To UBEC or Not to UBEC? That is the question.

[leccyflyer]

The thread discussing the post-mortem on the dead Seagull PC-9 has been closed, with the recommendation that any further discussion on the subject of power looms, UBECS, connectors and data loggers should be in a new thread.

 

I came to that thread rather late, but in reading through it found a number of items that were clearly of a wider interest to electric flyers. Those include the fragility of the power and radio control systems being dependent upon the linkages and connectors that we fit to the various components in our models.

 

My main question is - is the use of a UBEC a case of putting all our eggs in one basket, in a safety-critical part of the system?

 

I admit that reading the original posts in the threads relating to the big dead Seagull I was drawn to conclude that it makes a great case (IMO) for a completely separate power supply for one's radio system, especially in larger models, where the additional weight isn't detrimental. In the event of any sort of serious mishap in the power train, one does not then run the risk of losing radio control as well.

 

I think it's settled the matter for me in that I won't now be fitting a UBEC to my favourite model, since the potential disadvantages would seem to outweigh the advantages..

 

The other thing that I was enquiring about in that thread, regarding the wiring up of our power looms, connectors etc concered the additional complexity of in-flight monitoring equipment.

 

The introduction of extra gubbins into the power circuitry, whilst it produces some nice, useful data, is something that IMO needs to be looked at very carefully, lest that compromise the integrity of the overall system itself. As another poster pointed out, one is left with the little niggle as to whether the fitting of the data logger contributed to the situation at all.

 

I also asked the question whether the leads for the data logger (and UBEC) were added to the connectors in a second phase of soldering, which could conceivably have reheated, and weakened the original solder joint. I asked that for the sake of clarification and it appears that the datalogger was merely added in series in the circuit and there was no second phase of soldering to the 4mm bullet connectors used for the power circuit.

 

I stopped using my trusty Wattsmater years ago, in favour of a clamp meter, precisely to avoid the considerable extension of the power loom that results from using adaptors to fit that equipment in series, between battery and ESC. I could have lopped off the fitted ASTRO zero loss connectors. which would have shortened the extra cable length a bit, but there is still a fair length of extra cable to introduce into the circuit.

The other, similar, though opposite question is - in the interest of accruing data, are we in danger of over-egging the pudding and introducing extra points of failure in our models?

 

Both questions are egg-related, funnily enough. Hamlet and eggs - what more could you wish for?

 

Happy Easter.

 

Thoughts?

Edited By leccyflyer on 25/03/2011 10:23:09

[Craig Carr]

Posted by leccyflyer on 25/03/2011 10:18:54:

My main question is - is the use of a UBEC a case of putting all our eggs in one basket, in a safety-critical part of the system?

 

I admit that reading the original posts in the threads relating to the big dead Seagull I was drawn to conclude that it makes a great case (IMO) for a completely separate power supply for one's radio system, especially in larger models, where the additional weight isn't detrimental. In the event of any sort of serious mishap in the power train, one does not then run the risk of losing radio control as well.

 

Im sure others may have different points of view LF but I for one always fit a seperate RX battery on big electric projects. Just sort of makes sense to me personally not to have another electrical component in the works that can go wrong.

 

With a seperate rx battery you do have a bit more of a chance to bring the model back in if the motor battery goes

 

If the model can accomodate a bit more weight I wouldnt think twice about it.

 

Edited By Craig Carr on 25/03/2011 10:30:02

[leccyflyer]

Those are pretty much my thoughts on the matter.

 

Reading that particular post-mortem thread has only served to reinforce them, since I had a loss of power in the air on Sunday and was able to recover the model to a straightforward deadstick landing in the middle of the field with no more damage than one "How can you have a deadstick with an electric?" in my ear having called it.

 

I don;t think that loss of power would have led to loss of battery power to a UBEC (or an internal BEC) as it happens, and I do use BEC on several of my smaller models and am happy to do so. The equation there is a bit different though, where the extra weight of a receiver battery comes into the equation and where onboard integrated BEC on an ESC on 3s1p power makes sense to me.

 

The particular model in question needs a receiver battery to balance, with the 6s1p Lipos mounted on the CofG, enabling different packs to be used. The main disadvantage is that of keeping the RX battery charged, which is just the same as for an IC model. I usually fast charge the RC pack at the field, on arrival.

 

I suppose another option would be to fit a Lipo RX battery with a regulator, or the UBEC, but I'm not mad keen on keeping a Lipo permanently in the model and I don't charge my lipos at home anyway.

