Jump to content

Smoking....


Recommended Posts

Hi All,

I am a retired electrical/electronic guy and have recently been converting some of my IC models to electric. I am pretty new to the electric flight routine so a bit of a novice. As an experiment and whilst we are in lock-down I decided to static run two recently converted models to see how long the batteries lasted. I was using a ground restraint in the garden along with a tachometer and an electronic stopwatch. The 1st model lasted 6 minutes at full throttle before I closed the throttle to check the battery voltage which showed 10%. Everything was quite cool and I was more than happy. The 2nd model with the same setup lasted only 2.5 minutes before the motor suddenly wound down to zero with smoke bellowing out of the cowling. I quickly removed the battery prior to investigation. Motor was too hot to touch. Upon removing the cowling I expected to see a charred mass. But no, everything looked perfect as did the windings in the motor. Not a blemish anywhere. I also inspected the ESC and that was the same. I then reconnected the battery (which showed 50%) and tentatively opened the throttle. All was not well. The motor started okay, but when passing through about 1000 rpm started stutter and shudder so I abandoned the exercise.

So, I have a duff motor, ESC, or both. I haven't got a spare of either at the moment so I don't know which it is.

Motor: Turnigy Aerodrive SK3 3536 - 1050 kv, 505 Watts, Max 34 A

ESC: Aerostar 50A with BEC

Battery: 3S 3.0A

Propeller Master Airscrew (Elec) 10x7

Figures recorded from Wattmeter: 10.8V, 25,5A, 268W, 9300RPM

I'm thinking that it might be over-propped.

Your thoughts on the burn-out please gents (and ladies) as the parameters above look okay to me (as an electric flight novice).

Roger

Link to comment
Share on other sites

The first thing I would do is forget running the model statically on the ground at full throttle for extended periods.

The normal cooling airflow that would pass over the motor will be absent, plus minute after minute of wide open throttle is not representative of a normal flight. It is stressing the entire power system for no good reason.

Far better to use a watt meter to obtain the power, current and voltage reading for a short (say 10 or 15 seconds) ground run to prove the motor, ESC and battery are all within their ratings and then fly the model for a short flight and see how many milliamp hours the charger puts back into the battery when recharging. You can then calculate a safe flight time from there..

Link to comment
Share on other sites

I disagree with Alan on most points (sorry Alan!)

Full power, static, for the duration, is a reasonable worst case to try the power system with.

With 268W on 25.5A both your motor and ESC should be perfectly ok.

At full power the prop will drive air past the model faster than the model will fly - cooling airflow is present. Or at least should be. The question here for me is whether your installation effectively directs the air flow right past the motor and ESC and out the back of the model afterward.

As an aside, I have two similar setups.

The first is running a 3536 900kv on 4S setup, with a 10x6 prop, it shows about 450W on the wattmeter. A full power run on the ground, down to lipo empty, left it warm but not baking. This is completely cowled in and quite a "neat" install. The airflow past the motor and out the back of the model works and is sufficient for this "worst case".

The other setup is a Mega inrunner, on 10x5 and 4S, again close to 450W. Both motor and ESC have plenty of airflow and are happy, static, on the ground, at full power for about 3 minutes until the lipo is empty. In this case there however, is no airflow through the model - the bits are stuck out in the breeze right behind the prop.

I have found a full power run lasts about half as long as a normal flight will. YMMV.

Link to comment
Share on other sites

Running any motor / engine for a continuous period at maximum rpm is not a good idea as that is not what they are (normally) designed for.

Flight time on a given set up is dependent on many things, including the weather !

Alan is right in that the best way to get information about your set up is to use a watt meter and do the sums (I don't particularly enjoy doing it either) and factor in a safety margin for when flying.

I flew my Mini Robot during Xmas week. usual flight time about 8 minutes. It flew, but struggled for power for about 4 minutes. The only difference ? About 15 degrees C !

