BEC - Broken Electrical Connection.

[Peter Beeney]

This is the result of a small repair to a faulty ESC, and because I wondered if it might be of interest to anyone, I’ve attempted to make a chronicle of the proceedings.

A friend said that his ESC had no BEC output, which made me think the regulator had failed, but then he said that he thought it was still ok as a speed controller, which then made me think the regulator was probably all right, because the primary task of the BEC is to drive the main controller on the PCB. If the BEC failed completely the ESC would die completely.

Photo one

Photo one a

So, looking at photo one and one a, this is the guilty party, the small block, top right, is identified as the BEC, it’s bottom left connection is the input flight battery voltage input, the bottom right is the 5 volt output; the centre top soldered joint is the common, (negative), connection. We will return to this later. These are all SM (Surface Mount) components, the PCB (Printed Circuit Board) is multi-layered, very little chance indeed of being able to see or do anything much here. Connecting a battery, 3S, and checking the voltages, there are 12 volts at the input pin, and 5 volts at the output pin, and at the micro-processor/commutator driver, which instantly makes sense of the fact that the ESC is still working. However, at the red wire on the extreme right, which is the BEC supply to the receiver, there is zero volts, so it would appear the connection between this and the output pin has gone open circuit. I’ve no way of knowing, but I suspect it’s the PCB track that’s not up to the job, and it just may have been overloaded at some point. But for whatever reason I’d say it’s broken. So the easy answer is to provide a link, a piece of wire soldered in, as per photo two. Bingo, and so within a few seconds it’s connected to a receiver and it’s driving a motor, throttling ok and operating a servo.
I think the 3 units on the left are switches for the MOSFETS, to ensure they switch on and off as quickly as possible, and the single block is a converter for the throttle signal. But this is all a bit of a guess.

Photo two

I did have visions to start with of renewing the BEC, it wouldn’t be that difficult, and not very expensive, but as these are quite hardy and don’t fail very often, I thought this was slightly odd anyway. But as it happens, this little repair is even quicker, and as I’m a bit familiar with these components it does help. I’ve often had a go at repairing various bits and bobs in the past and more often than not it’s just a simple mechanical fault such as this, rather than an electrical fault in a component, where the problems lie.

[Peter Beeney]

PART TWO

As I’ve tried to show what I think are the main components, photo three is the underside, these are the MOSFETS that make up the inverter to change DC to AC to drive the motor. I know this is one of those very controversial topics with many forumites, many disagree with this idea, but nevertheless this is what happens. We might be able to explore this a bit further later. 6 transistors, configured as 3 half H bridges, make up a 3 phase inverter, here we have 12, so I suspect they are formed in parallel pairs. These had a piece of ally plate stuck on top, to act as a heatsink, I lifted this off.

Photo three

Talking of heatsinks, back at the BEC, photo one, you can see it’s well soldered to the main negative battery connection. I think this is significant. The BEC can be prone to overheating, and this is mostly due to the differential between the input and the output voltage, the higher this is, the hotter it tends to gets, and this can inclined to be on something of an exponential scale. The voltage range of the ESC is given as 5.6 - 26 volts, implying a max 6S pack, so at this level the BEC will need all the cooling it can get. I think this is a nice point, clever indeed, the negative battery wire is going to act like the good conductor it is, and is going to to (hopefully!) conduct any excess heat quickly away. Actually, in reality, I’m sure it really does, I think this would have been well tested; on the other hand, however, if it was the PCB track that gave up, it might not be that good.

I’ve now daubed some contact adhesive around the throttle cable and the capacitor to provide a spot of anchorage, refitted the heatsink on the FETS and taped the piece of yellow cover back; this is shown in photo four. Also shown is some solder I use, a small gauge variety as compared to 1mm sqd copper wire. This is nicely appropriate for this small work, I find that the heavy gauge stuff can sometimes get a bit much. I also have some that’s an even smaller gauge, almost as fine as cotton, but I don’t get to using that very much nowadays.

Photo four

So there we have it, back in action for more or less zero cost. Could now be operational for a long time. These things are now fairly cheap, though, and it wouldn’t be the end of the world to buy a buy a new one. Or you could use it with a rx battery. But I’ve always liked to keep things for going a long as possible, that probably stems from my working days long ago, we always had a go at repairing faulty equipment first whenever possible. Maybe someone else might also be inspired to cut the heatshrink off and start digging about inside when the is a bit of a malfunction… not much harm done if it’s not a success, there’s not much to lose anyway.

PB

[Simon Chaddock]

Peter

Well spotted! I am not sure I would have assumed it was as simple as making a 'jumper'.

I have pulled some apart but mainly to replace the sheet heat sink with one that has fins on it that stick out into the air stream.

A heat sink completely covered by thick shrink wrap always seems a bit of a thermodynamic nonsense to me.

Your description of the operation of the ESC would seem to suggest these motors are 3 phase alternating current variable frequency types running at a frequency set by the Tx throttle position.

[John Cole]

From your description, the BEC will be switched mode, and the heat output from these is quite small as they run typically at 75% efficiency.

