1088XEL Atari ITX Motherboard DIY Builders Thread

[+mytek]

When you guys get to installing the Mouse Port Select board in the Realan cases, here's the hardware I used that seemed to work pretty good.

 

M3 2 mm thick washer between the board and the inside of the case.

 

4-40x1/2" flat head screw.

 

4-40 KEPS nut (or nut with star washer)

 

And a 3 mm tall T1 LED spacer (optional)

 

You'll likely have to do a little bit of filing on either, the sides of the switch, or the inside of the hole to get it to fit. And if you are using the filler panel, it'll probably need a slight amount of sanding as well. If you do this a bit at a time, you can make it a press fit in the hole (I used the handle of a plastic knife as a tool to press it in, so as to not scratch the panel, worked real well).

 

 

I also changed to a different tact switch that I got from Jameco (the parts list on the schematic just recently got updated), because I just wasn't happy with the feel of the one I got from Digi-Key. Also the length is shorter which I think looks better (that's the Jameco switch in the photo above). Jameco P/N: 149948

[chevymad]

Ok, it seems i'm not quite functional. Not sure if it's user error or what though. Carts play fine now, I can access and make changes to U1MB no problem, and spartados will load. But I can't get to Basic or Missile Command. My Sdrive Nuxx will power up, but won't load at all. The stock system test will get a green block for the first rom, but when it starts to test the 2nd screen will either turn black or green and computer will freeze. The OS C jumper is the xegs is it not? With it unjumpered my display recieves no signal at all.

[+mytek]

The OSC header on the motherboard should always be jumpered unless you have a VBXE installed. I think you have it confused with the jumper on the U1MB which enables XEGS mode. I have no ideas on what is causing the memory test issues, but it sounds like it's time to do a very complete and detailed check of all the solder points on the board to see if something either got missed, or bridged. Since we just discovered that having the SparkFun board's slide switch set to VCC instead if 3.3V can create some issues with SIO, I would suggest checking that yours is set correctly.

[+DrVenkman]

Based on similar symptoms and my experiences diagnosing and repairing my troublesome 1200XL last year, I’d suggest checking the PIA socket closely and perhaps the MMU header on the U1MB and where that female header meets the 1088XEL motherboard.

[+mytek]

Based on what DrV mentioned, I have to ask what kind of U1MB board do you have, is it a black one or a newer white one that needed to have the pin headers changed from right angle to vertical? Reason I ask is that if that board required surgery to remove and replace the pin headers, there is the possibility that a plated through hole got damaged or a trace connection to a pad got disconnected during that process. So if you can test that U1MB board in another system, then any possible issues with it can quickly be ruled out.

 

 

Based on similar symptoms and my experiences diagnosing and repairing my troublesome 1200XL last year, Id suggest checking the PIA socket closely and perhaps the MMU header on the U1MB and where that female header meets the 1088XEL motherboard.

[chevymad]

Its a new white one, pins came out very easily. Theres a pic couple pages back. Does the black one plug directly in? One of my 800xls has an original black u1mb.

[Lavalamp]

Regarding the U1MB Installation

 

I saw Jazzcat's removal of the horizontal pins for the U1MB , but I would like to be 100% about the steps, I assume its:

 

1. Remove the horizontal pins on the U1MB
2. Soldering male pins to the U1MB
3. Fitting the headers on to these before fitting the assembly into place (to ensure correct position)
4. Soldering the headers
5. Don't upgrade U1MB to 1088XEL Firmware until all proved working. (URL to the latest version of this?)

Only waiting on the UAV and three components now to complete my build. Bought a 3D printer to prototype my case.

 

Thanks chaps!

Edited January 30, 2018 by Lavalamp

[flashjazzcat]

That sounds about right.

 

• Don't upgrade U1MB to 1088XEL Firmware until all proved working. (URL to the latest version of this?)

http://atariage.com/forums/topic/272817-1088xel-atari-itx-motherboard-diy-builders-thread/?view=findpost&p=3945961

Edited January 30, 2018 by flashjazzcat

[+mytek]

Its a new white one, pins came out very easily. Theres a pic couple pages back. Does the black one plug directly in? One of my 800xls has an original black u1mb.

 

Yep that black U1MB should pop right in. Might want to give that one a try. If it does work, then something isn't quite right with the white one, but you can always verify by plugging it temporarily into the 800XL.

