Where'd the current go? help understanding degraded performance

I could use some help understanding what is the limiting factor on my ride today was. Things got very slow. On metr I can see full PPM but the VESC outputting only 45% dutycycle then dropping down to 27% duty cycle.

the whole trip was a slight hill climb, this problem bit was the near the end and possibly the steepest. but I’ve done this bit before with more power.

in metr, at 11:16 I’m at a stop then try to climb the last bit of hill. pause at 11:18 and then complete the rest.

it’s worth noting that metr was collecting data with “dual vesc over can” enabled so it appears to add the two vescs current values together in these records bot motor and battery. the temperature values, I believe, come from the primary vesc.

here’s the metr record zoomed in on that portion.

image

here’s the whole metr record: https://metr.at/r/JGLXU

The only thing I can observe for that interval that is suspect is the temperature hitting 80c. which is in the beginning of l_temp_fet_start at 80c. would this really cause the such a dramatic cut to power l_temp_fet end is 100c ?

vesc settings, same on both vescs:

l_abs_current_max  130
l_current_max      70
l_in_current_max   35
l_temp_fet_end     100
l_temp_fet_start   80
l_temp_motor_end   100
l_temp_motor_start 80
l_watt_max         1.5e+06
l_watt_min         -1.5e+06

board details:

dual focbox 410 connected via can bus. 
dual 6374 190kv 
16/62 wheel ratio
200mm diameter kenda 909s 
10s5p  sanyo  20700b  pack.
1 Like

your vesc likely thermal throttled your motor current

2 Likes

right? that seems the only thing the data would indicate? but… it seems rather severe? or is it wrong to inerpret the l_temp_fet start/end as a gradient of gradual throttling? I barely got into that throttling zone. so that’ leaves me confused.

81 temp is higher than 80 temp fet start…

i believe end would mean complete shutdown

5 Likes

start:

The MOSFET temperature at which motor current starts to get reduced

end:

The MOSFET temperature above which motor current is not allowed and a fault is thrown

I had interpreted that as a throttling gradient with complete shutdown and a fault thrown at temp fet end.

hmm.

So what drives up FET temp? is it just elevated current for prolonged period of time? or anything else?

does my trip look like current levels high enough to drive higher temps?
should I suspect VESCs not adequately cooled? they are bolted to an aluminum plate that spans the width the ot the enclosure. ( Lacroix DSS50+) but are not otherwise cooled.

the heat production in the motors and fets is proportional to the square of the motor current via I^2R=W heat… double the current produces 4x the heat. on a hill in particular you will be using additional current.

technically in this case since you are using the metr app and it has improperly added the currents of both motors, the motor current reading isn’t actually amps… because it doesn’t follow ohms law or I^2R=W, and it isn’t the charge in coulombs per second through a cross section of the conductor in the motor… it’s as inappropriate as adding both motor temps together

1 Like

well at least the temps aren’t added together. :slight_smile:

but yeah, I had been hoping that metr record on the web might have had the individual vesc current and temperature values. [ cc @rpasichnyk :slight_smile: ]

fwiw, keeping an eye on the realtime data on the rides. the currents appeared roughly the same between both VESCs.

So if you if you half the current data. i am doing pulses near 35 battery amps per vesc max. and 70 motor amps. with maybe 50/50 off time vs on time over 15 minutes. Is that enough to push the limits of focbox 410s ?

with these views the temp rise with current draw is pretty obvious. and I guess the temperature is accumulating. so… perhaps I just don’t have addequate disipation in the current setup to handle pulsing 35 battery amps / 70 motor amps so frequently?

well i think on the flats you’d only do 70a in short bursts while accelerating then it would be way less to maintain constant speed, but on hills you might do 70a continuously, which could build up heat rather quickly

thanks. with your help i think I’m seeing things in the data more clearly. definitely pulsing heat with pulsing current. heat dissipates somewhat slowly.

still confused about the start / end cutoff thing.

if you’re doing lots of hill climbing, you can reduce the temps for the same hill climbing speed with a higher gear ratio or smaller tires but this will sacrifice some of your top speed

either of these options reduce the current required for the same constant speed… increasing the gear ratio increases the torque multiplication between the motor and the wheel, and reducing the tire size increases the thrust for a given wheel torque

you could also increase from 10s to 12s, and multiply the current gear ratio by 1.2, and you’ll have the same top speed & acceleration performance and reduce the motor/controller heating by about 30%

switching from 2wd to 4wd would cut your motor / controller heating to 1/4th per motor/controller for the same acceleration or constant speed

1 Like

Oh stop taking the piss. Its fine that you believe this but you cant start spreading this theory of yours around the forum

1 Like

the definition of the amps, ohms law etc aren’t my theory… it’s fact.

^adding the 2 motor currents doesn’t satisfy this equation so it isn’t current and it isn’t amps

I think @rpasichnyk may be the person here to clarify things.

Im an electrical engineer,im well capable of knowing how wrong he is. Ive tried explaining this in other threads and ways so i give up on trying to make him understand. But im not okay with him still going around the forum seriously miss-educating people on a matter im familliar with.

Dont listen to him,he for some reason uses ohms law and deliberately miss-interprets it so it fits his theory.

3 Likes

I am on your side mate, but maths is not my realm, anything that becomes more specific i have trouble understanding. I just summoned the Roman so we have a clear explanation from the guy who built and programed it instead of ‘‘theories’’ that no one ‘‘agrees’’ on :blush:

^well an amp by definition measures the electron flow or current “in an” electrical conductor (not 2 completely separate conductors) — & is the # of coulombs moving past a “specific point” (not two specific points added together)

@wasp calling yourself an electrical engineer simply doesn’t change the facts. can you explain how the duty cycle electrically isolates the 2 motors from each other, as shown below:

ps one of the people that liked your post earlier stated that in foc the motor current was determined by adding the currents of the 3 phases together, which is false— in a vesc it’s the peak current per phase @ducktaperules

The thing is you are right, but still far enough from being right that your final statement is completely wrong.

You are going about this the wrong way and doing it throug being pedantic
What you are going for here is how you measure amps, which in itself is correct with what you are saying.

Where you are wrong however is trying to say two sources that draw power can not add togheter to one total amp-amount. Which it absolutely can. 10 amps across one circuit and 10 amps across another gets us a total 20 amperage if we add them up so we for example can measure the total powerdraw from lets say an electrical skateboard. There is no disputing this no matter what you may think.

Your problem here is that you are barely educated enough to know the words but you dont understand the thing you are arguing for. And now you are sticking to your guns and digging a deeper hole. Believe whatever you wish man but dont spread your knowledge, its just wrong.

2 Likes

I have previously stated it’s acceptable adding together for the battery current but not the motor current. With the battery current one can reason there is a single point in the battery where that many amps are flowing by adding multiple sensors together. The point where your logic is flawed is extending that to the motor current, because there is no single point in the conductor where the stated number of coulombs per second is flowing past a specific point in the conductor.