Jump to content

Using 2 UBECS in parallel?


Tim Hooper
 Share

Recommended Posts

probably OK in a low load situation like the Tx - but for high drain use, I wont touch anything other than genuine A123 indudtries stuff. I have tried the soft pouch type Life packs before and was very disappointed.

Another thing to watch for on some of these dedicated packs, is the use of a PCM - I had one in tx pack from HK ( lipo ) and unfortunately it worked...EG: The whole damn thing just shut down dead as the voltage dropped. I switched to a home made 3 s pack and never looked back.

usual disclaimer applies.

Link to comment
Share on other sites

Advert


Another way of looking at it:

Use a 2S LiFe, and back this up with a UBEC run off the 5S LiPo, set to 6V output.

If the LiFe is well charged, the UBEC will not deliver any current - so will take litle from the 5S LiPo.

If the LiFe becomes discharged, the UBEC will provide the current needed.

You don't need a complicated switch: just put the UBEC output and LiFe in parallel.

Link to comment
Share on other sites


Martin, - I’m afraid I’d have to take exactly the opposite view. I have done some tests, on 4 of the Hextronic monitors, and I think they are very accurate. I did a thread on this, not too long back, and from this I would be more than happy to say that with such a monitor I would be able to instantly record the EXACT state of my receiver pack at any given moment. Without such a monitor, (or any other type of indicator), I could never really guarantee what the state is. Certainly just charging and flying has served aeromodellers well over many years, but as I’ve also having seen the presence of a monitor save a model on at least two occasions, and a number of pilots taking off with flat batteries, and the resultant totally out of control high speed crash, the ubiquitous ‘closed throttle half failsafe’ certainly doesn’t stand a cat in hells change of operating here, a proper failsafe operates when the system totally fails, under any conditions, I have suggested that we try and start using monitors in our little club. A monitor will definitely stop you taking off with a flat battery, it can’t possibly be reading on the top green one minute and on the bottom red the next, and now a number of pilots are really keen on the idea.
It’s probably all very much a personal choice thing, it’s what you are most happy with. But after pilot error as a source of crashes I think power induced faults might be next; and probably most being avoidable with a little bit of knowledge. So if I can go some way to reduce these misfortunes I certainly will, hopefully with gentle persuasion.

Chris - I’m still not sure how power is transferred with an opto-coupler. I thought it was only a changing signal condition that would be transmitted by the device.
If you have an opto ESC and some form of ohmmeter or continuity tester, just see if there is a circuit between the black wire that goes into the rx with the red and signal, and the main black to the battery. If there is no continuity then I shall have to start to re-consider how I think they work.
I may be a few days before I’m able to check one with one of our members, but thinking back our strange little happening with an ESC, which I related in a thread, this was an opto and it still ran ok even when the common wire to the UBEC was disconnected. So there must have been continuity there.

PB

Link to comment
Share on other sites

Peter very little power is transferred in an opto coupler, it is only used to transfer the PPM signal.

As far as I know the difference we are talking about is the difference between an Opto coupled ESC and an ESC that just doesn't have a BEC.

Having said that I've just measured a Hobbyking one that I have in stock which is labelled as "Opto only" and that does have a common connection between Power and Rx blacks. So I think even the manufacturers are at least lazy in their use of the term

I think that's why I used the term "proper opto" in my earlier post.

I had a google of Opto ESC and there were plenty of results to other forums. The ones I clicked on explained Opto as completely isolated just as I thought. But I'm going to do a little more research.

Link to comment
Share on other sites

Sorry Tim, we are off topic here aren't we.

But I knew I'd recently read confirmation of how I think a proper Opto speed controller works.

This is from a discussion on Spectrum telemetry and how connecting the telemetry sensor to the Flight battery would cause a common ground between power and receiver systems, thus defeating the purpose of the opto isolator. Single post on another forum

Oh, and the Jeti advance pro instructions say that their Opto ESC's provide "galvanic separation" between the power drive and Rx/servos.

Edited By Chris Bott on 20/02/2012 20:45:25

Link to comment
Share on other sites


Chris - as I see it functioning, the power supply from the rx lights a led, this is triggered by the signal pulse. This shines on a photo cell, powered by the ESC supply. So the wire has a gap in it, and only a signal pulse can pass, because it relies on the led being lit. However, this is only one leg, I think it will have to have a conductor between the two to provide the return path. When I find a mate with an opto in his model we will disconnect the black common, leaving just the red and the coloured signal, and see if it still works. I will be surprised if it does.
Certainly in the normal configuration, with the black return wire connected, I would consider this to be completely isolated, there is no physical solid electrical connection between the two. I’m not quite sure how connecting a sensor to the flight pack would defeat the object of the opto-coupler any more than connecting a servo would, say. Presumably this is a temperature sensor, and has no direct electrical connection to the battery? If it were a voltage sensor this may require some further consideration, but even this might be opto-coupled, for the same reasons.