[Danny Fenton]

On the face of it this seems an obvious conclusion, however you are now introducing a second battery, the looms associated with it, the switch harness, the need to charge the battery (correctly). and you could easily suffer black wire corrosion which would take out your Rx. I use a good high powered UBEC and feed it fromthe ESC connectors as Tim does and have not had an issue.

I also only use 4mm bullet connectors but would probably go up a size if the current was going to average more than 50A. Bearing in mind my 14lb Hurricane, capable of more than 2Kw averages only 30A.

 

Cheers

Danny

[Tim Mackey]

Almost the same title HERE lads, however I may be rethinking my original preferences

[leccyflyer]

Hi Danny

 

Agreed on the connectors. With 6mm connectors readily available these days I'd probably do the same. As you know I've seen your lovely Hurricane in action and it's prototypical flight pattern on just 30amps shows the advantages of a high voltage system (over one that is frying eggs on the motor casing).

 

As regards the introduction of a second system,

 

To my mind it's really just exactly the same as in every IC model that I've ever flown. With a very large one of those - which are not in my sphere of interest - then redundancy of RC systems, especially batteries, is a common feature.

 

Yes you do introduce extra equipment, but you also offer redundancy in that, in terms of a back-up.

 

That would be the ideal for very large models, but in lieu of that a well maintained conventional power source for the radio gear would seem to avoid the loss of all radio control, in the event of a failure in the rive-power chain.

 

cheers

 

Brian

[leccyflyer]

Posted by Tim Mackey - Administrator on 25/03/2011 11:51:20:
Almost the same title HERE lads, however I may be rethinking my original preferences

I only noticed that thread was there after I had created this thread, or I'd have given this one a different title.

 

I don't actually enter the site through the individual forums and hadn't seen that when I posted this thread. I use the View Latest Posts options and refresh that page.

[Steve W-O]

I don't think the answer is so clear cut.

 

Each way has pros and cons.

 

I usually take more than one model flying, but several of them use the same size lipo, so one bag of batteries to charge and change.

 

 

I think if I was convinced to use a seperate battery, I would set the BEC to 6V, and have a 6V battery wired in parallel with blocking diodes, with two diodes on the battery. That way I could use a small AAA receiver battery, and fit it to the model I was flying at the time.

 

That way the battery would not normally be supplying any power, and would solely be a back up.

 

In the lst year, I have damaged two planes, both small depron plane, one I ran out of talent rolling 6' off the ground, the other suddenly lost power at 30' just after take off, as I took off with full flaps and down elevator, it wasn't really flying in the true sense of the word, so it just fell to the ground. Had I have done a stick reset of the ESC, it would have regained power, so both times it was pilot error.

 

I have yet to be convinced that going to any extra trouble like having a seperate RX battery with its associated problems, is going to give me any real advantage.

 

The electronics, when used correctly are very reliable.

[Tom Wright 2]

Hi all.

This thread seems to to miss the main point when it comes to the reason for fitting a ubec .the usual bec built into the esc is prone to destruction when the esc is subject to over heating ,if a ubec is fitted the esc can be destroyed and the the ubec will continue to function if the supply is taken from a well considered point.

 

TW2.

 

[Steve W-O]

Posted by tom wright 2 on 25/03/2011 13:09:45:

Hi all.

This thread seems to to miss the main point when it comes to the reason for fitting a ubec .the usual bec built into the esc is prone to destruction when the esc is subject to over heating ,if a ubec is fitted the esc can be destroyed and the the ubec will continue to function if the supply is taken from a well considered point.

 

TW2.

 

 

 

If the ESC overheats, the drive to the motor mosfets is cut/reduced, not to the BEC.

 

Of course if it smokes, it may stop the BEC from working, if it shorts out the battery, but then thay may stop anything connected to that battery from working.

 

Any ESC over 30A, I use a seperate BEC anyway, but it is still powered by the same battery.

[Tom Wright 2]

Yes Steve a good esc should go into thermal shut down and leave the bec ok,but i have often seen other peoples models downed due to the cheaper esc types smoking.i have at least 40 ep models most with ubecs never had one fail.

 

TW2.

[Bob Cotsford]

my take on this is that if the model is too big/complex to run off a BEC, then I'd use a dedicated rx & servo battery.

I'm now having second thoughts about my Cularis with it's 6 (admittedly micro) servos running off a BEC. No problems to date, but IF there were a problem it could go an awful long way, bearing in mind that we're talking catastrophic failures here so the failsafe wouldn't work either.

[Tom Wright 2]

The other point not mentioned is that the most popular ESCs only have a 2-3amp bec,were as you can buy UBECS very cheaply with ratings of 7 amps ,this redundancy keeps the electronics running nice and cool , and more reliable.

