DIY: norns shield

[asg]

Not necessarily. The HDMI output is disabled in the Shield kernel, so you’re not going to get console or anything on a plugged in monitor.

Ah. Okay. Understood.

After you boot up, you can try to connect over SSH using Ethernet if you have a computer with that available. That will at least tell you if the system is running and you can run some command line tests like dmesg to see if the display is initializing, etc.

I can connect over SSH with an ethernet cable. I ran dmesg and got a LOT of lines of text. Is there something specific I should be looking for to know if the screen is initializing?

I’m seeing this error a lot: snd_soc_register_card() failed (-517) (although I see that’s my DAC…that wouldn’t keep the screen from working would it?)

[okyeron]

dmesg | grep snd to look for soundcard/DAC related messages - like the one you posted.

working shield output

[ 0.960158] snd-rpi-monome soc:sound: ASoC: CPU DAI (null) not registered - will retry
[ 0.960178] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 1.244416] snd-rpi-monome soc:sound: cs4270-hifi <-> 3f203000.i2s mapping ok
[ 1.246024] #0: snd_rpi_monome

dmesg | grep fb can look for fb/framebuffer related messages.

Good fb messages copied here

[ 0.965735] fbtft_of_value: buswidth = 8
[ 0.965750] fbtft_of_value: debug = 3
[ 0.965761] fbtft_of_value: rotate = 180
[ 0.965773] fbtft_of_value: fps = 20
[ 0.965808] fb_ssd1322 spi0.0: fbtft_gamma_parse_str() str=
[ 0.965822] fb_ssd1322 spi0.0: 1 1 1 1 1 2 2 3 3 4 4 5 5 6 6
[ 0.965880] fb_ssd1322 spi0.0: fbtft_request_one_gpio: ‘reset-gpios’ = GPIO6
[ 0.965908] fb_ssd1322 spi0.0: fbtft_request_one_gpio: ‘dc-gpios’ = GPIO5
[ 0.965947] fb_ssd1322 spi0.0: fbtft_verify_gpios()
[ 0.965960] fb_ssd1322 spi0.0: init_display()
[ 0.965972] fb_ssd1322 spi0.0: fbtft_reset()
[ 1.107704] fb_ssd1322 spi0.0: Display update: 3579 kB/s, fps=0
[ 1.107720] fb_ssd1322 spi0.0: set_gamma()
[ 1.107865] graphics fb0: fb_ssd1322 frame buffer, 128x64, 16 KiB video memory, 8 KiB buffer memory, fps=20, spi0.0 at 16 MHz

not 100% sure if the DAC not working would affect the screen.

Make sure all your solder points for the display are good, along with all the pins on the RasPi header.

[asg]

dmesg | grep snd gives me this:

[ 0.969631] snd-rpi-monome soc:sound: ASoC: CPU DAI (null) not registered - will retry
[ 0.969649] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 1.196790] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 1.196809] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 1.253423] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 1.253442] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 1.294918] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 1.294940] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 1.624409] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 1.624429] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.074186] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.074205] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.811793] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.811803] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.812035] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.812040] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.813983] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.813989] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.822160] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.822167] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.851280] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.851287] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.852577] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.852583] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.858631] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.858640] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 2.964267] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 2.964276] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 3.075392] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 3.075403] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 3.082606] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 3.082696] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 3.370732] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 3.370739] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)
[ 3.507636] snd-rpi-monome soc:sound: ASoC: CODEC DAI cs4270-hifi not registered - will retry
[ 3.507645] snd-rpi-monome soc:sound: snd_soc_register_card() failed (-517)

dmesg | grep fb looks like the good fb messages:

[ 0.974578] fbtft_of_value: buswidth = 8
[ 0.974592] fbtft_of_value: debug = 3
[ 0.974602] fbtft_of_value: rotate = 180
[ 0.974612] fbtft_of_value: fps = 20
[ 0.974643] fb_ssd1322 spi0.0: fbtft_gamma_parse_str() str=
[ 0.974655] fb_ssd1322 spi0.0: 1 1 1 1 1 2 2 3 3 4 4 5 5 6 6
[ 0.974703] fb_ssd1322 spi0.0: fbtft_request_one_gpio: ‘reset-gpios’ = GPIO6
[ 0.974727] fb_ssd1322 spi0.0: fbtft_request_one_gpio: ‘dc-gpios’ = GPIO5
[ 0.974761] fb_ssd1322 spi0.0: fbtft_verify_gpios()
[ 0.974773] fb_ssd1322 spi0.0: init_display()
[ 0.974784] fb_ssd1322 spi0.0: fbtft_reset()
[ 1.117683] fb_ssd1322 spi0.0: Display update: 3595 kB/s, fps=0
[ 1.117697] fb_ssd1322 spi0.0: set_gamma()
[ 1.117831] graphics fb0: fb_ssd1322 frame buffer, 128x64, 16 KiB video memory, 8 KiB buffer memory, fps=20, spi0.0 at 16 MHz

I went back over all the headers with a multimeter and then reflowed everything. I’ll check those again.

[cementdrums]

Is there a .cpl file available for the norns shield board for ordering smt assembled pcbs?

[forestcaver]

Generate it from eagle - it takes about 5 seconds… for a lot of the Chinese fabs eg jlcpcb you can even find scripts that generate the files in the correct format…

[asg]

DAC working now. I have the same messages as what you posted.

I’ve reflowed all the headers and still got nothing on screen. I don’t know what I’m missing.

Would it matter if I opted for the white instead of the yellow screen?

One more thing - when I retested continuity between the GPIO and the header for the screen, everything checks out, except pin 18 (which isn’t on the schematic) is connected to ground. The data sheet says

Active LOW Shutdown control pin for boost converter
(pulled HIGH via on-board 15kΩ resistor)
Can be made a No Connect by removing resistor R1

Not entirely sure I understand that, but I’m wondering if it being connected to ground is the problem. Maybe?

EDIT - that was it. I pulled R1 from the screen pcb and I have a screen that works! Now the next bug to squish: Supercollider Fail.

[asg]

Well… Almost.

• supercollider fail error message on start up
• encoders seem a little wonky
• no sound
• and trying to load an app, it stays stuck on “loading”

[raaaaaaaaaab]

These were all symptoms I had from bad soldering on the crystal

[asg]

These were all symptoms I had from bad soldering on the crystal

I took a look at the crystal. Reflowed the pads and then checked the connections. I couldn’t find any shorts. Everything’s connected to what it should be.

I got into maiden and on the SC tab it just says “Not Connected to SC.” Refreshing doesn’t do anything and the ;restart command doesn’t either.

On the matron tab ;restart tells me Script error: supercollider fail.

Is it possible I’m looking at a faulty DAC?

[j450n]

Hi; I’ve got one working norns DIY shield and one not working. I’m hoping people have advice.

[Update: I removed the screen cleanly, and the Pi still won’t boot with the shield on (now perhaps that never happens if the shield isn’t attached, but I don’t know yet.]

1. the same Pi for both
2. the red light comes on for the non-working shield, but not the flashing green one.
3. reflowed everything & done a manual comparison under a stereo optical microscope
4. I soldered the screen onto the shield rather than using removable headers because otherwise the screen’s too high off the shield, but I can desolder it (have a good soldering vacuum).

Picture attached. The soldering’s not as good as it was, due to the reflow.

Thanks for any thoughts/suggestions.

[kbra]

Just a quick thought without zooming: the fourth column of PI header counting from the left looks it’s pretty badly broken - does it just look like that or could that be bad?

[j450n]

THanks for looking. It is actually fine. I checked the continuity from the Pi through the header for them all. I actually had to take 6 pieces of header and sand them down to fit. The person I’m working with who did the ordering part of the project bought some wrong parts. I was teaching her how to order parts, and I guess I’m not a good teacher! Notice that lots of the caps and resistors are 0603 metric, not imperial; the super tiny ones. I’m just desoldering the screen and will see if I can get it working without the screen. And without the screen I can get at some of the tp on the other side of the board, though I doubt I’ll know enough to do more than just poke about.

