💪 EUC pull force test results - reviewers please do this!

[RagingGrandpa]

Results:

(More models coming soon; logfiles available here)

Why is this interesting?

Pull force shows the maximum acceleration that an EUC can create from a standstill. 

I hope we can get EUC reviewers to measure prototype wheels in the future. 
@Marty Backe @Hsiang @evX_Mick @colton @Jack ex-KS @Afeez Kay @Jason McNeil @Alien Rides @GoGeorgeGo @ray rokni @Mike Sacristan @U-Stride @Cutiyo 

 

Edited November 25 by RagingGrandpa
added more results

[RagingGrandpa]

FAQ

Will this hurt my EUC?

If your EUC is working properly, no.

If your EUC has defects, this could reveal them. It is a stress test. 
Mechanical or electrical failures could occur, so be sure to perform this test in a safe area outdoors, away from people and fragile objects. 
Your EUC needs to be able to tolerate its maximum current for a brief pulse, because this is exactly what can happen when riding the EUC. Better to discover a defect during a test like this, than by crashing in traffic!
 

What about voltage?

Battery voltage usually doesn't matter, because the motor speed is zero. With MSP, I saw equivalent results with the battery at 95%, and 25%. 
 

What about tire size?

It matters! 
Especially during riding! 

This is one of my favorite attributes of this type of test: it accounts for the effect of tire radius, even among EUCs with the same motor type.
 

What about geometry: handle length, hand position, pedal height?

It matters, and we account for most of it. 
When pulling, ideally your hand will be exerting a purely-horizontal force, directly vertically above the EUC axle. Of course perfection is impossible, but the good news is inaccuracy will come with the cosine of any angle away from vertical. E.g.: 10 degrees adds about 1.5% error. That's not much. 
And we'll account for handle length and pedal height with physical distance measurements (see next post, procedure step 6). 
 

Is this accurate?

For me, the result was repeatable within 10 lb. But you could certainly do better! 
My force gage is a cheap low-sample-rate device, and I can only look at its display to see the result. A high-sample-rate or peak-detecting gage that creates a data log file would be even better.

Also beware jerky motions and posture during the test- the objective is a smooth pull, and then holding a steady overlean angle. 

Cheers

 

Edited October 18, 2023 by RagingGrandpa

[RagingGrandpa]

HOWTO

1) Make some handles
As shown in my example below, we need something to attach to the EUC that allows us to overlean it consistently while tied to the gage. 
I used metal; wood might work too. Perhaps a hockey stick could work, with an additional hook bolted to its end.
Bear in mind, the handles need to not touch the ground underneath the EUC pedals- otherwise the force data is invalid.

2) Buy a tether rope
Thin synthetic ropes are best for not getting stuck inside the EUC shell. 
3mm "Amsteel" and "Dyneema" are great. Kevlar maybe. Nylon is bad. 
15ft of length will be plenty. 

3) Buy a gage and a measuring tape.
I used a $30 "crane scale"...
The "hold" function of these is useless, because it will only hold a stable force value, and won't hold a peak value.

4) Anchor the gage to something sturdy, near the height of the EUC axle. 
The anchor could be a tree, post, car, etc.
The objective is to match the axle height of the EUC motor. The rope should become level when the EUC is pulling. 
If you use a car, apply the parking brake firmly! Rocking of the transmission reduces test accuracy.

5) Feed the rope around the EUC's axle attachment
This is easiest with the EUC lying on its side. Hold the rope against the tire, and rotate the tire by hand until the rope comes out on the opposite side of the shell.
Then carefully upright the EUC and keep some tension on the rope with your hands, to prevent snags.
Then connect the rope to the gage.

6) Using a measuring tape and an assistant, measure the actual heights of the rope and handle:
 + Measure and record the vertical distance from the ground to the axle center (e.g.: the rope height, with a level pulling rope). 
 + Measure and record the vertical distance from the ground to the center of where your hand grips the handle.

