4 * 1050 boards fixup - general questions/strategy

[E474]

Hi,

 

I've recently picked up 4 1050 motherboards, see picture. Firstly, is there anything out of the ordinary with the motherboards? The top-left is obviously a bit different, but I don't know why (there are also some extra wires connecting points on the underside).

 

Some of them have the socketed chips missing, but I have access to a stock 1050. I'm guessing if I remove each of the remaining socketed chips, and try them in the stock 1050, that's the quickest way to test them? I'm hoping this should be OK, I guess the only problem would be if the ROM doesn't like the stock 1050? Obviously I want to avoid the "magic blue smoke".

 

If all the socketed chips are good, that gives me enough chips to test each of the motherboards in turn. All of the motherboards have damaged pin connectors at the bottom right hand connector, so fixing these is going to be the first bit of soldering.

 

Any comments on this, there's no obvious physical damage, but they were sold as defective/for parts, so this might take a while to fix?

 

 

[Paul Westphal]

Www.atarimax.com. Happy enhancement boards are $40. Good place to start.

[_The Doctor__]

top left need a lot of care, the component are smooshed down and need careful separation and standing up. I am not sure but it also appears some components are missing, and I think but can't be certain it's missing a transistor. A few of those boards saw themselves outside of Atari. I'm sure we've discussed that board on AA somewhere. so you might need to do a deep dive to find all the particulars about it. I am vaguely remembering something about it... but it's out of reach...

[E474]

Hi,

 

Thanks for the tips. The top left has extra connections (?) on the PCB, Q7, Q8 and Q6 - they're not present on the other PCBs.

 

I've attached close up pictures of the front and back of the top left board - there's also extra wiring on the back of the circuit board. The board is marked as Rev B on the back of the PCB.

 

How well would this board tolerate a later ROM, that is, how careful do I have to be with boards and ROM versions?

 

Also, quite a few of the components have been bent over, but they don't look damaged (cursory inspection so far).

 

 

 

 

[_The Doctor__]

the roms have more to do with the mech used than anything else, the pcb really ends up being near identical to the 1050 pcb when populated. the regulators should be about the the same from each but the spot on the pcb wasn't routed where it was convenient so the wires jump them to where they need to be until our work in progress managed to get traces routed on the pcb. It will be for the most part all the same stuff with jumpers and wires to get things that were not routed but needed a connection... I have one nagging thing about the power supply section in my head. Can't remember.

[+Nezgar]

Very nice boards in your acquisition! I'll comment on each individually... You should be able to verify/test all 4 boards as you can make a complete set of the main IC's from the chips in the bottom two boards. I bet there all mostly good, if you install the missing IC's.

Top Left: This is the most interesting, and looks just like a prototype/pre-production PCB that DavidMil posted about in a previous thread: http://atariage.com/forums/topic/275166-one-last-eprom/?p=3959593 The voltage regulators were also manually wired to the heat sink, maybe because the location was not finalized, or trace routing was not complete as __The Doctor__ noted. Also, the jumper wires under the SIO ports suggests that some or all the pins were not mirrored between each SIO port in that PCB layout. (Same as DavidMils). Even the trimpots for RPM, and DD write precompensation timing are in different locations than the standard one... The PCB's were probably manually retrofitted for use after the main production run started. The jumper configuration (JP 1/3 OPEN) indicates EPROM. DavidMil's had an EPROM that identified as Rev "H" (earlier than the more common Rev J/K/L). You can burn Any standard 1050 ROM revision that supports the Mech and WDC controller you match with it.

Top Right: JP 2/4 OPEN meaning Mask ROM configuration.

Bottom Left: Missing ROM and CPU. 2793A-PL WDC Controller present, 6810 RAM and 6532 RIOT present. JP 2/4 OPEN meaning Mask ROM configuration.

Bottom Right: Only 6810 RAM missing. Looks like 2793A-PL (Blurry?) WDC Controller. JP 2/4 OPEN meaning Mask ROM configuration.

Lots of upgrade possibilities for these boards, but most of the chips still can be easily sourced from BEST or eBay. You can change the ROM jumpers on the drives to make them work with EPROMs if needed, or use various upgrades. Happy or Speedy boards require removal of OS and CPU, and then plug into the CPU socket. For US Doubler you replace the ROM, and the 6810 RAM with either with a genuine ICD 'Module', or a homebrew stacked 6810 like I did here: http://atariage.com/forums/topic/220482-1050-us-doubler-speed/?p=4011333

When selecting a stock 1050 OS ROM, here's the differences in the most common stock firmware floating around:

• Revision J & K: Supports 2793 WDC controller with Tandon mech.
• Revision L: Supports 2793, and adds support for 2797 controller with Tandon mech.
• Revision "WST R5" - Modified version of Revision K, Supports 2793, NOT 2797, modified (faster) stepper timings for World Storage Technologies mech.

