Arc explosion on first time build test

30kv dual

1 µW
Joined
Apr 13, 2018
Messages
4
Hi everyone, i've been a lurker for a while but now the time has come to make a post.

First of all, this is my setup, wired it up for the first time today:

  • Sensored RC outrunner motor
    48V generic ebay brushless controller 38A
    12S (2 x 6S) Multistar 10000mAh Lipo connected in series
    throttle

TL;DR: When I connected the battery power, the connector made a huge spark and exploded with smoke and soot everywhere.

apologies for the quality, i had to use paint to convert the photos to 256 color because of slow internet
1.jpg
2.jpg
3.jpg
4.jpg

I had set up the motor, controller and battery on a table to test everything before installation. The motor was mounted on a CD with the shaft sticking through the hole, and the CD was placed on top of a roll of tape so that the outrunner could spin freely.

The motor phase wires (bullet connectors) were connected to the controller (hole end wires) just by putting the bullet partly through the hole and covering it temporarily with some large heatshrink tubes. The throttle connector had different wiring to the controller so I cut the connector off and twisted the correct wires together. For the Hall sensors I used some staples as jumpers to link the JST-ZH connector to the bigger controller connector. All of these were temporary measures to test if the controller and motor worked. Since the motor was unloaded for testing purposes, I assumed that the motor current would not be significant.

For power the two Multistar 10000mAh lipo batteries were connected in series, with a wiring harness that I made myself out of XT90 connectors and 10 gauge wire. I had difficulty in soldering the wire to the connectors because of a cheap and weak soldering iron and the heat shrink was not fully sealed. However I had tested the connections for short circuits before connecting the batteries and also tested the voltage output after connecting the batteries (about 48.9V). So the wiring harness was confirmed to be working.

To connect the batteries to the controller, I use a banana-XT60 cable for an iMAX B6 charger and an XT60-alligator cable. The banana plugs fit into the holes of the female XT90 connector. The wiring was as follows:
Battery XT90/Explosion/Banana---XT60/XT60---alligator clip/hole end wire-controller
The controller power connectors were hole end wires (The ones that you bolt down, I do not know what they are called) so I clamped them using the charger cable alligator clips. Again, I expected not to draw much current since the motor was running free.

I connected the positive banana plug first. An explosion occurred when I connected the negative. It gave off a very large spark (arc?), I think it was purple in color but I'm not sure since it was momentary. Burnt stuff got sprayed everywhere, and ground zero was covered n soot(?), as were my hands. A lot of smoke was released. Although it has been about one and a half hours since, the smell is still in the room. The outer part of the negative banana plug came off and became stuck to the XT90 connector, which was partly melted on the negative side. I think it may have welded itself to the inside of the connector. The heat-shrink failed and the wires got knocked off on the banana plug's XT60 connector. Some of the insulation on the banana wire has melted. As for the motor and controller, none of their wires have visual signs of damage. The smoke came from around the vicinity of the XT90 power connector.

Both batteries had low voltage alarms connected. These were rated for 3.3V per cell. Prior to the explosion, the total was measured at 48.9V. During the explosion the alarm briefly sounded but I don't know if it was one or both. I presume the large current caused a massive voltage sag? Both batteries seem to be ok as of now, they aren't warm or bulging from what I can tell and the voltage for each 6S block is about 24.6V. The XT90 connectors don't seem to be damaged. Are these safe to use? How much current would you expect to have flown through at the moment of explosion?

Also, the controller's manual said to connect the self-learning wire first, then the power, then the 'door lock' (this was a single thin red wire with the same hole end as the power so I simply connected this together with the positive power-please correct me if this is wrong), then the Hall sensor wires (from what I can make of the bad translation), then the phase wire and then the throttle. However as I was lacking in proper connections I found it easier to wire up everything before connecting the battery power. I don't know if this was a factor in what happened.

I would like to know what could have caused this explosion. I think it might have been my shoddy series wiring harness or the charge cable. Also if the controller is damaged can it damage or short circuit the batteries if i connect it again? In any case, I am going to purchase a premade wiring harness from ebay, and also get myself some crimp connectors and make proper connections to everything (throttle and hall sensors aren't high power so maybe not those yet) before I proceed further.

This is my setup:

  • Sensored RC outrunner motor
    48V generic ebay brushless controller 38A
    12S (2 x 6S) Multistar 10000mAh Lipo connected in series
    throttle

Thanks
30kv dual
 
Sounds like you connected up to a short circuit. No idea where, but its pretty definite you got a KFF. This is always when you connect positive terminal of a powerful battery to the negative. somehow.
 
