By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by CML

Wiring a Twin

All Topics | Latest Posts

Search for:  in Thread Title in  
G-JIMG09/10/2018 17:44:08
avatar
73 forum posts
9 photos

Until now I've always flown IC aircraft but my latest build will be a twin using:

2 x PO-3547-800 Motors
2 x PP-TESC45AV ESCs
2 x 4S 14.8V 3700mAh Batteries
1 x tbd UBEC

Question is, what's the best way to connect them? As two separate configurations with the UBEC connected to one of them (a), or with everything connected in parallel (b)?
wiring.jpg
Also, there is plenty of cooling in the nacelles so presumably it's a good place to put the ESC's, but the batteries will be in the fuselage requiring about 18" of wire to reach the ESC. I suspect that may be a problem, but I also suspect that two ESCs in an uncooled fuselage may also be an issue. Thoughts?

Jim.

Chris Walby09/10/2018 20:58:39
avatar
691 forum posts
128 photos

Hi, Best suggestion is to talk to George at 4Max, I have always found him very helpful + there are a few help guides on his website.

IMHO it won't make any difference either having longer battery leads or motor leads (tin hat on!) however other ESC manufacturers state that if battery leads are extended then additional capacitors are to be added.

If its a small model then option "a" is okay (although if you have a spare channel you could have differential throttle if you want (very helpful if you have no rudder control)).

I don't think option B helps unless diodes are used so the UBEC can be fed from either ESC (in case one battery goes flat). There are dual fed switches that can supply an RX from either battery so that's an option.

I have a couple of 62" models that use only 1 lipo 4S5000 for both ESC's and a separate 2S1000 lipo for the RX (via UBEC). That way you power up the RX first and can check controls, then power up the ESC's when ready to fly

Is there a reason why you want 2 main batteries?

Do you have spare RX batteries for your IC models you can use?

Tin hat on again! You may have a start up sequence issue with option A if you power up the lower ESC and don't have the RX powered up (check with George, as other makes have caught me out).

PM me if you want a wiring diagram

Allan Bennett09/10/2018 21:54:15
1403 forum posts
37 photos

If you use two batteries, for whatever reason, I would suggest that you connect their power leads together so they both feed both ESCs and motors. This avoids the embarrassment of one motor dying before the other, due to one battery not fully charged, assuming of course that you never try to run them down to less than, say, 20% capacity.

As for the BECs, it depends partly on what spec. the ones built into the ESCs are, and how big and complicated the model is. You certainly shouldn't have them both feeding into the receiver, unless you have an arrangement of diodes to prevent them back-feeding into each other. In some of my models I use the BEC in one of the ESCs to power the electric retracts and lights, and the other to power the receiver and servos. In others I've paralleled the BECs, using diodes to isolate them, so they both feed the receiver and servos. And in a couple of helis I've used a stand-alone BEC in parallel with the ESC's BEC, via diodes, to provide some kind of redundancy in case one BEC fails.

It is said that long wires between battery and ESC can cause the ESC's capacitors to break down over time, followed by the electronics. This can be prevented by adding more capacitors across the power wires, close to the ESC. Long wires between motor and ESC don't cause any problem other than, perhaps, extra weight, and the fact that the ESCs might overheat if installed in the fuselage.

G-JIMG11/10/2018 20:52:16
avatar
73 forum posts
9 photos

Thanks for the help guys.

I contacted George and he was very, very helpful. He determined that:

I do need two batteries because it will draw too much current with only one.
I shouldn't connect them in parallel because if there's a miss-match it could result in a high current condition.
Long wires between the battery and ESC is no longer an issue with modern ESCs.

Based on that, and using standard 'Y' harnesses as far as possible, the wiring will be as per the schematic:
ka wiring.jpg

As per Chris's comment, I will have to ensure the Right battery is always connected first to ensure the Rx is powered.

Jim.

Simon Chaddock11/10/2018 21:19:36
avatar
5249 forum posts
2738 photos

That schematic looks fine but I hope the advice is right about long battery to ESC leads being acceptable for 'modern' ESCs.

