1088XLD - Custom A8 Computer in 1050 Drive Case

[flashjazzcat]

No stock A8 requires stacked sockets under VBXEs of either configuration (and the 1088XEL does not require an extra socket under an XE VBXE adapter), but OK.

Edited February 25, 2019 by flashjazzcat

[+mytek]

No stock A8 requires stacked sockets under VBXEs of either configuration (and the 1088XEL does not require an extra socket under an XE VBXE adapter), but OK.

I stand corrected for the VBXE .

 

However I know there have been cases where Sophia has needed an additional socket in order for it to clear other components, and the more recent version of the TK-II piggyback board does as well. So suffice it to say whenever there is a clearance issue, stacking another machine pin socket has been a good solution.

[flashjazzcat]

It would be well worth test fitting some of these devices during the design phase in order to remove some of the guesswork, although it's a little difficult when the people with the upgrades aren't the same people with boards in their hands (although I guess the people with the upgrades would be happy to provide measurements if asked). I know you own more than one Sophia, while I have VBXE, Rapidus, etc, but no Sophia. VBXE is perhaps the most critical since it's designed to work with U1MB and is supported by the firmware (and the motherboard is designed with RGB-out in mind). I have certainly been entirely engrossed by the firmware in recent months, otherwise I would have mentioned the matter sooner. Not that it's even a big deal if one stacked socket takes care of business, although when more than one is required (think Rapidus in a 130XE, which requires four or five stacked sockets), things aren't necessarily robust. Pragmatically speaking, a stack of four or more sockets would probably allow an XL-format VBXE to clear Rapidus, but XE-style is definitely the one to get for the 1088XEL/XLD.

 

One purely cosmetic thing I wondered about is why the IC numbers on the silkscreen are rotated 180 degrees to the usual orientation in relation to the IC except for U22 (bearing in mind pin 1 is at the notched end of the chip). I'm guessing this is just so that the text is readable when looking at the board from the front?

[+mytek]

 

It would be well worth test fitting some of these devices during the design phase in order to remove some of the guesswork, although it's a little difficult when the people with the upgrades aren't the same people with boards in their hands (although I guess the people with the upgrades would be happy to provide measurements if asked). I know you own more than one Sophia, while I have VBXE, Rapidus, etc, but no Sophia.

 

The design was based on actual measurements of the XE-VBXE as well as the standard Rapidus so both should fit without overlapping each other. And there should also be room for a P-Covox to coexist at the same time. Of course if any of these boards have been designed to snuggle up closely to the motherboard, then yes there is the possibility that one extra stacking socket will be needed to clear adjacent ICs or other components, as is the case for the Sophia. The 1088XLD was designed with all of these upgrades in mind. And since height was not an issue inside of a 1050 case, I didn't go out of my way to clear the underlying areas, although I did try to keep things to a low profile which is evidenced by the horizontally mounted PAL crystal. However I did forget about J7's height with a shorting jumper in place, so that might need to be specified as a right angle header vs. the vertical one I have been using, assuming that interferes with the VBXE.

 

Jon as you know my intentions have always been to have you BETA test this board, so at some point you will have ample opportunity to check your VBXE and Rapidus for fit. If problems are discovered they will be fixed before a public release is ever made .

 

 

 

One purely cosmetic thing I wondered about is why the IC numbers on the silkscreen are rotated 180 degrees to the usual orientation in relation to the IC except for U22 (bearing in mind pin 1 is at the notched end of the chip). I'm guessing this is just so that the text is readable when looking at the board from the front?

 

I never gave that any thought, other then I wanted all the text oriented in the same direction where possible. Hopefully people will pay more attention to the notch part of the chip outline on the silkscreen, then which way the text ID is shown. Speaking for myself only, that's what I was taught to do when inserting ICs and sockets .

[flashjazzcat]

Thanks for the info. I keep forgetting that things aren't cast in stone at this point.

[+mytek]

Thanks for the info. I keep forgetting that things aren't cast in stone at this point.

No problem .

 

And due to the complexity of this project, and my limited time to provide technical support, even a public release is not set in stone.

[ebiguy]

Hi MyTek,

 

Now that I have a beautiful 1088XEL (I received my rear panel. It's clean and it fits perfectly !) I was looking at some other hardware projects that could be related with the 1088XEL.

I found dropcheck 1090XLR.

Reading the web page gave me an idea about the 1088XLD (my future project I hope).

i know that you consider this project almost finished but anyway, I tell you what I have in mind.

It would have been great to make some room in the 1088XLD by using more daughter boards near the front to be able to put several expansion slots at the back (like in 1090XLR for example).

Some new extensions could have been created and the 1088XLD could have been extended in a "standard way". Anyone would be able to design an expansion board.

