfPro Studio. I sold my old Flux S and bought a brand new Direto XRT and I was think it will be an improvement and that I am taking a step forward, but now I am confused.

Direto XR

13.10.2021. 20-14 - nenadjev.jpg (33.17 KiB) Viewed 14360 times

11.10.2021. 20-01 - nenadjev.jpg (31.43 KiB) Viewed 14360 times

Flux S

10.10.2021. 18-01 - nenadjev.jpg (59.74 KiB) Viewed 14360 times

[Dwhight]

I dont even bother using ERG on my Direto XR any more, its rubbish. My old Tacx Vortex managed to keep a consistent wattage in ERG mode, within a few percent. The Direto is way off, 10-20% out, its impossible to try and keep it consistent. I've tried the small front ring, smoothing of watts, tried pedalling as consistent as I can, yet it does the same thing, resistant and watts are just all over the place. So disappointed, I'll never buy an Elite trainer again. I can get more consistency trying to keep to a set wattage myself and thats basically how I train now. Forget about ERG.

[Elite Admin S]

@nenadjev

I confirm you that the overshooting that appears on large power differences seem abnormal.
But it's necessary to analyze the data to understand what happened exactly (we need to check also speed).
So, please contact our customer care specifying:
- trainer serial number
- latest calibration offset

Here's the contact form:
https://support.elite-it.com/en/ticket


@Dwhight

You can't compare a Direto with a Vortex.
Vortex doesn't have a power sensor inside. It doesn't track down all the power fluctuations as a trainer with an internal power sensor.
If you want to verify this, use an external power sensor and you will see that those oscillations are real and are present even when the motor is blocked. Even pro riders have 10-15W fluctuations when they pedal.
If you have an accuracy issue, instead, first of all check the calibration. Then contact us using the link above and we will ask you some specific test to check the accuracy

Regards