7) Connect to the EUC and start an EUC World or WheelLog app data recording (not a tour).
This records motor current during the test.

8) Start a video camera (e.g.: phone) viewing the gage display.
This will record the pull force values during the test.

9) Connect the handles.
I like to position the EUC with one foot, then wiggle the handles onto the pedals. 

10) Overlean the EUC a few times and read the rope force.
You must put some body weight on the pedals, to ensure the tire maintains traction and does not skid during the test. Either stand on the EUC and have a friend help balance you; or use one foot on the EUC and the other on the ground. 
High-torque EUC's need lots of pressure to make grip, and are easier to test with two people pulling (one person on each side).  
Take care to hold the handles in a way that they never touch the ground during the test. If the handles touch the pavement underneath the pedals, the result will be invalid.
Grip the handles and pull until you feel the EUC pedals dip, then hold the dip position for one full second to allow the gage to measure the force. 
Then relax and rest the EUC for at least 10 seconds before trying again. 
Don't do this 10 times in a row, to avoid heat buildup. 

(Test complete - report results)
You can download a copy of the formulas here to account for handle and rope height and derive tractive force; or just reply to this thread with your numbers and we'll be glad to help.

Pics of my test method: https://photos.app.goo.gl/JcGM38bwtuZV1pi58 

 

 

Comments appreciated

 

Edited October 18, 2023 by RagingGrandpa
(added step 6)

[Freestyler]

I created a table for you RagingGrandpa with information I extracted from the firmwares.

This value is used to calculate the PWM and I believe it is the max current.

I cannot confirm this, so take it with a grain of salt, but maybe you'll find it useful!

When wheel voltage is not specified assume the max model version (for example mten has 67v & 84v versions, I wrote for the 84v only). And in general data are for the latest black controllers.

Mten is the only outlier here. (I'm salty my mcm5 is last) 

 

Edited May 10, 2022 by Freestyler

[litewave]

7 hours ago, RagingGrandpa said:

<snip>

As shown in my example below ...

<snip>

Pics of my test method: https://photos.app.goo.gl/JcGM38bwtuZV1pi58 

 

 

Comments appreciated

 

 

Dude, your Gotway is Paink!

 

[Freestyler]

Update: A forum member contacted me and he told me that he reached 222 phase current on his ex.n c30, but did not feel a pedal dip.

He runs the "accelerated stability" firmware. It turns out that this version for this particular wheel has 240a limit instead of 220a for the regular version! (no changes in other models)

Edited May 11, 2022 by Freestyler

[Bizra6ot]

It's me 

Thank's to @Freestyler looking it up in the firmware to confirm the extended value 

 

@RagingGrandpa Do you know if it acts "like a fuse" (if the max current need to be crossed several milliseconds/second) or if this value is exceeded even on a short peak it will necessarily cut off?

Edited May 11, 2022 by Bizra6ot

[Chriull]

40 minutes ago, Bizra6ot said:

Do you know if it acts "like a fuse" (if the max current need to be crossed several milliseconds/second) or if this value is exceeded even on a short peak it will necessarily cut off?

It's the mosfets cutting off once some current limit is exceeded.

Like this ~50A limiting while stalling:

PS: if it's exceeded to long the mosfets act as fuse, indeed...

Edited May 11, 2022 by Chriull

[Freestyler]

I would also like to request you @RagingGrandpa to enable extended packets if you perform another test. (euc dash)

This way we can monitor the other current related values: RoatCurrMax, MiI and MaA. (Protect_Current is phase current / 10 as reported by main packets)

Unfortunately I don't have a log mechanism there though. @supercurio was working on that on his Euc alarm app I believe.

Otherwise I can print the values in the browser developer console if you want a quick & dirty logging mechanism.

Update: I added the "debug=true" url param to log the raw extended packets in the developer console. example: https://freestyl3r.github.io/euc-dash/begode.html?debug=true

Also be aware of the 600ms delay per packet vs 100ms of the main packets.