Anyway, as you say, we are wandering now, but at least the parallel UBECs are sorted! Hopefully……

PB

Link to comment
Share on other sites

Yes we really are wandering.

Just quickly, the power/signal (same thing) for the LED comes down the red wire from the Rx and the return is the black wire to the Rx. However, there is no need for that black wire to be connected to teh ESC circuitry.

1 and 2 are connected to the Rx. 3 and 4 are connected to the ESC processor.

Link to comment
Share on other sites

Gotta get up and do a little glass skinning. However this should be my last word on Opto.

Peter it seems we may both be right. I just found the following two posts on another forum.
Seems the term might be used for either. Maybe Jeti/Hacker.Castle use the term as opto isolated, and the Chineese use the term as meaning "no BEC".

1. Optically isolated.

Meaning there "should" be no direct copper path from the high voltage side to the low voltage side. I say "should" because I've seen reports that some chinese controllers use a common ground plane.

2. As stated - as it used these days it means the ESC has no on-board BEC.s

Link to comment
Share on other sites

vPosted by Peter Beeney on 20/02/2012 19:03:58:


Martin, - I’m afraid I’d have to take exactly the opposite view. I have done some tests, on 4 of the Hextronic monitors, and I think they are very accurate. I did a thread on this, not too long back, and from this I would be more than happy to say that with such a monitor I would be able to instantly record the EXACT state of my receiver pack at any given moment. Without such a monitor, (or any other type of indicator), I could never really guarantee what the state is.

What worries me about this approach is the need to assess the voltage under normal working load and factor in temperature (quite significant on this 1C discharge example - anyone have data at a much lower rate or for a non-loaded cell?), expected discharge rate, possibility of unexpected factors in flight and all on a very small voltage difference between an almost full pack and a substantially discharged one as per the (randomly chosen - HLP18650P) example below.

Admittedly, careful and knowledgeable monitoring with accurate equipment is going to give a pretty good indication of battery capacity but I think that for the average modeller (especially on a hot summer's day), the recharge well within capacity method as comprehensively outlined by Timbo is a safer bet - especially with the high charge rates permitted by the technology.

Edited By Martin Harris on 21/02/2012 00:48:28

Link to comment
Share on other sites


Martin. - This is exactly my point, the monitor is quite capable of registering the state of charge of a receiver pack under any conditions; hot or cold conditions or fast or slow discharge rates. In fact, from the battery theory and testing that I did years ago, I’d be more concerned with packs in cold weather, all batteries lose capacity when the temperature drops, as is evidenced by your graph. Which is why I’ve always chivvied our lot to keep the batteries well charged in the cold weather.

With regard to the accuracy, I just figure this is an expanded scale voltmeter, over perhaps less than a volt. They are certainly cheap enough, at about three dollars from HongKong, I think they made a good job of this one, much better than some other past examples have been. If you handed me a pack I could tell you within quite close limits what capacity was left, and certainly if it was capable of flying or not. As I said, I did give these a test drive, and I've now copied part of that thread here, it’s quite possible that only a few people read it; in fact there seems to be a number of ways you can connect rx batteries, with diodes and backers and the like. Everyone has their own idea.
I ignored the fully charged voltage, it’s the low end I’m interested in here; and the two red leds at the bottom eventually go ultra bright, which would be difficult to miss. It’s a nice touch. It would be an irresponsible pilot that flew again at this point, but even then there is a little “safety net,‘ there is still a small amount of capacity remaining below this level. For flying purposes I consider the cut off voltage is 3 V/cell but the charger says 2. I did the discharge at 1 amp, but I’d consider that the average 46 size sports plane consumes an average of 150 mA so this would be spread out a bit further. There are 5 green and 2 red leds, and it is specifically designed for 2 cell LiFe and Lipo packs only. Although the discharge cure is flat, rather like a nicad, this does differentiate between the levels very successfully.

As always, I tried to do this as carefully as I could, at least it’s then a best shot!

A little excerpt from the ‘Servo plug lock and Battery Monitors’ thread -

As promised above, I’ve now done a discharge with a battery with the monitor and the charger and a voltmeter all connected together. I had two LiFe two cell receiver packs, one at 2,300 mAh and a smaller one at 1,100 mAh. As it happened the smaller one is only actually 900 mAh capacity, this perhaps needs some more investigation. For the record, I did several cycles, deliberately without balancing; and when fully charged the cells in the large pack were within 3 millivolts of each other and in the small pack 2 millivolts. When fully discharged the figures were 40 and 30 millivolts. I used a receiver as a common bus so for interest I connected in all 4 monitors on the big battery to get a lit-up-like-a-christmas-tree style comparison. Also on this battery I used a 1 amp discharge current. Without going into lots of figures all four were very similar on the way down; there was one slower one and one faster one. Right at the top, the second green led soon became lit as the voltage stabilised and it then very gradually went through the other greens as the voltage slowly reduced. The interesting point, where the fifth green and first red were lit, after 110 minutes, at a voltage of 6.152V and a discharge of 1,830 mAh they all reached very close together and at the next stage, when both reds are brightly lit, eleven minutes later at 6.010 volts and at exactly 2,000 milliamp hours they all reached simultaneously! Quite remarkable. The discharge then terminated 7 minutes later at 4.025V and 2.145 mAh. If we consider that a monitor consumes an average of 25 milliamps/hour then over 2 hours the four will account for the other 200 mAh discharge current which would not have been clocked on the charger.

I discharged the smaller battery at 500 mA, and one monitor, and got a very similar result. Here both reds became incandescent at 6.000 volts and 848 mAh.

There is absolutely no reason for someone to recharge their packs regularly, after every flight if they want to. Monitor or no monitor. Indeed, opportunistic charging has been shown to be beneficial, and my tx and rx get charged before every session. Regardless; and they’ve all lasted for years and are still going on.
On that subject, a mate was standing about recently and it transpired his tx low volt alarm had started when he switched on, and so he was charging it. It was a lithium polymer and he’d not charged it for months; which is one of the advertising ploys, I believe. As our flying time is limited, this seemed a bit counter productive, but, significantly, I really cannot now believe that anyone would now seriously question the value of a tx low volt alarm. … So why not a receiver low volt alarm?

Humble apologies for the severe digression from track, but I would consider this a Safety item and I try and take this seriously. Aeromodelling has a good safety record, I suspect, as reflected by the low insurance rates; but I think if the number of crashes were taken into account and were considered as of any consequence at all then it might be amongst the most dangerous in the world!

PB

Link to comment
Share on other sites

I'd still contend that some dangerous assumptions could be made - note the 0.2v difference between a cold and fairly normal hot day represents the difference between an almost fully charged battery in the cold and one just about to fall over the cliff on a warm day...I'm sure you'd take that in to account but would one of the less electrically aware flyers in your club?

My concern with voltage monitoring in hot weather stems from the more rapid drop at the end of the discharge under these conditions giving less warning...

I'm sure there's nothing wrong with your method but in safety matters, I'm a believer in aiming at the lowest common denominator and the KISS principle.

Bottom line is the capacity should be more than sufficient for any normal flyer (and can be planned in at the fitting out stage) and recharging before a session (or topping up between flights if there's any concern) should give as close to a guarantee of sufficient capacity as practically possible. Add in telemetry of the receiver battery voltage and things look positively rosy to me!

Link to comment
Share on other sites

Martin - I would have to say that reading an on board voltage display is about as technical as reading the fuel gauge in your car. Or being aware of the low volt alarm on your tx. Again, that’s the whole point of it, you don’t have have any electrical knowledge at all to notice which leds are lit. If on a very cold day the battery voltage starts to descend quickly, and that’s quite likely, the pilot can soon put it back on charge, is it possible that someone can accurately mentally estimate this? He would put it on charge too, just to be sure, but he would have to be fully aware of that. I’m not quite sure how any voltage differences will affect anything, it’s automatically compensated for. Exactly as is the LVA in your tx, you don’t have to make any mental adjustments for that.

If you feel that it’s safer not to have a monitor rather than to actually have one, that’s perfectly ok with me, each to his own, and all aeromodellers do their own thing, but having seen a couple of near misses, one particularly so, a 10lb model diving in at full speed less than two minutes after taking off, just about as close to the pits as you can get without actually being in it, I’m inclined to be cautious. I know the battery was completely flat because I grabbed it from the wreckage and checked it out.
And there is absolutely no doubt in my mind that if the pilot had been using a LiFe with a monitor he’d have never got off the ground. Someone would have spotted the ultra bright red led from across the other side of of the pits. Certainly this has also happened a couple of models previously and avoided a couple of possible total disasters. All rx packs, of every variety and voltage, can have a monitor.

We could theorise about telemetry too, but I really think that ought to wait for another day…..