My understanding is the problem is due to the long leads generating voltage spikes due to the pulsed nature of the current flow. The spikes damage the ESC's electrolytic capacitors. The longer the lead the higher the potential voltage of any spike. Capacitors can stand the spikes for so long then fail almost at random. Without correctly functioning capacitors the rest of the ESC is likely to fail in short order.

Of course I may be wrong.

Chris Walby11/10/2018 21:27:55
avatar
691 forum posts
128 photos

Jim,

Looks good and if George suggests it with his experience I am sure he's spot on. His advice to me (twin IC with flaps and retracts) was to have one RX battery and a high power UBEC. His rational was that if a retract servo failed the UBEC voltage would not drop as the servo failed due to the high current performance of the UBEC.

As you have two batteries you might as well spread the load for the cost/weight of another UBEC so I can see that logic.

Only issue with UBEC's was with a HK one and some digital servos where if all three were used at the same time it "browned out" the RX voltage. Fortunately if found it on the bench and just bought a 4-max UBEC to solve the issue. It would have been a real gutter if the plane had been flown as my normal control checks are individual and slow so I would not have found the issue until a time of high stress panic which is not when you want to test the re-initialisation time of the RX!

Be great for the rest of us if you set a build log up and post a few pictures

PS all the best with the build

Trevor Crook11/10/2018 22:09:02
738 forum posts
58 photos

That's a very similar set-up to the one I use in my Black Horse Mosquito. In that, the esc's are in the fuselage as there is room. A bit of air gets in around the nose machine guns and I have a small air scoop hidden under the wing root. The esc's have heatsinks and small cooling fans on them. There is an air exit near the tail. I've used 3-way versions of XT60 connectors for each set of motor leads - the wing halves separate.

This arrangement was in the model when I bought it (secondhand) and has worked well. If I was starting from scratch I would probably put the esc's in the wing root radiators, which would be better for cooling and still keep the battery wires reasonably short. Depending on your model, you may be able to do something similar.

G-JIMG11/10/2018 22:16:02
avatar
73 forum posts
9 photos

Simon, I had the same understanding as you but couldn't find a definition of "long". I asked George at 4Max and his response was, " The long wires on ESCs used to be a problem, but most modern manufacturers have designed around this and now long wires are no longer a problem". I assume they've addressed the issue within the ESC itself.


Chris, Which 4Max UBEC did you purchase for your twin? He only has 4 for sale and the 20A version seems a bit of an overkill. I'm considering the 6A version.
I am taking photographs of the build and may well start a build log at some point, I just didn't want to be yet another guy that starts a log that peters out after a few months.

Jim.

G-JIMG11/10/2018 22:31:15
avatar
73 forum posts
9 photos

Trevor, The model is based on the King Air 350 Turboprop aircraft. Consequently, there is a large air intake directly below the prop hub and an even larger exit 'hole' in the underside of the nacelle. That's why I'd really like to mount the ESC directly beneath the motor, smack in the middle of the airflow.

The only concern I had was the length of wire to the fuselage mounted battery, but, according to George, that's no longer an issue.

Jim.

Trevor Crook12/10/2018 08:07:19
738 forum posts
58 photos

Jim, I read recently in an RCM&E article that the adverse effect of long leads can be mitigated by fitting capacitors every few inches. Unfortunately I can't remember the spacing or capacitor values, and I think I've passed the mag on now. I think it was in Brian Collins' series on escs, perhaps someone else can enlighten us. Better safe than sorry!

Allan Bennett12/10/2018 20:48:25
1403 forum posts
37 photos

There's a long thread about the long-battery-lead saga in another forum. Suggestions for overcoming the problem are in the first post. Presumably the advice that modern ESCs don't suffer this problem is based on the fact that modern ESCs may have more capacitors (those big round cylinders near the power wires) and thus be able to cope with the voltage spikes generated in long wires. I can't vouch for this though, based on my own experience, and the fact that some major ESC manufacturers offer supplemental capacitor packs for you to add.

