New 30AMP ESC burntout?

[Tony H]

Hi all,

 

I just got a 30AMP Airtek ESC for my little speed 400 glider. I have bought an equivalent 400 outrunner motor connected it all up with a 3s lipo and all was fine when I tested it.

 

Then I left it went back to test it again (a few seconds run only) and there was little power and the ESC started smoking, so I unpluged the lipo straight away.

 

After inspecting the motor wires I noticed the yellow middle wire had broken from its connector (clean off).

 

Could this be why it smoked?

 

I reconnected everything properly again and now the ESC has only about 1/3 thottle of power.

 

Tony

[andy watson]

I would guess the leccy bods will want some more details about your system, but I would advise you to phone them.  I bought a charger from them, and for some reason it went "pop" when I plugged it in.  I sent them an email, and they got a replacement out to me before they had received the faulty one back- so excellent service.  They also refunded the postage it cost me to return the faulty one.

 

Can't praise them highly enough.

[Tony H]

Thanks Andy. I would but it was a christmas present, which makes it more awkward. Would the yellow mddle wire breaking cause the overload/smoking?

 

My setup is a 930kv Airtek Outrunner Motor (Airtek AK/2830/930/15T) with a 3s 2200 Overlander sport battery and the ESC was an Airtek 30amp, all powering a 8x4.5 folding prop.

 

I was only testing it in my spare room at the time a feww seconds at a time. I have tried my 50amp arrow wind on the motor after the other ESC failed and it worked fine. 

 

 

[David Ashby - Moderator]

Sounds like the loose wire caused the problem Tony unless the system was pulling substantially more than 30 amps to cause the ESC to fry or suffer but, based on the info you've given, it shouldn't have done. I assume you've measured the current when the 50-amp ESC is in situ?

 

  

Edited By David Ashby - RCME Administrator on 03/01/2011 08:04:14

[Tony H]

Thanks David. What I don't understand is why would a broken motor cable burn out the ESC? the last time I used electic motors was in RC Cars about 15years ago on normal brushed 540 motors if the motor cable pulled out on them the motor would stop but that is all.

 

Why do out runners need 3 cables? Airtek recomend only a 18amp ESC for this motor. I'm still waiting for my watt meter in the post so I haven't been able to measure the current draw yet.

[David Ashby - Moderator]

Not saying it did Tony, bit of a long distance guess although perhaps intermittent shorting-out damaged some circuitry?

[ken anderson.]

where's sparky mackey(timbo)..when we need him ?..

 

 

      ken anderson   ne...1.

[Martin Harris - Moderator]

I'm not going to put it to the test but I'd guess that given that as there is virtually a dead short while the motor is stationary and the back emf is not being generated, if the motor isn't turning due to a wire off then it won't take long for a magic smoke pipe to burst!

 

The brushless motor (outrunner or inrunner) is a 3 phase motor which has bursts of energy fed to the relevent windings at the right time by the ESC - sensed by the back EMF generated by the motion of the magnets past the coils or vice versa.  A very different concept to a brushed motor which simply stops running if a wire breaks.

Edited By Martin Harris on 03/01/2011 13:37:59

[Tony H]

Thanks guys,

 

I think the motor did stop and after a few seconds smoke came out of the ESC. The ESC still works but at much reduced power, why is that?

 

EMF= Electro Magnet Field?

 

So in short if any of the three motor wire conectors disconect or break the motr will stop turning and any ESC attached is likely to smoke and die?

 

 

[Tony H]

Also the lipo and ESC is fitted with Deans Style plugs so I shoulded be able to plug them in the wrong way around. The do give a little spark when I attach them though.

 

How do you turn a Electric planes motor and RC equipment off, the only way I can is to disconect the battery as the switch you would use on IC can't be fitted as far as I can tell.

[John Privett]

Tony,  yes, with a brushless motor if one of the three wires breaks then the motor will stop running, and possibly just stutter back and forwards a little as the ESC tries to make it work.  What happens to the ESC in these cases is frankly anybody's guess!  I would suspect that cheaper ones may let the magic smoke out,  possibly more expensive, better-engineered ones might be more likely to survive?

 

The "back EMF" that Martin mentioned is the voltage generated by the motor whilst it is running.  A motor with no load on it will (theoretically) run at a speed at which the back EMF is equal to the voltage applied from the battery and no current will be drawn.  I did say theoretically,  as there is always some load from friction! 

