Welcome to Multi-Rotor UK. Please login or sign up.

Monday,November 25, 2024, 21:23:02

Login with username, password and session length

Shoutbox

Bad Raven:
12 May 2024 08:13:51
 I have some F1 Abusemark boards going spare,,,,,,,,,,,,,    ;)    :azn
DarkButterfly:
11 May 2024 22:12:29
And with oldskool parts  :D
DarkButterfly:
11 May 2024 22:11:57
I must be the only one doing tricopters right now  :laugh:
DarkButterfly:
11 May 2024 22:09:30
 :D
Gaza07:
11 May 2024 21:15:16
Domain has been renewed closure has been cancelled  :D
Gaza07:
02 May 2024 08:07:52
Who are most people ??? I think the person you are referring to has put in a lot of effort to keep things moving  :rolleyes:
hoverfly:
01 May 2024 10:16:12
Most people I have spoken to are pizzed off with the yellow peril  flooding the forum,go figure. :whistling:
Gaza07:
23 Apr 2024 08:09:45
The Domain expires for the forum in 60 days, I'm not going to renew it this time unless I see any activity  :beer2:
Gaza07:
20 Apr 2024 18:02:50
Is there anyone who would like to see this forum stay open ? :shrug:
hoverfly:
17 Apr 2024 17:15:13
 :rolleyes:
Members
Stats
  • Total Posts: 201,498
  • Total Topics: 20,274
  • Online today: 22
  • Online ever: 530
  • (Tuesday,June 26, 2012, 08:34:46 )
Users Online
Users: 1
Guests: 15
Total: 16

Theme Changer





3d - Printworx

F3 FC Magic smoke

Started by Powernumpty, Wednesday,October 19, 2016, 16:17:01

Previous topic - Next topic

Powernumpty

I built a wierd quad from a Flycolor FC with ESC';s on a PCB, not sure quite what I was thinking when I bought that but it was doing OK for about six batteries then the magic smoke came out of the FC lower side (BAT54C Shotky diode seems to have expired).
http://www.gearbest.com/multi-rotor-parts/pp_357756.html?wid=21

So people can call me an idiot I had a laptop and was connecting the USB while the battery was still connected to change PID settings.

Question does anybody else do that, am I mad to even have considered it?

Didn';t see any issue until snagged the bleeper switch on the Taranis while connecting the USB, was I just lucky up to that stage or was the bleep just the last straw?

I';ve ordered some spare diodes and will carefully go through testing everything but thought I tell the tale and wait for advice and/or derision.

Revs

Connecting the USB while the lipo is connected shouldn';t be an issue. I looks like an SPRF3 clone. I';ve seen loads of comments about these smoking, I don';t think it';s owt you';ve done buddy.

Powernumpty

Cheers for that, was one of those times when I had to step back and say "wait a minute, was that me".
Wanted to avoid cooking the laptop USB port as well.
Will be intersting to see if I can get it going again as otherwise I';ll be desoldering the esc';s from that PCB for use somewhere else (at least that is possible, so not total loss).

So you';ve heard of others, bugger.

Powernumpty

Just to fill in some more info, I replaced the diode without checking too much else, instantly smoked on USB conection.
Spent some time (I should have before replacing it) looking at what was connected to ground after that diode and the input of the voltage regulators after the diode seem to be grounded, now I';m not sure if replacing the voltage regs is worth doing but as they should be pennies and I';ll learn something from it I might give it a go if I can definitively work out what a DE=C03 sot23 5pin package is. I assume it is a fairly standard regulator but
don';t want to just burn the STM by fitting something else.

While looking with a lupe I noticed what I thought was a fine strand of wire near the STM32 pins so brushed that off thinking that may have been the cause and how careless of me to leave a microscopic strand like that on the board (couldn';t even work out what multi strand wire it could have come from as it was so thin) further inspection revealed more, like the board had been sprayed with thinner than human hair strands of solder about 1-2mm long, enough to short fine pitch SMD.
I';m guessing solder as they look silver, thinner at either end,  but I could not work out how to hold the lens and test continuity of the bits, either way the same was not on the underside of the board so not sure. Might try some sticky tape to pick one up and see if I can arc from it with a current limited power supply. Has spooked me really as I don';t bother to look that closely at bought flight controllers but from here on probably will.

Cheredanine

Ok so some bits:
As revs said,  to only should connecting the USB and battery at the same time be possible, it is necessary to calibrate escs.

These combined esc/pdb/FC boards - I stay away from them, blow one component and you are either trying to redo smb soldering (as you have), physical damage from a crash is expensive and more likely, and contrary to what EASA seem to think, Chinese mass production soldering is a lot less reliable than Self built

In my experience when a component blows like you had, it can produce those thin solder strands (this is peculation) you found, I have frequently found them on escs I have blown after the component blew,

I really think you are chasing something that will continue to give more problems, I would recomend binning it and using your own components in principle (if they fit)

Powernumpty

All noted, I just wanted to look at the pro and cons knowing I could recover both FC and ESC';s if I didn';t like it.

The bits on the board are from something other than the blown diode, those things blow under 1amp/5V so the energy involved is enough to raise the top and let the smoke out but not much else, the diode is on the underside of the board so any bits would need to get round and onto the top. This I think is just poor manufacturing.

Just tried to find an example but I spent some time blowing bits off the board was almost out of luck.
It looks like solder to me as you can almost make out the crystalline structure, the ones more in the light were shiny like solder.
I wish I';d have photographed the thing out of the antistatic bag.
This is one between the pins of the STM32F303


I understand this would be a questionable board but as I fly away from people and its a 200mm frame I';m prepared to flag it as unreliable and fly with that in mind.
I hate waste due to unknown failures and don';t want to walk away until it has taught me a little more.

Cheredanine

Yeah my speculation mate is that the blow out on the sd was caused by a lot of current which alsomelts the SMD joint holding it in place, particularly if there is still paste in place, when it blows the solder gets blown clear, in liquid form this rapidly solidifies, producing the hairs, but is is pure speculation, just noticed that effect on boards where I have had components go bang (mostly escs) and not seen on any of the brother boards

Powernumpty

On ESC I';m sure you are right, in a previous job (battery power grass cutter, wheelchairs etc) I';ve blown individual FETs good for about 150A + and had bits of the top cover embedded in the lab wall at about eight feet, needless to say as a fairly nervous sort I hated with a passion testing those things and at various prototype stages when testing software based "stall" current limit the boss would fire it up from underneath the bench, perspex shields are OK but a good thick bench is better. I still tend to cringe if I see someone connecting a couple of sizable lead acids to something beside me.

Sounds like old man talk that does, "don';t talk to me about splatter, back when I was a lad..."

I';m guessing this will come to nothing but I';d truly love to get it going again just so I know what happened and can hopefully avoid it happening next time.