Link to comment
Share on other sites

Posted by Roger Dyke on 05/05/2020 10:56:16:...................The 2nd model with the same setup lasted only 2.5 minutes before the motor suddenly wound down to zero with smoke bellowing out of the cowling.............. Motor was too hot to touch. ............... All was not well. The motor started okay, but when passing through about 1000 rpm started stutter and shudder so I abandoned the exercise.

So, I have a duff motor, ESC, or both. I haven't got a spare of either at the moment so I don't know which it is.

.........................

Figures recorded from Wattmeter: 10.8V, 25,5A, 268W, 9300RPM

I'm thinking that it might be over-propped. ...........

Roger

A few points spring to mind :-

You say you used two identical setups, so which setup did your Wattmeter figures come from – the good one or the bad one!

If you have two identical setups you can swap components to see if it is the motor or the ESC that is a problem.

You are thinking it might be over-propped, and it is possible that it is over-propped on the ground but would be fine in the air. With a 10x7 prop, current will almost certainly be lower in the air – one of the reasons I am wary of long full throttle ground runs.

Last but not least, magnets can lose their magnetism at high temperatures, so if the motor was “too hot to touch” it could mean the magnets are now degraded and the current will be a lot higher which could partly be the cause of the stutter problem.

Dick

Link to comment
Share on other sites

With the current and power well within spec. I would not have thought that it is over propped. The quoted 505W for these must be on 4s, not three and you are running at just over half this so OK there.

It could possibly be incorrect timing but I would agree with the inadequate cooling theory. It sounds as though the motor is damaged internally, i.e. a winding burnt through and shorting out which would explain the erratic attempt to start up again. Try getting a new one from HK as they rarely quibble about such things. A while ago I had a rather expensive 120A esc go up in smoke on its first test at only 20A and they sent a new one straight away.

If it is possible to hook up your esc to the other model (assuming that the parts are compatible) then you could try that. Some of their cheaper esc`s are, well, none too good.

I have one of the Aerodrive motors which came with an ARTF and it is lousy, but my Gliderdrive one is brilliant.

I have found the Propdrive series to be OK and use quite a few.

Hope this helps.

Link to comment
Share on other sites

The quoted max power of 505W isn't really the critical figure, it's the max current of 34a which is the defining figure. The specs don't say whether that is a burst rating (usually something like 15 seconds maximum) or constant. Either way I'd treat it as optimistic like most of the manufacturers ratings. Saying that I'm pretty sure I've propped my 3536 motors to run at over 300W as standard on 3S.

Second point, how well cooled is the motor? Is it exposed, cowled with good cooling or fully enclosed, this makes a lot of difference. Remember that the motor is at the centre of the prop arc where there's not so much air being moved!

I've not had any problems with the SK3 motors that I've used but there's always the chance that you could get a dud. It sounds as though the insulating lacquer on the winds has cooked somewhere if it was generating smoke, hot and there's no charring on any wiring or connector insulation.  Oscillating usually means one wind or one connector is dead, an overheated motor would point to it being a winding.

I suppose you have looked closely at the heat shrink on the esc? 

Edited By Bob Cotsford on 05/05/2020 13:57:00

Link to comment
Share on other sites

JD8: Thanks for the reply. All the connections have been checked and have perfect contact.

Alan: Thanks for your reply. I have heard before about static runs being inadvisable. But as an electrical guy, I would have thought that if the specification of the motor is 505 Watts with a maximum current rating of 34 Amps, then it ought to run comfortably with a current draw of only 25.5 Amps, and a power dissipation of 268 Watts. Or am I missing something here? I do agree that flight time will be different to static testing time though as there is lots of other factors involved. I did the static test as I considered it to probably be 'worst case' regarding battery time.

Nigel: Thanks for your reply. Thank you for your examples which were refreshing to hear. I think that the problem may be the cowl. I had to make a new cowl when converting from IC. There are two large vent holes in the bottom of it, and two vertical slots left and right of the propeller in the front. I am thinking that this is not enough. Although, what I mentioned about operating within the motor's parameters as explained above still holds true. Perhaps the motor was faulty and just give it's neck after 2.5 minutes flat out. Or maybe not?

