1088XLD - To Build or Not To Build?

[spookt]

When I first saw @mytek’s plans for the 1088XLD I got excited. I’d been interested in the XEL but the thought of a pumped up Atari 8bit in a 1050 shell is just really cool. 
 

I recently had a little good look and got a small tax rebate. If I chip in a little more I should have enough to buy the boards and components from TBA and the U1Mb etc from Lotharek. My question is, should I?

 

I’ve seen numerous posts here about how difficult the build is, and folks who’s ability I respect having issues with it. I’m not in a position to have this kind of money to ‘blow’ on my retro hobby very often, so before I do I’d like to get an idea of just how tricky a build it is.

 

I’d consider myself pretty competent with a soldering iron. I have good tools and have done SMD repairs (I have a decent stereo microscope) have hot air and a good iron, got an oscilloscope (though it’s about 20 years since I’ve needed to use one!) and all the usual bits of diagnostic kit. I also have an empty 1050 shell which I hung onto for this exact purpose.

 

Having just moved house it’ll be a little while before I get my bench set up (got a bathroom to fit first) but if folks who’ve actually built this thing could give me some feedback on the experience I’d be grateful.

[+Larry]

I was in the same quandary a year or so ago.  Ultimately, I looked at what I'd end up with and decided that I'd rather buy the EclaireXL mini.  That is not to take anything away from the 1088 kits -- nice idea/implementation.  But I have a slightly tricked 600XL that does pretty much the same thing, and it has a great regular Atari keyboard.  And I saved a whole lot of work!  But just my opinion.

[+DrVenkman]

Having built both, and been part of the beta team for the XLD before it was publicly released, I can say definitively that the XLD is a tougher build than the XEL. The actual soldering part isn’t really any harder but it is definitely trickier to get the mounting and spacing correct on the front panel I/O boards and of course the case bottom requires modification for the the DIN13 jack as well. If you go for one of Dropcheck’s rear-mounted MIDI connector I/O boards, then the back top of the case will need more work ...

 

Now with all that’s out of the way ... it’s incredibly rewarding to have built one.

[+mytek]

1 hour ago, spookt said:

I recently had a little good look and got a small tax rebate. If I chip in a little more I should have enough to buy the boards and components from TBA and the U1Mb etc from Lotharek. My question is, should I?

It is a very big soldering project, but if you've done surface mount soldering, then this should be a lot easier due to it being a through-hole only board. However to be objective based on how Covid affected the economy, and if there is a 2nd wave it could be even worse, I would say to hold off until next year, since building one of these is an expensive proposition.

 

32 minutes ago, Larry said:

I was in the same quandary a year or so ago.  Ultimately, I looked at what I'd end up with and decided that I'd rather buy the EclaireXL mini.  That is not to take anything away from the 1088 kits -- nice idea/implementation.  But I have a slightly tricked 600XL that does pretty much the same thing, and it has a great regular Atari keyboard.  And I saved a whole lot of work!  But just my opinion.

Yep this isn't everyone's cup of tea, nor was it meant to be. It sounds like you've already got things covered from a functionality aspect with your current crop of Ataris, so no real need would be fulfilled by the 1088XLD. The EclaireXL mini is actually much better, giving you a very full featured accelerated 8-Bit for far less than the sum of parts if you wanted to create the same thing via upgrades to either the XLD or a stock A8.

 

8 minutes ago, DrVenkman said:

Having built both, and been part of the beta team for the XLD before it was publicly released, I can say definitively that the XLD is a tougher build than the XEL. The actual soldering part isn’t really any harder but it is definitely trickier to get the mounting and spacing correct on the front panel I/O boards and of course the case bottom requires modification for the the DIN13 jack as well. If you go for one of Dropcheck’s rear-mounted MIDI connector I/O boards, then the back top of the case will need more work ...

 

Now with all that’s out of the way ... it’s incredibly rewarding to have built one.

