Votol EM-100 & EM-150 controllers
- Thread starter Stefanj.
- Start date
Miguel Angel
1 mW
- Joined
- Jan 20, 2020
- Messages
- 15
My motor is 72v 1500w, I have the sport cable deactivated, why does it do that to me at high speed? How can I send you a video?
larsb
1 MW
Bartosh said:
I can try it but i think your looking in the wrong direction:
-It is logical with more agressive ramp in sport mode.
-I don’t think the other issue is time limit either - i guess it’s a hardware temp limit that should be respected (controller has reached the limit for output) push it further and it will break - eventually.
Where is the temp sensor, can you locate it? Should be on a middle mosfet somewhere or close to the mosfet line.
Sure, you may be right. Actually two or even three "problems" here, one of which can be a normal "feature" for some users. Though, when I disable the "soft start", I expect it to be disabled. No one promised the votol would be the one in this regard, but I had some expectations. We can put it aside for now, there is "the sport mode". I don't mind using it all the time, so yeah.
I can't find anything on the pcb looking like a (regular) temp sensor, even smd type, but I'm no electronic engineer. The controller's temp reading is wrong, never @ ambient. There is also an additional "temp coefficient" shown in the app. This value changes, depending on how hard you operate the throttle, or for how long, or in a rpm-related way, hard to say... anyway it changes [; But also resets when the controller is turned off. So maybe a dead end.
I didn't have an opportunity to test it with a laptop when it actually happened. But I did it when I got home, the controller was still in this limited state, reporting 34 deg, motor 46, coeff 0. I let it cool down, motor 19 (ambient), controller 26, same voltage - it rips hard again. IMO the controller was not overheated at 34.
SilverLine said the problem might be lvc-related. Unfortunately I wasn't able to reproduce the problem tonight (no load). Though, the under voltage protection [the upper limit] works. It just cuts off the motor totally and reports an error, until the motor stops. Then you can continue without resetting the controller. This is not what I experienced before. It never cut off the motor completely, it just got weaker from the start. But this may be the case for some.
Not having a "regular" temp sensor... maybe they just measure it based on a piece of copper? Temp rises and so does the resistance. Maybe the temp coefficient represents this change? Maybe it's the big shunt (at B- BTW)? Anyway it seems inaccurate to me.
I will test it properly in couple days, to rule out the soft lvc limit. Got the hit = checking the temp ASAP + voltage sag on the go. Right now I'm just unhappy and pouring tears all over ES [; Even my good old svmc72150 endures much longer without breaking a sweat (but it's got some real temp and phase current sensors inside ;- ).
I can't find anything on the pcb looking like a (regular) temp sensor, even smd type, but I'm no electronic engineer. The controller's temp reading is wrong, never @ ambient. There is also an additional "temp coefficient" shown in the app. This value changes, depending on how hard you operate the throttle, or for how long, or in a rpm-related way, hard to say... anyway it changes [; But also resets when the controller is turned off. So maybe a dead end.
I didn't have an opportunity to test it with a laptop when it actually happened. But I did it when I got home, the controller was still in this limited state, reporting 34 deg, motor 46, coeff 0. I let it cool down, motor 19 (ambient), controller 26, same voltage - it rips hard again. IMO the controller was not overheated at 34.
SilverLine said the problem might be lvc-related. Unfortunately I wasn't able to reproduce the problem tonight (no load). Though, the under voltage protection [the upper limit] works. It just cuts off the motor totally and reports an error, until the motor stops. Then you can continue without resetting the controller. This is not what I experienced before. It never cut off the motor completely, it just got weaker from the start. But this may be the case for some.
Not having a "regular" temp sensor... maybe they just measure it based on a piece of copper? Temp rises and so does the resistance. Maybe the temp coefficient represents this change? Maybe it's the big shunt (at B- BTW)? Anyway it seems inaccurate to me.