My feeling about this is that anyone may choose or not choose to use an onboard display, but should anyone choose not to use one, and then take off and cause a serious incident due to a flat receiver battery, I would consider that to be completely avoidable accident. Now there really could be no excuses!

Chris - If you use a LiFe as the parallel pack, it can be seen from the above that this will do all the work and will be just about flat at 6 volts when the regulator clicks in. In itself this wouldn’t be a problem and will go completely unnoticed; until as or when eventually something goes wrong… … when it then becomes very apparent that this is what has been happening. Two regulators in parallel, the one with the very slightest highest output will take the lead, they always do, every time, one of the problems of paralleling diode or transistor outputs, but when the voltage falls a trifle under any loads the other will chip in and help out, they always do. Every time. And again, unnoticed. If you keep the lipo supply battery topped up, and the regs. remain largely in spec. they will last for years, they are very reliable. A good check is a contact thermometer, if they remain cool, (some heatsinking, perhaps?), not too much above 30 degrees C, then they’re just fine, however hard you’re thrashing them.

Happy Landings.

PB

Link to comment
Share on other sites

Posted by Peter Beeney on 21/02/2012 20:43:17:

Chris - If you use a LiFe as the parallel pack, i....

I think you mean Tim?

I think a voltage monitor is extremely useful to have. The only slight issue might come when the A123's are almost empty but still showing in the green, is it possible (on that graph) to be at a point where the monitor shows green so you take off, but then quite quickly they go flat? I'm not sure, as all my experiences with then are at much hihgher currents.

As an amusing asside, I once witnessed a club instructor stating that "these voltage monitors are useless, you only have to waggle the servos and they go right to the bottom red, but I know the battery is fine 'cos I charged it"

Link to comment
Share on other sites

Very sorry, Chris and Tim, I certainly didn’t intend to swap your names. I’ll just have to put it down to an age thing..


I would calibrate the display first, I always do. As I said, the 4 that I tried were absolutely identical at around this low point, and they were in the red just before they fell to 6 volts. They follow the voltage very closely, and in practise I think they would fall very slowly indeed, so you would always have plenty of warning where the voltage was. I think the situation you envisage would never happen. And even then, if it did, as I said there is still some spare capacity. They’ve thought of everything!
My colleague, that I did this for, changed entirely to LiFe’s sometime ago, and, as he says, he would be recharging long before it got to the red anyway; it’s so quick and easy. On the strip, though, they go into the second green and stay there forever, the capacity doesn’t even seem to get bent.


He has a large Tiger Moth, ASP 120 FS, which he fly’s but he also likes to watch it fly, so that’s a very good excuse for a bit of spirited stick stirring, 20 minutes at the time. And again after that, on a nice day. The pack doesn’t even notice.


Interestingly, it seems that rx LiFe packs now don’t have a balancing lead, so that can mean one of two options, possibly. Either they now consider they don’t require any balancing, or it’s a bit of built in obsolescence...

PB

Edited By Peter Beeney on 21/02/2012 22:24:30

Link to comment
Share on other sites

Peter,

Perhaps we're not really disagreeing here - I'm certainly not suggesting that anyone shouldn't have a monitor fitted and of course, if it was indicating a low voltage they'd be insane to take off without recharging and yes, in this case you'd save a model at the very least. I'm trying to make the point that the primary defence against running out of battery capacity should be the discipline of charging prior to a session and flying well within the capacity of the battery - easy to do now that we've moved on from 500 mAh NiCds even with multiple digital servos.

The point I'm trying to make is that there are significant variations in cell voltages due to temperature and although the monitors you've tested appear accurate, are all makes of these (or even all batches from Hextronic) as good? Do they stay in calibration as components age? I don't know the answers and short of checking each one with a calibrated DVM you'd be taking them very much on trust.

Most users would, these days, have access to a reasonably sophisticated charger which will easily allow a discharge after a flying session to quickly give a reasonably accurate indication of remaining capacity - plus or minus 10% here is a lot less critical than the same error on an expanded scale voltmeter...

Link to comment
Share on other sites

Martin,

I sure we’re not disagreeing, I never disagree with anyone, only perhaps succeed in getting a little bit crosswind.. But invariably a successful landing…

So with that in mind I would I certainly agree about charging, I’ve always been vociferous in proclaiming the necessity of proper charging. As said before, I charge everything, tx and model rx pack, before every outing, often in the car on the way. This I've always strongly advised, too, to charge before every session, some folks listen, some don’t. Even if in the future my kit, a tx say, has a lipo in it will still get charged every time. I’ve always done that. But I’m not everyone else, and I think it’s fair to try and help everyone that doesn’t have the same ideas, or interest.

