Hmm righto. I’ll try to confirm esc throttling today with RT data.
If I switch to focbox I assume I have to reprogram everything? New remote? Is 1.6 or 1.7 the best option here? What about newer VESC models.
Where would I find a 1.7?
yes (kinda) Many options can just be copy pasted
no, current one can be reused.
1.7 (or completely different esc)
Not new from shop, only used/new from private sellers. You need to ask if its a 1.6 or 1.7 (they need to open up the cover to see)
For a single drive, 1.6 and 1.7 are identical.
For a dual drive, 1.7 is slightly better in that it has a tiny bit more safety circuitry for tolerating a miswiring without frying the CAN chip. But if wired correctly, they are are the same.
So for a single drive, a 1.6 might be slightly easier to find and/or slightly cheaper, but we’re talking by razor thin margins here.
The thing I wouldn’t suggest is a 1.3. There are significant improvements between the 1.3 and 1.6. The 1.3 also wasn’t called FOCBOX, it was called VESCX 1.3.
If you meant to ask “Is FOCBOX the best option here?” the answer is no, probably not. Assuming cost is no factor, there are much better things available.
If cost is a factor, there are also a lot of other ESCs that will work. But there are several of these on the used market right now you could probably snag for less and they will work GREAT for what you need.
2 Likes
I wouldn’t say cost isn’t a factor but what ballpark are we in here? Given my very limited programming knowledge is there any option that would allow me to backup current settings in VESC tool, install new VESC and then restore settings to the new VESC? Or is that a no-no in general?
Before I start throwing money at this problem here’s some RT data from the ride home. The vertical line on the charts is the top of the hill in question, and I arrived home at the 500 second mark
Current settings are
Motor amp Max 50
Battery amp Max 45
Battery voltage cutoff start 36 end 34
Mosfet temp cutoff start 85 end 100
Motor temp cutoff start 85 end 100
What are you guys seeing here? The mosfet temp doesn’t seem to ever get over 81, so that doesn’t actually look like the issue? The motor temp reads as -86 so I don’t think it even has temp sensors? Is it battery voltage cutoffs that’s throttling performance? There seems to be a lot of voltage sag - is that a normal amount?
Also I’m wondering why battery amps never gets over low 30s when it’s set to 45? At the time I’m trying to climb the hill it’s down around 20A.
Just thinking about voltage cut-offs and sag- I note the voltage was right down towards 34 there near home, but the battery is indicating still half full and after resting at home for a few minutes voltage was back up to 37. So if you lower the end cut-off to say 28 knowing you have that level of sag, the cell voltage would actually still be about 31 when not under load? Am I thinking about this the right way?
Definitely temp issue. On the last screenshot you throttled after only 5 minutes.
2 Likes
Isn’t that just showing temp tho? To me the mosfet temp graph just plateaus suddenly but I can’t see what has actually caused the temp to stop rising.
I. E. you can’t see any distinct throttling or drop in any of the other graphs. Notice in the first screenshot there’s no discernible change in battery amp output at the same point in time for instance.
And the fact that the mosfet never actually hits the start temp cutoff of 85 - is that just an expected margin of error in the sensor or software or something?
I just found in settings something called ‘acceleration temp decrease’. It’s set at 15%. If I understand correctly, that will drop the cut-off start to about 72 degrees under acceleration?
If your battery was down to 34 volts you’d be experiencing soft cuttoff. Could also be heat on the controller though. Ur fets got pretty hot a few times. I’d also be concerned as to why your 10s3p pack is barely taking you 8km. Are you just gunning it the whole time? I used to be able to get like 20km out of my near identical pack driving somewhat conservatively on 2 x 6355 motors.
If I understand the graphs correctly the mossfet got to 80c in about 5 minutes. I was punching a 20 knot headwind for about 2 km right before the hill but even so it seems to get hot quick.
In terms of range I had already ridden about 10km or so on that pack earlier in the day before I recorded the data.
Should I be looking for a new esc or investigating settings a bit more do you think?
Focbox 1.6 seem hard to find…
And yes I am gunning it pretty much everywhere unless there’s a corner or stop sign etc.
There are lots of other controllers besides the focbox 1.6/1.7 that would suit your needs.
I’d personally grab one of these Torqueboards controllers if they’re still available:
There’s lots of options for single controllers though- makerx has a few:
Or, you could grab another motor and a dual controller and upgrade your ride.
1 Like
Just tying this one off for anyone who has the same issues - it is definitely the MOSfet temps causing throttling. Even though they never actually get to the start temp cutoff of 85, I just went for a long ride in the wind/uphill and watched the live RT data. Whenever it hit 80c it started losing power rapidly, and then by about 84 it was like riding it in safe mode or something. I dropped the voltage cutoffs to 32 and 30 just to eliminate that as a factor also.
So I guess I’m up for a new ESC. It’s strange they don’t seem to be designed with external cooling fins that you can through-mount on the enclosure - is that because most simply don’t have this overheating issue? I’m running ABEC 107s, 40 tooth rear 15 tooth front cog and climbing hills in head winds with a single ha ha. To be fair I only weigh about 68kg. Is there any esc that handles heat better than others? That maker x one looks really small - how does it keep cool?
There aren’t really any through-mount heat sinks for single controllers that I’m aware of but ultimately, thermal mgmt is about figuring out a way to heatsink your fets. Fins are overrated imo. It’s really all about thermal mass and ensuring heat transfer between the controller and the sink. Just buy a chunk of 5mm alu, cut it to size (scroll saw and WD40 w a metal blade) and make sure you have either a thermal pad, paste or tape between the controller and the sink.
You may even attempt to heatsink your current controller before upgrading for science. Alu is much cheaper than a new controller. In any case you’ll likely eventually upgrad the esc bc those Flipskys are pretty shite.
So I just pulled the board apart to see how I might go about adding a heat sink to the esc and turns out it already has one, with what looked like thermal paste joining it to parts of the esc (assume the fets?). It was all still wrapped in heat shrink tho, and I had it mounted with Velcro stuck to the heat sink side. So it would have had literally no air flow.
I cut a square out of the enclosure a bit smaller than the sink and then sikaflexed it in place so it gets direct airflow. I do recall b264 saying it was hard to sink these effectively so I won’t be surprised if it still overheats but worth a shot I guess. Still needs tidying up, and I might try to fabricate a little air scoop the airflow makes a difference. Waiting for sika to dry…
Well that made precisely zero difference damn it. Even with an air scoop, 5 minutes punching into a strong headwind then a hill and the esc hit 80c. Guess I’ll have to drop battery max down to 25 or so and shop for a new esc.
Guys I’ve just realised shopping around for ESCs that I’m pretty sure this is a FSESC 6.7 in my board already, not a 4.2.
Am I right in thinking this should be more than capable of 40A battery continuous, 50A motor without throttling?
Did you raise them back up?