Espritmodel.com Telemetry Radio
Reply
Thread Tools
Old Mar 15, 2012, 11:52 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Discussion
What's going on?

I've read of people hooking up their ESC to a motor and getting the magic smoke for no apparent reason, and I've read of ESC's losing timing, making a motor screetch.

Tonight I was piddling around with a 4s setup on one of my 450's, trying to figure out what throttle I would need to set in governor mode using an 80A Turnigy Superbrain ESC to have the motor run at the speed I wanted to achieve my desired HS.

For example, I know that setting my 40A Superbrain ESC to governor High mode, and using 75% throttle results in a motor speed of about 28200rpm on 3s with a 6 pole 3550kv motor, and using a 13t pinion onto a 150t maingear, that translates to about 2560HS.

So what speed would an 8 pole 2900kv motor run at on 4s? The answer seems to be 21200 at 75% throttle. That makes sense, since 21200/28200 = 6/8, pretty much.

The thing that suprised me was that the motor would run at the same speed whether the ESC was connected to a 3s or a 4s pack.

But my biggest surprise was when, on one test run, the motor started to run erratically, and I hit throttle hold to kill it. The motor had gotten hot. Upon reviewing the data log, there was a current spike of 98A! I'm not sure that my 4s 1800 25C pack will have enjoyed that!

In the attached log, you can see that the slow start takes the motor up to about 21200rpm in just over 6 seconds. The motor then runs fairly steady for about the next 7 seconds, before having a severe coniption that drops motor rpm to about 15600, and gives a massive current spike.

So what could possibly cause such a huge increase in current? The rise is more than twentyfold. Bear in mind the motor is running without any load. The ESC is supposedly rated at up to 210,000rpm for a 2 pole motor, which equates to 52,500rpm for an 8 pole motor, so I should be running well within it's capability.

Could it just be that some ESC's have "bogey" motors?

P.S. should have made this a "Help!" rather than a "Discussion".....
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Last edited by 2Doggs; Mar 16, 2012 at 07:21 AM. Reason: Category
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 16, 2012, 11:55 AM
DX5e fatal flaw- PM me!!!!
United States, NY, Cortland
Joined Sep 2010
2,839 Posts
Maybe the ESC gets a 'brain fart' and briefly (or longer) connects one phase across the batt. That would make the motor run erratically (as though a phase were dead).

Can FETs get into a state where they keep conducting when they shouldn't?
CNY_Dave is offline Find More Posts by CNY_Dave
Reply With Quote
Old Mar 16, 2012, 04:58 PM
We want... Information!
Bruce Abbott's Avatar
Hastings, New Zealand
Joined Jan 2001
5,161 Posts
Quote:
Originally Posted by 2Doggs View Post
the motor started to run erratically, and I hit throttle hold to kill it. The motor had gotten hot. Upon reviewing the data log, there was a current spike of 98A!
Your data log shows throttle going to 100% at the same time as the spike. Looks like the motor was seizing up, and the governor tried desperately to compensate.

The governor appears to have locked in at about 30% throttle. Is that right? (can't see a throttle scale on the graph). If so then how did it manage to get 21200rpm?
Bruce Abbott is offline Find More Posts by Bruce Abbott
Reply With Quote
Old Mar 16, 2012, 05:05 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Quote:
Originally Posted by Bruce Abbott View Post
Your data log shows throttle going to 100% at the same time as the spike. Looks like the motor was seizing up, and the governor tried desperately to compensate. ?
When it happened, it was quite dramatic. I could immediately sense radiated heat from the motor and ESC, and I am sure that the throttle was indeed at 100%.

Quote:
Originally Posted by Bruce Abbott View Post
The governor appears to have locked in at about 30% throttle. Is that right? (can't see a throttle scale on the graph). If so then how did it manage to get 21200rpm?
Having calibrated my TX throttle to the ESC, I started out setting the throttle curve - actually a flat line when using governor mode - to 75%. That doesn't mean the actual throttle will be 75% - it will just be whatever it takes to get the motor to run at the commanded rpm - in this case approx 21200. If I put load on, you'd see the motor rpm maintained, but the current draw and throttle in the data log increase.

Trouble is, I'm actually fearful of testing the motor now! I think if I'd been using a 40A ESC, there would definitely have been white smoke!