Flying within capacity. Our trainer has an old 800mAh nicad pack, in good order, and a GSW nickel monitor, also excellent for the job, but now priced at the thick end of 11 quid at our LMS; and even after a long session, all the afternoon and evening, it seldom moves off the top green. Admittedly it’s gentle flying, but flight loads are often very light, possibly a saving grace in many instances. I found this out years ago, when a trainer would run all day on a 500mAh nicad, the transmitter was the problem, they were also 500’s, and would only last about 4 hours. Hardwired in to boot, and no fast chargers.

The cell voltage variation is only because of the reduced capacity, and if you have a monitor that’s no problem whatsoever. Fortunately, though, most modellers only fly generally between about freezing and 25 degrees C, so they’re not greatly affected. These Hextronik displays I did test with a calibrated DVM, as I do everything else, but I’d certainly consider any DMM would be more than accurate enough for modelling purposes. Even the cheepies are very good, in my investigations anyway. I’d be more than happy to buy one of these voltage indicators and take it on trust. They are certainly not likely to deteriorate with age, any more than your radio or anything else will, they appear to be little more a chip, and the forecast for the working life of some of these is now halfway to infinity…

It’s the batteries I’m more concerned about, mostly good but there are rogue samples about. But there is little pre-first use testing to be seen here by all accounts, it appears they are just thrown in at the deep end and just have to get on with it. And definitely there has been grief, which so could easily have been avoided. Also the chargers, most folks are quite at home with these but others are not quite so good. I’ve had several calls for help with a driving lesson, indeed, one chap rang and said “Can you come over, I’m trying to charge a rx NiMH on my new charger, and it’s getting a bit hot!” I asked him how hot was hot and he replied “Well, it’s burnt my hand!”. Not seriously so, as it turned out, but the covering on the pack was well distorted by the heat. What can you say?

Related to the low temperature flying, a few years back some lads were flying powered gliders on a cold day, they said the thermometer registered -5, but it was a good day for flying. Eventually 4 lithium powered models had crashed, far too many to be coincidence, they were all experienced pilots, and significantly always towards the end of the flight When one of the pilots was telling me about this we came to the conclusion the the capacity was being significantly reduced, I’m sure the pack was getting quite cold in the fuz, mostly just powering the radio. It seems they wren’t fully aware this could happen, and I suggested the packs all needed wooly hats in the winter to keep them warm.

Anyway, I thought this might be of some interest, so I fired off a little communiqué to Headquarters, to which no one ever replied, but about two news letters later someone had written an article describing how lipos can lose capacity in the cold and that we should be aware of this, because it can cause model crashes. So I was pleased to see that at least someone else had experienced this condition.

So now, hopefully, I’ve successfully landed…

PB

Link to comment
Share on other sites


Bob - Yes, I can well believe that, it’s just that I’d recently seen a pack at the strip without the balance lead, but I don’t have any details. Even my single cells have balance leads, and I’m not quite sure how that works anyway.
And I certainly agree about the leds running up and down the scale when waggling the sticks, as you say, it may require investigating, but I’m afraid I’ll never be convinced about the monitor and the requirement for charging.

And, of course, there is always another angle, so perhaps it could be construed as an indicator as to when the battery doesn’t need charging, but a very simple experiment that anyone can do anywhere is to merely connect a Hextronik monitor to a LiFe receiver battery and then just leave it alone. The slight current consumption will gradually discharge the battery, it would probably take about a couple of days for a 1.1Ah pack, but the voltage would fall proportionally and would be indicated by whichever led is lit. To verify this, you could add a voltmeter and confirm the levels at which the leds become lit. When the reds at the bottom end go to ultra bright, at 6 volts, I’d say that was the best indicator yet that the pack needs charging.

However, as always, just a different set of ideas, perhaps of little consequence anyway, but as an addendum to my recent post about lipos in the cold, it seems that a lipo receiver pack is now quite a popular item. If by chance I was operating one in some very exceptionally cold conditions I’d be inclined to be a trifle cautious, and consider that the capacity may become severely depleted.

Just a thought in passing……

PB

Link to comment
Share on other sites

  • 3 months later...

Wow! I think you chaps might need a referee to give points for every round! Again I'm flummaxed while trying to follow the discussion on this thread. I have just set-up my super cub & am fitting a UBEC voltage control from seperate flight pack with "the red" wire disconnected (previous thread advice) & every thing seems to be A O K:there is a "ferrite" (I think) ring on the lead to the Rx. Seem to remember somewhere reading why its there & not to get it too close to nthe Rx? Can someone put me straight about this little magic circle. I have a Dualsky & Turnigy UBEC to use: any comments about these brands?

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...