Jump to content

Safety plug


Recommended Posts

Posted by Geoff Sleath on 13/04/2018 13:28:55:

That's very interesting, Bryan. Thanks for that.

One term I don't understand is the S factor of the esc. Is that the number of cells? ie the maximum input voltage? So the zener should be chosen to be the rated maximum voltage plus one extra cell - ie (say) 4v for LiPo?

The input capacitors on the esc I have on my desk (Black Mantis 60 amp) are 2x220mfd in parallel (so 440 mfd) 35v working.

You estimate a 4S LiPo as having an ESR of less than 20 milliohms. In the measurements I've made 5 milliohms per battery is quite low and they can go up as high as 10 milliohms/cell before they're virtually useless as sourcing a decent current. So I think your estimate is perhaps a little on the low side.

Geoff

Yes Geoff, I meant the S rating of the ESC. I do not know how different manufacturers decide on these ratings. I hope that 4.2 times the S factor is less than the voltage ratings of the capacitors and FETs. A few volts more for the zener above that S equivalent is all that it should take.

What is the S rating of that ESC? How does it compare to 35V (read off a capacitor?). If that is a 4-S rated ESC then it corresponds to twice the battery voltage. QED.

The ESR of a Lipo battery is very temperature dependent. It is high at low temperatures and low at high temperatures. My 2200 mAh 3S cells measure about 5 mOhms per cell at 25 C. My numbers are meant to be illustrative. YMMV. In any case, a good engineer should always consider the worst case.

 

Edited By Bryan Anderson 1 on 13/04/2018 14:26:23

Edited By Bryan Anderson 1 on 13/04/2018 14:47:43

Link to comment
Share on other sites

I measure the ESR of my LiPos each time I charge them with my iCharger or, probably more accurately, with my Wayne Giles meter but more often the former because it doesn't take any extra effort My new Graphene 2200 3S batteries measure less than 5milliohms (3or 4 IIRC). It's a point that the poorer the battery (ie higher ESR) then the longer the battery leads you can have because the Q is lower

However, I like the idea of installing a 35v zener to protect the esc. Very easy to do assuming it's fast enough in its voltage limiting - it's not something I've never needed to measure or consider.

Your explanation is very clear. I guess thicker wire has a lower inductance which reduces the resonant frequency but increases the Q.

Geoff

Link to comment
Share on other sites

Posted by Geoff Sleath on 13/04/2018 14:48:14:

I guess thicker wire has a lower inductance which reduces the resonant frequency but increases the Q.

Geoff

To a small extent. The impedance and hence inductance of a length of transmission line goes as the

log( 2 * s / d ) where log is the natural log, s is the conductor spacing and d is the conductor diameter. You need roughly 4 times the weight of copper to halve the inductance.

And, as I am sure you know, the smaller the inductance, the higher the resonant frequency and the Q if other things stay equal.

Link to comment
Share on other sites

Posted by Bryan Anderson 1 on 13/04/2018 13:08:54:

The reason why the turn-off transient is less stressful is because
a) the current from the battery is usually minimal when disconnecting
b) the energy stored in the capacitance would typically be an order of
magnitude greater than that in the inductance so the ability of the
latter to cause overshoot is minimal.

This is true for the turn off transient when disconnecting the battery, but during normal operation the ESC switches power between the outputs and, particularly at throttle settings less than full, switches the current on and off.
This switching therefore causes voltage transients in the wires between the battery and the ESC.

Since this switching also operates at something like 8kHz or higher, these transients occur at that frequency. The result is that the capacitors at the input of the ESC are repeatedly charged and discharged at this frequency by the voltage transients.

Longer wires result in a higher inductance and so a higher transient voltage. This higher voltage then causes a higher current into and out of the capacitors. For the capacitors to handle this alternating current they need to have a sufficiently high "Ripple Current" rating (which is helped by them having a low ESR rating).

If the capacitors don't have a high enough rating, then they heat up. Electrolytic capacitors tend to "dry out" over time and so lose some capacitance. If they heat up, this drying out occurs sooner. Eventually the capacitor fails completely at which point the transients are applied directly to the rest of the electronic circuitry, usually causing a catastrophic failure of the ESC.

Adding extra capacitors to long wires, reduces the voltage transients due to the higher capacitance, and also reduces the ripple current in the capacitors as the current changes are shared between them.