Reading other posts, I see that badly soldered connectors might cause the ESC to lose timing - but I used lots of heat from a 50w iron and thought I'd done a pretty good job of the connectors. If a connection was bad, would it have been possible to get that 98A peak?
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Old Mar 16, 2012, 06:08 PM
Registered User
icarus46's Avatar
Joined Apr 2010
280 Posts
Sounds like an esc issue.
icarus46 is offline Find More Posts by icarus46
Reply With Quote
Old Mar 16, 2012, 06:20 PM
Registered User
United States, NJ, Frenchtown
Joined Mar 2003
9,016 Posts
That is 1 phase. Correct ?
I am interested in the flat black line showing heat spikes occuring more & more frequently all the time. Why ?
The the intermittant jagged rising line. Looks like a drive being erratic as speed increases.
Hook up a dirt cheap LOW wattage motor. Strap the motor DOWN safely. See if it freaks out also. Then send back / buy a new ESC.

Might be the motor. Smell it VERY CLOSELY. See if there is any bearing tiny catches as you rotate it several revolutions in each direction. Slide the shaft in & out. Should be almost no play
cyclops2 is offline Find More Posts by cyclops2
Reply With Quote
Old Mar 16, 2012, 06:24 PM
Registered User
United States, NJ, Frenchtown
Joined Mar 2003
9,016 Posts
Poor solder joint cause THAT connection to almost become a broken lead.
That creates a almost dead short circuit in the other 2 wires.....Bingo 98 amps.
cyclops2 is offline Find More Posts by cyclops2
Reply With Quote
Old Mar 18, 2012, 12:40 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Update - some progress

Well, I'm making some progress on figuring out this problem.

First of all, to reduce my abject fear of the ESC I decided to conduct some more tests using a very weak, tired old Turnigy 3s 2200 pack. After all, if it struggled to supply even 15A in the 450, it couldn't possibly give rise to those frightening 98A peaks!

No matter what throttle percentage I set in governor mode, I could not get the motor to run above 21200 rpm, so I tried turning governor mode off. Same thing - the motor gets to 21200, and then just stutters - though without using massive current or dropping to half the rpm.

Now with a 3s pack, putting out a meagre 10.35v under the 5A or so load, by my calculations a 2900kv motor should be able to spin at just about 30,000rpm - so I reckon it could be the ESC. I don't want to rush in and load different firmware, since I don't have a copy of the existing firmware - and the new foirmware might result in the ESC not working at all.

So my next step will be to test the motor with my known good 40A Superbrain, on the feeble 3s pack. I reckon if I can get the motor up to 30,000 rpm on 3s with the 40A Superbrain, it will more or less confirm a problem with the 80A one.

If it behaves the same, I will re-solder all my connectors.
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Old Mar 18, 2012, 01:35 PM
Suspended Account
Joined Mar 2012
16 Posts
I have never seen governors work the way they are suppose to in these small helis. And I found that on a 450 heli it is a waste of time anyhow. The best thing is to be running 100% all the time, this is where I found the best performance and least problems.
Desert Rider is offline Find More Posts by Desert Rider
Reply With Quote  (Disabled)
Old Mar 18, 2012, 02:03 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Quote:
Originally Posted by Desert Rider View Post
I have never seen governors work the way they are suppose to in these small helis. And I found that on a 450 heli it is a waste of time anyhow. The best thing is to be running 100% all the time, this is where I found the best performance and least problems.
Whereas I've had great success with governor mode on my 450 using the 40A Superbrain, and the Hobbywing Platinum Pro. Of the two, the SB holds the rpm much more tightly, but if I switch from normal to stunt mode and have more than 1 or 2% difference in the throttle settings, the abrupt change causes the OWB to wear out in just a few flights.

The conventional wisdom is supposedly that you don't need governor mode on small helis - but in the end it just depends on how you like to fly. I love the fact that in governor mode the heli feels the same throughout the flight. I can drop the heli with full low collective without overspeeding the head, and I can use normal mode as one of my aerobatic modes. So it's just horses for courses.
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Old Mar 18, 2012, 09:20 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
More Progress

OK, maybe I should have made this a blog entry, but I'll post the results of further tests as I do them.

The next thing I did was to run the 2223 motor hooked up to my known-good 40A Superbrain, with a 3s pack.

Both in governor High mode and un-governed (confusingly called governor "auto" mode in the Superbrain) the maximum motor rpm I could achieve was 24600rpm - which, with the voltage under load of 11.8v, would suggest the motor was behaving as if it was 2100kv rather than 2900kv.

On one test run, there was an instantaneous rpm spike in the data log to 29700. I'm not sure if this was just a glitch, but if the motor did reach that speed, it would suggest it was 2500kv. (when I did the test run, I don't recall hearing the motor speed up, even for an instant, and on the data log there's no accompanying current spike, so I reckon it was merely a glitch in the logger)

As a check, I then ran my Typhoon 2215H motor with the 40A Superbrain, and at 11.7v under load, it reached 41600rpm. That tallies well with it's claimed 3500kv rating.