As stated, this can be a very rewarding process to build your own, and then seeing it operate after the fact is the icing on the cake. But it does take some extra fiddling to get all the pieces aligned to achieve that factory finished look. If done right, you will definitely have something that looks every bit like it could have come this way from Atari (old Atari, pre-Tramiel).

 

[+selgus]

29 minutes ago, mytek said:

As stated, this can be a very rewarding process to build your own, and then seeing it operate after the fact is the icing on the cake. But it does take some extra fiddling to get all the pieces aligned to achieve that factory finished look. If done right, you will definitely have something that looks every bit like it could have come this way from Atari (old Atari, pre-Tramiel).

 

I know everyone is different, but for me, the building is half the fun/enjoyment. You also know the machine much better, and helps if ever you need to troubleshoot. I prefer SMD parts, but many thru-hole components are pretty forgiving and getting the board to look "professional" is not as hard as it might seem.. just takes time and focus (IMHO).

[+Firedawg]

I'm prepping to build one myself and not because I need another Atari but for the fun, challenges, and some learning it provides.  The concerns by the mytek and others of the difficulties that the 1088XLD would present to anyone wanting to build this kit was not to scare folks off but to give them thought before they leaped into the build.  The main build issues for anyone, like myself, lacking circuit design understanding and component knowledge may find attempting to perform the necessary diagnostics to find the problem and fix it a bit difficult without some help;).          

[+orpheuswaking]

With your soldering experience I would say that as long as you are aware that we the beta group is happy to help troubleshoot, but ultimately you are on your own if the board doesn't boot up then go for it. The main thing is to go slowly and triple check your work. 

Edited June 10, 2020 by orpheuswaking

[chevymad]

Build it. Even after my troubles and being told I was the reason they shouldn't be offered in kit form, i'd still do it again. If I hadn't roasted my original board, I'd build that one up too. I've built an XEL and an XLD both now. The XEL went fairly easy using the complete kit from TBA. The XLD I apparently had 1 bad transistor the whole time. Otherwise I bet my original build would have worked from the get go. If you buy a complete BOM it should be even easier. I picked my parts from several sources which can cause some errors. Using resistors I allready had was one reason I managed to get the wrong ones in a couple places. Also the BOM changed from when I first started ordering parts. Would probably have worked better to order everything at once.

[_The Doctor__]

How roasted and toasted could it be? Trace repair kits (copper foil tape) or hard wire could be the ticket.

[spookt]

Thanks for the feedback folks. I'm still sorely tempted so I'll give it some thought over the next few weeks while we sort the new place out.

[+mytek]

11 hours ago, _The Doctor__ said:

How roasted and toasted could it be? Trace repair kits (copper foil tape) or hard wire could be the ticket.

I'd love to see you repair 7 mil traces having 8 to 10 mil spacing with copper tape ? . That would require a very steady hand, fantastic vision, and a very sharp xacto knife. Probably be better off using wire jumpers made from 30 gauge kynar wire between vias or pads where conductive paths have disintegrated between them. This is how I fix and/or modify boards all the time.

 

15 hours ago, chevymad said:

I apparently had 1 bad transistor the whole time.

I strongly suspect excessive heat caused the damage during the soldering process, especially considering that you had two transistors fail in a similar way. I know the 2nd one was a different part number, but for all intents and purposes it should have worked fine. In fact just about any NPN signal transistor will work in this particular application, since its only being used as a switch and not as an analog device. But as I mentioned earlier, I'm glad to see you got your board up and working ? .

 

[chevymad]

I actually tried 4 replacements before receiving another pn2222. All worked the same.

[chevymad]

14 hours ago, _The Doctor__ said:

How roasted and toasted could it be? Trace repair kits (copper foil tape) or hard wire could be the ticket.

2" delamination from trying to remove a 40 pin socket with hot air.

[Mr Robot]

3 hours ago, spookt said:

Thanks for the feedback folks. I'm still sorely tempted so I'll give it some thought over the next few weeks while we sort the new place out.