 

Just a thought.

[+mytek]

It would have been great to make some room in the 1088XLD by using more daughter boards near the front to be able to put several expansion slots at the back (like in 1090XLR for example).

Some new extensions could have been created and the 1088XLD could have been extended in a "standard way". Anyone would be able to design an expansion board.

 

There actually is a way to do that right now. It's called the Cart/ECI parallel expansion bus, and is common to the XE/XEL/XLD (well not all XE's), and very similar in nature to the XL PBI. This is an under utilized expansion bus, that unfortunately has mainly been focused on RAM expansion and Hard Drive interfaces, all of which already exists in the XEL and XLD in the form of the U1MB and the CF/IDE support. But the things that people have been craving for such as buffered RS232, buffered printer port, 80 column video, and Ethernet/WiFi are still waiting in the wings. Of course some of this is possible thru the SIO port as well, but only at reduced speed and efficiency. Also ideas like resurrecting the ill fated 1090 have been tossed around, which can also avail itself of the Cart/ECI port or the XL PBI, and then provide multiple parallel expansion slots.

 

It just seems like parallel expansion keeps getting pushed to the wayside, and instead people have focused on SIO or joystick connected devices. I think in large part this is due to not all machines having parallel capability built-in, whereas all of them do have SIO and joystick ports. This is also the case why so little stuff was developed for the right cartridge slot in the original 800, since it's cousin the 400 didn't have one. Developers have shown a tendency to try to appeal to ALL machines in order to reap the greatest possible sales.

 

Well at least we are seeing some creative uses of the common left cart slot .

[+mytek]

Prototype Serial Number Labels arrived.

 

 

Updated image files in zipped archive on this page.

[+Stephen]

Great - my case has no sticker. Speaking of - look what I did tonight. 266 joints (not that I am keeping track this build).

 

Long winded explanation of some soldering trouble below:

 

 

I was having trouble getting solder to flow from bottom to top layer of this board. I ended up upping my solder station from 650 all the way to 800 and still went slower than my usual 650 speed. Not sure if I am just out of practice, but I am still fairly proficient TIG welder and can do the "lay stack of dimes" bead. This is how I usually do my large items - drag the tip slowly across pins, while tapping/dabing solder as I do when TIG welding. But even when I went super slow and did a pin by pin - just couldn't make it all the way through. I am using my usual solder & flux and everything. I am not in any way saying I think the electrical connection is bad at any joint, but the shape and amount of each solder on the pin is upsetting me. I'll only post good pics.

 

 

 

[+DrVenkman]

Stephen - you issues are the same as I experienced building my XEL. I had a HELL of a time getting the solder to go all the way through the via. I had some questionable joints that had to be reflowed with additional solder added as a result, which took a shit ton of time after I thought I was complete. I put it down to my inexperience with soldering thicker 4-layer boards that have integral ground planes. That plane has got to be a giant heat sink. I would suggest that as an isuse for you as well, but I know you've built at least one XEL yourself and know very well what you're doing. Fortunately, the through-plated vias make up for a lot in that regard.

Edited March 1, 2019 by DrVenkman

[_The Doctor__]

that is part of the reason some board have the internal plain connect with the the plus sign looking through hole... it's late I don't know my words but you know what I mean..

 

pre heating the entire pcb will help though... a prep oven if you will, not hot enough to melt anything. around 115 degrees Fahrenheit usually helps and wont burn you.

[+Stephen]

Thanks guys. I'm a middle age guy with only 2 layer PCB experience. I just wanted to point out my troubles because I've had a shit ton of experience with this stuff, so I figure a beginner could really get a surprise. I'll keep experimenting, and we;ll see how things are at the end of the build.

 

Like I said - it would function just fine, but I am such a perfectionist - I want every damn solder joint to be perfect. My brother calls me "Rain Man" for this habbit. I think he is lazy and sloppy

[_The Doctor__]

filling the thru hole is important... it stands up overtime. the stress of vibration and hot cold cycles can turn a working connection bad over the years.

[+mytek]

It's a 4-Layer board, so it needs more heat due to the mid layer power planes. Doesn't matter that those planes are not connected to all of the pins, still acts like a heat sink for the entire board area. But yes wherever it does connect to a pin, it has been done with thermal relief which like Doc pointed out, is like a cross if you could see inside of the board.

[+Firedawg]

This board looks great and as always kudos to mytek for sharing and allowing everyone an opportunity to get involved. I'm in the middle of another house move as the one from a few years ago was for down sizing. I guess this one is for right sizing according to the wife. Any how, I'm hoping to get my space back up, operational, and ready for this boards release from Michael.