The bent 4-pin headers in the bottom right should be a simple correction with needle nose pliers... but if they do break, pin header strips are common and cheap. Get a long strip of 40 or so, and you just break off the amount that you need. (In this case 4) Hopefully available at a local electronics shop, or wait forever to come from china: https://www.ebay.ca/itm/40Pin-2-54mm-Male-PCB-Single-Row-Straight-Header-Strip-Connector-for-Arduino-2pc/273135453542?epid=2289076391&hash=item3f98244566:g:RaUAAOSwaadavcYF

[+Nezgar]

Top left board Q7 looks missing? On production boards that is the 12V voltage regulator. Needed to spin the motor in the drive mech. If you wire one in, you should put it on the back right side of the heat sink to even out temperature distribution. In your picture, it even looks like there's a hole in that heatsink there, and an outline of where such a regulator may once have been....

[E474]

Hi,

 

Thanks very much for the help, I have made a little progress.

 

Top left board indeed has a missing voltage regulator (I assume that was mounted on the heat-sink (?)), also the potentiometer (?) at VR2 is missing. See photos.

 

The bottom left board has a damaged socket at U9, so probably I need to replace the socket or piggy-back a new one on. See photo. On close inspection of the other boards they all seem fine.

 

I tried moving the 6810 at U8 from the bottom left board into the bottom right board, as that was the simplest first step. When I powered up the bottom right board, nothing happened, not even the red power light at the front lit up. Any ideas?

 

After that I thought I would check all the socketed chips using my stock 1050. All the chips worked fine, so I have 1 spare set of stock chips already. That's as far as I have got.

 

I have another quick question. Is it OK to power up the motherboards without any socketed chips or mechanism? I just want to see if there is any sign of life, that is, does the red power light come on or not?

 

I would like to make sure the boards are all in working condition before I start to upgrade them, but I am also trying to be a bit cautious about my "repairs".

 

 

 

 

[_The Doctor__]

make sure the fuse isn't blown in the transformer wall supply

 

on board power supply section

check diodes

check capacitors

check transistors

report back your voltages

 

grab the 1050 fsm from atarimania

[E474]

Hi,

 

Thanks! If you mean the PSU (external power supply block), it's the same one I used for the stock 1050, so that's not to blame (no fuse problem, etc.).

 

All the socketed chips have now been removed, can I try powering up the boards with a known good external PSU, even though the socketed chips have been removed? I think this would be the quickest test to see which board is closest to being good, but I don't know if it's a good idea to try this without any socketed chips present.

 

Otherwise it is FSM + checks listed (*4)!

[+Nezgar]

Red power light should still come on with no main IC's... if that's not lighting, check if there's any 5V coming from the 7805 regulator. And maybe if not, look further up the chain, ie the big ac smoothing caps, the rectifier diodes, and then maybe the barrel power connector itself has a cold / broken solder joint.

[E474]

Hi,

 

Thanks for the help. I tested each board without socketed chips:

 

- top left: no power light

- top right: power light

- bottom left: power light

- bottom right: no power light

 

The bottom left has a damaged U9 socket, so I am leaving that one alone until I can get a replacement socket.

 

I put the spare (tested) 1050 chips in the top right, and connected a good mechanism. When I powered up the drive, the stepper motor positioned the head OK, but the motor didn't spin the drive. When I closed the drive door, the motor also didn't spin. I'm guessing this board is closest to being fixed - does anyone have any idea if this is an obvious fix, or is it checking components one by one?

 

Thanks!

 

[_The Doctor__]

make sure the wire blocks are attached correctly for which mech you have... some need to be reversed in the spot they plug in....

[xrbrevin]

also, make sure you have the correct ROM/EPROM type that matches your jumpers.

Incorrect ROM arrangement can cause odd results!

[+Nezgar]

Stepper motor positioning the head indicates the ROM is probably working, running POST. You can confirm if the drive responds to computer trying to boot. (Boot error with drive on).

 

One of the 4-pin header blocks is specifically for the 12V motor power, and tachometer feedback. Trace the wires to see where they go) As the doctor mentioned, ensure it's not connected backwards. If still no go, the motor may be dead in that particular mech - have you verified that mech in a known good controller board?

