Author Topic: Motor Failure Mode? (Read 1761 times)

Robert Redmon · « **on:** August 24, 2012, 05:21:57 PM »

The other day, I smoked a Phoenix Lite 50 esc when I nosed over in (errr, suddenly) high grass on take off (grumble). I also lost the E-flite Power 25 motor. (No other damage, didn't even break the prop.). Motor doesn't have any obvious (visible) burned places on the windings and resistance checks nominal (no dead shorts, no opens). However, it doesn't feel the same when I turn it over in my hand. It almost seems like the brake is on....but, of course, it is not hooked up to anything. That could mean that the windings were shorted somewhere, but I cannot find a short. Just curious and still smarting after losing $150 in equipment in a simple nose over. Theories, explanations?

Bob

Tim Wescott · « **Reply #1 on:** August 24, 2012, 06:57:32 PM »

If one turn of one winding is shorted then the motor will show a lot of drag (and make your ESC profoundly unhappy), but would have a winding resistance close to normal. It could well be that.

Theoretically it'd show up as one leg having a lower resistance than the other two, but that would require a better meter than most electronics types have, much less most modelers.

John Rist · « **Reply #2 on:** August 24, 2012, 10:08:27 PM »

If you were using a KR timer I don't think you would have lost anything. I have tested mine by grabbing the motor with a rag. It simply turns off.

Andrew Borgogna · « **Reply #3 on:** August 24, 2012, 10:32:39 PM »

As John said get a KR timer/governor and put an end to this kind of loss. I did exactly the same thing on the first electric stunt plane I built. Cost me a motor, ESC and a battery. The only thing that didn't smoke was the timer. After that I switched to KR and have not lost a system since and I have had several nose overs and prop strikes. It really is an unnecessary waste of good hardware.

Andy

John Cralley · « **Reply #4 on:** August 25, 2012, 09:25:05 AM »

Others have had ICE 50s blow in similar situations.

See this thread for more information:

http://stunthanger.com/smf/index.php?topic=27929.0

Mark Scarborough · « **Reply #5 on:** August 25, 2012, 11:38:17 AM »

I lost an ICE lite 50 in the air,, its not the timer,, I suspect that something in the mode we are using the 50 is vulnerable,, though I have heard of a couple toasting them in the RC world, again, not overloaded,, just letting the smoke out.
I am going to upgrade to the 75 amp controller,, I have used them in RC planes as well as CL, if the parameters are set right in the current limiting, they will shut off with a prop strike. its not the timers fault.

Robert Redmon · « **Reply #6 on:** August 25, 2012, 02:30:23 PM »

Thanks for the replies, guys. Tim, I suspect you are correct. I am able to measure .1 ohms with reasonable accuracy, but one turn would be less than that. I imagine it is hidden near the laminations. Others....yes, I am familiar with the many similar failures suffered by owners of the Ice Lite 50 esc. I own three of the things and have followed the trials of others over the past year and a half since I bought mine. This is the third prop strike I have suffered, but the first to cause failure. Also, I have a couple of Keith's timers on the way. I may learn slowly, but I do learn.

For the time being, I will use KR timers with my Ice Lite 50s; however, it seems that a much less sophisticated esc is needed with the KR, so is anyone using a less expensive but robust esc with their KR timers? If so, what?

Bob

Andrew Borgogna · « **Reply #7 on:** August 25, 2012, 05:33:23 PM »

Yes. I use CC Thunderbird series ESCs, ZTW ESCs, Turnigy ESCs, and Hobby King ESCs they all work with the KR timer/governor. There maybe others but these are ESCs I have personally used and are currently installed in my planes. They all cost less and coupled with the KR timer they give you protestion against nose overs.
Andy

Dennis Adamisin · « **Reply #8 on:** August 26, 2012, 07:51:57 PM »

Bob
Circle back to your original post: You said the motor "feels" different, I do not know how burning up windings could cause that (and I could be all wet too) After a prop strike it COULD be possible that either the motor shaft is bent, the bearing tower is bent or bearings jogged out of place.

Admittedly any of those scenarios assume a pretty stiff SMACK - but since the only thing the motor turns on is the shaft in the bearings, it is hard to see how anything ELSE could come into play... Make any sense?