[+mytek]

My goal this time around was to better capture the look of an Atari, since the last project although very nice, has a totally different feel to it. At the same time I also wanted to take what I had learned and make it a little better. And as the project progressed I got MIDI fever, which is highly evident in its final form.

 

 

Because of the MIDI research and the unique PIC MIDI Controller firmware that was developed as a result, we now have a very easy and inexpensive way to add MIDI ports to any machine. What would be nice to see because of this break-through, would be to have more networked multiplayer first-person-shooter games. I know there were several created for the game-link system, so perhaps those can be modified for use with MIDI networking as well.

 

Also having the built-in Stereo MIDI Synthesizer brings a new dimension to the sound capabilities. I am hopeful that a new MIDI song player will be created, giving us something more akin to a true media player for playing back MIDI song files. Sporting song lists, shuffle mode, and perhaps incorporate music visualizations as well.

[+Stephen]

That is so damn pretty. Here's a before and after on the case I am using for this drive:

 

[tschak909]

retrobrighting?

 

-Thom

[+Stephen]

retrobrighting?

 

-Thom

Yes - I hope it doesn't re-yellow on me. Oddly, none of the gear I owned ever turned colour. Non smoker, and always covered, out of sunlight, and temp stored. I am a "night owl". Light sensitive, prefer the dark, etc. This was an ebay item, broken, that I was trying to fix.

 

Oh - no tricks. 40 vol cream lightener, saran wrap, and sunlight. Re-applied every 30 minutes for 3 hours, and I had this perfect result. Ambient temp was 73, moderate humidity when I did this (outdoors).

[_The Doctor__]

be advised, none of my stuff yellowed until I started using LED lighting...it's yellows stuff super quick... it's got to be putting out near uv ligh or something... high energy bright blue light must be in their dna

[+DrVenkman]

be advised, none of my stuff yellowed until I started using LED lighting...it's yellows stuff super quick... it's got to be putting out near uv ligh or something... high energy bright blue light must be in their dna

 

LED's typically have more blue than natural light, but less blue than incandescent bulbs. They emit nearly all their energy in the visible spectrum. There is no UV emission from a consumer LED - the reason LEDs are so energy efficient is because they are emitting light ONLY in the spectrum range for which they're targeted, not wasting it off on either end where it is wasted.

 

https://www.comsol.com/blogs/calculating-the-emission-spectra-from-common-light-sources/

 

You can buy a spectrometer and measure output yourself if you like, or build one but you'll have to use a wide-range photocell, which most consumer grade sensors are not. Of course, you can GET UV LEDs if you want, but you have to seek them out, usually from specialized vendors for specific (usually industrial) applications.

 

I think it's pretty well established now that yellowing is more complicated than just "UV light causes it!" because after all, UV light plus peroxide helps reduce it. In fact, per some experiments you can find online, it's not even really the UV light per se - it's the heat that builds up under the plastic film, warming the yellowed plastics and helping "pump" the chemical reactions between the bromine in the plastics (which cause the yellowing) and the peroxide which is breaking them down and removing the color. And in fact, I have a NIB 410 Program Recorder that looks like it was never removed from its box before I bought it - it's more yellowed than one I have that was used in a school room setting for years. I think the volatiles in the styrofoam packaging affected that one since it obviously couldn't have been sunlight.

 

The only large thing I've retrobrighted is my "Ugly Duckling" 1200XL, which I did about 20 months ago - it's been fine ever since, though my Atari stuff is in a darker area of my house - I tend to use my machines with task lighting (reading lamps) and the glow of a CRT monitor, not out in a brightly lit room.

[AtariNerd]

Yeah, from what I've read, blue LEDs are easier to create and are a bit more efficient than pure white LEDs. Most of the "white" LEDs on the market are created by using a blue LED, but having a phosphor coating on the inside, that through some process of photon-bumping, it transforms into what we perceive as closer to pure white light, though not quite a full, continuous-spectrum version of it. There be gaps. Some principal, that if I understood quantum-mechanics, I might be able to fathom, but since I don't, I have to scratch my chin and go "neato".

[_The Doctor__]

the blue light appears to be bad for my eyes and my old computers..

Edited March 2, 2019 by _The Doctor__

[+Stephen]

Well I got more work done on the board today. I had every socket except for 18 pins for U15 and U20. I also found a 13-pin PCB jack that I ordered way back in 2005 Current pin count is 525.

 

Just curious Mike - can you easily get a count of the "through hole" pins for this and the 2 auxiliary boards from your design software? It's easy to keep track of on the big stuff like this, but I think I'll get tired of counting when I start dropping in all the 2 pin components.

 

I'll post some more pics as soon as I get get my camera to cooperate.