I don't agree with the advice about not connecting the batteries in parallel: If there is a mismatch (one charged less than the other, I presume), then one motor will cut before you expect it to, unless you have separate voltage monitoring telemetry on each pack so you can land when the voltage gets too low. With them connected in parallel you can be reasonably sure that both motors will lose power at the same time, maybe resulting in a dead-stick landing, but not in a spiral dive. In fact, with them in parallel you can ensure that both motors will cut at the same time if lvc is reached, by using a LipoShield between the receiver and the two ESCs.

G-JIMG12/10/2018 22:28:38
avatar
73 forum posts
9 photos

George (4Max) can obviously only speak for his ESCs, which he guarantees will operate correctly with long ESC to Battery wire lengths (20"+). Apparently it's not just the capacitors they now use, it's also that the Power Triacs are now more resistant to spikes.

Regarding connecting the batteries in parallel; it's not recommended because if the voltage is different (more that 0.03V per pack) the current transfer from the high voltage pack can be huge and end up damaging both batteries.

PatMc12/10/2018 22:50:53
avatar
3928 forum posts
492 photos
Posted by G-JIMG on 12/10/2018 22:28:38:


Regarding connecting the batteries in parallel; it's not recommended because if the voltage is different (more that 0.03V per pack) the current transfer from the high voltage pack can be huge and end up damaging both batteries.

That's baloney, there's more chance of one battery failing before the other & causing an unexpected single motor spiral dive.

What does George's guarantee cover if your model is written off due to one ESC failing ?
A replacement ESC ?

Personaly I'd go with parallel batteries & 2x 45A - 60A opto ESCs installed in the fuselage with extended ESC to motor leads. There must be plenty of room in the fuselage to allow for adequate ventilation for batteries & ESCs.

Allan Bennett13/10/2018 11:18:46
1403 forum posts
37 photos
Posted by G-JIMG on 12/10/2018 22:28:38:

.....

Regarding connecting the batteries in parallel; it's not recommended because if the voltage is different (more that 0.03V per pack) the current transfer from the high voltage pack can be huge and end up damaging both batteries.

I don't have any numbers for exactly how much resistance is in the wiring and connectors, but even a total of 0.003 ohms (including wires, connectors, and batteries' own internal resistance) would only result in a momentary current of 10A if the voltage difference between parallel packs was 0.03v. That's not 'huge' and, I'm sure, is far less than the momentary current one gets when the ESC's capacitors charge up when the battery is connected, and it's much more preferable to me than one motor shutting down suddenly.

Peter Beeney13/10/2018 20:39:54
1531 forum posts
59 photos

Back in the day, when lipos first appeared on the scene and then balancing the cells when charging suddenly became necessary, a lot of discussion started up which suggested that before paralleling the packs together all the voltages must be exactly equal otherwise as stated the current transfer would be high and damage would occur.

I wasn’t really convinced about this so in the usual pb ‘poke and hope’ tradition I tried a little experiment. Using two 3S 2200mAh packs, both in good order, I fully charged one to12.6 volts, and discharged the other until the open circuit voltage read 9 volts; it was definitely flat! I then just paralleled them directly together. I made a sacrificial item for the final connection; a short piece of heavy duty cable and a large croc clip. I didn’t particularly want to burn the sockets, I thought that perhaps this was the biggest danger.

I used a clip on power meter to measure the amps and when I closed the croc clip on the wire the instantaneous current flow read 18 amps. This then began to tick down at about 1 amp per second until it reached a rate of about 8 amps. It then started slow further and when it fell to the 3 amp level I disconnected, nothing was going to change much from there. The two packs would eventually reach parity, cells in a simple parallel circuit will always eventually level out at the same voltage, whatever the circumstances; plus my power meter might not always be strictly accurate below 4 amps anyway.