[Peter Beeney]

Simon, Thanks for your reply. Yes indeed, as it happens, this was very straightforward anyway, and the fact that I was not totally convinced the BEC was really AWOL helped. And this takes all of 2 seconds to check out…
With regard to the heatsink, I suppose anything is better than nothing, but I have seen some of my club mates with ESC’s that come with some nice big ribbed/finned heatsinks; so I’d consider that is an excellent idea; and my good friend that owns this always brings the ESC to the outside of the fuselage, at least on his sports models. Then they get the benefit of any wind-chill factor in the slipstream.

Your phraseology of the ESC operation is not quite how I’d put it, exactly, to make it 100% clear. They are certainly 3 phase motors, but all electric motors are AC internally; although it’s possible perhaps that sometimes a stepper motor might be in something of a ‘grey area‘ here. So to all intents and purposes, because it uses a DC source, it would be termed a DC motor. Also if by frequency you mean ‘speed’, the motor’s rpm, then yes, this is set by the throttle position. But the speed, or frequency, of the inverter, is controlled entirely by the speed, or frequency, of the motor, not the other way round.

However, this is really a little bit like starting at the top of the tree and working down, a brief description of the modus operandi from the beginning might throw at least a little light on the matter. At the end of the day, once the motor has done one revolution, all the rest are the same!

John, You may well be right, but this BEC is marked 78M05, implying that it might be one of the standard LM 78 series of linear regulators. I very much suspect that the 3A rating might also be ‘in house’, as calculated by the ESC’s manufacturer. Simply because they hope it will keep cool enough.

If my colleague is not desperate to get it back I will load it up a bit and measure the resulting temperature. It’ll also be a check on my handiwork, better to fail whilst testing, rather than when in service, as my boss use to say.
Might be an excuse here to get into a bit of linear v switching theory too, perhaps…….

PB

Edited By Peter Beeney on 05/01/2013 14:31:11

[John Cole]

This thread suggests that the red-wire disconnect may have been at manufacture.

I searched this for a totally different reason: the use of the name Pentium.

It does not look like a HobbyWing ESC, but they have switched BECs @ 40A.

[Peter Beeney]

Having now done some more investigation into this MYSTERY ESC, I’m fast coming to the conclusion that there are more appropriate descriptions for this device, but they are rather unprintable; it’s no wonder that anyone trying to start in electrics finds it literally impossible to comprehend… But having said all that, it appears to be a success, thus far!

It is debatable whether it comes with a BEC connected, or not, I found a short YouTube clip which says that it doesn’t, and showed how to install a receiver battery in place of. But, very oddly, this already seems to have been removed. The label implies that it’s capable of handling 26 volts, yet the adverts say max 3S. Whatever you look at, any form of information on this doesn’t seem to tie up with any information anywhere else. But all the adverts seem to say that it does have a 3A BEC…

However, I’ve at least partly put it to the test; I gave it a heat run. Only the BEC so far, that is. All temperatures are in degrees C, monitored with a RS contact thermometer, the ambient temperature being 14 as a starting point. Because of the voltage difference between the input and output, the BEC has to dissipate all the excess heat, and if it gets hot enough, 150 degrees, it shuts down to protect itself. I connected it to a receiver, with three full size digital servos also plugged in; taking it at face value, max 26 volts, on the label, I connected two fully charged 3S in series, giving me in excess of 25 volts. I’ve also removed the heatsink on the FETS, although I’m not connecting it to a motor at this stage. The BEC output reads 5.041 volts, that’s slightly unusual in itself, it’s invariably just under 5 volts, but a quick waggle on the sticks to get the 3 servos moving soon showed 20 degrees. I used an old SkySport tx, to save wear and tear on a decent box. So then a two minute, (timed), stick stirring, all three servo constantly moving, and the temp steadied at around 78 - 80 degrees. Then I got the heat gun out, an infinitely variable temperature type, from cold up to soldering heat, and warmed the ESC to 50 degrees. This time after two minutes the temp steadied at around 107 -110 degrees, quite warm now, but still a long way off that fatal 150 degrees.

Then I tried a full repeat of tests using only 3S, 12 volts, which is how the owner is going to run it, and the temp was lower, this to be expected, the two minute run at 50 degrees resulted in a settled temp of 77 degrees. At no time at all did the BEC show any signs of flagging; and it was certainly working harder than it ever would in the air. So for the moment I think I would be more than happy to use this as it is.

But I shall now try connecting to a motor, at 3S, loading it up to at least 40 amps and noting what sort of temperatures we get from this. If this raises the overall temperature of the ESC to considerably above 50 then we might still have a problem. At around 135 - 140 the BEC could begin to approach the danger area, but as soon as the load comes off a bit the temp starts to fall rapidly.

My colleague is fairly confident that the BEC did originally work on this ESC, but he can’t be 100% sure, so I think we may well consider buying another new one, then we can really get a handle on the situation; I did see one ad on eBay for £9, so they’re probably not that expensive. It does point to one thing, though, the importance of making sure that the ESC can get as much cooling as possible, not always an easy task, but perhaps worth making some effort to help this. It also seems to prove that in the aeromodelling world there are still many enigmas, and sometimes you’re only really sure of what you’ve got when you prove it!

Final test results to follow shortly.

PB

[Tim Mackey]

Nice work Peter, and thanks for sharing with us.