Page 14 of 17

Re: Drivo Erg Mode

Posted: Wed Dec 25, 2019 9:46 am
by humphuk
Right - getting more fed-up with the ERG mode now I am doing structured workouts again. I will set up a workout with the steps mentioned. Then run it via several apps/devices to check a possible signal/app issue. Keep the cadence as locked as possible at 90. As my technical background is in dynamic process simulation and control engineering - I will be happy to engage in some discussion. Either the algorithms they are using are set incorrectly (which I presume can be set via the app - or updated in firmware) - or the device is not capable. So it should be "fix or return". However - do Elite make any claims on performance in this respect? Not sure....This is a real pity - as in other aspects I love the Drivo and would recommend it to those not using ERG mode training

Re: Drivo Erg Mode

Posted: Wed Dec 25, 2019 10:30 pm
by humphuk
Tests done - using MyETraining - once on a PC (using a USB lead to ensure proximity of the sensor) - and once on a tablet - both using ANT+. I tried keeping to 90 cadences. It's not very pretty. A bit of research shows it is covered by GP Lama in his recent video. So it looks like in an attempt to make the system react quickly - it overshoots - and starts to oscillate. Some tuning could be done (I assume Elite can modify this in software) - but I also assume the system itself has limitations such as the speed and "digital" nature of the stepper motor control. I think we will also make it worse - as we react to the numbers .... and end up fighting the controller. So - all very interesting - but it just means that in its current state - this turbo trainer cannot cope with short ERG intervals

Re: Drivo Erg Mode

Posted: Thu Dec 26, 2019 10:05 pm
by aldesr
Excuse my poor english as my mother tongue is french.

The 1st test to make is free run of eg 200W 1', 300W 1', 400W 1', 500W 1', etc, several times. This will kind of profile the athlete, as Elite is definitely right on 1 point -> The driver athlete is 'oscillating' himself.

The 2nd test is to execute the same plateaus in ERG mode.

Use a power meter to record the ('real') watts.

This can be done eg in Perfpro Studio. Then Excel or another tool for data analysis.

The LOAD (ERG) p-p numbers should not exceed significantly the UNLOADED ones. As per the tests I made, it will happen the Elite control will cause adjustments of 20-30W/second during several seconds (sometimes 3-5 secs). Sorry Elite. Yes, athletes power will vary. But not -100W p-p during a 300W level test.

I fear the real problem could named robust control, while adaptive control has beeen used for decades in the industry. The trainer has to adapt to the athlete profile, not the contrary.

Elite says also the trainers were tested by elite bikers. For sure, Tom Dumoulin ? Point is most athlets will choppier than Dumoulin, Froome and the likes from 300W - 400W. The closer the athlete comes to his own limits, the more he is 'oscillating'. Yes, a 300W - 400W - 500W will ALSO prove that. A decent trainer controller has to cope with that. The Direto/Drivo as they are today are not capable of short intervals, they are not capable of MAP/IPN/PWC etc submaximal test either, ERG just does not work.

Come on. Is this product suitable for indoor "training and racing with a powermeter" ?

Re: Drivo Erg Mode

Posted: Wed Jan 08, 2020 11:40 pm
by pryswilliams
Just came across this thread after trying a Zwift workout with short intervals and noticing my new Drivo I completely failed to hold the required power in time, meaning I failed to complete the interval. When reverting to longer intervals, the Drivo was taking at least 15-20 seconds to hit the required power. Really gutted. I changed a Tacx Flux S for the Drivo due to issues with noise from the Flux. Really thought I was on to a winner with the Drivo but it seems I've now got something incapable of operating Zwift ERG sessions.

Whinge over, I suppose my real question is has anyone managed to achieve any meaningful improvement with their Drivo I? Is it worth pursuing the support route?

Thanks

Prys

Re: Drivo Erg Mode

Posted: Wed Jan 15, 2020 1:31 pm
by sergey.maximenko
I'm trying to squeeze some sense from Elite support. My Drivo ramps up power in 7s avg in ERG mode, Ant+ connection. First 3s it won't even move the stepping motor!
I'm angry because it fails short intervals in Zwift training session. That can't be tolerated for the trainer that is advertised as Zwift compatible.
Elite support asked me to do 100-200-100-300.... training program in myEtraining app. Is showed the same lagginess and delays. Elite support guy looked on the CSV file (containing time stamp, power, cadence, speed reported by the trainer) and said it is due to some 'interference'!
It is a miracle. Just look into the file that doesn't contain any diagnostic information regarding packet loss, quality of wireless communication etc. etc. and say 'it is due to interference'. At the same time pretend there will be new firmware at the beginning of the year that will fix trainer's responsiveness.
Damn.
Elite support:
1. Make claims that I've issues because of interference(?)
2. Pretend that interference issues could be resolved by firmware update??

I also diagnosed other problems with my trainer that weren't obvious at first (Drivo is my first interactive trainer). There is an obvious problem with resistance curve of the trainer in FE-C SIM mode. At one point it becomes to steep making is uncomfortable to ride anything that isn't supposed to be an interval training. To pinpoint the problem I've used myEtraining for video ride yesterday, reached 1% slope and got ~100W on 21.5 km/h, ~170W on 25 km/h :)

The issue obviously looks like buggy firmware or defective configuration variables written. However Elite acts dumb - no diagnostic, no configuration, no f/w updates.