 

Oh, simple test - see if the 7812 regulator is actually outputting 12V.

[E474]

Hi,

 

I have had a little bit more time for this project. I put a socket in the damaged 6507 socket (piggy-backed the socket) PCB (bottom left of the 4 boards), then a good 6507 in the piggy-backed socket. I attached a known good Tandon mechanism, and powered up. The drive motor started spinning, but did not stop (at least 30 - 40 seconds). The stepper motor didn't move at all. I tried booting, but the 800XL dropped in to self test mode, so the drive isn't responding to SIO commands.

 

I then tried the PCB using the 6507 directly in the damaged socket, but the motor didn't spin at all. I tried a few different combinations, but it seems like the motor will only spin if the barrel connector of the power supply is removed, then re-inserted (sometimes I had to wait a few seconds to a minute), and only when there is a piggy backed socket in place. If the 6507 is directly in the damaged socket, the motor never spins.

 

If the 6507 is in the piggy-backed socket, and I use the power switch to power cycle the board, then the motor doesn't spin up on subsequent power cycles. If I unplug the barrel power connector, wait a minute, re-insert the barrel connector, power up, the motor spins (forever).

 

This is all with a known good Tandon mechanism and good chips.

 

I'm going to try the other board that lights up the power light next, but thought this one would be a quick test.

 

Any help and advice would be appreciated.

[_The Doctor__]

make sure all voltages from power supply sections are correct before pluggin in those good chips... make sure the mask jumpers match the same as the working board you got the chips from.

 

You can't trust piggy back into a bad socket to work... you must replace the socket...

[E474]

OK, though it looked like there was more activity with the piggy backed configuration than with the 6507 going in to the damaged socket. Either way, I will replace the damaged socket as a next step.

[E474]

Just tested the other board that showed signs of life. The stepper motor activates, but the motor is still not spinning. However, when it is plugged into an 800 XL and booted, I get "Boot Error", they come up very quickly, e.g. 1 "Boot Error" / second, or quicker.

[_The Doctor__]

great news..that's super close to being there, should be easier to use the fsm to track it down now...

[+Nezgar]

Just tested the other board that showed signs of life. The stepper motor activates, but the motor is still not spinning. However, when it is plugged into an 800 XL and booted, I get "Boot Error", they come up very quickly, e.g. 1 "Boot Error" / second, or quicker.

This is very good. Means the logic board is running, and and that is the expected behaviour of the drive latch open. Its immediately NAKing any I/O request. Check your 12V regulator output, and see if the BOOT ERROR behavior changes with the drive latch closed.

[DavidMil]

Another thing to note is that Q6 (at least on my 1050 board that looks like your upper left board) needs to be isolated from the aluminum

heat sink with a plastic insulator kit to prevent it from grounding. If you plug it in and Q6 starts to very hot, very quickly turn off the power

and loosen the chip from the aluminum plate and look for the plastic. When I tested mine the insulator kit was missing and man, Q6 got

hot really fast! You can test this without any chips in the board.

 

David

Edited October 21, 2018 by DavidMil

[1050]

You've confused Q6 with Q7 (+12) missing in upper left photo

against the back right side of the heat sink. The leads

for which are also missing in the center of the board,

clearly marked Q7 (+12) and in a horizontal line with Q6

and Q8 (+5).

Q6 lives on the left front of the heat sink and is

the only one that does require a mica insulator be

mounted between it and the heat sink, without the

heat sink, 12 volt buss is shorted to ground and the

drive motor will never spin up.

 

Don't worry about it, I mess up all the time myself.

[DavidMil]

You've confused Q6 with Q7 (+12) missing in upper left photo

against the back right side of the heat sink. The leads

for which are also missing in the center of the board,

clearly marked Q7 (+12) and in a horizontal line with Q6

and Q8 (+5).

Q6 lives on the left front of the heat sink and is

the only one that does require a mica insulator be

mounted between it and the heat sink, without the

heat sink, 12 volt buss is shorted to ground and the

drive motor will never spin up.

 

Don't worry about it, I mess up all the time myself.

 

If you are replying to me, I think you may have misread my post. We are saying the exact same thing.

 

David

[1050]

I was and I was sure you mentioned Q6 as the one missing

in the vacant spot, perhaps something similar was

misunderstood in the version before edit that I

happened upon? The current version is solid stuff.

 

At any rate proof positive that I can be a basket

case at times.