EDIT:
Forgot to mention this: I have KR timers flying with Matrix and Turnigy Sentry ESC's. I others RTF with Arrowind (blue label), I had a small set-up issue because of those ESC's "Super Soft Start" mode (see the thread on that subject) but that was easily corrected and we are now good to go.

Tim Wescott · « **Reply #9 on:** August 26, 2012, 10:00:30 PM »

Quote from: Dennis Adamisin on August 26, 2012, 07:51:57 PM

Circle back to your original post: You said the motor "feels" different, I do not know how burning up windings could cause that (and I could be all wet too) After a prop strike it COULD be possible that either the motor shaft is bent, the bearing tower is bent or bearings jogged out of place.

Bob didn't say how the motor felt, and I made an unstated assumption in my answer.

A shorted winding will make the motor more draggy in a way that feels like viscous friction. You can feel this yourself if you have a motor handy: short two (or three) motor wires together, then give it a twirl. You'll probably need to repeat this with the motor un-shorted, unless you twirl it good and fast, or have better feel for this sort of thing than I.

A shorted winding won't present a perceptible drag with the motor turned slowly, while a dinged bearing or something bent should be readily apparent when you turn the motor dead slow.

Robert Redmon · « **Reply #10 on:** August 27, 2012, 11:46:22 AM »

Dennis and Tim,

The motor does indeed turn sluggishly, like the windings are shorted together (brake mode). That is why I indicated I thought you had identified the problem, Tim. I have satisfied myself that I have a short near the laminations.

Dennis, the prop never struck anything harder that thick grass a maybe a bit of soft turf. Never got off the ground, and only rolled about ten feet all together. We had just had a couple days of rain and the grass on my circle decided to pop up a couple inches overnight (it seemed)....Thought I could get it off anyway....bad decision. Anyway, shaft and bearings all okay.

I am a bit disappointed with the Castle "Ice Lite 50" esc, though. I had thought that the current defined cutoff would protect against such happenings. Obviously, I (along with others it seems) was wrong. Anyway, a couple of Keith's timers should arrive today from Eric, so I am (hopefully) on my way to avoiding future such events.

Heard from Castle, and apparently, they are going to charge me to repair the esc ($50 flat rate)....even though it was still under warranty (had no sales receipt....bought on ebay and failed to preserve paperwork). Episode ended up being quite expensive.

Bob

Andrew Borgogna · « **Reply #11 on:** August 27, 2012, 01:49:48 PM »

Dennis, the prop never struck anything harder that thick grass a maybe a bit of soft turf.

Bob
The first thing you must understand regarding electric motors, current goes up as the RPM is dragged down. The current is maximum at zero RPM. You don't need to stop the motor to generate current values way above the safety limit of the ESC. If the MOSFETs in the ESC short out then the full potential of the battery is placed across a set of windings in the motor, and then the motor, ESC and if you are not lucky the battery are all history.

What needs to happen is the throttle pulse from the timer to the ESC must go down to 1 ms. as the RPM is dropping before the current level reaches a point where it can blow windings. Or the ESC senses a rise in current or reduction in RPM and shuts down the current to the motor windings. This requires very fast action by the either the ESC or the timer. For the timer to do this it must know the RPM of the motor. To know motor RPM the timer must have a feedback path from the motor itself. The ESC can monitor the same back EMF pulses from the motor that the KR timer does, so in theory it should be able to detect a sudden drop in RPM or rise in current, and before things start to smoke it should be able to shut down the current. It would seem the ESC you were using either does not sense RPM dropping below a safety threshold and shut the current down, or it doesn't do it fast enough.

Also remember that when a governor is involved the first thing it will try to do as the revs start to drop is to raise them back up, that means add current. So there has to be a threshold established where the revs are detected to be in a area heading to motor/ESC destruction and shut the current down before the destruction can happen. This takes quick timing and quick sensing, and very quick action.
Andy

Robert Redmon · « **Reply #12 on:** August 27, 2012, 05:29:23 PM »

Thanks for the concise and cogent explanation, Andy. Makes perfect sense and a strong case for not allowing the esc to govern motor rpm unless it also monitors for and shuts down in the event of dramatic changes in that rpm (like Keith's timer) in addition to sensing for overcurrent. Looking through the clear shrink tubing that Castle uses, it appears that they use some kind of a fusible link and that it melted, shutting things down before destroying the battery and associated wiring. To be honest, I was very surprised to not find visible damage to the motor windings....hence my original posting.

