Methodology proposal to establish the assistance levels in an E Bike

[AntonioAlfaro]

I was able to sideload the Ki2 workaround (highlighted by DC Rainmaker and Shane Miller) onto the Karoo 2. It shows the front and rear chain positioning very nicely. Not sure of the origin, but I also get a nice two beep alert when the next rear gear would trigger a concurrent Synchro 2 front and rear shift.

Hammerhead a also added an e-bike specific set or parameters, with mixed success so far for the Trek Domane+ SLR6. The K2 reports very nicely on speed and main battery status, but not combined power. I haven't looked tried cadence.

Surprising about the Garmin pedals, I thought the hangup was only with Shimano.

Thanks rdv, I'll check out DC Rainmaker and Shane Miller. The K2/Garmin problem has to do with the Cycling Dynamics parameters, K2 cannot collect that data.

I like the Hammerhead dashboard page, in it you can see different data of the ebike after the ride, especially the % of battery capacity consumed and remaining. With these data you can make certain estimates for future tours. It would be very valuable to be able to collect in K2 the personal power that is displayed on the Domane+ screen. We would have a lot to analyze.
Look at the next picture, lot of info there...

 

[mschwett]

… It would be very valuable to be able to collect in K2 the personal power that is displayed on the Domane+ screen. We would have a lot to analyze.

this is definitely possible, very routine. i record power on every ride with my creo or aethos and the domane+ manual shows the same possibilities.

 

[rdv]

It doesn't work well for me as yet. While Rider Power and Motor Power are visible on the TQ display (as suspiciously variable as the latter is), the Karoo 2 only reports Rider Power, drawing a blank for its Combined Power metric. Hammerhead punted on it back in March (see attachment). I hope that with major players like Trek, Pinarello, BMC, Scott adopting the HPR50, a solution will come along before long.

 

[mschwett]

It doesn't work well for me as yet. While Rider Power and Motor Power are visible on the TQ display (as suspiciously variable as the latter is), the Karoo 2 only reports Rider Power, drawing a blank for its Combined Power metric. Hammerhead punted on it back in March (see attachment). I hope that with major players like Trek, Pinarello, BMC, Scott adopting the HPR50, a solution will come along before long.

right - rider power is the one we’re really interested in here, but it would be nice to have both. i’m not aware of many/any bike+computer/phone combos that log both other than the manufacturers proprietary apps.

but rider power should be easy.

 

[AntonioAlfaro]

I have done 3 medium-distance routes (in the range of 40Km to 60Km) to feel the levels of personalized assistance according to my methodology.
My custom levels in which I have concluded are:
ECO= 136x50%=84W
MID=136X75%=111W
HIGH=136X100%=156W
My first verification from the quantitative point of view is the behavior of the power delivered by the motor as a function of: assistance level (Mode) and the power delivered by the cyclist.
I have taken the data of a 51.5Km route to graph information extracted from the Trek app.
Despite the fact that the route has a slope of up to 16%, the distribution of use of the different Modes was:
OFF: 34%
ECHO: 62%
MIDDLE: 3%
HIGH: 1%
And the total battery consumption was 36% for 51Km.
Qualitatively, my first impression is that the Domane+ rolls freely with no drag, making every W delivered by the rider 100% applied to moving the bike.
It is without a doubt a light bike that helps a lot on climbs.
I must admit that in none of my 3 test rides have I felt at the limit of what I can deliver, which makes me think that the assistance values in each Mode can still be adjusted downwards.This would allow to increase the autonomy of the battery.
I will keep you informed

 

[AntonioAlfaro]

The following information corresponds to a test ride of 44Km with a maximum slope of 16%.
For this tour, only MODE 1 (ECO) was used:
ECO= 136x50%=84W
An image is attached with general information of the route
The purpose of the test was to have a better feeling of the assistance for different terrain conditions. I have considered that using a single level of assistance would best serve this purpose.
An image of the behavior of assistance for the entire route is attached, with data extracted from the Trek app. It can be seen that 70% of the time the assistance was used and 30% remained OFF. Total battery consumption was 43%.
As far as the feeling is concerned, and to try to find quantitative reasons, I have attached an image that is a segment of time and data from the route.
It can be seen from this image that there are moments of time in which the assistance drops to zero even when there is cyclist power.
I have prepared an image in which with green dotted lines I focus on curious behaviors that in my opinion affect the feeling during pedaling.
It can be seen from the dotted boxes numbered 1,2,3 and 4 that the motor drop to zero always occurs when the rider's power is decreasing but always below 86W (ECO) and above 45W.
The dotted box 5 shows a very interesting condition. The motor cuts from 45W down towards the rider, and it engages again from the same value approximately if the cyclist's power goes up. This behavior produces an instant drag feeling.
The way to mitigate this drag feeling is to make proper use of the gear combination, when you feel your pedaling release. Try to keep the cyclist's power constant or increasing, but never decreasing.

