You can use any heatsink, even a dual size one, which works even better when it’s only got one ESC on it.
6 days ago I suggested turning that down as it will run for longer before suffering from heat soak.
If you’ve not tried that yet, you’re doing yourself a disservice.
Try 20A battery max and see what happens. Nothing is stopping you from turning it up again later.
Absolutely agree with you- the more thermal mass the better. Only concern was space constraints
According to this chart, that controller should be able to do 60 amps continuous for your phase current but these controllers aren’t very robust. I think the issue is that you’re pulling a ton of current through a single controller to get your desired speed and torque etc… @b264 has given you a good suggestion re reducing the battery max. If that addresses your issue then great. Personally, I’d grab another controller and motor and ugrade your ride. For your needs I’d think 2 motors is required. Or maybe you’ve outgrown your board and it’s time to DIY a high-performance murder board
How many amps a controller can handle is mostly based off of time. There is some peak upper limit where the mosfets will short/burn, but for the most part what separates good controllers from bad controllers is mostly their ability to cool properly. Unfortunately there is no standard for current ratings on ESCs so manufacturers just put out whatever value they want and that’s what I used in that ESC Wiki.
For example, a VESC 4.12 and a VESC 6 can both be set to 80A motor current and run reliably for a long time… but the 4.12 will thermal throttle a lot quicker.
I’ve tested a few ESCs from room temp up to thermal throttle on the bench with no airflow using foc_openloop. The hope was to get an apples to apples comparison between controllers, and the results are pretty surprising.
The original VESC 4.12 has mosfets on either side of the PCB which feed heat into each other, and no heatsink by default. It heats up almost instantly and overall is a really disappointing performer.
The Flipsky 75100 boxy version uses TO-220 mosfets attached to a small aluminum rail, which is then attached to an aluminum box. Also a very disappointing controller.
The VESC 6 MK3 uses direct fets, which have a metal package compared to most mosfets which use an epoxy package, and is cooled from the top of the mosfet directly to a decent sized aluminum heatsink. It is a decent performer.
The Stormcores (and Unity/Xenith) went a different route. They used smaller mosfets but more of them, and primarily cooled from the bottom of the PCB. This allows the copper in the PCB to help spread out the heat, then it gets moved into a sizable heatsink. Keep in mind those numbers are with both sides running at the same time. The 100D heats up faster because higher voltage mofsets have a higher internal resistance… however 80A at 12s is ~3,500w and 80A at 18s is ~5,200W so it will still perform better if the voltage is higher. A 100D at 12s will perform worse than a 60D at 12s.
The Flipsky 75100 Aluminum PCB was a real surprise. They did the same thing by cooling from the bottom of the PCB directly onto a heatsink. It’s pretty cheap too
No not yet. I read your values in that link you posted and realised the values I had of 36/34 were pretty conservative and given I was getting about 3v voltage sag I figured 32/30 would be closer to 35/33 in reality?
I dropped the amps to 25, and it barely made a difference. As b264 originally suggested 15 amps but also said the performance will be bad, and I know it won’t climb the hill I need it to even at 25, let alone 20 or 15. So that’s not a long term solution. I will try 15 tho for trouble shooting purposes.
But I’m confused now about what would be a long term solution apart from going dual. Earlier everyone thought I had an earlier VESC 4.2 or 4.12 and said buy a new ESC. But it’s actually a 6.7 pro mini with a heat sink, and it’s not like 30 amps battery is a lot right?
Before I realised I had a 6.7 I actually picked up a cheap unused FSESC 6.6 over here in Oz that someone was selling nearby but now I’m wondering if that was a waste of money. Are they better at handling heat than the 6.7 mini? Is it even worth trying it?
Is it really normal for ANY 6 series VESC to heat up that fast running only 30 amps? I really didn’t think I was asking for that extreme a level of performance.
I’m only telling you because I’ve junked batteries before and it’s no fun at all. Do whatever you want, though. In a perfect world, running it down to 30V won’t hurt the battery. In a perfect world.
What are your phase current settings?
Phase current? I have no idea mate. The board is right here and I can bluetooth it but where do I find that setting?