Roger

Link to comment
Share on other sites

Kevin: Thank you for the useful information. I am very puzzled though as to why someone would design a motor that couldn't be ran to it's full specified potential.

Dick: The Wattmeter figures came from the bad one. The one that died. I cannot swap them over easily as they are sort of built into the plane. Also they are of different ratings. Today I have sent for a new ESC to try and decide whether it's a motor or ESC problem (or both). The thinking behind the over-propping was that with less pitch (or diameter, the motor would be using less current hence dissipating less power. That is all. I still believe that the motor should operate correctly within it's specified parameters continuously. Thanks for the tip about the magnets getting hot. I'd not thought of that one.

Martin: I agree. There doesn't look like there is anything wrong with the parameters I was running and would have thought that the motor was quite comfortable with them. As in my last reply to Nigel, I'm a little concerned about the amount of ventilation in the new cowl I've manufactured. Even so, 2.5 minutes from cold is not a long time to run when a motor well inside it's spec. Each side of the propeller, the cowl has a 1/4" wide slot each side almost top to bottom and two large 'D' shaped holes at the back underneath to let the through-flow of the air out. I'm thinking it probably needs more ventilation. Cannot get a new motor from HK at the moment as they currently have none at the European warehouse. Also the items are not interchangeable with my other aircraft. I have a new ESC on order. Thanks for your tips.

Bob: The motor was only running at 2/3 of the maximum stated current so was well in hand. Also it was only dissipating 268 Watts which is almost half of it's stated 505 Watts. Even with a total lack of ventilation I would have expected it to do better than 2.5 minutes from cold. I am looking at the cowl though to provide better ventilation. Absolutely no sign of burning or charring whatsoever either on the motor windings or the outer housing. Absolutely no sign of overheating on the ESC either. Wiring is also good. As the motor now judders when passing through about 1000 rpm I feel it has to be something to do with the internal windings of the motor.

Roger

Link to comment
Share on other sites

Roger if your cooling is poor to the motor as you admit yourself then perhaps you should consider the maximum rated current is given assuming the motor will be able to have it's excess heat dissipated by cooling airflow. If the cooling IS poor then you need to derate and perhaps you haven't derated enough, therefore the motor has overheated and melted the insulating lacquer on one or more winding.

My original point about not ground running continuously was based around cooling and I do still disagree with Nigel that you will get a similar cooling airflow on the ground as you would in the air. However, you are hammering the battery pack by running at full power.

If the model needs full power to stay in the air then I would suggest it's power system is undersized and should be adjusted.

Even Nigel admits that continuous full power running doesn't actually give you a realistic figure for the flight time you might achieve, so I'll stand by my comments about performing a calculation to give this.

 

Edited By Alan Gorham_ on 05/05/2020 17:44:39

Edited By Alan Gorham_ on 05/05/2020 17:45:21

Edited By Alan Gorham_ on 05/05/2020 17:45:51

Link to comment
Share on other sites

Roger

If significant smoke issued from the cowl then all is most definitely not well.

Smoke from an ESC is normally terminal. Smoke from a motor rather depends. If it was just the windings over heating and cooking the polyurethane insulation you may still be all right but if the cooking was serious enough to actually create a short between adjacent wires then the motor is finished.

The obvious test is to replace the ESC with one that is known to behave normally and check the motor runs up to speed - with the prop off! If all is ok then it is worth buying a replacement ESC. If it still judders then buy another motor.

Remember a motor whirls round so does a reasonable job of spreading the heat about from the hotter parts. The ESC generates significant amounts of heat and from small components encased in plastic and in many cases the whole thing is then surrounded by substantial shrink wrap. The ESC needs a positive cooling air flow.

As ESCs are sensitive to over loads the 'rule of thumb' even with a cooling air flow is to ensure full power amps are at least 25% below manufacturers limits just in case the claims are more to do with 'marketing' than factual.