I will test it properly in couple days, to rule out the soft lvc limit. Got the hit = checking the temp ASAP + voltage sag on the go. Right now I'm just unhappy and pouring tears all over ES [; Even my good old svmc72150 endures much longer without breaking a sweat (but it's got some real temp and phase current sensors inside ;- ).
larsb
1 MW
Never had this issue, but since i ride on road the loads are not comparable.
Some undervoltage protection is unlikely, don’t you test sometime at full voltage?
Some undervoltage protection is unlikely, don’t you test sometime at full voltage?
SilentEnduro
10 W
Hi folks, regarding shunt-mod.
While @Bartosh and @SilverLine are trying to find program way to don't let pull down the power, we are looking how to cheat controller's logic.
So for now in config we have 150A battery current, however with shunt-mod it's actually 180A.
Next step increase shunt to reach 210A then decrease to 100A in software so in the end we should have 160A of battery current but controller will think that it is 100A.
But again this is theory =)
will see...
While @Bartosh and @SilverLine are trying to find program way to don't let pull down the power, we are looking how to cheat controller's logic.
So for now in config we have 150A battery current, however with shunt-mod it's actually 180A.
Next step increase shunt to reach 210A then decrease to 100A in software so in the end we should have 160A of battery current but controller will think that it is 100A.
But again this is theory =)
will see...
and now imagine how this piece of sh!% pissed me off =) hard enduro is all about climbing))))Bartosh said:
larsb
1 MW
You don’t need shunt mod since controller is unlocked. Set whatever current you like for the busbar current. I am running 300A, this is also shown on my bms to be the rough output. Difference to 150A is huge.
If I get it right, SilentEnduro's trying to get rid of the cut offs, making an assumption the problem (or the controller's reaction) may be dc current- or shunt- related. If the controller limits the batt current in the event we're talking about, he may be successful. Problem I see here: there is no phase current sensor inside. If it's calculated based on that shunt, which I believe it is, you can overshoot and burn the mosfets. But I know nothing.
Anyway fingers crossed bud. It needs to start climbing.
Yeah, maybe it's easier to set 300A and check it out [; The question is what the controller actually does. Limits the input current, phase current, gain. It needs to be measured when it hits.
BTW The guy from Votol responded, asking for a link to this thread [one sentence, nothing more]. Maybe he will shed some light on what's going on here.
Anyway fingers crossed bud. It needs to start climbing.
Yeah, maybe it's easier to set 300A and check it out [; The question is what the controller actually does. Limits the input current, phase current, gain. It needs to be measured when it hits.
BTW The guy from Votol responded, asking for a link to this thread [one sentence, nothing more]. Maybe he will shed some light on what's going on here.
SilentEnduro
10 W
yep, you got it)Bartosh said:
p.s. have 2 spare controllers, so have a chance for mistake
larsb
1 MW
I’f it’s a hardware protection you’re trying to remove then how do you think it’ll work?
Let’s say root cause is overtemp then you remove temp sensor, take the risk. Should work.
If you instead adjust shunt then only thing you’ll do is offset the current sensing, still protection will occur from the overtemp.
If root cause is hard coded time limit for sport mode, same thing, it won’t be solved by offsetting current limit. Just reprogram controller in “high” mode instead
If root cause are voltage spikes (not likely) then you might succeed in shunt modding but risk blowing the controller
Let’s say root cause is overtemp then you remove temp sensor, take the risk. Should work.
If you instead adjust shunt then only thing you’ll do is offset the current sensing, still protection will occur from the overtemp.
If root cause is hard coded time limit for sport mode, same thing, it won’t be solved by offsetting current limit. Just reprogram controller in “high” mode instead
If root cause are voltage spikes (not likely) then you might succeed in shunt modding but risk blowing the controller
SilverLine
100 W
But how warm do you guys votol get ? Mine dosnt even get temperatur at all, when touching the heatblock. So i find it hard to believe its a over-heating problem ? That massive heat block, looks pretty overrated compared to many other controllers on the market.
alek.palmero
10 µW
- Joined
- Jan 21, 2020
- Messages
- 5
Hi.Bartosh said:
Can "15:cruise" setting be placed to the "Low Brake" that I don't use. Which "PB" corresponds to the "Low Brake" plug?