So, my next test was to run the 2215H motor using a 3s pack and the 80A Superbrain - and I've attached the data log. The scale on the left side is set to show the motor rpm, and you can see that the maximum achieved was only around 34000, and that the response became erratic with increasing throttle (the thin blue line) so that at about 73% throttle, at around 6 seconds into the run, the current spiked at 27.4A, with the motor rpm dipping to around 26000 rpm. Note that with the ESC in "auto" mode, the throttle was changing according to my TX stick movements.

So it seems like there is definitely a problem with the 80A Superbrain, and I may have to try other firmware.

It's also interesting to see that the minimum current draw shown in the 80A SB logs is always 5A, whereas the 40A SB shows the 2215H to be drawing only 2.6A at 41600rpm.

I'm not sure if the 2223 motor was damaged by the 98A current spike that it had to endure, and if that could make it behave as if it was a lower kv motor. As far as I know, the higher the motor kv, the less turns there are in the stator windings, so any shorting of windings, which would also have to be unfeasibly symmetrical, would result in the motor kv rising rather than falling.

I have more tests that I can do - I also have a 2218 motor, 2500kv, which, like the 2223 2900kv one is a re-branded 8-pole Suppo motor, and I might even try running the 2223 off another two 40A ESC's I have.

At least I am beginning to regain confidence in my soldering skills, and the 98A peaks in the earlier tests tend to suggest that the 4s 1800 25C Gens Ace packs are going to be very strong packs!
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Old Mar 18, 2012, 09:24 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
More Progress

___
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Last edited by 2Doggs; Mar 18, 2012 at 09:25 PM. Reason: Duplicate!
Reply With Quote
Old Mar 20, 2012, 09:05 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Sorted - sort of!

Well, loading new firmware sorted the ESC out! Even though I could not find a copy of the firmware that came with my SB to use as a backup, I decided to load the DTuA A108 12MB 8000 firmware that can be found in the "files" section of the HK SB80 product page.

The first motor I tried with the new firmware was my 2218H 2500kv, since it's also an eight pole re-branded Suppo motor like the problem 2223. It ran fine, on both 3s and 4s, although it appears to be effectively 2600kv, not 2500kv. (43,000rpm @ 16.5v)

I next tried the 2223 2900kv motor - and that ran fine too, but effectively as a 2080kv motor, rather than 2900kv. (33,300rpm @ 16.0v)

It's really odd - Suppo don't list anything other than 4, 5 and 6 turn motors in the 2223 range, with 4400kv, 3500kv and 2900kv respectively. To achieve 2100kv, the motor would have to use 8 turns.

I might still be able to use it with a 12t helical pinion and 121t helical main gear, and can use the 2218 with a 12 / 150 combination.

So at the end of all of this bother, I'm relieved to have sorted it out, but dismayed that HK should send out ESC's in a non-functional state, and a state that might well have resulted in a burned out motor, ESC and battery pack to boot.
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Old Mar 21, 2012, 06:56 PM
just gotta mess with it!
2Doggs's Avatar
North West Louisiana
Joined Nov 2009
4,583 Posts
Governor Mode - deeply flawed!

One of the main reasons I bought the 80A SB was that the governor mode on my 40A SB was so good.

I am therefore shocked to see that governor mode on the 80A SB is deeply flawed.

It seems to have been set so that at 100% throttle in governor mode, you get more or less the same rpm as you get with 100% throttle with governor mode off, when running a 3s pack.

But that maximum governed rpm stays the same whether you use a 3s, 4s, 5s or 6s pack!

As an example, I can get my 2600kv motor to run at 43000 ungoverned on 4s, but only 31200 governed.

Unless you happen to be running a 3s setup, therefore, governor mode is effectively useless - at least with the DTuA firmware I loaded. I can't say if the DTKA firmware I originally had treated governor mode differently, since anything above 50% throttle was likely to burn up the motor and ESC.
2Doggs is offline Find More Posts by 2Doggs
RCG Plus Member
Latest blog entry: Hk500 tt fbl
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Suggestion What's going on with the moderators? HappyKillmore Site Suggestions / Complaints 0 Mar 02, 2012 10:15 PM
Discussion What's going on with this UBEC? Berkie Power Systems 5 Jan 03, 2012 03:46 AM
Discussion What's going on at Virginia Tech?? Marten Life, The Universe, and Politics 9 Dec 08, 2011 01:33 PM
Rant What's going on in is country ?? 335driver Life, The Universe, and Politics 4 Sep 29, 2011 10:16 AM