This would imply that the total circuit resistance would be around 200 milliohms, certainly the internal resistance of the fully discharged pack would be at a maximum. Nothing warmed up, at least to the touch. Both packs recharged ok and just carried on powering models as if nothing had happened. As it happened, I was not in the least bit surprised about any of these events.

Not really a comment on how to parallel packs and cells, this is just a statement of what I recorded when I tried the worst case situation. If I tried it with say 6S packs the voltage differential would be twice as much but the internal resistance would also be higher so maybe there would not be much difference in the current flow rates.

PB

andyh18/10/2018 15:20:23
avatar
419 forum posts
22 photos

another vote here for parallel batteries feeding both ESC's (& both uBECs), so long as they're well matched I don't think you'll have any issues.

one other thing to note, I see on your diagram that you've disconnected the +ve red wire on the "trunk" of the Y-lead from the Rx to the ESC's as you're using a uBEC. personally, I would (& do) disconnect the red wires on the two "branches" of the Y such that the two ESC +ve's aren't connected, rather than on the "trunk".

Nigel R18/10/2018 16:14:58
avatar
1969 forum posts
366 photos

Absolutely disconnect the red wire at each ESC, if using a separate UBEC (or battery).

"I do need two batteries because it will draw too much current with only one."

Say what? You could fit one single larger battery. Of around double the capacity of the individual pack size. 4S 8000mAh. Here's one for £56. With max rated discharge of 240A. I'd de-rate that to about half - being a hobbyking effort - but I would think 120A would more than cover your needs.

On the other hand, 4S4000 is quite a useful size pack and you might want to use a single one in another project. Your call of course.


"I shouldn't connect them in parallel because if there's a miss-match it could result in a high current condition"

I would personally absolutely connect them in parallel.

G-JIMG18/10/2018 18:32:41
avatar
73 forum posts
9 photos

Thanks, I'd already figured out I need to disconnect both Red wires at the ESCs but it's nice to have it confirmed.

I already have 4S 3700mAh batteries, hence the statement that I need two of them. The model uses scale 10.5" 4-bladed props and the Battery/ESC/Motor combination has already been tested and proven to provide the required thrust.

I understand the rationale behind parallel wiring of the batteries and the benefits it can bring, but there is also battery life to consider, these things are not cheap! Although a slight mismatch is unlikely to destroy the battery(s) I understand that it could severely shorten their life.

Peter Beeney18/10/2018 19:00:28
1531 forum posts
59 photos

G-JIMG - With respect, but may I ask how you are considering, or describing, a mismatch, please? And indeed, how it might shorten a particular battery’s life?

I’m only just asking out of interest…

PB

Simon Chaddock18/10/2018 19:29:15
avatar
5249 forum posts
2738 photos

I understood that one advantage of a LiPo cell is that it obtains its nominal charge voltage very quickly from a discharge state. This means that connecting batteries in parallel but with a different charge state will not result in a large long term current flow. This is exactly what Peter Beeney identified even with batteries at the extremes of charge level...

This is the same situation when you use multiple charging board where in effect several batteries are connected in parallel and you don't have to worry about them all being at exactly the same charge level. They all quickly achieve a common cell voltage and then share the charge current proportional to minute variations in cell voltage until they all reach the full 4.20 V state.

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!

Find RCM&E! 

Latest Forum Posts
Support Our Partners
CML
Slec
Expo Tools 14 July
Motion RC
Gliders Distribution
electricwingman 2017
Overlander
Airtek Hobbies
Wings & Wheels 2018
Sarik
Advertise With Us
Latest "For Sale" Ads
Which part of building a new traditional balsa aeroplane do you enjoy the most?
Q: Which part of building a new traditional balsa aeroplane do you enjoy the most?

 Research & choosing the model
 Building the fuselage
 Installing the engine and radio systems
 Building the flying surfaces
 Covering/painting/finishing
 All of it!
 None of it. I'd rather someone else did it!
 Other

Latest Reviews
Digital Back Issues

RCM&E Digital Back Issues

Contact us

Contact us