 

I just checked out the pics you posted earlier. Nothing apparently wrong that I can readily see, but then I can only blow it up so big before things start to get blurry. BTW, don't forget to add some screws and nuts to hold down that SIO connector .

[+DrVenkman]

Regarding the U1MB Installation

 

I saw Jazzcat's removal of the horizontal pins for the U1MB , but I would like to be 100% about the steps, I assume its:

 

1. Remove the horizontal pins on the U1MB
2. Soldering male pins to the U1MB
3. Fitting the headers on to these before fitting the assembly into place (to ensure correct position)
4. Soldering the headers
5. Don't upgrade U1MB to 1088XEL Firmware until all proved working. (URL to the latest version of this?)

Only waiting on the UAV and three components now to complete my build. Bought a 3D printer to prototype my case.

 

Thanks chaps!

 

 

The steps I've bolded, #'s 2 - 4, I did all in one step. I inserted the male headers into the U1IMB at both ends pointing up, sitting on my desk. I fitted the three female header pieces on top of the exposed U1MB pins. I used a few small pieces of 3M brand reusable putty (what Jon refers to in his videos as "Blue Tack") to hold the headers in place in the U1MB. Then I rotated the entire thing around to slide the pins of the male headers through the 1088XEL. Then, and only then, I tacked two opposing pins on either side of the main U1MB headers into place. I verified everything was as trued-up and perpendicular as possible, then I soldered all the remaining U1MB pins.

 

I then rotated the entire 1088XEL upside down using the U1MB's three fully-soldered header blocks to ensure alignment of the female headers before soldering those pins. Again, I soldered two pins each on the bigger ones to verify everything lined up as close to perfectly true as possible, then finished up the rest of the pins.

[Lavalamp]

 

 

The steps I've bolded, #'s 2 - 4, I did all in one step. I inserted the male headers into the U1IMB at both ends pointing up, sitting on my desk. I fitted the three female header pieces on top of the exposed U1MB pins. I used a few small pieces of 3M brand reusable putty (what Jon refers to in his videos as "Blue Tack") to hold the headers in place in the U1MB. Then I rotated the entire thing around to slide the pins of the male headers through the 1088XEL. Then, and only then, I tacked two opposing pins on either side of the main U1MB headers into place. I verified everything was as trued-up and perpendicular as possible, then I soldered all the remaining U1MB pins.

 

I then rotated the entire 1088XEL upside down using the U1MB's three fully-soldered header blocks to ensure alignment of the female headers before soldering those pins. Again, I soldered two pins each on the bigger ones to verify everything lined up as close to perfectly true as possible, then finished up the rest of the pins.

 

good old Blutack, great suggestion!

[+Firedawg]

The DVI port modification on the Realan case has been completed. I would not recommend this type of modding unless you have a strong stomach, have a good bit of patience, handy with small tools. I'm not saying that I possess any of these attributes but they help . Michael was spot on with the over hang for the DVI connector board which I had to trim back so I could get a flush contact with the case.

 

 

 

[+DrVenkman]

Nowhere near as far along as Firedawg, but I got my board at least sitting in its bare case. Nice to not have to short the two jumper wires I had plugged into the power headers to turn it on and off.

 

 

I also verified that the Sparkfun board can talk at Divisor 0 speeds with my RPi Zero W. That tiny little board makes a great little RespeQt server. I've got it plugged into the same monitor I use for my MAME setup which is right next to my Atari stuff (which I use CRTs for).

 

 

And while I was at it, I also confirmed that my standalone SIO2PC-USB device works with the 1088XEL too now that I've got that switch on the Sparkfun board properly configured, though I didn't test it in conjunction with anything connected to that board.

[_The Doctor__]

Looks like the tides have turned and the the ship has come in. Glad it's all starting to head your way!

[chevymad]

Both U1Mb's act the same. Actually the Candle version won't even boot into Spartados. I found that the pins on Q9 were cut too long, and moving the board around on my desktop caused them to bend and touch. I've trimmed them back more but have I damaged something? No change afterwards. Going over the rest of the board with a magnifying glass, but haven't found much yet. I also noticed that the Atari fart startup noise sounds like static, but when the sdrive is plugged into the sio it sounds correct? No action from the sdrive however. Installed the XEGS jumper on the U1mb.. still no missile command or basic. Machine can be booted into U1mb setup, test mode, spartados, or a cartridge. Memory tests failed at the 2nd rom block with both U1mb.