 

TW2.

[leccyflyer]

Posted by Bob Cotsford on 25/03/2011 13:49:39:

my take on this is that if the model is too big/complex to run off a BEC, then I'd use a dedicated rx & servo battery.

I'm now having second thoughts about my Cularis with it's 6 (admittedly micro) servos running off a BEC. No problems to date, but IF there were a problem it could go an awful long way, bearing in mind that we're talking catastrophic failures here so the failsafe wouldn't work either.

Bob

 

When we did some tests a number of years ago some microservos drew more current than standard servos. There were some real amp hogs in there with a couple of microservos drawing upwards of 1 amp.

[Peter Beeney]

I’m following this thread with some degree of interest, as I think this is subject that’s been in the news before; and as I’ve been Thinking Safety for about the last thirty five years there is only one option for me. For anything other than a foam model or a smallish powered glider I would just fit a separate receiver pack. For me, the reasons are sound enough.

I have done the diode/backup battery trick a couple of times in the past, but this was i/c, when I was experimenting with some batteries, and it was backing up a receiver pack, rather than a regulator. I would consider that, whilst it is an alternative way of providing some safeguard, it’s still not entirely without it’s drawbacks. I’ve also done a lot of bench testing with battery and regulator (BEC) combinations. Here you can load the lash-up much higher and for far longer than you can in the air, all night for instance, and I’ve never seen the slightest problem. I’m a member of ORWISC (Our Red Wire Is Still Connected), but you have to consider that’s only slightly tongue-in-cheek. It’s more or less irrelevant in any case. Until you get a fault, that is!

Funnily enough, I seen a really good practical example of this, when a model was totally stuffed because the pilot changed his procedure slightly on one occasion, and on inspection of the remains it became apparent as to just how severe the conditions can get and still the electronics continue to function. As it so happened, in this little episode he’d got it connected correctly! Or, at least, conventionally! But, sadly, fatally! I’ve related this particular story before.

I’m a bit of a belt-and-braces-and-a-bit-of string man, meself! I hate watching models crash.

PB

Edited By Peter Beeney on 25/03/2011 14:15:46

Edited By Peter Beeney on 25/03/2011 14:19:00

[Tom Wright 2]

Here here! Peter.

 

TW2.

[Tony Smith 7]

I have this feeling that ESCs are more likely to fail than other bits of airborne electronics. Is there any special reason that this would be the case? Its really that feeling that made me fit a separate UBEC to my Cub, rather than any calculation about load from the servos etc.

[Tim Mackey]

There is absolutely no way to guarantee that an overheating ESC will not affect the BEC circuitry - they almost always share the same PCB, and probably several tracks and components. I remember well flying my mates Ripmax Mustang on a slope session, and upon just trying the motor to check it was all working in case it was needed for the landing etc, and found no increase in speed. I questioned this with him, to be told that he had left the large rubber band on the cowl to stop the folding blades flapping around !

A superbly realistic plume of white smoke started bellowing out from the cowl, and as i was already turning back to the slope face wit full control, I thought we might get away with it. Unfortunately, controls were lost on the final approach, and the model "arrived" just short of the stone wall. Upon examination, the ESC was burnt to a crisp.

[Tom Wright 2]

Here here Tim.

 

TW2.

[Tom Wright 2]

At the field today i watched a guy landing an MX2 that promptly wiped the U/C off "not hard to do on that model" it was sat on the runway with the prop stalled ,while the pilot was making excuses , i noticed the tx throttle was still at near half way ,and promptly closed it for him ,to late that tell tale smell was already drifting across,and on investigation guess what no bec output ! I have seen this sort of thing happen so many times yet pilots just blame the esc,until i sit them down to explain that stalled motors draw horrendous currents that will often destroy the esc and the bec.

 

TW2.

Edited By tom wright 2 on 25/03/2011 23:39:49

[Peter Beeney]

Actually it’s quite difficult to electrically harm or destroy a regulator (BEC), you have to physically kill them, as in a fire perhaps, or with a sharp clout from a small club hammer. They are pretty resilient creatures. You might do it if you maybe applied a high enough reversed polarity voltage, but I’ve never tried that, because it would be just simply wanton vandalism. Even the very cheapest of the cheap appear to have efficient current overload and thermal protection; this works very well in protecting the device, but maybe not so good at the expense of your model. This is one explanation for the situation where the model suddenly has no control but when you get to the wreckage it all appears to mysteriously work ok again.


