PDA

View Full Version : Possible root cause for some servo failures, check your servos


brianp6621
08-03-2011, 02:20 AM
So like many of us, I've had the mysterious sudden servo failure where it drives to a limit and stays there until it burns out. Some have hypothesized that this is due to the dirty/oxidized contact strip/fingers causing the servo to loose its position, and this is certainly plausible.

Tonight I may have found another failure mode for some of our servos.

I had some spare time so I was servicing some servos and decided to look really closely at the one servo I've had that burnt out.

First I noticed that I couldn't identify ANY component that looked burnt. How exactly do these servos fail and which part? The motor, one of the chips?

Second, I shortly noticed something very peculiar.
http://www.pbase.com/image/136932963/large.jpg
What you're seeing in the red circle is an extra tinned length of one wire from the connector leaning over and making contact/shorting with another one of the wires' solder point. On top of that, the pad it is laying on, also has an extra length of its' wire, that seems to almost be, if not actually touching, the lead from the yellow chip next to it.

So to me it seems that at least in this board's case, improper assembly allowed an extra wire end to remain, which through heavy forces exerted on it through normal flight or more likely crashes, to short to another point on the board and fry it.

I checked my other servos and saw some slight extra lengths, but nothing as obvious or concerning as on this servo.

I'm wondering if people who still have their failed servos could look at these locations and am also suggesting that some sampling of people check their functional/replacement servos for the same type of quality issue.

Who knows, this could be a rare one-off event, but I thought with all the servo issues going on with this bird, it would be worth putting it out there.

Hope this helps someone.
Brian

arabidkoala
08-03-2011, 09:30 AM
Hah. Looks like one of my soldering jobs. A bit of soldering braid + flux would clear that right up. It also looks like that QFN chip has some bridged pads on it... is that intentional?

That joint you pointed out can cause issues for your mainboard too, since it shorts one of the GPIOs from the processor to ground, causing excessive current draw. So on top of making your servo go wonky, it could also cause intermittent or permanent failures on your mainboard.

Thanks for the heads up, I haven't yet noticed any problems, but its a good thing to check.

CX2Pilot
08-03-2011, 10:35 AM
Good post, should be made a sticky.
Two of my three servos were almost like that. I cut them with a sharp nail clippers
which works very well. Cuts clean and very close to the solder.

My other 3 spare servos are fine. No excess wire at all.


Tom

brianp6621
08-03-2011, 11:54 AM
Good post, should be made a sticky.
Two of my three servos were almost like that. I cut them with a sharp nail clippers
which works very well. Cuts clean and very close to the solder.

My other 3 spare servos are fine. No excess wire at all.


Tom

Glad I could help. I did similar to you and cut the excess wire close to the solder with a fine pair of pointed wire snips.

For giggles I tried the servo again and still dead, so the short definitely kills something on the servo (rather than just making it non-functional while the short was present). I'd still like to know what actually dies so I know which part of this servo can be used to service other servos, is the motor bad or the board?

brianp6621
08-03-2011, 11:56 AM
Hah. Looks like one of my soldering jobs. A bit of soldering braid + flux would clear that right up. It also looks like that QFN chip has some bridged pads on it... is that intentional.

Actually all I had to do was snip the end off like it should have been originally.

Good catch on the other chip, but I'm assuming that is intentional. I'll have to check my other servos again when I get home.

This servo was working perfectly for a while (stock elevator servo on my mcpx) and died after 1.5 months or so, so I'm assuming the chip bridged legs are intentional and the short was the cause of my failure.

arabidkoala
08-03-2011, 10:05 PM
Actually all I had to do was snip the end off like it should have been originally.

Good catch on the other chip, but I'm assuming that is intentional. I'll have to check my other servos again when I get home.

Nippers work too, and are perhaps a better idea for a problem like this.

Looking at the chip again, it looks like those bridged pads connect to a large (ground?) trace, so I guess there is really no problem there (:

Finless
08-04-2011, 11:36 AM
Great find! I would also like to point out another reason the servo can fail. Fine dirt grains can get in the gears and can cause a hard or a bind spot. I had that happen to me last weekend. I found the problem quick enough because the heli wasnt flying quite right but when I landed the servo was hot and had I kept flying, good chance it would have burned out.