Given the choice between the XEL and the XLD, I'd go for the XLD, purely because there is so much extra room in there for adding stuff. I've added MIDI, VBXE and an XEL-CFIII to my XEL and its very 'snug'

 

Maybe think about costing out the upgrading of an Atari to the same spec level, you'll get a more authentic 'feel' that way.

 

@mytek is there an XEL-CFIII for XL/XE computers yet? I know you were looking into that.

 

[_The Doctor__]

Yeah, that's why I also suggested hard wiring as well.... beings we didn't know where the damage was / or how big an area was affected.

I just went back and read my post and it does say 'or hard wire'.  The link to the wire could be useful though. probably the extra step I should have included.

Edited June 11, 2020 by _The Doctor__

[+mytek]

2 hours ago, Mr Robot said:

@mytek is there an XEL-CFIII for XL/XE computers yet? I know you were looking into that.

No not yet.

 

[+mytek]

1 hour ago, _The Doctor__ said:

Yeah, that's why I also suggested hard wiring as well.... beings we didn't know where the damage was / or how big an area was affected.

I just went back and read my post and it does say 'or hard wire'.  The link to the wire could be useful though. probably the extra step I should have included.

I never meant to suggest that you didn't mention hard wiring. I simply picked up on the copper foil approach that you also suggested, and wanted to rule that one out for this particular case due to how small the traces and spacing between is. Perhaps I could have said it a bit differently, sorry if it came across in a negative way .

 

[chevymad]

5 hours ago, _The Doctor__ said:

Yeah, that's why I also suggested hard wiring as well.... beings we didn't know where the damage was / or how big an area was affected.

I just went back and read my post and it does say 'or hard wire'.  The link to the wire could be useful though. probably the extra step I should have included.

But at $30 for a new set of boards it's much more cost effective to toss the damaged one. Now if the board was rare or something, sure i'd attempt to fix it. I also find it hard to translate the schema to the board. It's not quite like the wiring diagrams i'm used to working with everyday. When the schema says a chip is connected to A1 for example, i'd rather know where exactly the trace goes. The diagrams I work with, show say pin 42 on the ecu goes out to splice 50, through bulkhead connector pin 65, then onwards to temp sensor pin 3. Splice 50 also has 3 other sensors connected to it. This shows me several places I can jumper between. With the schema I have to wonder if the broken trace might have dead ended more of the circuit.  That's more just what i've gotten used to though. 

 

Anyway, to the important part of this thread, if anything, the trouble I had shows there's plenty of help around and you can fix most anything. Until the beta testers and Mytek get tired of us anyway.

[_The Doctor__]

True, for a one off fix, it might not be cost effective that's for sure. I'd say based on the fun your having with it now...

Your' vote is to build yes

Edited June 12, 2020 by _The Doctor__

[Kyle22]

4 hours ago, mytek said:

No not yet.

 

I would love one to use w/ my 800XL. I need the cart port not taken by SIDE.

 

[mdivancic]

I’m thinking of building an 1088xld i/o mod myself. One question I have is how do I go about programming the chips if I don’t currently have an Atari 8-bit computer? Can the Joy-to-PIC be used with other computers such as a ST or Commodore 64?

[+DrVenkman]

6 minutes ago, mdivancic said:

One question I have is how do I go about programming the chips if I don’t currently have an Atari 8-bit computer? Can the Joy-to-PIC be used with other computers such as a ST or Commodore 64?

The ATR files provided by Mytek are self-booting Atari disk images. He does provide the hex files as well though you'd have to write C64 or Apple code to interface with the device and then write those hex files.

 

Far easier to wire up a little breadboard to plug the PIC chips into that has the programming, power and ground pins connected, and then use a little USB programmer to write those hex files to the chip.

[chevymad]

I used a pickit3 and one of their adapter boards to do mine. Since my pickit is a clone there's some foibles, but it seems to work.

[mdivancic]

So this would work?

[chevymad]

That looks like what I use. There are some peculiarities with voltage setting and verification. Found a youtube video that helped out. Always seems to do the job anyway.