A continuous full power run to 10% battery capacity in 6 minutes is not exactly kind to a LiPo either. The claimed performance figures of LiPos can also suffer from 'marketing'. wink 2

Link to comment
Share on other sites

Hi Alan,

Having spent a lifetime in the electronics industry as an engineer I am very aware about components and machinery being 'fit for purpose'. I agree that my cowl leaves quite a lot to be desired regarding cooling. But I have never seen from any model motor manufacturer any sort of motor specification which specifies that the motor is intended for either intermittent use or requires a certain amount of ventilation for it to operate to it's full specification. Take an electronic component like a power regulator or a triac switching device for instance. The manufacturers of those items specify that to use the devices at certain power levels, a heatsink is required of certain dimensions to disipate any heat build up during their use. That's fine as it's specified. All I am doing, is using the specification offered by the manufacturer and was just puzzled as to why they needed caviats to be usable. It's probably my background creeping in but can't understand why it's bad for a motor to be ran constantly at full speed. I've worked with motors all my life and had them working flat our all day for months. Why are these any different? Is there something special about them I don't know? Anyway, sorry about going on, but I'm just a bit confused.

I agree totally, that the run time in the air is totally different to that on the ground. It is is just that I see a run on the ground as probably a good benchmark as to the shortest battery time the model will have. These aren't new electric models. They are converts from IC so are generally heavier that electric models and are always a problem as to what size battery can be accommodated (amount of room). So there is a problem between the amount of electric power available and the weight of the model. It borders on the 'not worth the trouble'. Thanks anyway for your input.

Roger

Link to comment
Share on other sites

Posted by Peter Miller on 05/05/2020 18:29:29:

I claim to be a complete ignoramus when it cmmes to electric power but one thing did occur to me reading all this.

When a model is in flight the load must be reduced. I know that ic engines do unload in the air.

Could that have any effect?

yes

These can smoke badly to with high heat and are approx £4

27202.jpg

Edited By Denis Watkins on 05/05/2020 18:54:53

Link to comment
Share on other sites

Hi Simon,

Just what I thought, and there was lots of it. So much in fact that I thought the front was on fire. That's what prompted me to quickly remove the battery. The amazing thing was, that when I removed the cowl there was no sign whatsoever of anything that had been getting hot. No discolouration, no stains, nothing. With a small torch I looked in at the windings. They all looked golden and perfect. I then looked at the ESC. That also looked perfect and showed no sign at of stress, discoloured plastic, or anything. The motor was too hot to touch and the ESC was cool. After cooling down, I reconnected the battery and gingerly advance the throttle. The motor started away like normal. But as I advanced the throttle through about 1000 rpm the motor started to judder and kick. I tried this several times but it was the same. I would guess that the motor is stuffed. I have a new ESC coming in the next couple of days (I hope) and I will try that. but I think that I'm in for a new motor. I'm still puzzled though as to the cause. The ESC was running at about 50% of it's capability.

The other aircraft that did 6 minutes was discharging at about 8C from a 30C battery which I didn't think was too bad at all. Also the final terminal voltage was 11.02 volts or 3.67 volts per cell which I thought was also within decent parameters.

Roger

Link to comment
Share on other sites

Can I add a few points. Firstly running a test like you did where the lipo gets down to 10% is a very bad idea. You will certainly shorten the life of the lipo discharging to that level. Down to 20% at worst case, 30% better.

Secondly, model motors are not designed to be run flat out all the time. That may be your experience with industrial motors which are, but then I suspect they are physically much larger and heavier for the same power. You may need full power for take-offs and for some manoeuvres, but all of mine are happy taking off on half to 2/3 power. If you need full power to take off you need a larger motor/battery setup.

Thirdly its a very bad thing to run an electric model flat out all the time, unlike IC. Up to about 2/3 full power you will find that there is a linear increase in thrust and current. However, in the last third of the throttle, the current rises dramatically with surprisingly little increase in thrust. You will find you can double the flight time if you run the motor at no more than 2/3 power, and you will probably never notice any speed difference. With IC you hear the motor noise and assume its actually going a lot faster at full throttle. Before you ask, yes, I have checked this out in flight with full telemetry including an air speed indicator.