Bob

GonzoBonzo · « **Reply #13 on:** August 27, 2012, 08:42:19 PM »

Wonder if a 45 amp blade automobile fuse would work. Might make a great arming plug too!

John Rist · « **Reply #14 on:** August 28, 2012, 06:59:23 AM »

Quote from: GonzoBonzo on August 27, 2012, 08:42:19 PM

Wonder if a 45 amp blade automobile fuse would work. Might make a great arming plug too!

I am guessing no. Fuses have resistance to make them work. It seem reasonable that adding a fuse is the same as adding battery wire length. It has been pointed out that that long battery leads can destroy speed controllers. Also fuses are slow to react and can require currents greater the rating to blow. Having said all this fuses are not a common pratice - so reason tells me it probably is not a useful solution.

GonzoBonzo · « **Reply #15 on:** August 28, 2012, 12:21:31 PM »

Hi John,

Is the resistace that much more than the wire? It's probably not common practice due to the fact people assume a reputable ECU like the Lite 50 would shut it self off before it fries.

Tim Wescott · « **Reply #16 on:** August 28, 2012, 12:36:19 PM »

Quote from: GonzoBonzo on August 28, 2012, 12:21:31 PM

Is the resistace that much more than the wire? It's probably not common practice due to the fact people assume a reputable ECU like the Lite 50 would shut it self off before it fries.

The real problem is that fuses take time to blow. Automotive fuses are sold into a market where the two biggest constraints are price and a desire to protect the vehicle from bursting into flame. Price aside, what they're protecting against is the wires heating up, and that takes some time. Because a wire that's designed to carry 45 amps continuously in an automotive application will be perfectly happy carrying 200 for tens or even hundreds of milliseconds, you don't want the fuse to blow quickly. On the other hand, the actual working part of a FET is teeny, and will fry with overloads lasting in the ones or tens of milliseconds. So the sequence of events with a fuse goes overload -> dead FETs -> fuse blows. That protects your wiring and maybe battery, but not your ESC. If the fuse blows slowly enough it won't protect your motor wiring (which is designed for light weight, and has less thermal mass) frying, either, although the motor would probably fry after the ESC.

GonzoBonzo · « **Reply #17 on:** August 28, 2012, 03:50:14 PM »

Sure they take time to blow, but seem to do a good job at protecting sensitive electronics. (radios, power amps). I would guess these nose over, over amp cases are more of a slow roast. Probably took 10 secs to get to the plane.

I'm curious, what type of battery and C rating?

Robert Redmon · « **Reply #18 on:** August 28, 2012, 04:35:24 PM »

Battery 3.3ah 25c

Andrew Borgogna · « **Reply #19 on:** August 30, 2012, 11:11:02 AM »

The R/C guys tried fuses in the early days of R/C electric and finally gave up on them. It seems the only thing that got protected was the fuse.
Andy

Robert Redmon · « **Reply #20 on:** August 30, 2012, 06:31:26 PM »

I can confirm that. I was one of those guys. In the early days (fuse full of nicads and Astroflight motors with brushes) it was quite common to use an automotive blade fuse for an arming switch. Actually, they introduced less resistance into the circuit than the cheap (gold plated) Molex connectors.

Bob

Tim Wescott · « **Reply #21 on:** August 30, 2012, 06:50:28 PM »

Quote from: Robert Redmon on August 30, 2012, 06:31:26 PM

I can confirm that. I was one of those guys. In the early days (fuse full of nicads and Astroflight motors with brushes) it was quite common to use an automotive blade fuse for an arming switch. Actually, they introduced less resistance into the circuit than the cheap (gold plated) Molex connectors.

I was one of those guys in those days, although I was just learning RC. My biggest success with an Astro 020 was to crash the plane into a tree above my head. Oh, sweet victory!