 

Switches - no, generally no switches are used.  If the battery is connected the model must be treated as "live" and the motor could potentially burst into life at any moment.  So only plug the battery in when you're ready to fly,  and disconnect it after landing.

[Tony H]

Thanks John, Still not fully understanding the EMF as I did have the 8X4.5 Folding prop on it when I tried it, so I would have thought there would have been fairly normal load on the motor.

 

The motor did stutter jolt and stop then the smoke came, I guess that was the frankly pathetic motor bell wire snaping.

Edited By Tony Hamer on 03/01/2011 14:05:51

Edited By Tony Hamer on 03/01/2011 14:06:37

[John Privett]

OK Tony,  all Martin was effectively saying is that any electric motor will draw a lot more current when stationary - for whatever reason it's stationary.  That could cause the ESC to burn out. 

 

The back EMF bit is just an explanation of why the current drawn is much greater - the important bit is that it is!

 

[Peter Beeney]

   Tony, It just so happens I had a little try at writing a fairy story very recently in the thread: High voltage low current by Simon Chaddock 02/01/2011. This is my version of how a servo motor might operate.
   Now we need something similar for brushless motors and ESCs.

   EMF does get a mention there, and if there were a a little discussion on the brushless motor subject in general we might get a fair indication of what is actually going on. Without getting too complicated.

   Just an idea.

   PB

                    

[Tony H]

Sounds like a good idea Peter

 

So when you check a motor with a watt meter as a bench test the current drawn is not the same as it would be when flying? So my next question is by how much? as the ESC I was using was around 50% more rated amps than was needed for the motor.

 

Also when you use a smaller prop does that usually make the current draw and watts of power less? as it would suggest it does on some of the motor specs i.e.

 

10X4.5 might be 130W and 12amps

9X4.5 might be 110W and 10amps

8X4.5 might be 90W and 8amps

 

Is this the trend?  The prop I had on was a 8X4.5 folding prop which the motor spec says it should be less than 10X4.5

 

It's a shame my Watt meter hasn't turned up from Hobby King yet over 3 weeks now

[Erfolg]

I now only use cheap ESC's. I have stalled a number of them prior to flying. They bounce back and forward, until the throttle is closed. I have seen others do this also, with much the same result. If allowed to continue above a few seconds I would not like to predict the outcome though.

 

In my limited experience, to date watching others, the magic smoke appears to occur when running a set up outside or close to its rating. It did and does surprise me, the number of people who have told me, that it was all within rating, but have no idea what the set up was actually pulling. Not having a watt meter.

 

I myself had a problem about a year or so ago, when relying on data sheet information. It was only after some issues and finally using the watt meter to find the error of my ways and shame.

[John Privett]

Posted by Tony Hamer on 03/01/2011 14:29:18:

Also when you use a smaller prop does that usually make the current draw and watts of power less? as it would suggest it does on some of the motor specs i.e.

 

10X4.5 might be 130W and 12amps

9X4.5 might be 110W and 10amps

8X4.5 might be 90W and 8amps

 

Is this the trend?  The prop I had on was a 8X4.5 folding prop which the motor spec says it should be less than 10X4.5

 

 

Tony,  yes you've got the idea. 

 

Electric motors have very different characteristics to i/c motors.  The bigger the load you put on them (by way of a larger prop) the harder they work to maintain their natural speed - and to do that they draw more current.  If you put too much load on then something will go "pop".

[Peter Beeney]

   Tony,
        To try and answer completely honestly might take a while! And in my case even longer!
   I think that from what has generally been said before, the electric motor does not greatly unload in flight. So maybe that might not be a big consideration anyway. But there might be more details to come from out there.

   One snag as I see it, we are only ever measuring the input to the motor, in terms of volts and amps. The mechanical output, that’s spinning the prop, is in terms of torque and rpm’s. (But that can still be converted into watts). Ideally, we need to be able to equate the different units. Fitting a smaller prop will allow the motor to spin faster and reduce the amps, that will reduce the input, certainly in watts, it might also result in an increased output at the shaft but also it might not. Rpm’s we can measure ok, the torque is measured by a dynamometer, and as far as I know this is not easily done, at least for model motors.
   Watching what works for other folks, that’s a database, is perhaps what is gradually happening. I have a friend that has flown only small fast electric models for years, he’s got it about right. Now he has very, very few failures, he got there due to trial and error.

