What might cause an xt-90 anti spark mechanism to fail

The title pretty much says it all. I've been using an xt-90 antispark connection on one of my 20s packs for about a year. It failed a while back (a few weeks after I had started charing to 4.2v/cell from 4.1v). The contacts appear to be clean on both the receptickle and the dingus end. Any thoughts boys?

Can you tell exactly what failed? There is a separate contact patch for the precharge and some kind of resistor. You should be able to measure the resistance between the precharge and main contacts. Maybe the resistor was a little under sized?

Luckily these are pretty cheap and not hard to replace.

What does failing mean? Does it make a spark now?

Yeah it was sparking but still functional, I guess I thought it was clean but I guess I didn't really get into it too well. After cleaning with rubbing alcohol and a light filing it works again. No sparks! I probably prematurely posted but I guess now people will have a search reference if they encounter the same problem.

I guess its also worth noting that you should definitely desolder the connector (with a mating piece attached to keep the contacts lined up in case the heat starts to melt the nylon). I don't know the conductivity of alcohol but I certainly didn't want to find out that way. Thanks for the replies fellas, reminds me that ES is one of the best online communities around!

t_tberg said:

worth noting that you should definitely desolder the connector (with a mating piece attached to keep the contacts lined up in case the heat starts to melt the nylon).

Click to expand...

I have melted a few and the male/female "pins" go abit askew. But you make a great point in connecting m + f up but do try to still solder them quickly. I just fill the pool with solder, tin the wire, heat the pin from the backside with the wire ready to go and connector firmly in place.

markz said:

I just fill the pool with solder, tin the wire, heat the pin from the backside with the wire ready to go and connector firmly in place.

Click to expand...

Yeah I get my soldering iron as hot as possible and try to get in and out quick. I actually don't tin the tip with 10g wire cause I find it's hard to fit if there are any bulges after tinning. I have an angled tip that can push the wire down and sit flush against the flat parts of the pin so then I start on the flat edges when it's hot enough and solder until the wire is saturated and pools up on the the bottom side of the "u". Doing it this way definitely gets the connector really hot so keeping a mating connection connected is imperative to hold the pins in place.

t_tberg said:

bulges after tinning.

Click to expand...

I just do it because then the wire doesnt soak up too much solder in the pool, then maybe you are left with not enough solder connection. I dont mind if the pool overflows with solder, thats a good sign actually, it doesnt mess anything up unlike 4mm bullets.

Also I heat up the wire first and make sure its hot hot, then heat up the pin. Then its for sure a solid connection. I have a fat tip, angled a bit but I can do it on a pointy solder tip too. If the price was right, I'd buy the already wired up XT-90's then do a butte mechanical connection between the two wires. But that requires a clamping tool to do it right, more cost. Or wrap wire around both and solder. I dont mind soldering XT-90's they are big enough for me fat fingers. Unlike trying to do the controller sensor wires, still haven't found a good connector for those yet. Tried Justins, they are too small, tried Molex 0.093" but again need the termination tool, I just solder them on. Need to be sure the wires is a decent size, used small guage and it did not go well. Ended up just twisting them together, and taping them for an on-road repair.

I may just go for small pig-tail connectors from Hobby King then twist the wires and solder and elec tape. I dont like to get them all on one connector, I like each connector separate so I can move it to different motors/cont combos.

What might cause an xt-90 anti spark mechanism to fail

Click to expand...

too high battery voltage in combination with too high controller capacitance

I did a test a few years ago (there is already a thread about those connectors) and i would not use them on batteries with more than 13-14s. The small 5 ohm resistor will blow sooner or later!

You better go with the Jeti antispark connectors. Those are available in 4mm, 5,5mm and 8mm (XT 90 has 5,5mm bullet connectors) and work well up to 100V.

Thanks for the heads up madin, I've definitely had some fail in the past for seemingly no particular reason, always on the higher voltage packs too, my 15s pack antispark connectors have never needed to be replaced. It's too bad those jeti connectors don't come in a similar form factor as the xt-90, I like to use computer cable sleeving on all of my wiring. I'd have to find some kind of work around to maintain a similar aesthetic. View attachment 1

Actually, the sure way to cook them is simple. The antispark function is accomplished with a 2 watt resistor. All you need to do is plug the connector in slowly and the resistor will overheat. Plug it in quickly and they last much longer.
The other thing to contribute to the issue is if there is too much current draw on the load side. I used these connectors on a high current application for a radio control car with a very powerful 1/5 scale steering servo that pulled a few amps and a slow connection would heat up the resistor enough to melt its solder joint. (not the wire solder joint)

Aerialpixels said:

Actually, the sure way to cook them is simple. The antispark function is accomplished with a 2 watt resistor. All you need to do is plug the connector in slowly and the resistor will overheat. Plug it in quickly and they last much longer.
The other thing to contribute to the issue is if there is too much current draw on the load side. I used these connectors on a high current application for a radio control car with a very powerful 1/5 scale steering servo that pulled a few amps and a slow connection would heat up the resistor enough to melt its solder joint. (not the wire solder joint)

Click to expand...

Seems sensible at first. But not so much when I think it through. I thought the whole idea was to pre-charge the capacitors (through the resistor) to reduce or eliminate the rush of current that would occur with a low resistance connection. If so, then if you quickly insert the plug fast enough to protect the resistor from passing all the current that the capacitors could draw and store, then won't the capacitor be partially empty and thus prone to spark when the non-resistive connection is made? And if the capacitors get full or nearly full with a quick insert, then how can there be much increase in current flow (that might damage the resistor) if you insert slowly? Seems like the resistor is either sized right (can handle the amount of current the capacitors will initially draw) for your gear, or it isn't. What am I missing?

wturber said:

Seems like the resistor is either sized right (can handle the amount of current the capacitors will initially draw) for your gear, or it isn't.

Click to expand...

Thats how it is.
XT90s had beed designed for RC model controllers, so for max 14s or 60V hot off charger. Maximum! And added to this the capacitance is usually much lower as it is on e-bike controllers.

madin88 said:

wturber said:

Seems like the resistor is either sized right (can handle the amount of current the capacitors will initially draw) for your gear, or it isn't.

Click to expand...

Thats how it is.
XT90s had beed designed for RC model controllers, so for max 14s or 60V hot off charger. Maximum! And added to this the capacitance is usually much lower as it is on e-bike controllers.

Click to expand...

madin, you were right. The connector's antispark function is actually probably toast still, I don't think cleaning it did anything, the resistor is likely blown. I just thought it was working because I was using it only to connect the charger and the batteries, I guess that means that the charger (when plugged in) has no capacitance?

The anti-spark resistor must absorb 1/2 * C * V * V each time it precharges from 0 to V volts. So the Voltage (especially) and the Capacitance are the key factors, as well as how often it must perform. If it has not had time to cool off (but the capacitors have discharged) the heat can stack on top of what it already has. So using precharge twice in a short time could push it over the top and damage it. RC voltages and capacitances are of a much smaller magnitude, as has been mentioned.