Freeplaytech Forum

Just right off the bat to ease and speed up other troubleshooting steps, I've already gone through the troubleshooting section.

• I'm using the Freeplay Zero
  
• I've formatted and reflashed the sd card.
  
• I've disassembled and reseated everything.
  
• The freeplay is currently outside of the case while troubleshooting
  
• The ribbon cable and gpio header (both sides) are free of any noticeable defects and look good. There are no solder bridges or visibly damaged components.
  
• I can hook up to an external monitor and see everything just fine, and I can use a keyboard.
  

I posted on here earlier regarding buttons not working, and I was able to narrow it down and solve the issue thanks to this forum, so I figured I would post in here about this as well. I was able to play games fine, and took a break to make dinner. When I turned the freeplay on after dinner, the screen stayed white for about 3-4 minutes before I finally turned it off (by holding the power button for a bit).

I believe it's almost certainly a problem with either the raspberry pi, or the freeplay zero board. Has anyone else ran into this problem before? Freeplay display full white, but external displays and other peripherals working fine? I've been trying to get it back in working order for about 3 hours now...

This could be caused by multiple things. We recently had another builder experience a similar problem. I told him to try this, and it fixed it for him.

"I’d maybe try hooking it up to HDMI. Then, at the RetroPie menu, look for "Freeplay Change Display Driver" settings or some such. See if changing that helps anything."

That will likely work if it's a software issue.

If it's a hardware issue, that probably means that one or more of the connections between the Pi and the LCD aren't making a good connection.

An all white LCD basically means that the LCD is getting power but not getting initialized with any data.

The first/easiest thing to check is the LCD's ribbon cable attachment to the Freeplay board. I know that this is an old/poor video, but I think it gets the point across well. Check it out and make sure that the ribbon is fully inserted and the shoulders are locked down all the way. I have helped some people troubleshoot this issue where the shoulders weren't locked down. AND, it's a bit of a problem that sometimes it will work that way, so people have built it up and had a working unit and then it just stopped working all of a sudden. I think it was making a little contact and then something shifted and stopped working. Anyway, locking the shoulders should prevent any intermittent connections.
https://www.youtube.com/watch?v=sch5ZW1idz4

With the Freeplay Zero, the connections between the Pi and the Freeplay board are always a top suspect. In fact, even some problems can be solved by just reseating the Pi on a the Freeplay CM3 builds.

For the Freeplay Zero, I would always closely inspect the GPIO header. For anyone troubleshooting this stuff, I almost always ask for a photo of the pins and solder on the Pi Zero.

For you, StonyYodelers, because if your other thread about buttons, I would not assume that the solder is a problem. For anyone else that comes along reading this, though, it's the most typical problem.

Your mention if it working for a while and then going white makes me wonder if it's a problem with the ribbon cable's shoulders not being locked. That's what I've seen in the past, anyway.

(07-08-2021, 02:13 AM)Flavor Wrote: [ -> ]This could be caused by multiple things.  We recently had another builder experience a similar problem.  I told him to try this, and it fixed it for him.

"I’d maybe try hooking it up to HDMI.  Then, at the RetroPie menu, look for "Freeplay Change Display Driver" settings or some such.  See if changing that helps anything."

That will likely work if it's a software issue.

If it's a hardware issue, that probably means that one or more of the connections between the Pi and the LCD aren't making a good connection.

An all white LCD basically means that the LCD is getting power but not getting initialized with any data.

The first/easiest thing to check is the LCD's ribbon cable attachment to the Freeplay board.  I know that this is an old/poor video, but I think it gets the point across well.  Check it out and make sure that the ribbon is fully inserted and the shoulders are locked down all the way.  I have helped some people troubleshoot this issue where the shoulders weren't locked down.  AND, it's a bit of a problem that sometimes it will work that way, so people have built it up and had a working unit and then it just stopped working all of a sudden.  I think it was making a little contact and then something shifted and stopped working.  Anyway, locking the shoulders should prevent any intermittent connections.
https://www.youtube.com/watch?v=sch5ZW1idz4   

With the Freeplay Zero, the connections between the Pi and the Freeplay board are always a top suspect.  In fact, even some problems can be solved by just reseating the Pi on a the Freeplay CM3 builds. 

For the Freeplay Zero, I would always closely inspect the GPIO header.  For anyone troubleshooting this stuff, I almost always ask for a photo of the pins and solder on the Pi Zero. 

For you, StonyYodelers, because if your other thread about buttons, I would not assume that the solder is a problem.  For anyone else that comes along reading this, though, it's the most typical problem.

Your mention if it working for a while and then going white makes me wonder if it's a problem with the ribbon cable's shoulders not being locked.  That's what I've seen in the past, anyway.

The ribbon cable and gpio headers were some of my initial checks before I reflashed the sd card. In my day job I deal with a lot of circuit boards and the like, so I've seen a lot of ribbon cables that I thought I had secured actually be loose or not fully seated, so that's what I thought was it for sure lol. 

The solder could be an issue, but I've got a few active IPC certs, I'm pretty competent at soldering so I would be surprised if it were the issue. Just as an extra troubleshooting step however, I'll probably still touch up the solder on the pi and freeplay's gpio headers (but visual inspection under magnification looks good, no reason to think either is a problem). 

When I get home from work today I'll check the display drivers. I have low confidence in this doing anything just because it had been working initially and I don't know why it would have lost the drivers.

I'm going to try and look into it more at a component level soon as well and see if maybe something busted. I'm thinking if it's hardware it's probably on the pi end. I'll post back with any results if I find anything. Will probably be a day or two tops.

