WattFlyer RC Electric Flight Forums - Discuss radio control eflight

WattFlyer RC Electric Flight Forums - Discuss radio control eflight (http://www.Wattflyer.com/forums/index.php)
-   Power Systems (http://www.Wattflyer.com/forums/forumdisplay.php?f=17)
-   -   Why do ESCs smoke when battery is connected? (http://www.Wattflyer.com/forums/showthread.php?t=73176)

abborgogna 02-11-2014 05:27 PM

Why do ESCs smoke when battery is connected?
 
This has now happened to me twice, once on a control line plane and yesterday on a new R/C plane. After an very normal flight I let the plane rest for a few minutes then inserted a new battery and the ESC smoked on me. The only load on the ESC was the BEC. A very similar even happened when I was flying one of my C/L stunt planes, plug in the battery and instantly the smell of "Death Smoke". I would have thought that if anything was going to smoke it would be the MOSFETs that drive the motor, but in both cases the motor was not running or started. There appears to be a weakness in the front end of some of the less expensive (read cheap) ESCs. I replaced the burned one in my R/C plane with a new CC 36amp ESC, I hope it does better than the old one. It only had about 15 flights on it before taking a dive.????
Andy

thepiper92 02-11-2014 05:39 PM

Are you sure the second battery doesn't have the polarity mixed up. If you plug in a battery with the negative pole making contact with the positive of the esc, then magic smoke shall appear...I have made that mistake. The other potential problem could be the capacitor, it could hold a charge and for some reason sends a charge back through the esc. Just a guess, I've only had smoking from plugging the battery in wrong. Perhaps the esc is too weak for the setup, and you are destroying it as it is flying.

abborgogna 02-11-2014 07:51 PM

No on the mixed battery polarity. These are batteries I have been using for a couple of years in my electric control line planes. Standard 2200mha 3 cell batteries. I spent nearly 45 years in the electronic arena and I can tell you for sure the battery was built correctly.

One thing to note, I was using a 1500mha 3 cell batteries prior to yesterday. On the time before yesterday I tried one of my 2200 mha batteries and it dropped right in and worked fine. The first flight yesterday used a 2200mha batteries without any problem. It was on the second flight attempt that the ESC smoked. Voltage and current can lead to bad things but battery capacity should not be a problem.

Maybe I am just snake bit!

thepiper92 02-11-2014 08:00 PM

Perchance it was just bad luck. Unless the esc is being pushed too hard it should be fine. I've never had an issue with a dead esc, other than accidentally reversing polarity. On cars I have pushed escs to the limit as well as motors, running at nearly destructive temps. I haven't for planes, due to the fact that an esc dying in flight means no more planes. What plane and motor? Also which esc was in it?

DHC Beaver 02-11-2014 08:27 PM

Also be aware that the cheaper esc's are usually overrated by a considerable margin.I factor in a 20% buffer when choosing an esc.e.g.if you need 40A,use a 50A.

CHELLIE 02-11-2014 08:29 PM

I have had that happen too with the really super cheapo ESC, Buy the Turnigy plush and turnigy sentry ESC and you should never have that problem again, its all i use and never had any problems with them. Hope that helps, Chellie

CHELLIE 02-11-2014 08:30 PM

Quote:

Originally Posted by DHC Beaver (Post 939750)
Also be aware that the cheaper esc's are usually overrated by a considerable margin.I factor in a 20% buffer when choosing an esc.e.g.if you need 40A,use a 50A.

+1 :ws: use a ESC thats 20 to 25% larger than whats needed, it will stay cooler and last longer.

thepiper92 02-11-2014 09:07 PM

Also, based on my experience with Castle Creations in the land rc realm, they don't tend to handle heat all too well, unless it was a $200 dollar "Mamba Monster". I really don't know if the same goes for the CC air esc lineup, nor do I know if the esc you used that burnt up was a CC.

flydiver 02-11-2014 09:10 PM

Theoretically a larger battery should only affect capacity. In reality a larger (better?) lipo will support more throughput and sag less. Therefore more load on the ESC. If the ESC was running on the ragged edge, maybe that was enough extra stress to take it out.

