#1518

I would take the crystal off the board gently, this pcb finish can only take so much action before pads start popping so be quick but relaxed.

Wick the 4 pad areas and get as much solder out as possible, give it a test then, may do the trick, if not, try safely remove the crystal, wick everything off the pads and the crystal, replace / resolder.

I would also apply some flux on the codec legs and even out the excess solder, it should be smooth and not “craggy”, keep an eye out for bridges

Turn up the heat and work quick.

Cheers!

#1519

I cleaned up the DAC legs, but I can’t get the crystal off the board.

I tried wicking away as much of the solder as I could (but this solder wick has never worked well for me). This is the first time I’ve soldered something with the pads under the component and I’m afraid I’m either going to destroy the crystal, the pads or both if I keep trying to remove it. So I stopped and checked the connections to the crystal with my DMM. Everything is connected where it should be. No bridges that I could find either by eye or with the meter.

#1520

Is the connection between c12 and c13 ok? Also are c11 and c13 bridged over by the via?

#1521

Those crystals can be an absolute beast to solder correctly. I had a problem with a Mutable Plaits build where everything seemed perfect, but there was obviously something awry. I checked the crystal under a microscope and with a multimeter and everything looked okay. It was only when using a loupe and viewing the crystal side-on that I could see a tiny gap on one corner of the crystal. A liberal application of flux and a bit of extra solder and it was perfect.

2 Likes
#1522

This has been a learning process for me, but I’m determined to get it working. I’ll revisit the crystal when my hot air rework station arrives tomorrow (because it’s clear to me I’m not going to manage this with just my iron :slight_smile: )

2 Likes
#1523

Hot air is magic for these purposes. I would recommend grabbing some Kapton tape to cover the Pi header / 3.5mm jacks.

Use some type of shield to prevent the plastic from getting melty.

2 Likes
#1524

The crystal was the culprit. Once I got it off the board (btw, SUPER easy with the hot air), I could see that one pad had barely any solder on it. Cleaned it up, got an appropriate amount of solder on the pads, stuck it down again, and presto! No more Supercollider error.

Next…sound. I plugged headphones into the output and got no sound. Does it need to be amplified or should I be able to hear something with just headphones?

1 Like
#1525

I run directly into headphone with no problem and plenty of volume…

#1526

understood. I’ll be looking at the output section next then.

#1527

just as a sanity check: which patch are you running - are you sure it should be making sound - and did you check that the output volume in the norns mixer is >0? if you’re new to the device, it’s probably a good idea to confirm whatever you’re doing should make sound in the first place.

1 Like
#1528

Yes, good call. I did switch to the Awake script before I plugged in headphones. The bars on out are moving like something’s playing but I don’t hear anything. This is what the mixer looks like:

IMG_5866
IMG_58663024×4032 1.54 MB

1 Like
#1529

Give it a SYSTEM->RESET

1 Like
#1530

I did a reset & that didn’t change anything.

I traced the connections. Everything checked out, but even so I reflowed the output components. I heard the Awake script for about 10 seconds before it started getting really quiet and then it was just static.The volume didn’t actually change the volume anymore except when it was all the way down, but even then I could hear kind of a hissing noise. I reset again after that and got the same static & hissing.

P.S. - thank you all for your help and patience while you help me figure this out. It’s immensely appreciated.

1 Like
#1531

Getting closer! Still sounds like it could be Codec related.

Now that you have the hot air, I would reflow the codec (if you haven’t already)

Preheat the board in the area by holding the nozzle at a bit of a distance for a little while, get some flux on the codec. Hold the codec down with the tweezers by applying a bit of pressure using the point of the tweezers, line the legs up nice. When the chip starts dancing, it’s flowin’

Keep an eye on your heat and the surrounding components to make sure they stay put.

#1532

Nailed it. I have a working Norns Shield now.

I looked more closely at the codec, got the hot air on it and realized that some of the legs didn’t really have enough solder to make a really good connection (kinda like what happened to my crystal). I guess in my fear of getting too much solder, I didn’t put enough. Once I got that sorted out, everything works great!

My takeaway: If the crystal and the codec ain’t happy, ain’t nothing gonna work :joy:

Thank you again for all the help.

IMG_5867
IMG_58673024×4032 2.38 MB

7 Likes
#1533

Silly question: when I plug a mono signal into my Norns shield input, it only turns up in one channel. How can I make it turn up in both?

1 Like
#1534

Hi;

I’ve built 2 working norns (which work fine), but have trouble with the first one I tried to build. It ‘works’, but if I turn E1 the Pi reboots. If anyone’s ran into a problem like this, I’d love to hear about it.

Note, that depending on where you turn E1 the signal goes to ground. Don’t know why, but it happens in my working norns. However if E1 is in the ground part of the turn, the Pi won’t reboot, and stays off. So when you adjust E1 out of the ground state, it starts up again. I’ve reflowed my heart out especially around those R2, 26, 28, 29 and C 35, 36. but not sure what to do next. Should I start swapping them out?

Screen Shot 2021-01-09 at 5.01.35 PM

1 Like
#1535

Howdy! I have had this happen with a switch a few times, typically reflowing the pi header thoroughly has solved the issue.

Go nuts with the multimeter in continuity mode and make sure the everything is groovy at the pi header.

Next step would be to take off the encoder and take a look at the through hole pads. If you are using an Oshpark board, those can usually take a beating before any issues arise but it is still a possibility.

I have had an instance where one SMD pad for a switch cap had a tiny crack on the corner that was shorting to ground. Could have been due to hot air or a defect.

Best bet would be to take off the SMD components, take out the encoder, clean everything up nice for inspection, and then reset

Good luck! Cheers!

#1536

Thanks. I’m using an Oshpark board. Strangely enough, it can survive my beating. Out of the 20+ module kits I’ve built, this was the first time that I had a pad/trace detach. Had to add a wire from the pin to the via.

You’ve at least motivated me to put it aside and start building a fresh shield with the parts. I have more shields than I need, since i had to make up a batch. Cheers!

1 Like
#1537

So, I got my heart set on a DYI Norns Shield but it seems nearly impossible to get hold of a CS4270 codec. I emailed Mouser and they gave me a date in late july 2021. Does anyone here have any good leads on any sites which might me be able to sell me one? I’m based in europe (Sweden)

Thanks in advance