   Just to try and make the point, let’s consider two extreme cases. At the moment of start-up, the motor is stationary. The only resistance to the in-rushing current is the coil resistance, which is very low. So the current will be at it’s max, assuming the battery voltage holds up. Maximum power transfer takes place when the resistance of the source, (battery), equals the resistance of the load. So at this point there is a maximum of watts input, but no watts output. Now we load the motor until it stops, and we have the same situation. Max input, no output. Result, discord! Somewhere in between these two the output will equal the output, result, harmony!

   Or will it? Nothing is perfect, there will always be losses across the system. Again, for the moment we can only estimate these, suppose we assume an efficiency of 90%. Then if we have an input of 10 volts and 10 amps, 100 watts, we would have 90 watts in terms of pulling power at the propellor. Anything that is generating heat is probably making the system less efficient in terms of the motor being able to turn the prop, so that needs to be avoided as far as possible as well.
   Regarding your ESC, I would personally think that any ESC worth it’s salt would have some inbuilt safety features. However, we also have to consider that all these things are aimed at aeromodellers, the consumer market; so we don’t if they are on board or not. Even the cheapest of electronic components, i.e. voltage regulators, have inbuilt thermal and current overload protection. But I’m afraid this failure rate is something we may have to live with.

   I’m sure that the broken motor wire would have unbalanced the ESC, at the very least, and in doing so might caused a modicum of discomfort somewhere. As John P said, who knows what goes on! Really, the broken wire is perhaps of some concern, too, if this can happen very easily then it’s yet another area that we need to perhaps be aware of.

   PB  

                 

[Erfolg]

Peter

 

If we were to take a Timbo type test rig and turn the motor propeller unit through 90 degrees, could we not measure the torque?

 

Knowing that P = 2 Pi nt could we not obtain a figure for power out. As we would now know torque and revs.

 

We know power in from the meter. the difference being the efficiency (of the motor and ESC).

 

Of course how this equates to thrust is another matter.

[Peter Beeney]

   Erfolg,
            I’m afraid I’m not familiar with Tim’s rig, but maybe it’s possible to devise some sort of testing torque beam. But you would need something really accurate to calibrate it against. Once upon a time I think they tried to test full size aero engines like this, but the results were so inconsistent they had to abandon it. An eddy current brake dynamometer is today’s ‘must have’, I know nothing about them, and I expect they are very expensive to employ for any testing. There was a small one made long ago, by some aeromodellers, for testing some of the multitude of small i/c engines of yesteryear, Mike Rolls had much nostalgic knowledge in a thread a while back.

   I certainly think the power would be a direct readout now, plus the efficiency by measuring the power in. (Testing the whole system in one go!). Then all we want is a ‘motor speed against different prop sizes’ chart, and we could fit the prop for the best performance on the first flight!
   However, if this happened on a large scale, would it not soon start to become apparent which were the best combinations? And would the ‘trade’ be perfectly happy with that? It may already happen, to some extent, but just idly browsing some electric sites there is definitely a vast variety to choose from.

   The all important criteria still seems to be the quality!

   PB   

                          

[Boots]

I have a selection of cheap escs and some of them do get hot when no motor is connected, some dont, I guess they require the load and feedback from the motor coils to operate correctly.

There is a set of mos fets which connect the motor wire to + and another set which connect the motor wire to -  ( yes 6 sets of mosfets in total ) if by way of a fault both the positive and negative sets switch on together the esc effectivly shorts out the lipo and with out the loading of a correctly connected motor anything is possible , especially if it is a cheapy

[Erfolg]

Peter

 

My suggestion is the typical approach of simple lab works, of our youth, warts and all.

 

The degree of accuracy is one of those areas where we all could fill half a page on why theory did not reflect results there being an error of about 1%. Later on having access to a proper lab, we would all explain the differences along the same lines, now lamenting our accuracy indicated an error of 0.5%. In many ways the absolute accuracy matters little, all we want to confirm that the efficiency is, I do not know, but would guess circa +80%, for motor and ESC.

 

My own interest would be in, is there a real, meaningful difference between, budget equipment , compared to a comparable expensive set ups.

 

If I am truthful, I tend not to care, as long as the set up works to my satisfaction.