 

[rdv]

The following information corresponds to a test ride of 44Km with a maximum slope of 16%.
For this tour, only MODE 1 (ECO) was used:
ECO= 136x50%=84W
An image is attached with general information of the route
The purpose of the test was to have a better feeling of the assistance for different terrain conditions. I have considered that using a single level of assistance would best serve this purpose.
An image of the behavior of assistance for the entire route is attached, with data extracted from the Trek app. It can be seen that 70% of the time the assistance was used and 30% remained OFF. Total battery consumption was 43%.
As far as the feeling is concerned, and to try to find quantitative reasons, I have attached an image that is a segment of time and data from the route.
It can be seen from this image that there are moments of time in which the assistance drops to zero even when there is cyclist power.
I have prepared an image in which with green dotted lines I focus on curious behaviors that in my opinion affect the feeling during pedaling.
It can be seen from the dotted boxes numbered 1,2,3 and 4 that the motor drop to zero always occurs when the rider's power is decreasing but always below 86W (ECO) and above 45W.
The dotted box 5 shows a very interesting condition. The motor cuts from 45W down towards the rider, and it engages again from the same value approximately if the cyclist's power goes up. This behavior produces an instant drag feeling.
The way to mitigate this drag feeling is to make proper use of the gear combination, when you feel your pedaling release. Try to keep the cyclist's power constant or increasing, but never decreasing.

 

[rdv]

Will look into your rides tomorrow. This is good information that others will appreciate in the larger forum. You should post both rides there.

 

[WillShootPhotos]

Hello everybody,
There is a concern about the discrepancy in the power level reading on the TQ system display, attached response received from TQ support.

View attachment 154866

Some of my posts in the Domane+ thread illustrate some of the data issues I've seen w/ the power numbers coming out of the system. I've started to do a deeper dive into crunching the differences between what the TQ reports as "human power" vs what my power reading pedals report - I think most of the folks in this thread are also on the other, so I'll not repeat myself, but will drop a link to that thread... Domane+ Thread

Since my posts there, I've dug a bit more, but haven't yet found a consistent correlation between the TQ number and my Garmin Vector 2 numbers for power. One specific climb I did I used the reported average for a strava segment against the average of the readings for that same set of data from the TREK App data file - then reduced the TQ numbers down by that scale, and they seemed pretty close to what I saw... BUT - in other rides, where it wasn't a steady 3 mile climb (the aforementioned strava segment) - the power seems to match up well for a bit, until the human eases off a bit where it appears the TQ wants to "help" by sustaining the input power number until a steady decrease in actual human power occurs. I think, while it feels great on a hard effort, it may make analysis of the data - esp re: calculated optimization - a bit tricky.

More as I know it.

 

[WillShootPhotos]

i've heard quite a few complaints about the power readings of the TQ system. hope they get it sorted.

I sort of "get" why TQ isn't too bothered to make sure that their numbers aren't full on accurate to what a calibrated power meter would generate. For their needs, they just need (for lack of a better term) and "area play"... so long as they know they're close (ish), and if the rider is increasing or decreasing their input level... so long as the seat of the pants "feel" of the system as a whole is making riders happy - I think they'll chalk it up as a win.

Yes - it would be better if their system was more accurate, but would it be enough of a marginal gain that they end users would see benefit from it? Personally, I don't like looking at the TQ display for more than a quick glance, it doesn't let me keep an eye on the road sufficiently if I'm really trying to see if I can see the diff between what my power pedals say vs TQ for power... so I'm relying more on data exports/etc.

I've yet to try to see if my Garmin will let me display two power numbers on the same screen - I think I may be able to pair just the "power" number (as opposed to the "eBike" bundle which appears to provide more than the three that the TQ manual says are avail - power, cadence and speed... Garmin (connected via eBike) will also display remaining battery, remaining distance, assist level, and some other numbers I've not used because they don't seem to change - I've even forgotten their names). If I can get both "human power" numbers on the same screen, not sure which it'll feed into Strava, etc... but may be worth a ride or two where I can see how they change while riding.