I've also ordered another zero that should be here Friday, and I'll test that out as well when it arrives

Update.

I double checked all the solder, looked good. Went through and checked all the capacitors and resistors and none stood out. Further component level testing would take longer than I care to work on for a $25 board lol. Some hardware component-level issue isn't ruled out, but it seems less likely to me now.

I checked the display drivers through the retropie menu, and when selecting the topmost option (iirc it is just an update drivers option) the first few lines of the print log show the lines "Failed to stop fbcp.service: Unit fbcp.service not loaded.
fbcpZero: no process found
update-rc.d: error: cannot find a LSB script for fbcp.sh
Failed to start fbcp.service: Unit fbcp.service not found"

I tried to manually create the service in the system directory and while it was partially successful, I got a new error:
"fbcpZero: no process found
update-rc.d: error: cannot find a LSB script for fbcp.sh"


That said, I believe the fpzero is using a different fbcp that is located in one of the Freeplay directories anyway, so I stopped messing with it in case I was doing a bunch of work for nothing lol.

Also (and somewhat unrelated) I just got the okay from the Mrs to get the Freeplay CM3, so that's exciting. To be entirely honest, the only reason I'm messing with the fpzero is because the cm3 was out of stock when I placed my order lol

I'm fairly baffled at this. It's possible that the LCD is bad, but that's very unlikely. LCD problems (which are rare) typically exhibit other behaviors than just a white screen.

If you get a Freeplay CM3, then you could use the LCD included in that kit to test on your Freeplay Zero. You could use your current LCD to test on the new Freeplay CM3 kit. If you find that the LCD is, in fact, bad, then go to https://www.freeplaytech.com/contact/ and submit a SUPPORT inquiry.

The Freeplay Zero should be using https://github.com/TheFlav/rpi-fbcp/ and it is likely located in /home/pi/Freeplay
The "error" messages that you got about failing to stop a service may be a red herring here. It will try to stop any other fbcp services that may be running, and it'll get that fail message if it can't (which is totally normal when it tries to stop other stuff that shouldn't have been running anyway).

By the way, we do have plenty of Freeplay CM3 circuit boards, now, but we do have a rather limited supply of Raspberry Pi Compute Module 3+ Lite boards. That's the main limiting factor at the moment.

Also, when you said, "I've formatted and reflashed the sd card" are you using Freeplay_Zero_20051301.img.gz or Freeplay_Zero_Test-RetroPie-46-20051302.img.gz
from https://drive.google.com/drive/u/0/folde...dwp_RqwzsQ ?

I'd also be surprised if the LCD was bad. It would be odd for it to be working fine and then just immediately go full white with no previous indication of issue or anything. When the cm3 board shows up I'll try out the old LCD though just in case.

I'm willing to bet the errors I'm getting are just generic errors that are about unneeded services. My freeplay is using rpi-fbcp and it is located in the directory you mentioned.

As for the flashed image, I used the base version, not the test one. Should I be using the test one? I also tested my setup using the most recent image in the Archive (If I inferred the date codes correctly, the version I pulled from the archive is from March 13th of 2020 rev1)

I'm leaning most towards it being something with the pi zero itself. With it only affecting the LCD but not HDMI out, I'm assuming if it *is* the pi, it'll be somewhere off the gpio header (looks like pins 7-11, or 19-21, not sure which you guys use but wouldn't add much time to test it all just to be sure), but with the cm3 version coming I'll probably just buy a new zero at some point rather than do any component replacements. I'll just throw the current zero in like a small hdmi-screened bartop cabinet for pacman or something lol.

I really appreciate all the help you've been offering over these last few days since I started working on this. Based off what little I was able to play before it bricked, the system is really great and I'm just floored at how well you guys were able to implement something that fits so well in the gba. I can't wait to see what a more powerful version runs like.

I've got usps tracking for the cm3 set already, you guys move fast! Still isn't at the post office yet, but I'm pumped to get it! Ordered some psp1000 sticks and the boxy pixel housing as well, so hopefully everything shows up around the same time and I'll have the better one up and running in like a week or two!

I just had an idea that I can try when I get home from work no rule out or confirm it's the zero. As far as I'm aware, the gpio pinout between the pi's are identical. I have a pi 3b just collecting dust (not literally). I can add an extension to the header on it and should be able to run the freeplay that way.

Would that work with one of the images in the google drive?

I think it'll work with the Freeplay Zero image, actually. I haven't done this in a while, though. You might need the Freeplay CM3 image. Anyway, we've done this with a ribbon cable for testing before. It's a good idea to try. I just don't recall if there's anything else that needs to be considered.

Also, make sure you get the pins connected the right way. It's easy to get things backward or upsidedown when you're using a ribbon.

Plugged in the pi 3 this morning (using the zero image). Works perfect. So there's a component off the header on the pi zero that's causing the issues with the screen. I don't think its the processor considering everything else works fine, including hdmi video. No labels on the board itself, so I might be looking into the schematics if I get bored, but as far as I'm concerned this is solved. I can either repair the zero or replace it and it should go back to working good.

Thanks for all your help!

No problem. I'm glad this info is all here for anyone else to read if they have similar issues.

I've seen all kinds of odd problems. We've seen Pi's that were bad from the factory. We've seen hairline solder bridges that are almost undetectable. Traces can get damaged from soldering or accidental scratching. All that stuff is super rare, but it can happen.

From an engineering standpoint, I want to track this stuff down. From a time/money perspective, sometimes it's hard to want to put a ton of effort into solving a $10 (cost of a new Pi Zero W) problem.