Motor current in the VESC tool typically inputted at a positive and then a negative (brake) current.
My recommendation is still the same. I’ve built a lot of single drives, in fact I build mostly single drives, and have had this problem before.
It’s my opinion that if you put a FOCBOX 1.6 in it it, will perform a lot better. These can be found on the used market.
Adding external airflow (going through the enclosure) with the heatsink would be even better.
Sometimes you don’t even need to expose the heatsink.
This thin piece of aluminum is only thermally connected to the heatsink (on a FOCBOX 1.6) by those two M4 bolts, and yet it actually makes a lot of difference being there versus not being there. And this is a lot easier to waterproof and more robust than exposing the ESC through the enclosure.
This is like “half heatsinked”, not as good as exposing the metal on the ESC case, but not as bad as being bundled up inside the enclosure. It gives the ESC a way to cool off and bleed heat to the outside, through the two steel bolts. It’s also lifted slightly off the enclosure with washers.
Ah motor current? It’s 50 and - 50.
So focbox 1.6 definitely handles heat A LOT better than a FSESC 6.7?
Should I try the 6.6 or is that unlikely to be any different?
The only experience I have with a FSESC 6.7 is at low battery current. (not on an esk8) so I’m not sure how that one performs. Maybe someone else knows.
Just to add to the picture I rode home at 20 amps battery and it was quite a lot better. Only got to 62c at the top of the hill (instead of 80) and 67 by the time I got home. Performance tho, as predicted, was pretty lame.
Can a focbox handle 45A battery / 50A motor continuous, on a hot day, for a whole ride, climb hills etc and stay under 70c? Can any ESC?
From all the reading I’ve done I would have thought any VESC 6 could at least handle 30A standing on its head, and yet…
I rode a single motor board for quite a while.
I used a Torqueboards 4.12 and then a Trampa VESC 6 - 60A battery / 80A motor on both.
The 4.12 overheated and hit thermal throttle all the time.
The VESC 6 would only do it on super steep hills, or lots of continuous high speed stuff (need more power to push through the wind).
The Focbox won’t be as good as the VESC 6. It has the same mosfets (the electronic part that gets hot), but a smaller/lighter heatsink.
The mini FSESC 6.7 you have is a VESC 6 design, but all that means is that the layout of the electronic components matches the VESC 6 schematic. You can build a VESC 6 to handle 500A, or 20A. The mini FSESC 6.7 uses the same mosfets as some of the Stormcore 60D+, but only half of them. That means 4x the heating just by internal resistance alone. Then it’s sitting on a tiny PCB with an anemic heatsink cooled from the top.
I would guess that the mini FSESC 6.7 would fall somewhere around the 45-60 second range on the chart above.
If you can find a Focbox somewhere, it will work better than what you have but may not be enough. If you want a sub $100 single motor controller that will probably never thermal throttle on you, go for the 75100 Aluminum PCB. It does need to be opened up and the capacitors glued down, along with either grommets on the wire exits or extra shrink tubing to make it reliable… but it will handle 80A all day long without breaking a sweat.
The esc you linked was branded “Makerbase”. Is this the OEM for the Flipsky 71500?
I’m not sure. Based on their other offerings, I don’t think Makerbase made it.
I don’t like Flipsky as a company overall. They have stolen designs from people/not honored agreements in the past (DaveGa, original 75100) so this may also be a stolen design, or something they are putting their name on… but it could also be a Flipsky original.
It’s frustrating because this is such a good controller hardware wise.
Mate thanks for the detailed comparison / description between the 6.7 and others and where it sits in the mix - that was really the missing piece of the puzzle for me.
Still can’t work out why my VESC 6 behaves exactly as you describe yours did but with only half the amps.
I have no problem doing things like gluing and fitting grommets, but the programming side of things is where I quickly run out of knowledge. I read your thread actually on reprogramming the 75100 and it was like you guys were speaking a different language.
B264 walked me through setting the current esc up with VESC tool actually quite a while ago - is it essentially the same process? If so I could refer back to those notes, or is this a step too far for a beginner diy-er? We didn’t use the wizard in VESC tool but would it work on the 75100 essentially?