• Please review our updated Terms and Rules here

Pet 4040 drive blowing fuses

B

Bagpuss

Member
Joined
May 14, 2023
Messages
39
Location
United Kingdom
Hi, resurected my second PET how (bad A7 buffer and blown ROM). My first pet had to be a donor but it's getting a mini pet board now as the PCB was very poor.

But I am in the process of recovering a 4040 drive. I refurbished the disk drives and it did a proper under 3 seconds on led start up. I think though it's had a wobbler on me.

On my third power up it blew the fuse. I think it may be the RF filter has failed short as testing the on/off switch I get - On - near 0 ohms, off about 190 ohms. On the UK plug I also get 190 ohms L-N and open circuit in L-E and N-E which figures on blowing the fuse.

Removing the fuse and with the main switch off I still get 190ohms across the on/off switch so I think I am getting a short On the filter or transformer.

The filter is some whacky German thing with 2 blue wired and 2 black ones, no explanation of wiring.
I'll add pictures in my first update post.

I expect I can just cut the out but I get conflicting Google searches the it's blue input, black output Vs blue to blue & black to black (I suspect it's blue in black out from the wiring).

Now I have a second ratty 4040 as a donor but it looks like someone heath robinson'ed a repair but removing the RF filter, screw capped a connection and tucked a transformer wire out the way (which seems an odd thing to do).


So asking the hive mind here, are the RF filters like the old rifas, that let the burnt marshmallow smell out, nd just bypass it (and later get a replacement in there)?

I was thinking of pulling the whole transformer assembly from the donor but it looks a bit bodge repair to me.
 
Last edited:
Pictures for the 4040 drive.
 

Attachments

  • IMG_20240920_214851075.jpg
    IMG_20240920_214851075.jpg
    1.9 MB · Views: 10
  • IMG_20240920_214857549.jpg
    IMG_20240920_214857549.jpg
    2 MB · Views: 10
  • IMG_20240920_214857549.jpg
    IMG_20240920_214857549.jpg
    2 MB · Views: 10
  • IMG_20240920_214909357.jpg
    IMG_20240920_214909357.jpg
    2.1 MB · Views: 10
  • IMG_20240920_214932114.jpg
    IMG_20240920_214932114.jpg
    2.3 MB · Views: 11
Pictures from the donor with some other brand of tranformer
 

Attachments

  • IMG_20240920_214939787.jpg
    IMG_20240920_214939787.jpg
    2.8 MB · Views: 2
  • IMG_20240920_214945218.jpg
    IMG_20240920_214945218.jpg
    2.3 MB · Views: 2
  • 1726866046244313811375713897322.jpg
    1726866046244313811375713897322.jpg
    626.4 KB · Views: 2
Well, the transformer should be after the switch and fuse. So, if the switch is OFF and the fuse removed, it is unlikely to be the transformer that is causing the problem.

Can you sketch out the mains side of the power supply and post the sketch?

Dave
 
Yeah. I found two faults so far. The rocker switch had a slight contact whilst off. Just needed a little tweak of the output terminal (ordered a replacement). The rfi was showing 190ohms across the channels. I wasn't sure if that was right or not so have bypassed that and I have a new chassis rfi filter on the way.

Tomorrow I have a box of fuses turning up so I can give it another go. I'll draw up a diagram tomorrow, but essentially Negative goes to transformer terminal 2. Positive now goes to the fuse, the the switch the to transformer terminal 6. Switched off L E is open circuit and around. 30omhs closed ( will double check that again) which was the resistance across the transformer terminals.
 
I simply remove the line filter caps and discard them. One reason is that I have been replacing these things in a lot of equipment now for over 40 years and some of the ones I originally replaced are failing now. The power supplies in the PET computer neither puts out RFI onto the line, or is susceptabe to RFI on the line, this is the beauty of a line transformer based analog supply.
 
Here is the 4040 diagram. After removing the RFI it now powers up and does diagnostics correctly but I am having an issue that any command other than open freezes the PET (i presume this is because it is waiting on the drive response)
Using an SD2PET device the PET works fine had a few hours on it with no issues. I have a PET to IEEE adaptor running at IEEE GPIB cable to the drive unit. I'm using BASIC 4.0 and tried print DS$, DIRECTORY etc and they all just hang.
I assume for the cable there is nothing fancy going on internally, all lines on the cable test with a multimeter fine and I didn't find any cross cable shorts. I can see both drives spin up fine, and the heads move to relocate.
I also tried a listing to flash the drive LEDs and it just hung. I tried sending a NEW command to drive 0 and that hung, no activity on the 4040. Not sure what to try next. My first thought is a bad IO chip on the 4040 side. On the PET repair (a 3032) I had to replace the A7 buffer as all 4 upper bits were high on testing GPIB data but all the other diagnostics checked out fine. Once the buffer was replaced of course the SD2PET works fine as well.
 

Attachments

  • 4040 Power.png
    4040 Power.png
    4.8 KB · Views: 6
So the RFI filter was definitely kaput then!

One thing I have done is to manually drive all of the IEEE488 signals one at a time from the PET (using POKEs) and observing the effect on the disk drive on the 'far side' of the buffers. That will rule in or out the IEEE488 buffers in the disk drive.

See http://primrosebank.net/computers/pet/documents/PET_and_the_IEEE488_Bus_text.pdf page 28 onwards. You want to look for the POKE statements to drive the IEEE488 signals.

Dave
 
Cool. I'll try and push data out and see what happens. I guess I can take the assumption if the board starts up clean but them comms hang, a buffer is most likely .
 
Probably, but don't forget it could be the buffers from the drive to the PET...

Do one side first (PET to drive) and we can worry about the other direction later...

Dave
 
Yep. I saw a post from a few years back on testing by poking the ports and checking with a logic probe.

Sods law, I can't find my logic probe so got another on the way tomorrow.
 
As noted it is a good book that Pet and the IEEE 488 Bus (GPIB) book by Eugine Fisher - C.W Jensen mentioned in post #8.

Before I read this book I didn't have a clue how the GPIB interface worked, but afterwards I was able to figure things out, for example how the Taylor Wilson GPIB to serial printer interface worked. I got the real paper book.
 
Him with the it all plugged in and setting
POKE 59456, 255
POKE 59456, 251
POKE 59456, 255

ATN does not change on the 4040 side it's fine in the pet side)
I can see ua4 pin 1 stuck low and ub2 pin 2 stuck high.
I've tested all lines on through cable and they are good so I don't suspect the cable (I'm using a regular gpib cable with the pet adaptor and tested all that for continuity on all pins).

The drive goes through the flags 3 LEDs then the disk drive sets to home so I am assuming that it's starting up OK.
Thinking that is the io chips likley at fault but also will look at dumping the roms to test them at the weekend.
 
You must log in or register to reply here.
Back
Top