Edited May 11, 2022 by Freestyler

[RagingGrandpa]

On 5/11/2022 at 5:12 AM, Bizra6ot said:

Do you know if the max current need to be crossed several milliseconds/second?
Or if this value is exceeded even on a short peak it will necessarily cut off? 

It's a strict limit with no time delay.
The controller will act immediately, to regulate current, within 0.2 seconds of the limit being reached.

Sherman logs show it nicely: 

On 5/10/2022 at 8:19 PM, litewave said:

Dude, your Gotway is Paink!

Alas, the pink RS isn't mine... my wheels roll around in the dirt and could never be that clean  

Edited September 22, 2023 by RagingGrandpa
(img url)

[supercurio]

4 hours ago, Freestyler said:

I would also like to request you @RagingGrandpa to enable extended packets if you perform another test. (euc dash)

This way we can monitor the other current related values: RoatCurrMax, MiI and MaA. (Protect_Current is phase current / 10 as reported by main packets)

Unfortunately I don't have a log mechanism there though. @supercurio was working on that on his Euc alarm app I believe.

Otherwise I can print the values in the browser developer console if you want a quick & dirty logging mechanism.

Update: I added the "debug=true" url param to log the raw extended packets in the developer console. example: https://freestyl3r.github.io/euc-dash/begode.html?debug=true

Also be aware of the 600ms delay per packet vs 100ms of the main packets.

Yes I can add CSV logging for the extended packet data quickly if useful, just let me know @RagingGrandpa @Freestyler

[RagingGrandpa]

The chrome interface looks fine to me, will do a first log that way...

@Freestyler it's here.

 

Edited May 12, 2022 by RagingGrandpa

[supercurio]

22 hours ago, RagingGrandpa said:

Why is this interesting?

Pull force shows the maximum acceleration that an EUC can create from a standstill. 

I like that! It's very useful to know for technical off-road use.
Like: when you stop to balance and then re-accelerate to go over a root or rock.

@RagingGrandpa, it would be useful to add a column with the wheel weight, and pull force vs weight ratio, since it will determine a lot of how responsive and zippy a wheel feels at low speed.

[US69]

just a small heads up:

IMHO test is only usable for Gotway on EucWorld, if at all.

First reason is that KS is showing battery amps.

Then also: @Seba , the EucWorld developer, to my knowledge, is trying to calculate “kind of battery amps” when a GW is connected to EucWorld, but best is he comments himself on that topic.

That would also mean these numbers are NOT accurate!

Motorphase Amps on SUCH a test in my view should be even way higher.

As motorphase Voltage on a standingstill Euc is near zero/null, the motorphase amps on a peak 5000-7000Watt wheel can be up to 500 Amps. This not only endangerous your Mosfets, but if to often repeated, also your motor wires.

dont get me wrong: its a nice test….i am just saying the numbers are not real, and that its just GW, which are comparable this way

 

Edited May 12, 2022 by US69

[Seba]

1 hour ago, US69 said:

Then also: @Seba , the EucWorld developer, to my knowledge, is trying to calculate “kind of battery amps” when a GW is connected to EucWorld, but best is he comments himself on that topic.

That would also mean these numbers are NOT accurate!

@RagingGrandpa is using motor phase current, which is recorded as received from the wheel. As during this test there's no regenerative braking nor plugging, it's a good value for wheel comparision. Of course, as you wrote, this is only applicable to Begode/Gotway/Veteran (as they use the same circuitry and motor control algorithms).

Anyway, we all would benefit from battery current measurement in these wheels. In certain situations current flowing thru one phase is cancelled out by the currents flowing thru remaining phases, so phase current may be high while battery current is near zero. As battery current is the best indicator of real power output, it could be used for real world testing an wheel performance comparisions, regardless of manufacturer.

When I was working on battery current calculation formulas for Begode wheels, I recorded huge amounts of data. I used Hioki LR8515 portable data logger recording battery current every 100 ms. For this purpose, I've built a custom shunt device that was connected between battery and mainboard. Output voltage was recorded by the logger, which was also time-synchronized with Android device running EUC World. Current log and EUCW data log was then matched and merged, then curve fitting was applied to obtain the polynomial that could be used to estimate battery current. Lot of riding under different conditions and with different wheels, lot of data to analyze and validate. It was a really hard work... 