Otherwise I'll be merging my FIT file data (after extracting it from Garmin's format) and the TREK app GPX data... a bit cross-eye inducing in excel, but I may build a tool to chunk it all into a DB and start crunching on it from there.

I'm curious as to how frequently (or infrequently) TQ does a firmware update; how we (as Domane+ or any other bike manufacturer that uses the TQ platform) get notified to that update. Are the updates pushed to all systems, or does each bike brand "review/tweak/push" the updates on their own schedules - much like the various iterations of the Android OS for phones per manufacturer and model.

Time will tell... watch this (and other) space(s).

 

[WillShootPhotos]

It can be seen from this image that there are moments of time in which the assistance drops to zero even when there is cyclist power.
I have prepared an image in which with green dotted lines I focus on curious behaviors that in my opinion affect the feeling during pedaling.
It can be seen from the dotted boxes numbered 1,2,3 and 4 that the motor drop to zero always occurs when the rider's power is decreasing but always below 86W (ECO) and above 45W.
The dotted box 5 shows a very interesting condition. The motor cuts from 45W down towards the rider, and it engages again from the same value approximately if the cyclist's power goes up. This behavior produces an instant drag feeling.
The way to mitigate this drag feeling is to make proper use of the gear combination, when you feel your pedaling release. Try to keep the cyclist's power constant or increasing, but never decreasing.

Antonio,

I'm wondering if there is a min power that the motor can produce... if it can only engage at 40W, it may explain some of what you're seeing here... if your % of input power is low enough, it may cut in and out... I've not looked at my data to see what I've gathered, but will dig into that early next week when I jump into the data pool again...

 

[rdv]

Will look into your rides tomorrow. This is good information that others will appreciate in the larger forum. You should post both rides there.

Apparently 96% Off/Eco was too easy for Antonio, had to go to 100%, and came out thinking he may need to trim back the support! Then again, he did make pretty good use of the available 84W and acknowledged finding the help on the climbs.

The support drops below ~40W is a meaningful insight worth tracking. More on having to figure this out on our own below.

 

[rdv]

Antonio,

I'm wondering if there is a min power that the motor can produce... if it can only engage at 40W, it may explain some of what you're seeing here... if your % of input power is low enough, it may cut in and out... I've not looked at my data to see what I've gathered, but will dig into that early next week when I jump into the data pool again...

Thanks to Antonio and Will for adding to the dearth of information from TQ on the HPR50. It's a shame that we have to sort these parameters on our own. Because my maximum sustainable heart rate and power are lower than Antonio's minimums, I'm very interested in figuring out how to get the most assistance from the motor. With that in mind I sent TQ's ebike group an email asking whether Trek can increase the maximum assistance in High mode to 300% (keeping motor maximum power and protections unchanged) and quickly heard back: no, but may consider in future. I followed up saying rather non-committal and complained about the lack of published technical information regarding motor's algorithms, intentions, upcoming firmware schedule and content. Response: silence.

 

[Jeremy McCreary]

Hello, allow me to share with you a first draft document that I have prepared to guide in the determination of the levels of assistance in an EBike. It is not a final document, as soon as I receive my Domane+ SLR6 I will finish field checking and correct it as necessary.
If you cycle consistently and also use the assistance of the E Bike as a complement to your personal strength, you will surely want to use just enough of your EBike battery. This will allow you to go further on your rides or take more demanding rides.
I hope it's of your interest

Your PDF doesn't mention gearing or cadence. I see that at max assist, you're down to 6 kph on a 20% grade.

Q1. Do you have a low gear that allows you to maintain a tolerable cadence at that speed?

Q2. Can you actually produce your personal FTP at that cadence?

 

[AntonioAlfaro]

Antonio,

I'm wondering if there is a min power that the motor can produce... if it can only engage at 40W, it may explain some of what you're seeing here... if your % of input power is low enough, it may cut in and out... I've not looked at my data to see what I've gathered, but will dig into that early next week when I jump into the data pool again...

Hi Will,
I haven't confirmed it yet, but I think around 30W. We will have it on the checklist.

 

[AntonioAlfaro]

Your PDF doesn't mention gearing or cadence. I see that at max assist, you're down to 6 kph on a 20% grade.

Q1. Do you have a low gear that allows you to maintain a tolerable cadence at that speed?

Q2. Can you actually produce your personal FTP at that cadence?

Hello Jeremy,
Normally my average cadence for 140W is between 70 to 80 RPM, the motor likes those revs too.
I don't use any special pinion, the normal thing during the tour. 34 forward to climb and moving between 11 to 34 back.

