guus.assmann

Hello, From a company in Germany, you don't need to worry at all. I've ordered many parts from this particular company and never had any complaints. And the prices are also reasonable. Not the very lowest, but ok. And they have a lot in stock as well. BR/ Guus

Hello, Have a look at Reichelt Elektronik in Germany. They're asking 12,35 Euro for the XC 95144XL TQ100. BR/ Guus

The prototypes I made, were also soldered by hand. And at some point, I replaced the CPLD 9572 by a 95144 with more logic inside. This was also done by hand on 8 or 10 units. Soldering 150 units all by hand was a bit too much. And the PCB manufacturer did deliver original CPLD's, I'm sure. I had them assemble all SMD parts onto the top of the PCB, they wouldn't do the parts on the bottom side as well. And the Ram I also did myself. The sockets for the Flash were also soldered in China and that was not so good. (About 15 failures if I remember correctly) For price reasons, the flat cable to the display was soldered/crimped on one side. BR/ Guus

Back when I made the units, I got the CPU's and Ram-chips from Chinese vendors. For the processors, I got 2 or 3 failures and got some refunds. These parts were about 5 Euro on average ! The RAM chips were all ok. Flash-roms came from Farnell. The CPLD´s were assembled and delivered by the PCB manufacturer. Not sure, but it could have been JLCPCB. I had about 10 failures, that were fixed by manual resoldering. I made 10 prototypes, fully by hand. On the 100 + 50 units in the serial production, I had 3 that I couldn´t get to work at all. And only 4 or 5 waranty claims, where the unit could be fixed. (They had bad solder joints) The rotary encoders and displays were a problem. Distrilec and Farnell. So I removed a lot of displays from defective PCB's that I got from my employer. This also saved some cost of course. All of this took a lot of work, but saved the buyers a lot of money. And I had ABBUC to refund all parts quite fast. BR/ Guus Assmann

Hello Larry, Please see attachments. This should answer the questions I think. The PCB is very simple and using the PDF files of the design, you should be able to make PCB's yourself. Even the toner transfer method could do the trick I think. And Yes, I do have the design files. It was made using Ultiboard and Ulticap. These programs may not be available anymore. Using KiCad, you should be able to redo the design in a few hours..... Best regards, Guus anleitung_1050_turbo.pdf 1050TRUB.zip 1050TRUB.zip scan_1050_turbo_2.zip soft-trb.zip software_1050_turbo.zip

Shortly after B.Engl gave the information and rights to ABBUC, I became aware of it. And I made some PCB's after the design that was available. Next, I made a version that was even much smaller, it had most parts below the Eprom. Some units were made. Much more important (from my point of view) was the next step. The Freezer data also was available. And together with Hiass, this product was recreated and modernised I made a PCB design for the freezer as well, using bigger memory. And Hias made the CPLD content. Once the first version was working, with homemade PCB's by me, the design was extended a bit. Hiass changed the CPLD to also include cartridge emulations. This idea was triggered by some discussions over E-Mail. This was made possible because there was much more program space in both Ram and EEprom. Having an EEprom also opened the door to easy updates and changes to cartridge emulations. The banking mechanism used by CSS was also included at some point. This opened the door for XL/XE basic and an Assembler/Debugger. I have build and tested quite a number of units, with funding from ABBUC. Much later, the Turbo-Freezer was redone. And in between, the Mega-1050 saw the day of light. First version by Mega-Hz and Hias. Later I took over from Mega-Hz and together with Hias, the PCB's were developed further. Here as well, ABBUC did the funding and I did the builds and tests. Hias came up with some great testing tool and program. And I did all the (hard) hardware work. (With on-line help from Hias) P.S. Of course there were some Beta-testers as well. This prooved the viability of the product. BR/ Guus

That disk is only usefull for: Alignment. This means making sure the stepper motor sends the head to the correct place. It's no use to diagnose any hardware faults. Manufacter if memory serves well is Dynatec. The disk has a trace that has been written a-symetrical and will display a "cat-eye" on an oscilloscope. In the many drives I had to repair, I used it once or twice. BR/ Guus

Hello, It's also possible to use a 25Mhz oscillator and devide the frequncy by 3. BR/ Guus

And also, yes, the computer won't start if you have them wrong. But it will not do any harm. BR/ Guus

Hello, The component Q3 is the voltage reference. It's a TL431. (Datasheet on internet...) And the resistors around that component determine the 5V reference. Resistors can be checked with an Ohm meter. Best regards, Guus

Yes, I agree. A higher working voltage is always ok, as long as you can get it to fit. Once you can power up the unit again, check the other Tantaliums as well. The method is very simple, just feel if they don't get hot. A failing Tantalium capacitor very often gets a low resistance. BR/ Guus

Hello, My advice, just cut out the capacitor and retest the unit. It's a Tantalium capacitor and most likely used for decoupling / buffering of the supply. And in that application it's mostly redundant. This also means the value is not so importat, only the working voltage. (and the polarity of course) Be aware that more of these Tantaliums may fail or have failed. BR/ Guus Assmann P.S. Test my assumprion by following the traces or measure with an Ohm-Meter / Buzzer for low resistance.