Don't know the weight of your models, but if they are around the 2kg mark on a 46 IC setup, I'd say a Turnigy Aerodrive SK3 3536 is rather underpowered, especially on 3S. I generally use a SK3 3548 with a 4S setup giving 750 watts.

Lastly, its not just cooling slots in the cowl, but you need to consider airflow over the motor itself, and on an electric model this invariably means having decent holes through the motor bulkhead behind the motor and somewhere for the air to exit further down the fuselage.

Link to comment
Share on other sites

Posted by Roger Dyke on 05/05/2020 18:51:42:

Peter: I can only think that the load would be less as it cuts it's way through the air, whereby taking less current from the battery.

Is that right?

Doesn't make all that much difference, and again I have tested this.

Link to comment
Share on other sites

Dennis: Bearings yes. I suppose the grease in them would smoke a bit once they go very hot.

Andy: I take on board most of your points. But where will I find that model motors are not designed to be ran flat out all the time (regardless of desire). Since I have come back into this hobby I have found that there is a lot of folklore. Some of it true some of it not. And also lots of whispers that get passed from person to person that is complete rubbish. I am not saying for a moment that what you are telling me is not true. I have just not seen anything written down by any of the motor manufacturers. If you could point me there it would be appreciated.

Regarding the model and the motor size, it is a bit of a 50-50 project and I agree that the motor needs to be bigger. But as it's a very narrow fuselage I am very limited as to the battery size I can use, hence the motor size against discharge time in the air. The battery and motor are about as big as I can go. Otherwise it's to convert it all back to IC again and I don't really want to do that.

I admit that I have got to do something with the cowl. I don't know what as we speak, but I'll think of something.

Roger

Link to comment
Share on other sites

Posted by Roger Dyke on 05/05/2020 19:53:35:

................................

Andy: I take on board most of your points. But where will I find that model motors are not designed to be ran flat out all the time (regardless of desire). .................... I have just not seen anything written down by any of the motor manufacturers. If you could point me there it would be appreciated.......................

Roger

Not exactly the same thing, but at least one motor manufacturer points out that the power handling capability is related to the rpm you run at. So a blanket watt rating is more of a marketing "guideline" from most manufacturers.

I believe Kontronik to be one of the more honest manufacturers - but perhaps at their prices they are looking at a different market smiley.

Dick

Link to comment
Share on other sites

Could you post a pic. of the cowl? It would help a lot in trying to decide if that was the culprit.

You say that the fus. is very narrow so an SK3 Gliderdrive may be your answer. They are made for very confined spaces and very powerful. Mine is fine on a 3s 2200 but I now use a Multistar 4s 1400 although the run time on glider climbs is limited. There is no cooling air whatsoever through the very slim model, in fact many of my electric ones have no cooling to the esc or battery because the model types will not allow it and I fly them enthusiastically. I am talking about a Vulcan, Lancaster, Mosquito etc. The packs and esc`s get warm but not alarmingly so and I always use a way over rated esc.

Go back to i/c, it is very much simpler!

Link to comment
Share on other sites

Dick: Thanks for the info and the chart. I don't quite understand what the chart is trying to say but that's me probably being a bit thick. I understand that the RPM is controlled and maintained by the ESC which it tries to maintain regardless of load. Obviously, the larger the load, the larger the current.

Martin: The cowl is very simple and is square fronted 2.5" wide and 2.25" high. In the middle is a 7/8" hole for the propeller and a vertical slot 1.5" high and 0.25" wide each side of it. Underneath it has two elongated 'D' shapes, each 1" long and 0.5" wide. The cowl is 3" long. I don't think that glider type motors etc., would do it. The main problem is the weight as it is a converted IC plane. It weighs in at 5.0 lb. I have very limited space for the battery as it has to fit basically in the old (modified) fuel tank bay. All space is very limited. You are so right about IC being much simpler. I am on a knife edge whether to convert it back or not.

Roger

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...