[ngwese]

@j450n

Thanks for any thoughts/suggestions.

With the caveat that it has been awhile since I’ve looked at the circuit layout, stuff which sticks out to me include:

• R18 - looks like it is completely bridged, is there a resistor under there?
• the via near the lower right corner of the DAC (left of C6 silk screen) looks like it has enough solder to possibly be bridging two of the legs on the DAC
• C6 - hard to tell from the picture if it is connected to pad on the C6 side

Given that you mention it doesn’t boot the first thing I would do is remove the shield entirely and verify that the Pi will boot by itself with the norns image on the sdcard. If it does boot then that establishes that there is a fault on the board significant enough that it prevents the Pi from even getting off the ground - which would lead me to the header and looking for places where solder bridges are shorting things to ground which shouldn’t be and anything related to the power plane.

[j450n]

Thanks for your thoughts/comments.

The extra soldering is when I reflowed everything. Had the problem with the original minimalistic soldering.

Regarding the DAC, you mean pins just above C6? They’re not bridged, but it wouldn’t matter as they both go to the via next to it. I just confirmed that on a new PCB I’ve not touched yet.

6C is connected on both sides. Just checked again with the stereo microscope, and reflowed it anyway.

R18 IS completely bridged. It’s a zero ohm resistor, so a bridge is just as good. I checked with my electronics gods on that one.

I did test the Pi on my other working shield and it boots fine and runs.

You’ve confirmed my thinking. Thanks. I’m going to remove the header and start again. Though for my sanity, I’m going to do the third board I wanted to build and hopefully I’ll get that working. I had to make up 10 boards, so if I have to just toss this one (remove the pots/buttons and DAC) I should survive. I’m sad because I did a way better soldering job on this one than the first one, and the first one works.

THanks again for going over it and getting me to check things.

[kyanrinner]

i’m having an issue with my norns shield turning on, not sure what’s going on. it’s been running fine with no problems, hadn’t used it for a couple days and when i plugged it into the power bank there’s no response from the screen at all. the red light on the raspberry pi is on so i’m not sure what the issue is. i just recently got a midi fighter twister which i had used with the norns for the first time last time i used it, otherwise i haven’t done anything different. i can’t imagine that breaking the norns? any idea as to what happened/how to fix?

[kyanrinner]

just did a full reset and the norns seems to be working fine now, although when i first tried to re-flash the drive i got an error message. second try worked fine. still curious as to why this may have happened…

[SPIKE]

been having an issue with my norns Shield.

it tends to just randomly reboot itself.

i’ve multiple micro USB power cables and such but still no luck.

any ideas on how to fix it?

thanks for any help!

[asg]

I went back over all the solder points and reflowed them. I plugged it in and no more Supercollider error. I even managed to open the oooooo script successfully.

But then I boxed it up, plugged it back in and it went back to the same Supercollider error as before. I’ve got a spare pcb and the schematic and am tracing out connections looking for shorts or solder bridges, but I haven’t found anything yet.

I’ve tried rebooting to no effect. I’ve tried reflashing the microSD, to no avail. The fact that it worked once makes me think that the DAC is okay and that this is probably a soldering issue.

Here’s a photo of the board. Maybe someone sees something I don’t.

[alley_cam]

Hi there! Do you have a pic of the Pi header?
Was the system responsive when it did work or did it feel sluggish?

I would give everything a good clean with q tips and iso alcohol if you have some around the house. Makes spotting sneaky solder issues much easier.

[asg]

The one time it worked, it seemed to work great. OOOOOO started right up. I didn’t have any headphones plugged in (that was my mistake…I got excited it was working and wanted to put it in the case).

You’re right. I should have cleaned off the flux. Here’s a better picture and one of the Pi header solder joints.