I did try to work out the likely ripple current in the capacitors once, and came up with a value that suggested the capacitors on most ESCs were being run well beyond their ratings.

Mike

Link to comment
Share on other sites

Mike, I agree that the ripple current rating of the capacitors is also
a concern. There is no doubt that using minimal-length battery leads
is desirable but not always possible. The extra inductance of longer
leads would exacerbate the capacitor heating problem and this would
not be helped by an added zener diode which only helps protect from
over voltage. Any added capacitors should have a voltage rating of at
least twice the peak battery voltage.

Using longer leads and added capacitance is not without another
problem. The inductance and the total capacitance form a parallel
resonant circuit as far as the FET drivers are concerned. The extra
capacitance and inductance lowers the resonant frequency of this
circuit and there is the added danger that it will get closer to the
rate at which the FETS are pumping currents into it. Driving such a
circuit at resonant frequency increases the currents and voltages by
the Q factor and the heating effects by Q squared. Not good.

A belt and braces approach would be to decrease also the lead
inductance by using more copper. Rather than just fatten the
conductors which is not very efficient, it would be better to use
another "twisted pair" in parallel to halve the inductance. In fact,
more pairs would achieve even more reduction. Not very elegant and not
a nice solution though it would help to reduce resistance losses in
the leads.

Link to comment
Share on other sites

  • 3 weeks later...

Rather than use an XT60 arming plug I thought simply having a 4mm bullet connector of the type I use for battery connections any way (at least other than 3S 2200 which have XT60 connectors) on a flying lead from the esc that plugs into a panel mounting gold plated 4mm socket such as these which are actually intended for HiFi connections (probably by the sort of HiFi obsessive who also uses gold plated mains plugs!) which look quite beefy. The only problem no current specification is offered.

In my application it would a neat solution because the plug is intended to be inside an open cockpit and there'd be no loose connector to misplace.

Has anyone tried this idea? I'm tempted to buy and test anyway.

Geoff

Link to comment
Share on other sites

Posted by Geoff Sleath on 29/04/2018 21:18:02:

Rather than use an XT60 arming plug I thought simply having a 4mm bullet connector of the type I use for battery connections any way (at least other than 3S 2200 which have XT60 connectors) on a flying lead from the esc that plugs into a panel mounting gold plated 4mm socket such as these which are actually intended for HiFi connections (probably by the sort of HiFi obsessive who also uses gold plated mains plugs!) which look quite beefy. The only problem no current specification is offered.

In my application it would a neat solution because the plug is intended to be inside an open cockpit and there'd be no loose connector to misplace.

Has anyone tried this idea? I'm tempted to buy and test anyway.

Geoff

I doubt if the current rating of those hifi sockets is a problem, but you could achieve the same thing using a standard 4mm female bullet connector: Just solder a suitable washer to it to act as a collar, then epoxy it into a suitable location in your cockpit. That way you can get whatever size you want, to suit your amps load.

Link to comment
Share on other sites

Well, I ordered a couple last night (Sunday) and before I went to bed (on Monday am - I'm a night bird!) I got an email that they'd been despatched so I should get them soon. They were only £2.99 for 2 so worth a punt. Perhaps I'll get this model finished and test flown ... eventually.

I'll report on my success (or otherwise) with the connector. It's just that XT60 connectors are so difficult to remove I was afraid I'd rip the dashboard out of the cockpit trying to remove it after a flight I thought it worth examining an alternative.

Geoff

Link to comment
Share on other sites

Assuming I've understood you correctly as to how you plan to use those connectors you've ordered - the one on the dashboard is connected to battery +ve and a flying lead from the ESC plugs into it? I'd not consider that a 100% safe setup, as the battery +ve terminal isn't insulated and could easily be touched by either the ESC lead or something else, Maybe put some shrink sleeve around the socket?

Link to comment
Share on other sites

Thanks for pointing that out, Gordon. I'll certainly insulate everything by some means - probably shrink on. I wasn't going to rely on a tag held by a nut in any case. It may be OK for speaker connections but not for 4S LiPo.

It's unlikely to be touched by the esc lead inadvertently because what would be the bare part of the socket will be behind the panel and the flying lead at the front but it's always better to have stuff well insulated in any case.

I haven't actually seen the sockets yet so I still may not use them. It's just an idea at the moment.