Thanks!
Enviado desde mi Redmi Note 7 mediante Tapatalk
Miguel Angel
1 mW
- Joined
- Jan 20, 2020
- Messages
- 15
alek.palmero said:
The (low brake) is the pb4
Hey guys. It's a freaking blackbox. Please point a temp sensor on the pcb and I'll do it in the blink of an eye (-;larsb said:
larsb said:
This is exactly what he (SilentEnduro) wants to do. IMO not to make the current reading inaccurate just for kicks... but to "soften" the limiting factor when the hit happens. Assuming theres only one current sensor (@dc), all the rest is calculated based on that. Whatever triggers this suck-mode, it's worth trying. No one knows how this actually works. Sure, it's risky (-; Especially the phase current val in the config is expressed in such an enigmatic way.
larsb said:
Just reprogram you say (-; Yeah, piece of cake.... Not. However I already have over 1MB of stm disassembly waiting on my desktop. But it's going to be a freaking nightmare without a src. If the limit/max phase current possible/current rise speed/temp is hardcoded somewhere in the firmware, it will require a lot of effort and motivation to find it. Tens of hours or more, everytime risking blowing the controller. But it's just me. I think I will look at the communication protocol 1st, sniff some data, maybe will try to change the soft start value, assuming it's filtered by the app itself.
IMO? I think it's temp coeff related. Maybe there's a constant, like the temp measured is always 25+coeff. Coeff based on some shunt, sensor, current reading? but the reading is bogus. Has anyone ever seen the controller's temp below 25? There must be a reason they print out this coeff in the app.
Another idea - some kind of traction control? I was surprised by my CA3 once on loose sand. Needed to turn this off. Unlikely though.
SilverLine said:
SilverLine, not even warm. App showing 34 deg but no power.
Too many questions. What we can do now is looking for the trigger that's responsible for the mess or waiting for the votol guys to comment on that.
SilverLine
100 W
I think we have more luck in you, than with the QS / votol guys :wink:
larsb
1 MW
It’s not really clear for me what the main issue is and what has been tried so far. Based only on what is written here i cannot say that it’s ready to put it on the shelf. I bet you have tried a lot but not written about it..
(Except for the frustration part off course :wink: )
Basic issue: output limiting occurs when hard driving for a longer time.(or?)
Is it a complete shutdown or just lower output?
How long time before output is back to normal?
Can you remove limiting by switching on and off? (Then it’s not overtemp error)
Does it happen the same way every time it’s tested (not cutoff due to random electronics error)
Did you readout controller errors when it happens? First thing to do..
Do you have BMS that could trip in overheat error? Did you try without BMS?
Did you remove all voltage protections in program? Does this affect limiting? Remove bms if you can before this test.
Can you try to preheat controller with a heat gun? Does it affect limiting? (It’s overtemp)
Did you measure the time it takes to limiting? Is it always the same? (Aka there’s a time limit in sport mode)
Did you try in both high mode and sport mode? Is limiting the same? (General limit in program)
Did you increase busbar current? Is limiting the same? If it’s consistently shorter then protection is likely overtemp in controller or bms.
Some hours of test would answer these and then it’s off to the shelf if there’s a bad conclusion
(Except for the frustration part off course :wink: )
Basic issue: output limiting occurs when hard driving for a longer time.(or?)
Is it a complete shutdown or just lower output?
How long time before output is back to normal?
Can you remove limiting by switching on and off? (Then it’s not overtemp error)
Does it happen the same way every time it’s tested (not cutoff due to random electronics error)
Did you readout controller errors when it happens? First thing to do..
Do you have BMS that could trip in overheat error? Did you try without BMS?
Did you remove all voltage protections in program? Does this affect limiting? Remove bms if you can before this test.