 

[AntonioAlfaro]

Hello everyone,
I'm nearing the end of my Domane+ SLR6 fit testing.
After multiple tests of assistance levels (230Km routes test), I have arrived at personalized values that make me feel good on any route.
These final values are very close to those theoretically produced by my proposed methodology in this forum. I can tell you that just as it worked with the Orbea Gain D50, it also works with the Domane+.
I am attaching a comparative image for the same 51Km route, where you can see the results for the final adjustment versus the previous adjustment.
In the final ECO setting, it remains at 75W, which is about 10% compared to the previous setting. As the ECO level is used for most of the course, this 10% will translate into a battery saving to go further or more challenging.
I want to recognize that the Domane+ SLR6 is a light, fast, reliable ebike, a good climber and that produces a pedaling feeling where every watt delivered by the cyclist feels like it goes directly to the wheel. Overall I am satisfied with what I am getting for the relatively high price of this machine. I hope to write later if I find something new, for now I hope you enjoy your rides on the Domane+ SLR6 very much.

 

[AntonioAlfaro]

As a result of several value-added conversations with RDV about the above graphics, 2 question arise:
1) Why do you always use the ECO mode, even though could could manage with your own watts?
This is the result of tests, observations, and feelings during the adjustment rides. I came to realize that while on slopes from 0% to 3%, I didn't need any assistance at all. However, it was annoying and uncomfortable to keep turning the assistance system on and off. Pressing a button for about 5 seconds to turn it off completely is not to my liking and becomes a distraction during the ride.
For that range of slopes, and given that the motor cuts at around 40W, I noticed that I could easily keep the ECO level always active without much demand on the motor's power since I provide most of the total power, always staying below the ECO's max threshold of 76W. I believe that the few watts I would save by turning the assistance on and off would be negligible. Please refer to the graph.
When the slope goes up a bit, ECO increases and stays at 76W, and the rest is up to me.
In summary, it's a decision I made to make my cycling easier and it doesn't represent a significant waste of power.
(will continue with second question )

 

[AntonioAlfaro]

Second comment/question comes from : "My comments had more to do with the % assist settings and what I perceived to be an illusory use of anything above 100%."

I believe we are both agreeing that the % of assistance parameter has a lot of background. I have a lot of experience and many years of using the Orbea Gain D50 ebike with a rear hub motor (first generation), and I always thought that the feeling with that type of system had to be improved somehow. That's why I ended up acquiring the Domane+.
Let me put things into context: When you use an ebike with a rear hub motor, the only sensor in the system is for RPM. If you move the pedal, there is assistance; if you stop pedaling, there is no assistance. When there is assistance, it comes in at 100% of the preset level. You feel the kick, and it remains constant as long as you keep pedaling. That feeling is not natural, but you get used to it.
In the third generation of ebikes like the Domane+, all of this changes, and it is supposed that with the help of AI, they have developed a firmware that makes pedaling very natural.
I have spent time trying to understand the 3 adjustment parameters available in the TQ system, and I concluded that the max power setting is related to the cyclist's physical condition (similar to first-generation motors), but the % of assistance and pedal response are related to the feeling while pedaling (absent in first-generation motors).
Some time ago, I made some simple simulations of % assistance versus cyclist's power for a fixed ECHO level of 75W (similar to my first-generation motor), which is shown in the attached graph. Then I normalized the % assistance relative to 0% (meaning the cyclist's power), and those results I ended up being called "Feeling Gain." That's something I didn't have in the first-generation system. Please refer to the attached graph.
I believe the true purpose for which the designer of the TQ system has included this parameter is for each cyclist to find their point of satisfaction. I also think that they have lacked further explanation to the user, and many users do not fully understand it.

 

[AntonioAlfaro]

One of the concerns when planning a route on an EBike is knowing whether the battery's charging capacity will be sufficient for that journey. Naturally, the battery's capacity will depend on many factors that we are all already familiar with. Knowing that the 2023 Domane+ SLR has a battery with a total capacity of 360Wh, when we read the consumed capacity on the display, we might be tempted to say, for example, that 20% consumption of that capacity is 72Wh or that 70% consumption of that capacity is 252Wh. After recording enough journeys and carefully calculating the recharge capacity (i.e., actual consumption) and using an appropriate projection algorithm, I share a graph that displays the % consumed on the X-axis and the actual consumed Wh on the Y-axis. I hope it proves useful in your route planning on your Domane+ SLR.