[+mytek]

Both U1Mb's act the same. Actually the Candle version won't even boot into Spartados. I found that the pins on Q9 were cut too long, and moving the board around on my desktop caused them to bend and touch. I've trimmed them back more but have I damaged something? No change afterwards. Going over the rest of the board with a magnifying glass, but haven't found much yet. I also noticed that the Atari fart startup noise sounds like static, but when the sdrive is plugged into the sio it sounds correct? No action from the sdrive however. Installed the XEGS jumper on the U1mb.. still no missile command or basic. Machine can be booted into U1mb setup, test mode, spartados, or a cartridge. Memory tests failed at the 2nd rom block with both U1mb.

 

Q9 is the buffer for the color adjust pot. If you killed it, then there would be no color, or the wrong colors on screen. Next step would be to start swapping things such as the CPU or PIA, both of which could affect the memory test. Since you have video, Antic and GTIA are likely ok. You could also try swapping the two Pokey's with each other, although if you have audio and keyboard they're probably ok as well. And I'll assume you've done a very careful check of all the solder points.

 

Good luck.

[+DrVenkman]

Those symptoms still sound a lot like a bad PIA or MMU header issue. I’d check those solder points and traces very closely and look for debris around and under the header pins on top of the board not just underneath.

 

And yeah, good luck figuring it out. My 1200XL issues took 5 weeks to completely run to ground.

[flashjazzcat]

Donation incoming.

 

PM sent, but I'd also like to say thank you here too.

[+MacRorie]

I just wanted to post these here. It is the first fully cased 1088XEL order to go out the door. The rest are not far behind:

[+DrVenkman]

That looks fantastic! The recipient will be very happy, Im sure.

 

Question for those of you finishing up the Realan cases (and Michael of course). How is the back panel held in place? I see in the design files Michael makes available that they have holes for corner screws but none of the assembled cases Ive seen photos of are using them.

[+Firedawg]

DrV, the rear panel received by Front Panel Express is thick, sturdy, and holds in place nicely just with the connectors in place.

 

 

 

[+MacRorie]

DrV, the rear panel received by Front Panel Express is thick, sturdy, and holds in place nicely just with the connectors in place.

 

 

Quite correct. Once the screws holding the motherboard in place, it is not going anywhere. That, in conjunction with the strain relief screw on the video connector and it does not shift at all. Opening it back up to, for example, upgrade the TKOS is straightforward if a little annoying (because of the indicator wires, screws, etc )

[+mytek]

That looks fantastic! The recipient will be very happy, Im sure.

Question for those of you finishing up the Realan cases (and Michael of course). How is the back panel held in place? I see in the design files Michael makes available that they have holes for corner screws but none of the assembled cases Ive seen photos of are using them.

That's because the 4 corner screw holes are a recent addition (I uploaded the new files last week after verifying that the idea would work in my last H60 case).

[chevymad]

Changed all the atari chips and no dice. Then refluxed and flowed all solder points, cleaned and examined with a magnifying glass. I will say it's not as "flaky" as it was before. But still no disk access via sio. Haven't tried hooking to the pc yet. I've tried 6 or 7 carts. All of them seem to play, good audio/visual. Don't have a way to hook a controller up yet, but they all start. Plugged in a basic cart and that works too. Wrote small program and that worked. Printed free memory and shows 37 something.. I noticed that if I go into the Side menu, then exit, then run the system test, the rom and ram pass. Tomorrow I'll see if I can get my win tablet hooked up to it and have any access that way. Would like to update the u1mb and see what happens after that.

 

What are the different jumpers on the board? Where should they all be set? I'm sure that info is somewhere in one of these threads.

 

I will say the picture quality and sound quality is just awesome. Better then my UAV 800xl for sure.

[+DrVenkman]

No SIO access still points to either POKEY or PIA chips, or something in those signal paths. Have you tested sound? Find the appropriate version (NTSC or PAL) or something like YOOMP! on Atarimania to test both left and right POKEY. If you have speakers, can you hear keyclicks and the usual startup sounds as the Atari tries to access the SIO bus? And have you tested access via the Sparkfun board (assuming you have one installed)?