Agree with the 20% (extra) rule and not to believe most of the specs on cheap electronics.

kyleservicetech 02-12-2014 01:59 AM

1 Attachment(s)
Quote:

Originally Posted by abborgogna (Post 939726)
This has now happened to me twice, once on a control line plane and yesterday on a new R/C plane. After an very normal flight I let the plane rest for a few minutes then inserted a new battery and the ESC smoked on me. The only load on the ESC was the BEC. A very similar even happened when I was flying one of my C/L stunt planes, plug in the battery and instantly the smell of "Death Smoke". I would have thought that if anything was going to smoke it would be the MOSFETs that drive the motor, but in both cases the motor was not running or started. There appears to be a weakness in the front end of some of the less expensive (read cheap) ESCs. I replaced the burned one in my R/C plane with a new CC 36amp ESC, I hope it does better than the old one. It only had about 15 flights on it before taking a dive.????
Andy

H'mmm
These ESC's consist of a three phase bridge type power driver to the three phase brushless motors we use on our electric models.

Electrically it looks like the attached JPG.
Problem is, if any two MosFets directly in series turn on at the same time, you've got instant dead short, and instant smoke. That would only happen with a very poor design of the ESC.

Take a look at the two drawings on the attached schematic. Normally, as shown on the left drawing, only one switch (MosFet) on the top, and one alternate switch (MosFet) on the bottom would be turned on at the same time. These switches electronically turn on and off at a rate of thousands of times a second. (That would be the selected switching rate)

The right drawing shows two MosFets directly in series turned on at the same time. This is a dead short on the battery, resulting in hundreds of Amps through those Fets. And, instant failure, and instant smoke. Again, proper ESC design absolutely prevents this from ever happening, even for extremely short periods of time. Programmed into the microcontroller driving those MosFets is a very short period of time during the switch process where all six switches are open, kind of a dead zone or no man's land.

This might be one reason that Castle Creations does not allow anti-spark resistor networks to be used in their ESC's. The possibility exists that the Microprocessor could initiate where the MosFets are not properly turned off during the plugging in the battery start up process.

thepiper92 02-12-2014 02:24 AM

So essentially if the first MOSFET does not prevent electron flow, and the second does, the first one will overload?

kyleservicetech 02-12-2014 02:35 AM

Quote:

Originally Posted by thepiper92 (Post 939782)
So essentially if the first MOSFET does not prevent electron flow, and the second does, the first one will overload?

Normally, the MosFet switching will always put the top MosFet "On", connecting to a motor winding. Then a second motor winding will connect to a different MosFet turned "On" on the bottom half of the drawing.

So, there is ALWAYS a motor winding connected between any top and any bottom MosFet. The problem occurs when two MosFets directly in line with each other, as shown in the right side of the second drawing are turned on. That is a dead short, and will blast the MosFets in milliseconds.

Here is a somewhat technical description of what happens inside our ESC's.

For those that are interested:
http://ww1.microchip.com/downloads/e...tes/01083a.pdf

Here is a video
http://www.youtube.com/watch?v=ZAY5JInyHXY

thepiper92 02-12-2014 03:00 AM

I think I understand. Essentially the current flows through the MOSFET, to the motor, then through a second MOSFET after that. In the second, there is no current to the motors, only a direct path through two mosfets to complete the circuit. With that heat would be instant, the same as connecting a 9v to merely a wire. I'm no pro at this stuff. Based on the pics, it's seems as one path goes to two connections on the battery, or is that done to simplify thing.

abborgogna 02-12-2014 04:29 AM

Sounds like the cheap ESC could have a hole in the firmware that could allow two MosFets in series to turn on long enough to fry. Given the throttle sick was full off there should not have been a command to the firmware to start turning on the MosFets, they all should have been off.

I would think there should be a power on clear that prevents the firmware from starting up in code that could start turning on the fets. Like I said maybe I am just unlucky, but I am leaning toward the notion that the ESCs have bugs in them. This should just never happen, if it can (and clearly it can) then the ESC design is suspect. Poorly designed power up reset circuit?

The best solution is just use good quality ESCs, since the one that smoked came with the plane I will let it go at that. The replacement unit is a quality Castle Creation ESC. I have never had one of these fail when used properly.

Thanks everybody for your ideas.:)
Andy

thepiper92 02-12-2014 04:36 AM

Hopefully the cc escs for planes are better than the ones for land. I personally trust hobbywing. Just ordered a 40a one, 35 bucks, full programming and all. All this time liking hobbywing and just now realizing them don't only make escs for land vehicles.


All times are GMT +1. The time now is 05:02 PM.

Powered by: vBulletin
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 2005 WattfFlyer.com
RCU Eflight HQ

Page generated in 0.11923 seconds with 13 queries