#202

This is the first routing line I’ve drawn in Eagle. Did I do it right? I don’t know how to find where the 3.3V is coming from but I assume the other side of the board? In Eagle there’s a thin yellow dotted line that seems to pass through all the 3.3V contacts but it doesn’t look like the other traces. Is this some secret trace in between the board? I also noticed there are a few other resistors that aren’t connected to anything, in addition to the ones remarked on here.

22
Screen Shot 2019-11-02 at 19.17.22.png1688×502 18.2 KB

#203

no, you don’t need to connect the two 3V3 pads.

The yellow lines are ‘airwires’, indicating things that should be connected connecting. Except: there’s a power-plane on the board that will connect all the 3V3 pads. Hit RATSNEST:

image

in the control panel, and you’ll watch all the 3V3s connect when the power plane fills in. (You should always run ratsnest in EAGLE before running a rules check, to make sure you’re not missing connections).

So, how do you find what’s missing?

If you run DRC - “Design Rules Check”, Tools Menu -> DRC - you can run a rules check. There are bunch of errors depending on the rules you use, and not all of these are important. However, “Airwire” ones are - they indicate things not connected. You can click on the individual airwire errors to see them highlighted.

When I run a ratsnest, then a DRC, I’ve got two unconnected airwires in the latest git master - between R26 and R2, and between R3 and R32. So those probably need fixing, @tehn.

5 Likes
#204

Thanks so much. I never would have intuitively discovered that I need to “enable a ratsnest to show the airwire connections” but I work professionally in software and sometimes speak in regular expressions and expect others to understand what I’m saying. Learning is fun!

2 Likes
#205

The giveaway is the dotted outline on the board, indicating there’s a fill that’s not filled in. But it’s not immediately obvious.

Every EDA tool has its own eccentricities. That is one of EAGLEs.

Broadly, they are all at best as intuitive as Perl-compatible regular expressions.

3 Likes
#206

fixed, will push and update the oshpark

1 Like
#207

Can someone document a fix for the boards with this issue? I’ve got some on the way, and I don’t use eagle. Looks like it should just be a jumper between some resistors?

Also, has the image that will work with these been shared yet?

#208

Which version board do you have?

#209

git and osh links updated

fix is very simple, I can make a photo demonstrating

1 Like
#210

Mine will be the 191021 version I think. With the added resistors missing some routing. Thanks

#211

@tehn yes please pics for a fix.
Also, is there any beta image? I have all the components and RPI4 ready to build it, but no image to mount…

#212

Just realized I have some boards made on 20191007. Should I thrash them or can they be fixed as well?

#213

We don’t have a published beta image at the moment, there’s still a bit of work to do, a list of which can be found here https://github.com/simonvanderveldt/norns-image/milestone/1
(the project is currently still under my name and not the monome org because we’re still testing out this approach).

The code in the linked repo currently works on the Raspberry Pi 3 and norns, there is no Raspberry Pi 4 support in there yet.
But @ngwese has a Raspberry Pi4 and has been working on adding support for it which AFAIK is pretty much working. He can tell you more about that than I can though :slight_smile:
Note that we’re all quiet busy with multiple/different things, so it might take a bit of time till it’s finished.

2 Likes
#214

both previous boards will very likely work perfectly well without any modifications— part of the reason we didn’t notice the error in the first place.

for 20191007 you need to add three resistors tied to 3v3. i can make a diagram for this.

there are actually 3 missing traces on 191021. just use a tiny wire to bridge the solder pads. they may even be close enough for a solder blob.

23%20PM
Screen Shot 2019-11-06 at 4.07.23 PM.png1547×989 166 KB

45%20AM
Screen Shot 2019-11-04 at 8.58.45 AM.png1547×989 181 KB

newest/proper version is 191106. i’m finalizing the order of SMD-populated boards now.

osh link fixed on top post with newest.

4 Likes
DIY Norns - output is quiet [SOLVED]
#215

@tehn, I am building it right now, a diagram for 20191007 would be handy :wink:

#216

we’re working on a buildroot-based image, but in the meantime i’m going to post a raspbian-based version.

i just need to pull the kernel/libs/configs onto our existing build. hoping to get this done tomorrow.

#217

for 191007, 10k resistors between these three points

27%20PM
Screen Shot 2019-11-06 at 4.19.27 PM.png1547×989 186 KB
21%20PM
Screen Shot 2019-11-06 at 4.19.21 PM.png1547×989 185 KB

2 Likes
DIY Norns - output is quiet [SOLVED]
#218

thanks a lot, I was starting to scratch my head…“where are those R2, R3 and R4???” :slight_smile:

#219

currently struggling with freecad to cut out a 1590bbs hammond enclosure. freecad is not very user-friendly :confused:
image

5 Likes
#220

Board is built. i miss a couple of 47u but I guess I can test it withouth them, hopefully

Today I woke up and the norns.img was created. (it took hours…code and instructions taken from here: https://github.com/simonvanderveldt/norns-image)

I flashed it using balena, but the RPI3 doesn’t even boot, I even tested it with a standard HDMI cable connected to a monitor and no shield. The same RPI works and boots normally using a Raspbian image…

Is the norns.img file enough for it to boot? should I copy something else?

#221

Oh and BTW, the silkscreen for the 2x40 RPI connector is probably on the wrong side of the board, at least on my 190909 board.
It seems it’s still in the 191106 board:
(it’s pure joy when you realize that while soldering your last pins…) :slight_smile::cold_face: