Midipal is dead after assembling

Dear all,

just finished assembling the midipal.
Unfortunately it shows no signs of life. LEDs dead when connecting Midi In and Out, display also dead.
I have measured 3.3 volt at the points described in the assembling instructions but this is all I can do.
Is there a step by step trouble shooting way in order to check the parts and soldering points?
I believe my soldering skills are not that bad but I have no knowlede in microelectronics, so I do not know how to find the mistake in the built.
I would be grateful if somebody could help.

I found the most important thing was to use no pressure at all on the MCU legs. Just the weight of the iron. Let it gently drift across the legs drawing in the solder as you go. After a couple of passes, you should have clear solder joints.

If you press on them, they buckle and come away from the pads, and that causes more troubles. Low heat, no pressure, lots of flux from a flux pen. Thin solder too!

I should add that I have spent so much time and used so much solder wick on this vector board/perf board project of mine. I don’t even want to talk about it. But I think there is a special place set aside for people who perservere through projects like this. Don’t give up!

I think pin 19 an 20 could be shorted as well. You should see my FrankenPal after replacing the whole MCU (the first was dead, still don’t know why) without hot air smd rework tool, there are I wires… With time, a good soldering station, a good wick and good flux one can do amazing things. :wink:

For smd works, 30w is fine. Using flux is pretty vital there.

For repairs try a High Power computer controlled Iron with low temperature, so you can be shure it holds the low temperature no matter what/how long you touch with it. I work at 300° - but i have a rather good 60W Soldering Station…

Nope! Just hit it (carefully) with another drop of solder. Don’t overheat it, don’t make a cold solder joint. If that doesn’t help, let’s find the trace where it leads to and carefully solder a thin wire to there.

what temperature do you recommend for soldering this kind of thing?

thank you all for your help so far.
anyway, I will get a thinner solder iron tip tomorrow and continue working on the thing.
have a nice evening!

had it partially running! but no display backlight. showed the monitor app and registred incoming notes.
after 10 sec it broke down again. now completely dead again. I think that on ethe pads (below pin 3 from bottom left) has lost contact to the board because of too much too hot soldering iron penetration.

is this the end of the story?

Ouch. The two pins on the bottom left are for the crystal - with them shorted the MCU cannot boot. To confirm - because sometimes it’s hard to probe the tiny pins of the MCU - check that there is no short between the left and right pads of the crystal.

The pins on the top left of the MCU are the LCD enable pin and the MIDI in/out respectively. Shorting them will cause serious problems on both the display and MIDI transmission.

it will be pretty tricky not to destroy the display trying to remove it. You better try with the display in place.
Put some flux in here and use solder wick, this should be enough.

Damn! Checked the IC2 again and found some shorts! When assembling I did not find them, strange.
I have a meter set to buzzer mode which I use for checking the pins. Is that the right way?
I guess there are no internal connection in the IC that make the buzzer beep? Power is not attached.
Please see below a picture of the infected pins.
How can I access the upper ones, the display is in the way. Is it possible to remove the display without destroying it?

How about getting a solder wick in between the display and the CPU and gently trying to remove some of the excess solder?

Some pins could be the same, at least on a very large IC package. You would have to check the data sheet to be sure. However, let’s assume that’s not the case here for this small CPU.

Check the lower pins on the right side of IC2. The third one from the bottom looks a bit funky, maybe it’s just the picture.

That is good news!
I will check the soldering points. Hopefully this helps.
I attach a picture so you can have look at my work.

Yes, the sqares are filled.

Ok… Then it means that the most likely cause of the problem is that the processor just doesn’t boot. Check the soldering of the crystal, check for shorts between adjacent pins of the MCU.

Encoder shouldn’t be a cause of the problem you describe. At the very worst you can get inverted rotation or click / rotation swaps with some encoders with different pinouts ; but shouldn’t explain a MCU that doesn’t boot or an unresponsive MIDI interface.

One thing: if you turn the contrast trimmer to one of its extreme positions, do you get filled squares on the LCD?

I will make a picture and post it, so you can have a look.
One thing I forgot to mention: I destroyed the encoder by soldering it to the wrong side of the board (don´t ask why i did this!).
I have replaced it with the the following item, to me it is excactly the same as the original one, but maby this is not the case?


just in case, vincent, could you send a photo of your assembled board?