Hello all, A little back to the topic of the failing CPU's. During the introduction of the 600XL and 800XL in Europe, I worked for Atari in the Benelux. And I did repair many XL units. There was an issue with very specific production dates of CPU's produced in Mexico. (Ony two or three dates if I remember correctly.) Mind you, not all would fail. The test I devised was simple. Just open the unit and lay the CPU bare. Start the unit and let Basic come up. Enter the line: For x=0 to 100000: print x; "Atari test": next x. When this was running, use a hair dryer to heat up the CPU. The CPU never got so hot that it couldn't be touched ! When the CPU was warm, the display would freeze if it had the issue. However a Reset would restart the unit without fail. Sometimes it would pass this test and then there was another defect in the unit. I never found other CPU's with this behaviour. In my opinion it had to do with the timing of the Halt input. As this is not present in any 400 or 800 board, those were not affected at all. This is my experience as technician. BR/ Guus Assmann P.S. This discussion is quite arbitrary anyway... 🙂

Hello reifsnyderb, The schematics are either from cloned boards or reverse engineerd from a clone board. If you look at the memory map that is decoded by any of the address decoders, you'll find they all do the same. And to the best of my knowledge / experience, the track display was a hack by a private person. The displays are just put on a address in the memory map. (With lots of "shadow" addresses. In fact, all the "shadows" are a result of incomplete address decoding, based on chip-count. Either because of cost, or because of board-size. The use of a track-decoder is in analysing disk-protection scemes. And of course it looks way cool 😃 You could use a GAL device to do a decoder that can be changed. I've once made that design and a board, but never got to make the logic in the GAL. I do own one original Happy board. (Got it as a gift) And I never got to use it. It arrived "in the middle" of creating the MegaSpeedy1050 boards. Best regards, Guus Assmann

Hello reifsnyderd, Software is not my strongsuit. I have no idea why any Rom works or doesn't. I didn't look into that. What I do know, is that the display is only a memory location to write to. reading it has no value. So I think it's very unlikely that not having the display prohibits the unit from working. You can try to ask Hias. He made the software for the MegaSpeedy and analized a lot of Eproms in drive units. BR/ Guus Assmann

Hello, Have not read this tread fully. However, here's my schematics of Happy's. (With a track display, but that's irrelevant) All of these versions do work, I've also made PCB's and build a few. As for write protect on the floppy, the drives I know (Shugart bus) have a write protect that cannot be influenced by any program in a floppy controller. So the write protect on the disk activated/deactivates a signal that's commonly called "write gate". If inactive, the write amplifier to the head is deactivated. If I remember correctly, this mechanism is also in the 1050 hardware. The extra Ram on the PCB consisted of 3 pieces 2K ram in the very first PCB's. Once the 8K ram became available and affordable, this was used. BR/ Guus Assmann HAPPY2.pdf HAPPY2a.pdf HAPPY4.pdf

Hello Larryleffaovell, Thanks for posting the Zip with the Gerber data. Can you please also post the KiCad design file? Then I can see if I can panalize it to fit some in a 100 x 100mm PCB. This size is very cheap with JLCPCB and I also found how I can get the V-Scoring to work. Of course I'll post the results here again.... BR/ Guus Assmann

In my opinion, it should also be possible to run this OS from the Freezer. Of course one needs to have that piece of hardware.... Major detail of course. 🙂 BR/ Guus

I've looked but cannot find the original ordering numbers anymore. Obviously, the WDC parts is called WDC65C22. But the suffix is about the shape of the housing. They do exist in DIL40 and Quad-housing. (PLCC-44) Looking at Mouser, the number I get is W65C22NTPG-14 Place 955- in front of this and you have the Mouser order number. For me, they're listed at 10,21 Euro and more than 1700 in Stock. Not sure about the hardware mod, but I think it's only changing the clock speed to the Uart (65C51) For that, 955-65C51N6TPG-14 at 7,31 Euro and 228 in stock

Also, other than the change to the 6551 circuit, there is no possibillity to have a higher speed. The computer is just not fast enough. The only way would be, to use a 65C816 at 14Mhz and have that handle a Uart at this epeed. All Data to the BB I had, including the PCB design has been given to Fox. The 1488 and 1489 can e exchanged by a MAX 232 very simple. The only thing is, that one MAX has 2 input and 2 output channels. So a MAX 238 may be needed. BR/ Guus

Hello, Some while back, I bought the WDC part and tested it in a black box. It worked well. Buy them from Mouser and you'll be sure they are genuine. BR/ Guus

Hello, Sorry. you're so right. (And I'm getting rusty 😞 ) The original Eprom (or Rom) is only 4k x 8. And that's a 2732. BR/ Guus

Thanks, will do this in some time, so the MegaSpeedy retains it's value 🙂 Guus

Hello ryswoj2, Nice to see that it works ok. The IS Plate, I don't have that one and am not familiar with it. Can you please post it as well, so maybe I can get it into my units as well. Still haven't updated them to the latest CPLD and other firmware actually 🙂 BR/ Guus Assmann

Hello, Not sure what's in this file exactly, but it seems to be a bit more than standard. As far as I know, this has not been put in any units I've made, but not 100% sure. Els, have a look for Nir Dairy. BR/ Guus megaspeedy-2015-jan[1].rom