 

A few things have surprised, amused and left me perplexed. My first brushless was a Kontronik set up, There was a guy with a Axi and castle creation set up. There have been a few other similar combinations. My Kontroniks could not be said to be any better than the cheap set up, I bought shortly after. The Axi combination, failed in stages, releasing acrid stench and in one case a small fire. The number of excuses as to why it was not a problem, and was to be expected, caused me some consternation. Particularly when a very cheap set up caught fire after arriving, the throttle then being left partial on. This was the unacceptable consequence of poor quality, or so it was said. I am amazed that high cost failure never seems to attract the odium of contempt, that budget equipment achieves. For me it is perplexing

[Peter Beeney]

   Erfolg,
           You might like to experiment with a torque beam, I’m sure it not that difficult to construct, but they can have discrepancies of up to 50%, by all accounts. That’s one of the reasons why they were abandoned. The efficiency of the whole unit, battery, ESC, motor, I’m not sure of, this must be variable, but I think that a max of 80 + , at the normal running level might be in the general order.
   For me, I think the accuracy is more important, at least as a base line. I read a while back a report in a modelling journal about some tests of different motors and prop sizes and the listed results. In one table the revolution figures given for the smallest prop tested were higher than the no-load kv of the motor under test! How does that work? I think perhaps someone just accepted the given figures. If I were publishing the results of such a test, aimed at being the benchmark for other modellers, I would apply an accurate known voltage to the motor and then also accurately read the revolutions resulting; and then published these figures first. You then know the kv precisely.

   In my experience at least, the budget items can often be excellent value. Just a chance you take. In terms of standard operator usage it might be very difficult to tell the difference; instruments might be the only way of separating them.
   And also like you, I just want everything to work without too much attention. Although it’s often quite interesting trying to solve some of the various problems that occasionally arise with the other pilot’s gear. I learnt many years ago that things are not always perhaps what they seem at first glance, particularly in the electrical world.

   Re the Axi, could that have been overloaded? I’ve always found it difficult to understand how people can play with an electric aeroplane without a voltmeter and an ammeter. But they do, and unless they stick implicitly to what someone has told them, and that might not always be entirely reliable, then how would they know what is going on? I use a contact thermometer quite a lot, too, components warming up can also be a good a pointer. Electronic components don’t generally mind what they do, provided they don’t heat up too much. However, if they get to a point where they think someone is extracting too much of the michael then they can get very hot under the collar, ultimately exploding in a shower of sparks. I’ve seen that happen.
   The motor that caught fire? Presumably this was a stalled motor that was turned on? It might be argued that a expensive motor might catch fire even quicker, the copper resistance of the coil might be even lower, and thus the current being higher, and things warming up that much faster. This is all guess work, mixed with some speculation though! But it does sound like finger troubles, rather than a fault on the mechanical bits and bobs.

   I have to admit, I don’t think I’ve seen quite the difference in attitude between high cost and low cost equipment that you remark about. But then it also appears to me that many clubs seem to behave in a completely different way to each other anyway…….

   Good Luck with any torque reaction beam trials you might do!

   PB      

                         

[Tony H]

Thanks for the advice guys. I rang Peter at Airtek and he was excellent, he told me not to send the old one back and they will send me out a new one. Brilliant service, I am sending it today with a not inside to jog his memory. I will keep you posted.

Edited By Tony Hamer on 05/01/2011 09:01:56

[Erfolg]

Peter

 

Absolute accuracy is not something I can get to excited about. To achieve it you need standard test conditions, your measurement equipment needs to have all the right characteristics, in respect to sensitivity, repeatability, accuracy of measurement, all the other things we were all taught in "Instrumentation and Control". Also as taught the experiment needed good design, both in set up and understanding what you were measuring, why, and all the other goof. Of course no lab work was complete without copious mathematics to determine system response.

 

All in the real world of hobbies a waste of time.

 

I cannot isolate with my knowledge and equipment the ESC from the Motor for the purposes of experiment. For me that is how it is. I am only interested in the overall efficiency, to get an idea how efficient the combination approximately is.

 

There could well be better ways using all the data logging hardware and software now marketed by people such as Eagle Tree.

 

From the perspective of failed ESc's, I expect all my controllers to eventually fail. Partly because of how they are used, both as intended and occasionally inadvertently abused.

 

My only real point is expensive stuff fails as does cheap. Often the cheap stuff is either a clone or built of the back of the expensive stuff. As an Engineer I have always climbed on the back of others works. Often the hardware/component supplier, would supply generic designs, known to work. Those original suppliers who shout foul, often overlook where their ideas and IT really came from.

 

No not all budget is good, nor is it necessarily poor.