Bob

Rickesco
08-04-2011, 12:57 PM
Great Post Brian,
I had a servo go out this morning for no reason. I touched the motor and it was extremely hot. I saw your post and took it apart and had the same issue. I snipped the wire, but I think the servo motor is gone.It's intermittent. What do you think should I get just the motor assembly or do you think it may have burned out the board too and should buy the complete servo assembly?
Thanks again for your post...

Rick

brianp6621
08-04-2011, 01:02 PM
Great Post Brian,
I had a servo go out this morning for no reason. I touched the motor and it was extremely hot. I saw your post and took it apart and had the same issue. I snipped the wire, but I think the servo motor is gone.It's intermittent. What do you think should I get just the motor assembly or do you think it may have burned out the board too and should buy the complete servo assembly?
Thanks again for your post...

Rick

Sorry you lost your servo. Based on this small sampling so far it seems like more people should be pre-inspecting their servos to prevent this failure mode.

I STILL don't know what is actually failing when then happens, the board or the motor. I have yet to pursue it further to swap the motor to a known good board/or known good motor to the suspect board.

Others have indicated that it is the board because when the servos were in tight supply, people bought the mechanics which include the motor but not the board, and that didn't fix the issue, however it is still possible for both/either to fail I would think.

Rickesco
08-04-2011, 01:11 PM
I think I'll just order the entire assembly just to be on the safe side.most places are out of stock but I did find a HS that has them in stock.
Thanks again brother.. good find.

Rick

brianp6621
08-04-2011, 01:21 PM
I think I'll just order the entire assembly just to be on the safe side.most places are out of stock but I did find a HS that has them in stock.
Thanks again brother.. good find.

Rick

Just curious, did the wire with the extra end look like it was actually touching (in my case it was) or just very close to the other solder point?

Rickesco
08-04-2011, 01:27 PM
It did look like it was touching. Yeah i cleaned the silver strip and the contact points and the servo works but it's really sluggish but like I said it's intermittent.Still getting hot..

brianp6621
08-04-2011, 01:36 PM
It did look like it was touching. Yeah i cleaned the silver strip and the contact points and the servo works but it's really sluggish but like I said it's intermittent.Still getting hot..

So yours didn't fail completely dead, but only mostly dead :)

Rickesco
08-04-2011, 01:43 PM
Not completly, but wow that little motor sure gets hot. Oh well.. S*** happens.Now I wait for my new servo.

RedOctobyr
08-04-2011, 06:45 PM
So yours didn't fail completely dead, but only mostly dead :)

Is that a Princess Bride reference I spy? :) Such a great movie, with so many good lines.

brianp6621
08-04-2011, 06:50 PM
Is that a Princess Bride reference I spy? :) Such a great movie, with so many good lines.
Nice catch

rototiller
08-09-2011, 06:54 AM
I checked mine and two out of 3 had an extra long tail on them. I have a week old heli with b grips on it if that helps.

brianp6621
08-09-2011, 10:48 AM
Seems like the possibility for this to be a widespread source of the failures is growing.

Thanks for all the feedback everyone.

ki4bbl
08-09-2011, 11:23 AM
I put mine in a couple of weeks ago. Went to buy a new servo, and I was told that the assembly had to be purchased(servo, board, and worm gear). I had it sit for a while, well, until my rappy went in(I actually fly pretty well, it's just the parts keep failing ;) ). I pulled the mcpx off the shelf and started lookin at the board. A wire was barely hanging onto the elevator servo board. I cut it, used my good solder to reattach, and it works like a champ. This seems to be a QA issue with the soldering.

Greg

Uncle Chop Chop
08-10-2011, 02:05 AM
While this may not be a soldering issue, it may fix minor issues on some servos:

Noticed my heli suddenly went tumbling down to the right yesterday when I was definite I didn't put any right aileron input. So after checking it out and found the servo deveIoped a lot of slop, I tightened the nut (which I didn't even know) thinking it may work. Well, I found out I tightened it too much that the servo wouldn't work:wow2:. So I backed it off a bit, and she's all sweet.

brianp6621
08-10-2011, 12:11 PM
While this may not be a soldering issue, it may fix minor issues on some servos:

Noticed my heli suddenly went tumbling down to the right yesterday when I was definite I didn't put any right aileron input. So after checking it out and found the servo deveIoped a lot of slop, I tightened the nut (which I didn't even know) thinking it may work. Well, I found out I tightened it too much that the servo wouldn't work:wow2:. So I backed it off a bit, and she's all sweet.

I received a new servo with that nut already loose. Make sure you locktite/CA it (VERY carefully) or it will back out during flight.