Good day all.
Hope everyone is safe and sound. I had a question about ESC temperature today.
I was tuning a quad on the weekend. IT went perfectly except after the roll tune I swapped out the battery and got a pre arm fail on Bat2 voltage. Normally I don’t use Bat2 for measuring, but since it’s a new build and I am looking for issues. I have is set.
Anyway long story short I got a Bat2 lo voltage warning. So I pulled the battery and tried a third…same result. So I packed up and when home.
I checked the logs and found that there was no ESC telemetry at that point. So in diagnosing the issue I found Motor 2 ESC while working and spinning the motor was no longer providing telemetry and in fact had ground out the telemetry line so no other esc was providing it as well.
So I am replacing that faulty unit nce more arrive.
But it sparked a question.
What is the normal temperature range the community see’s on their ESC and is it even accurate.
I ask because all I can think of is an over heated device, which struck me as odd as these ESC are https://shop.holybro.com/tekko32-f3-metal-esc-65a_p1124.html
Which handle 65 amps and on a bad day I pull. Less the 18 amps during a tune and even less liek 5 amps during a normal flight so I assume they can handle this current with out cracking a sweat.
I once bought a frsky nuron 60amp esc and it was getting to high a temp for comfort, until i did some research. Are you sure the degrees are in Celsius? I did some research to check to see what is the highest temp an esc can stand and without knowing every specific to each esc, I believe 250 degree Fahrenheit is the normal cpu temp that will lead to catastrophic failure if everything else is good in electronics. 250 Fahrenheit is a number I keep in the back of my brain. Finding on a rc helicopter forum I found an extreme case, a guy running his esc temps in excess of 220 degrees Fahrenheit. My personal record i have seen my esc temp get as hot as 178 Fahrenheit before it leveled off as I eased off the gas. I changed the esc out with one blheli_32 esc and now I rarely see it get passed 100 Fahrenheit.
I’ve done a lot of ESC temp testing and management over the past year or so. First of all, if you manage to get up to around 120c, the F3 (or whatever MCU) will shut down automatically and in the process reboot your ESC. Second of all, air flow is utterly critical. The 4in1 we run is relatively enclosed. We use an RC car ESC fan on it and it make a big difference in temps. From 80 down to 65.
Normal temp range is tricky. Really you just need to get it as cool as possible with heatsinks and air flow. How are they currently mounted? What are your ESC settings otherwise for BLheli32?
@AeRiAL_PeRpLeXiOn I can confirm all ESC temp telem is reported in celsius.
So these ESC are running at 90C or about 194F…thats pretty warm.
They currently are mounted in the arm tube which I think is trapping the heat and of course preventing air flow… I used high amperage ESC to make sure there was lots of head room and I naively thought the high amperage capabilities would mean a cooler running ESC. Guess not.
Here is the other few things that confuse me.
In all cases the ESC start temperature is around 50c to 70c which is already high which makes me wonder how accurate the temperature measurement is. Also even in sub zero weather leaving the drone outside all night. IT still starts in this temp range.
Also oddly one esc runs 20c coolers yet they are all mounted the same way. I will have a look at ESC 3 to see if there is anything odd about it’s mounting. But this also makes me wonder about the accuracy of the temp measurement
One observable difference between ESC 3 and the rest is that it was mounted face up.
So each ESC is inserted into the tube on a 3d printed frame that holds the ESC in the center of the tube. In the case of ESC 3 that mount was rotated such that it was facing up while the rest are faced down. They are all shrink wrapped and conformal coated for water protection. I wonder if I should pull the shrink wrap off.
ESCs get pretty warm at idle as they are holding all the amps back till you start using them. I found that as I pushed the ESCs we were using harder, they actually got cooler as they got close to their peak efficiency!
You MIGHT be able to get away with some heat sinks, but active air flow would be best. 90 sounds hot, but’s it’s not terrible for an ESC, so you could just continue as is. Definitely remove the shrink. The logs show that positioning is important as well so those metal canned ESC can cool off better. As for the accuracy of the temp sensors in BLheli32 ESC, I’ve found it to be pretty close to reality, within a few degrees.
@vosair
I am running 6s on this machine.
I am looking at perhaps additional heat sinks. But I agree airflow is important.
The one ESc that is running cooler while sitting upright in the arm is interesting. I did notice one esc was at a 45 degree angle in the arm. Yet tht appeared not not be any cooler. I have since turned then so they are all facing up wiht the exception of the one I am replacing. It’s in the trash. lol
I have found some small 11x11x5 heat sinks that I could add to the ESC or at least to one for testing. I have some double sided thermal tape.
I will remove the heat shrink on the ESC and test the temperature difference.
I am thinking I will build a few test cases.
Heat Sink on one.
No Heat Shrink on another.
One just facing up.
It will be interesting to see if any of this makes a difference. I have also considered a hole or two in the top of the arm or perhaps even on on top and the other below.
I am concerned with weakening the arm. But I might need to do this regardless.
Thanks again for your input.
I am waiting on the replacement ESC to arrive and while I am waiting I am printing a new tube mount for it. One that I can put in place and with a little hot glue make sure it doesn’t turn in the tube to keep it facing up.
Then I need to figure out what other items I will test out and see what behaves the best before I start drilling holes. Actually I am thinking of a small slot in the top and bottom. I could 3d print a jig and with a router cut a simple slot into both sides. something maybe 6mm wide at 25mm long.
But lets try the other stuff first.
I had slots in mind as well. I will say nothing beats a little air flow. You can probably get by without it, but it’s certainly best practice for any ESC.
ya I will look into it.
I don’t imagine it would b hard to do.
I have other things as well that I just don’t get like why one motor runs at lower RPMs then all the rest.
I have three motors that run at say 3000 rpm but one is running at 2600…that makes no sense to me.
Hmmmm I should check they are all the same propeller pitch…but apart from that . I don’t’ get it. Unless the telemetry is off. When the new ESC arrives I might flash them with the most current version of BLheli32.
sorry I mean that the new one is likely a new version and the old are back leveled.
This is the config today, I think the latest is 32.8 and these are 32.6.
Also the 32.8 has some little code changes that may be helpful
I’ve got similar settings on the tekko32 65A metal 4in1. These settings have been stress tested.
Main differences would be:
96 kHz (higher, default)
Rampup 24% (lower)
Sine modulation On
Our Demag and Timing are defaults, but I think you did have a desync previously and you have slightly bigger motors and props than us.
Hey Shawn
I can see your peak temp isn’t over 70 so your running cooler then I am.
Thanks for the input. I am going to look into ways of improving cooling.
May have stumbled on one by accident. Going to test it out once the replacement ESC is here.
Hey good morning.
Yes I would love to be able to do that and have kicked around some idea.
As I really would like to understand the accuracy of the temp measurement. I will keep thinking on it.