Re: Drivo Erg Mode

Posted: Fri Jan 31, 2020 9:31 pm
by pryswilliams
Update

After contacting support with the ERG mode responsiveness issue, they sent me a link to a JSON config file which I uploaded to the Drivo (mk 1) using the Upgrado app. This has had a very noticeable effect on the responsiveness of the trainer with ERG mode power changes being applied relatively quickly. Quickly enough indeed for 10 second power jump intervals in Zwift to be successfully completed which wasn't the case beforehand. There is also however a side-effect whereby the trainer's stability at maintaining a power level is noticeably more erratic. So it seems the Drivo uses a software-configured buffering of required trainer power levels, with responsiveness being achieved at the cost of stability. Ah well, I can live with the more erratic power-level stability if it means the trainer doesn't take at least 20 seconds to get to a required power level.

If you want to give it a go, I've attached the link to the json file here https://elitesrl.zendesk.com/attachments/token/gLqi13F3ofNgaD2sbaGcUz9zR/?name=PID+Drivo+2.json, but Massive Disclaimer I have absolutely no idea what effect this will have on your trainer or even whether it will break it completely! My Upgrado app reports my Drivo (mk1) having Firmware: 058 / Hardware Revision: 8. Unless you have exactly the same I would probably steer well clear and try to get a specific config file for your trainer via the support guys.

Hope this helps.

Re: Drivo Erg Mode

Posted: Fri Feb 14, 2020 5:39 pm
by 4169
pryswilliams wrote:Update


If you want to give it a go, I've attached the link to the json file here https://elitesrl.zendesk.com/attachments/token/gLqi13F3ofNgaD2sbaGcUz9zR/?name=PID+Drivo+2.json, but Massive Disclaimer I have absolutely no idea what effect this will have on your trainer or even whether it will break it completely! My Upgrado app reports my Drivo (mk1) having Firmware: 058 / Hardware Revision: 8. Unless you have exactly the same I would probably steer well clear and try to get a specific config file for your trainer via the support guys.

Hope this helps.


Hi. I took a chance and tried. I liked the result. The power at intervals changes very quickly . 1 second. I didn't notice that it takes more effort to maintain power. Smooth by default 2. I'll post the charts later.

Re: Drivo Erg Mode

Posted: Fri Feb 14, 2020 9:17 pm
by Axel
14 pages on and into 2020, is there still no official fix for this issue?
I've have this same problem for over a year and while there is now finally an app, there still is no firmware update.
This is not an isolated problem for a few users (just check the comment section of Ray Maker's review of the Drivo 2 for additional evidence). And the idea that the Drivo II wouldn't need any firmware update, ever, seems unlikely in itself. Why is there no firmware support for a high end piece of equipment like this?

We need an official, supported firmware update. Not a 'try this, but it might break things' link. Come on Elite, support your product, please.

Re: Drivo Erg Mode

Posted: Wed Feb 19, 2020 8:22 pm
by 4169
4169 wrote:
pryswilliams wrote:Update


If you want to give it a go, I've attached the link to the json file here https://elitesrl.zendesk.com/attachments/token/gLqi13F3ofNgaD2sbaGcUz9zR/?name=PID+Drivo+2.json, but Massive Disclaimer I have absolutely no idea what effect this will have on your trainer or even whether it will break it completely! My Upgrado app reports my Drivo (mk1) having Firmware: 058 / Hardware Revision: 8. Unless you have exactly the same I would probably steer well clear and try to get a specific config file for your trainer via the support guys.

Hope this helps.


Hi. I took a chance and tried. I liked the result. The power at intervals changes very quickly . 1 second. I didn't notice that it takes more effort to maintain power. Smooth by default 2. I'll post the charts later.

Image
jpg image

Re: Drivo Erg Mode

Posted: Tue Feb 25, 2020 9:24 am
by stephen.r.gunn
pryswilliams wrote:Update

After contacting support with the ERG mode responsiveness issue, they sent me a link to a JSON config file which I uploaded to the Drivo (mk 1) using the Upgrado app. This has had a very noticeable effect on the responsiveness of the trainer with ERG mode power changes being applied relatively quickly. Quickly enough indeed for 10 second power jump intervals in Zwift to be successfully completed which wasn't the case beforehand. There is also however a side-effect whereby the trainer's stability at maintaining a power level is noticeably more erratic. So it seems the Drivo uses a software-configured buffering of required trainer power levels, with responsiveness being achieved at the cost of stability. Ah well, I can live with the more erratic power-level stability if it means the trainer doesn't take at least 20 seconds to get to a required power level.

If you want to give it a go, I've attached the link to the json file here https://elitesrl.zendesk.com/attachments/token/gLqi13F3ofNgaD2sbaGcUz9zR/?name=PID+Drivo+2.json, but Massive Disclaimer I have absolutely no idea what effect this will have on your trainer or even whether it will break it completely! My Upgrado app reports my Drivo (mk1) having Firmware: 058 / Hardware Revision: 8. Unless you have exactly the same I would probably steer well clear and try to get a specific config file for your trainer via the support guys.

Hope this helps.


Thanks for posting that link. My Drivo has the same firmware and hardware revision as yours so I've taken a punt and upgraded to that file. Like you, I found that the power changes are much faster than previously (tested using Sufferfest Downward Spiral in tandem with PerfPro Studio software). Also, like you, I noticed that there was a bit more fluctuation in power while supposedly maintaining a specific level, but that was not too bad. The final observation was that the cadence reading from the trainer was quite erratic whereas it was previously reliable. I'm not too bothered about that as I have a separate cadence sensor on the bike, so have switched to taking measurements from that. All in all, I'm pleased with the improvements that this file has provided. Now, we just need Elite to issue it officially :?