The motor phase wires, and also the Hall wires can be connected in the "wrong" orientation without damage. Such a motor would run porly, run backwards, or not run at all.

As to the throttle wires, those are typically 5V, and if they fail in the full on position, it would not produce a huge spark at the battery connector.

The problem when you encounter a spark that big on a battery harness is likely to be an incorrect arrangement on the wires or connectors (I agree with dogman in the post above). If you are lucky, the blown connector acted as a fuse, and it did that well enough that the batteries have not been permanently damaged...or the controller.
 
If it's a new (unused by you) controller, there's a possibility that some kind of manufacturing fault resulted in a short inside the controller.

One of my BBS02 controller failures resulted in a near dead short across the input leads, blew one of the pins right off my battery connector as I attempted to connect it.

The battery should be fine, the relatively weak connections to it would have prevented prolonged excessive current from flowing, by burning/blowing/melting the circuit open.

If you're sure your wiring wasn't the culprit, I'd look next at the controller.
 
dogman dan said:
Sounds like you connected up to a short circuit. No idea where, but its pretty definite you got a KFF. This is always when you connect positive terminal of a powerful battery to the negative. somehow.

Sorry, but what's a KFF?
Also, is there a common/standard way to connect the 'electric door lock' wire in these controllers? Even if they're all different, are there any common configurations? I have absolutely no idea where to connect it and apparently it's necessary to be able to run the motor so I just connected that wire to the positive battery terminal, since it was red and had the same big connector as the main power wire. Now I feel so stupid for doing that...

dustNbone said:
If you're sure your wiring wasn't the culprit, I'd look next at the controller.

Can I test the controller for a short circuit by measuring the resistance with a multimeter? With the power positive and negative I measured values in the mega ohm range (keeps changing) and with the thin red and the negative power I measured about 200k.

spinningmagnets said:
The problem when you encounter a spark that big on a battery harness is likely to be an incorrect arrangement on the wires or connectors (I agree with dogman in the post above). If you are lucky, the blown connector acted as a fuse, and it did that well enough that the batteries have not been permanently damaged...or the controller.

Yeah... I'm going to ditch the homemade harness and get a proper one. No more soldering, it's also killing my lungs. Seems like it was either the banana/bullet connectors or the XT90 since the controller wires and alligator clips don't show any sign of burning. Though the XT60's heat shrink fell off it also appears intact otherwise. Is it possible that the arc had jumped the distance between the banana plugs or the XT90 terminals while I was connecting? I'm testing all of the wires involed with a multimeter.
 
KFF = Kentucky Fried Fingers

"What's the best treatment for KFF?" (2 pages, PICs)
https://endless-sphere.com/forums/viewtopic.php?f=1&t=46713&start=25

file.php
 
Use a precharge resistor to limit current when first connecting. Like a big 100 ohm wirewound. Connect this in series and see if the voltage on the controller ramps up to full battery voltage in a second or two. If there's a short this will limit the current to half an amp. After the voltage comes up have a connector or switch that will short the resistor to allow full current later. The resistor is for starting smoothly without taking arc damage on the connectors.

Temporary lashups are dangerous, triple check everything and make sure things cannot move and short. Make sure the motor is restrained, recoil from accelerating the rotor inertia can make it move a lot. Reversing polarity on controller power will make it look like a short due to the inherent body diodes in the FETs.

The current during a short can be a thousand amps or more at 12V. At 48V it can be a lot more. Think battery welding.

KFF is for kentucky fried fingers. Welcome to the club. Ron beat me to that.

WEAR EYE PROTECTION and leather gloves are a good idea too.

SAFETY FIRST.
 
30kv dual said:
I connected the positive banana plug first. An explosion occurred when I connected the negative. It gave off a very large spark (arc?), I think it was purple in color but I'm not sure since it was momentary. Burnt stuff got sprayed everywhere, and ground zero was covered n soot(?), as were my hands. A lot of smoke was released. Although it has been about one and a half hours since, the smell is still in the room. The outer part of the negative banana plug came off and became stuck to the XT90 connector, which was partly melted on the negative side. I think it may have welded itself to the inside of the connector.
So, why do I see burnt positive pole on the XT90?
 
Update:

I was absolutely brain-dead, and connected the wrong polarity. O_O
Fortunately, the controller still functions with the correct polarity and the motor turns now.
Now to test out phase timing....
 
Back
Top