Geoff

Link to comment
Share on other sites

Geoff . In the fast electric boats we would epoxy 2 female 4mm gold connectors into the hull deck and make a loop from 2 male connectors . A quick pull pull of the loop and system - dead !😁 .Never failed despite being hit in smashes which were frequent ! We were killing 12 cell nimh 3000 mah in 5 minutes full throttle ! Surely that would work ? Colin

Link to comment
Share on other sites

Posted by IanR on 01/05/2018 21:54:55:

Would something like this do? **LINK**

Ian

That's the same thing I already have with an XT60 connector. The main problem with them is that XT60s are so tight I'd be in danger of destroying the model trying to get it out. A standard 4mm gold-plated bullet type connector is more likely to be actually used and will cope easily with the anticipated current draw.

I like Colin Carpenters idea, too. Motor/esc wiring is next on the agenda so I'm thinking mode right now on arming plug options.

Geoff

Link to comment
Share on other sites

Isn't this the answer to my earlier question? A ready made switching device with, apparently, enough capacity; bit of vaseline on the conductors, lots of epoxy on the back; may well get a couple for my son to hand-deliver from LA as the pound seems a little less weak against the dollar.

Link to comment
Share on other sites

Posted by Bruce Collinson on 02/05/2018 08:33:21:

Isn't this the answer to my earlier question? A ready made switching device with, apparently, enough capacity; bit of vaseline on the conductors, lots of epoxy on the back; may well get a couple for my son to hand-deliver from LA as the pound seems a little less weak against the dollar.

Not sure that vaseline is a good idea since it will reduce the effectiveness of the contacts. I use powdered graphite, as sold by locksmiths. Even rubbing the contacts with a pencil 'lead' (graphite usually) will help. So far I've never damaged my airframe when using my Deans Ultra arming plug.

Link to comment
Share on other sites

Posted by Allan Bennett on 02/05/2018 10:39:05:

Not sure that vaseline is a good idea since it will reduce the effectiveness of the contacts. I use powdered graphite, as sold by locksmiths. Even rubbing the contacts with a pencil 'lead' (graphite usually) will help. So far I've never damaged my airframe when using my Deans Ultra arming plug.

As an aside, do people widely use the powdered graphite approach for 'normal' battery-ESC connections?

Link to comment
Share on other sites

Posted by The Wright Stuff on 02/05/2018 11:19:05:
Posted by Allan Bennett on 02/05/2018 10:39:05:

Not sure that vaseline is a good idea since it will reduce the effectiveness of the contacts. I use powdered graphite, as sold by locksmiths. Even rubbing the contacts with a pencil 'lead' (graphite usually) will help. So far I've never damaged my airframe when using my Deans Ultra arming plug.

As an aside, do people widely use the powdered graphite approach for 'normal' battery-ESC connections?

Graphite is a conductor and best kept away from anything electrical.

Link to comment
Share on other sites

Posted by Gordon Tarling on 03/05/2018 10:22:04:
Posted by The Wright Stuff on 02/05/2018 11:19:05:
Posted by Allan Bennett on 02/05/2018 10:39:05:

Not sure that vaseline is a good idea since it will reduce the effectiveness of the contacts. I use powdered graphite, as sold by locksmiths. Even rubbing the contacts with a pencil 'lead' (graphite usually) will help. So far I've never damaged my airframe when using my Deans Ultra arming plug.

As an aside, do people widely use the powdered graphite approach for 'normal' battery-ESC connections?

Graphite is a conductor and best kept away from anything electrical.

But that's the whole point -- it lubricates the mating surfaces without affecting their conductivity. It does have to be applied very carefully and sparingly, but I'm sure if any residual powder did manage to form a track between the + and - pins, it would vaporise instantly without causing a problem. Using a graphite pencil avoids that potential problem, but I've got a bottle of powder so that's what I use smiley

Link to comment
Share on other sites

If I use anything on electrical contacts I use Electrolube. It's been around a long time and I still have a can and a half Years ago when I was repairing radios and TVs we actually used to take apart noisy volume control pots and lubricate the track with Vaseline (not the resistance track but the metal pick up track) and they out-lasted new replacements. Can't think why we did it because it was very time consuming but Dad was old school throw nothing away if it can be repaired and I think I've caught a little of that

I've never thought of using graphite (pencil lead?) on things like 4mm bullet type connectors but I'm sure it might do the trick. Something to make XT60 connectors easier to part without losing the low resistance would be welcome.

Geoff

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...