I’ve always thought that any out of control crashes are potentially dangerous, invariably they happen without any mishaps, but there’s no guarantee on that. To just repeat what has been pointed out before, if the radio current is taken from the main battery supply and that supply is lost, for whatever reason, then an out of control crash is inevitable! Although a separate supply is not an instant panacea, it does give you something of a second chance. I’ve lost the main supply twice in the past due to an open circuit battery but in each case no harm was done at all, I was able to land safely.

I’m of the opinion that a rx battery is quite simple, it has to be if I can do it, and there is a choice. A Ni-MH, or perhaps a couple of small A123’s; or how about a two cell Lipo with a 2A regulator, that will drive anything, too. If you add one of the pack monitors that scrolls constantly through a little menu, giving you the total voltage and then the individual cells, what could be better? I’ve have thought the cost, in terms of what might be a fairly expensive piece of kit, is low. If, on the other hand, subsequently that piece of kit is lost completely, all for the want of a separate rx pack………..

As with many other things, I guess this will never be a issue with most people until there is an actual occurrence of some absent amps; and for many this might never be; however, at our strip, and we are only small in number, there is a reasonable interest in electric, as there is everywhere; and it sometimes seems to me though, that the proportion of problems are a bit high, in the overall scheme of things as it often appears to be an electric model that’s suffered an unexplained tangle.

In some cases, I don’t think we’ve completely cracked it yet!

PB

[Bob Cotsford]

the BEC may survive, but if the ESC gets hot enough to melt the solder on the battery leads a live BEC won't do you much good. I've had three ESCs go (2 in WOT4 Foam-Es) and two of those melted the battery lead solder.

 

All three were on the ground at the time, either testing a setup or just after landing.  That last one was just plain weird.

Edited By Bob Cotsford on 26/03/2011 16:51:06

[Tony Smith 7]

That's exactly what happened to mine, something went wrong within the ESC so it started "frying" and the heat disconnected the battery lead.

[Peter Beeney]

Bob,

I think you are rather convincingly substantiating what I said in my first post in this thread, (25/03/2011 14:12:46). If the rx power is supplied by the main battery and that power fails, for any reason, then all is lost. A separate power supply, e.g. from a small battery, might at least give you some possible control. I do appreciate this is not always practicable in some models, perhaps many of the small ones particularly; and in a small lightweight model perhaps hardly necessary at all. Do you think it would be possible, or necessary, in the WOT4 Foam-E? Bearing in mind you might have lost it twice?

It would certainly seem as though there are plenty of anomalies lurking in the wings. It would be interesting to investigate the reasons why there are so many failures. If the current is sufficient to generate this amount of heat then I’d have thought the fault must be a very low impedance short circuit. A direct connection positive to negative. Serious problems on the ESC. I know it can happen, the most spectacular failure I observed was a small, very fast foam model, 4 cell, high revving in-runner motor with a very small prop on; and I’m sure the system was grossly overloaded. Suddenly in full flight the ESC just exploded in a red hot mass and just exited through the side of the model. The model performed a graceful downwards arc trailing sparks and smoke! In the perverse ways of aeromodellers, there was much laughter and merriment. Briefly it was thought that a new comet, Molten-Watt, had been sighted. There was nothing left of the ESC, just the burnt ends of the wires, but the pilot said later that the battery and motor were unharmed!

Talking to a flying mate today, he does a lot of powered gliding at another site, by coincidence he remarked that recently they’ve had a great many problems with out-runner motors. Too many! Looks like there is still a little way to go on the quality inspections…………...

PB

[Tom Wright 2]

Some very common sense observations from Peter,with perhaps a hint of IC is best,but i do feel that the vast majority or EP system failures are due to lack of knowledge or lack of due care and attention , after all you would not put paraffin in an IC engine ,would you?

But i have seen just about every EP installation sin you can think of ,and most get away with it most of the time ,even using cheap components i have found high levels of reliability can be achieved ,and i do fly a lot ,well my models do,my arms got tired when i tried it.Perhaps the the old trick of bec or ubec with a back up battery with diode is a good idea if you want to maximise system reliability . But we also have to consider that most probably the cause of most crashes is due to pilot error ,and all the many situations that could have been avoided.

To me the situation is quite clear BEC....UBEC ....UBEC with back up battery and diode.represent the worst to the best options for achieving the most reliable supply to the RX .

 

TW2..

 

 NB  if any one comes up with the back up battery black wire argument i would say you should have rewired it or chucked it away long before the problem occurred.

Edited By tom wright 2 on 28/03/2011 03:39:09

Edited By tom wright 2 on 28/03/2011 03:46:11