Can you try to preheat controller with a heat gun? Does it affect limiting? (It’s overtemp)
Did you measure the time it takes to limiting? Is it always the same? (Aka there’s a time limit in sport mode)
Did you try in both high mode and sport mode? Is limiting the same? (General limit in program)
Did you increase busbar current? Is limiting the same? If it’s consistently shorter then protection is likely overtemp in controller or bms.
Some hours of test would answer these and then it’s off to the shelf if there’s a bad conclusion
Hey. I don't have much time atm to describe it in details, but I've just done some more testing. A heat gun wasn't needed at all to "overheat" the controller. You can try it at standstill. Brake pressed, I did about 10-15 WOTs and got to the point where it started to loose "power". Initial phase current at 0 rpm started to drop. FYI there's a substantial difference between sport and normal modes in terms of stall phase current. Temp of the controller shown about 36, temp coefficient 8000-9000. IMO it's 100% temp coefficient related and this one is some kind of derivative of the current flowing (+rpms/time maybe). Based on that (plus the current reading, plus rpms?) the controller regulates the phase amps. I took some pictures, will describe it better in couple hours. Anyway I think the shunt mod is worth trying, assuming this is the place the reading comes from. I believe so. There's a place in the app (>) where you can calibrate the current reading. Changing this value affects the temp coefficient proportionally. It can't be saved but the controller keeps the settings until it's turned off. I changed it (significantly) which resulted in a beautiful current shot to start a wheelie but then the coeff rised and I got cut (-; Will keep you posted.
Larsb, please test you controller at standstill. You got an old version. I wonder if they didn't messed it up with an update. This is what I once got from qs, asking for s vs sp differences:
"3. Unlocked version is not available anymore, mosfet will burnt out easily."
"Hi, old controller is 470a, but for new controller, we have to limit as 380a for controller safty."
You also got the temp coeff printout in the app?
Larsb, please test you controller at standstill. You got an old version. I wonder if they didn't messed it up with an update. This is what I once got from qs, asking for s vs sp differences:
"3. Unlocked version is not available anymore, mosfet will burnt out easily."
"Hi, old controller is 470a, but for new controller, we have to limit as 380a for controller safty."
You also got the temp coeff printout in the app?
SilentEnduro
10 W
Bartosh said:
not sure about tepm, but it might be some kind of triger to cut-off current
IMO it may be like this - (Throttle Volts / Time) * RPM.
Lets imagine that you have full-opened throttle for 2 sec and no rotation = error, cut power (4/2)*0
Battery current should be throttle related, cause it's turque-controled, not RPM.
In this case shunt mod might help getting more Amps in same time and give some rotation to avoid error. risky, yep.
But this is confusing and might make no sense, but again this is Sparta China...
What controller can (basic mode):
measure Amps
measure time
get RPM info
calculate phase current
get temp? (under question)
get throttle position info
correct me if i'm wrong:
- it's like you open a throttle but nothing is happening. for example going long uphill, losing power on top but reaching flat surface, not releasing a throttle and it goes normat again.
in a second, no need to stay
nope
yep, and on different versions of Votol S/SP/CAN
no errors
same config with another controller (kelly) goes better, deffinitely not BMS related
yes removed, no changes
no need, can reproduce on cold controller
Different time, based on situation, but each time hard climbing
onestly, not tested on sport mode
limiting is not related to current changes in software
Just a quick question. You did test it in high mode only or there's no comma where it should be? (-; asking since the normal (high) mode is so laggy in my controller that it's basically useless. That's why I only use the sport mode.SilentEnduro said:
SilverLine
100 W
Bartosh said:
I have only used S-mode myself also. But when you say "laggy" is it on the bench, or also on the road ? Because sometimes the behaviour on the bench, is not the same on road and vice versa.
SilverLine
100 W
Bartosh said:
Okay damn...
Sport mode is all that i want then
Hoping for a solution.... The moment I have fixed my battery, i´m going to se if i can replicate the "problem"