[RagingGrandpa]

9 hours ago, US69 said:

IMHO test is only usable for Gotway

What? 

This is a pull force test!  

Easily repeatable on any EUC. 

The additional log data is just for curiosity; the force is the ultimate result. 
If we don't know motor amps for Kingsong, so be it. We still know force! 
 

Edited May 12, 2022 by RagingGrandpa

[Freestyler]

On 5/11/2022 at 5:49 PM, RagingGrandpa said:

The chrome interface looks fine to me, will do a first log that way...

@Freestyler it's here.

 

Thanks for the data! I plotted them here: https://plotly.com/~freestyler7/1/

text form: https://snippi.com/s/9bqh7n6

 

The main take away for me here is that PWM is the true duty cycle. It accounts for current & RPM. (it goes up to 1050 for 100v wheels and 980 for 84v wheels)

MiI has a maximum value of 9000 (verified from firmware)

Edited May 12, 2022 by Freestyler

[RagingGrandpa]

Added the mighty 16X and the whimpy V11  

[Tawpie]

3 hours ago, RagingGrandpa said:

Added the mighty 16X and the whimpy V11  

How mighty and wimpy? Enquiring minds…

[supercurio]

Now I really wonder if a dolly could draw a torque curve at all speeds, from 0 RPM.
Rolling resistance would certainly affect the result but that would be pretty cool to know about the torque in more situations than stalled as well!

(I wonder if the V11 limits stall torque for protection, I think I've seen something of the sort on my V10F also)

[RagingGrandpa]

27 minutes ago, Tawpie said:

How mighty and wimpy? Enquiring minds…

(I revised the first post in the thread w/ latest results)
 

21 minutes ago, supercurio said:

I really wonder if a dolly could draw a torque curve at all speeds

Many ways to build a dynamometer... but few of them are controlled, standardized, and affordable enough to be reproduced by others. 

No one man is likely to have rapid access to every machine, and I'm encouraged by the simplicity of the pull test that other people in other places could conduct it and gather meaningful data for comparison (with demo wheels, etc).

[Tawpie]

I am quite saddened to discover that I am likely to be looking at a downgrade of overall thrust from my learner. Who knew that I had selected one of the thrustiest wheels as my entrée into this sport/hobby/endeavor/time sink/money pit. 

Edited May 25, 2022 by Tawpie

[RagingGrandpa]

7 minutes ago, Tawpie said:

saddened to discover that I am likely to be looking at a downgrade [vs 16x]

I think 84V 16x was a good example of "too much compromise" - the speed range of the motor was chosen for more low-end pull than most people can use, at the expense of high-end power that most people do use. Very common for 16x riders to find 32mph and eat pavement.

Glad that we have an answer for more pull at all speeds: HT motor + higher voltage! (100V+ systems)

[supercurio]

@Tawpie at 0 rpm yes, but slightly above that, Adam / Wrong Way found that the S22 provided more torque before being over-powered on a dirt climb than his 16X in a direct comparison. Now let's convince someone with a S22 to run the pull force test 

Edited May 25, 2022 by supercurio

[Tawpie]

10 minutes ago, supercurio said:

Now let's convince someone with a S22 to run the pull force test 

I wonder if shibby_time has repeated his "push the S20 into a wall" test on newer firmware? The AlienRides demo of that test using Master had it losing traction with the floor, whereas the first gen RevRides demo S20 almost immediately went 'soft'. And I mean, almost immediately.

Oh well, makes no never mind to me actually. My riding weight would qualify me to perform the 200 km S22 range test, so I have a dramatically lower thrust requirement than most riders. (it's so nice to discover a 'sport' where brawn isn't necessarily a significant advantage)

Edited May 25, 2022 by Tawpie