4 * 1050 boards fixup - general questions/strategy

[E474]

Just for the record, I am assuming that none of the testing should be done on a "live" (plugged in) PCB?

[DavidMil]

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.

 

No, I wouldn't do something like that. I would have replied again and apologized for the error. God knows I make enough of them!

If you'll notice the times of post and edit, both are done before your reply. I always read something after I've posted it to make sure

that my fingers stay up with my mind when I'm typing. I can see from rereading my post that I still missed some words. The gray

matter in my head has always run ahead of my muscles...

 

David

[E474]

For the record, some of the tests had to be done on a powered PCB.

 

One PCB (bottom right) has been fixed, drive speed has been calibrated, and passes all the 1050 SALT tests, apart from 2 hour enhanced density burn in test, as I haven't had time to run it (format in enhanced density, write dos files, boot dos files worked ok).

 

Top right is almost there - but is failing salt test with:

 

Motor start test: fails

Motor speed: fails: too-high

Track 00 sensor test: fails

 

(Other tests - write protect, etc., pass).

 

When I run the motor speed test/loop calibration, I am getting 0ms.

 

Drive spins floppy and stepper motor is good, I'm testing with a known good mechanism.

 

Any ideas on this one?

 

The other two boards are either waiting on parts, or need a socket replaced.

[Kyle22]

Does the mech that you tested and got bad track zero sensor, work on a known good controller board using the same test?

 

If the mech works, then look for missing / bad parts in the associated circuits on the bad board. Look at 1050 schematics.

 

If the mech still fails the test, then the problem is in the mech. Try a known good mech on the (maybe not) bad board.

[E474]

Ok, thanks very much. I've been using a known good mechanism, so it is definitely a problem on the PCB. I should get a chance to test it in the next day or so with the help of someone who understands circuit diagrams a lot better than I do.

 

The fixed PCB (bottom right) also has passed the 2 hour enhanced density test, so it has been restored to a fully working condition!

 

Also planning on replacing the damaged socket next, which should allow repairs to start on that PCB too.

[E474]

Hi,

 

I replaced the socket on the PCB and the drive works fine - passes all the 1050 SALT tests, apart from the 2 hour enhanced density burn-in which I haven't tried yet, so I am counting that drive as fixed.

 

The top right board still isn't working properly. Q8 (12 volt voltage regulator is fine). The potentiometer at VR2 is probably OK, because when I turned it by quite a lot you could hear the motor spinning the disk slower or faster.

 

The transistor at Q5 (2N4124) got replaced with a 337 (I don't have the exact spec/serial number, but can check). This didn't make any difference. I've been told the most likely problem is the 2917 at U5, so have sent off for some replacements.

 

BTW, Is there a simple way to test resistors or diodes without unsoldering them from the PCB?

 

Also, does anyone have a good source for WD2793 FDCs, ideally in China due to price factors?

 

All tests have been done with known good parts.

 

Any help would be appreciated, but I am expecting that not much more can be done without parts being delivered.

[+Nezgar]

You could just lift 1 leg out for resistors/diodes... Not ideal, but maybe better than fully removing...

You can use WD2793 with any 1050 ROM, or WD2797 with pin 25 bent out of the socket or cut, with Rev L ROM only.

WD2793 - lots on ebay. This one looks good, US seller so quicker shipping than from China, $5.95 + $3.75 shipping. ie: https://www.ebay.com/itm/382442442867

WD2797 - also lots on ebay - ie $4.29+$5.00 shipping from Poland. ie: https://www.ebay.com/itm/151779398055

TMS279X are also compatible direct replacements for the WD chips:

TMS2793 - only 2 on eBay, not great price
TMS2797 - Lots on ebay, as cheap as $2.50 each+$5.00 Shipping. ie: https://www.ebay.com/itm/153104717740

[E474]

Hi Nezgar,

 

Thanks for the tip on removing one leg, I can see how that would work. I'm relieved the socket replacement worked, it was a bit of a fiddly job, and I have to remove 9 DIP chips when I have time to upgrade my 130XE, so I am not looking forward to that.

 

I found a company called utsource.net that seems to have some in stock, and also maybe acts as a distribution channel. I'm not in a hurry to fix the drives (I never have enough free time, and also plenty of other things to do), so I don't mind ordering the parts, and then fixing whatever it is a while later.

 

It is actually nice to work on something that can be repaired, so much of what I have nowadays simply can't be fixed once it breaks.

[+Nezgar]

I found a company called utsource.net that seems to have some in stock, and also maybe acts as a distribution channel.

 

New site to me.... interesting, was initially excited they had a 'Canadian' site, but it looks like another Chinese distributor like AliExpress more focused on electronics. Thought it might be Indian, but 0086 country code contact phone number = China.

[xrbrevin]

use the 2797 and bend up the pin for double sided access (think its 25)

[BillC]

Something to be aware of is that the only stock 1050 firmware that's supposed to be compatible with the 2797 is Rev.L