Help needed for a SMT easy fix(did I fry my board?)

#1

Hi everyone!
I accidentally touched R9 with enough strength to desolder one side of it... I bent it back in place and apparently everything is working. :sweat_smile:
Now, does someone know if it's a good idea to drop some superglue over it? I would avoid resoldering it as I feel my skills are not up to the task... any "easy fix" suggestion is greatly appreciated!

#2

Such a fix is likely to break down over time, and you'll run the risk of unpredictable performance/behaviour. You'll also make it more difficult to fix properly in the future.

Resistors are relatively easy to solder. If you have flux, apply a little dab to either side of the resistor. Make sure your soldering iron tip has been cleaned (e.g. on a sponge or scrubber) and lightly tinned with a small amount of fresh solder. Using a pair of tweezers in your off-hand, apply light pressure to the middle of the resistor, then very gently touch the solder on the side of the resistor where it's broken with the side of the tip of the iron. After a few seconds, you should see the solder liquefy - maintain pressure with the tweezers, but gently take the iron away. After another 10 seconds, the solder should have solidified and you can remove the tweezers. Because you have bent the resistor back, you will have fatigued the solder and stressed the PCB trace on the other side. You can fix this by repeating the process on that side of the resistor, after which the circuit will be almost as good as new.

The important part is to be patient and gentle. Don't do this when you're tired or over-caffeinated. Don't stab with the iron - there are some very helpful videos on youtube if you need to build your confidence.

#3

Thanks for the tips!

Ehm, that's now my normal condition as a father of a newborn baby... and probably the reason of the mistake in the beginning.

That said do you discourage me from any alternative way? I solder tons of audio cable with a big and cheap iron, and that's about everything I'm confident doing... If you and anyone with more experience than me confirm the low reliability of glue or any other way I'll try, but I'd LOVE not to...

#4

Also I'd love to know if R9 is something the board won't work without...

#5

From the schematic:

R9 is a 47K pulldown resistor from BOOT0 to GND. It sounds important. I wouldn't trust it to reliably not float. It might work today, but temperature and humidity are going to be your enemy. If this is in your studio, I would just risk it, but if you want to take this unit on the road, I recommend fixing it.

Here is a video for soldering a resistor:

If you're hesitant, you can buy circuit boards to practice soldering. I think you can pick it up in less than 10 tries. You can find SMT Practice Kits very cheaply. It is very intimidating at first, but you can fix your board in less than an afternoon of training.

One of my favorite SMT tools is SMT No Tack Flux No Tack Flux. This will be messy, so it's not recommended on this board, but if you have the time and money, try it on a practice board. It makes a stressful experience really sloppily, easy. You have to clean it with 99% IPA or distilled water, but it's a great help with SMT soldering.

#6

Do you mean Indian Pale Ale ? :smiley:

Ok, sorry... thanks for all the info.
I think i have some broken SMD circuit lying around. Tomorrow afternoon I'll try remove and resolder some resistor... thank you both for your help!

#7

After some experimenting with a spare fried circuit I felt brave enough and did the soldering.
To to the eye it's just a tiny bit off axis, but looks good enough, but actually it's not, 'cause now the board does not power on! I really don't think I melt the resistor, I can still see clearly 473 written on, and I was quick, no more then 3 or 4 seconds on the spot... I tried to take some pictures to post for an opinion but resolution on my phone wasn't high enough. What would you suggest as the next recovery attempt?

#8

That's a bummer. I don't think you have damaged the component, SMD resistors are quite resiliant and a soldering time of 3 or 4 seconds shouldn't cause any harm. Besides, they are made of some ceramic compound so they won't melt the way ordinary plastic would under those circumstances for instance.

Do you have any way of taking a sharp close up photograph and posting it? That way we might be able to see what has happened.

I'm also a bit confused as to how you caused the resistor to lift from the board in the first place. Were you soldering nearby on the PCB?

#9

I'll try to find someone willing to lend me a good camera, maybe tomorrow.

As stated earlier in the post I'm the proud father of a month old baby, and lack of sleep was the main cause of the accident: I was simply bolting the board into the case of my new multiboard (sigh) set up and touched the resistor turning the wrench to fasten the nut. What a dumbass...

My concern now is: may I have fried the board turning it on when R9 was not properly in place?

As for now, thank you all for your time!

#10

SAM_1216.JPG4320x3240 2.6 MB
SAM_1219.JPG4320x3240 2.64 MB
SAM_1218.JPG4320x3240 2.56 MB

Those are the sharpest pics I could make.
The soldering looks ok to my unexpert eye. I'm tagging @johannes in the hope he chimes in with answer to my previous question (now added to the title): is it possible that I fried the board turning it on while R9 was not touching properly or maybe shortened?

#11

If R9 is not properly connected, Axoloti Core may (it is an undefined condition) boot in "DFU mode". In "DFU Mode" there is no LED activity, but Axoloti Core enumerates as a "STM DFU bootloader" device on USB, and should show up in the dialog when you go to the "Board" menu and choose "select device".

No, without resistor R9 the logic level of Boot0 is just undefined, and this selects between booting into DFU mode or booting regularly.

#12

It does! I had tried to connect it to the patcher but I didn't check the devices list.
So, despite the look either R9 is not soldered properly or it's damaged, right?
If that's confirmed I'm going ahead with my brand new SMD hand soldering skills and try to replace it.:sweat:

#13

the right side of r9 looks like it is shorting out, do you have a multimeter? try measuring the resistance on the two solder points. if you get 0, you have a short there.

#14

Hi there :slight_smile:
the resistance I measure is 46.8k, so I guess no shortening there. On the other hand isn't it strange that I get the exact value of the component while it in the circuit? May this mean it's not connected? Sorry if the questions are silly, my knowledge in electronics is limited to the pre WWII technology you find on electric guitars..

#15

so next step would be to measure from gnd to the other side of the resistor and see if you get a reading of 47k as well which would indicate that that side of the resistor is properly soldered. not sure if you can actually put your probe on the boot0 :slight_smile:

#16

I assume left/right related to the writing "R9":
left to gnd is 0, right to gnd is 46.8k. Is this expected or left is not soldered properly?
Man, your helping me a lot!

#17

And BTW I'm right now including a mod of your octaver patch in my live setup.

#18

this is to be expected, one side is connected directly to gnd, so should read 0, the other side has a resistance of 46.8k to gnd (since the resistor is between gnd and your multimeter tip) so the right side (as per your post) seems to be not soldered properly. was this the one you did solder?

#19

I ended up removing the resistor and resoldering both sides, so chances are 11 10 01 00.
Anyhow I'll try to resolder tomorrow morning, thanks a lot for now!

#20

Great, that supports the idea that nothing is fried.

Damage to the resistor itself is unlikely, I'd not try to replace it, but just re-melt the solder with a soldering iron, and make sure it forms a nice solder joint.