HP 97 card reader - tantalum capacitors

[Dreato]

I refurbed an old HP 97 card reader - O rings for drive wheel, superglued the worm gear. Cards fed well, but ALWAYS with errors. I found a post here that mentioned a problem with the tantalum capacitors in the power supply section on the main board. Tantalum caps usually last a long, however, mine were bad. I replaced the 47 uF and 22 uF caps and voila! all worked well. BTW, when I removed the old caps, the leads fell off - these caps were degraded.

Although it's almost always a mechanical issue with the card readers, in this case it was electrical, too.

[Francois Lanciault]

Hi,

Thanks for the hint.

I will try it with my non-working HP-97 card reader. Like you it works like a charm mechanically (after refurbishment) but will always shows errors when reading cards. (Writing works however).

François

[AndiGer]

François,
Please keep as informed ...
Have a 97 that writes but does not read the written card.
Sometimes reads one side of a card from the standard pac.
Thank you
Andi

[AndiGer]

François,
Did you do the job successfully?
Regards

[AndiGer]

After analyzing a different (guess newer) layout of the main board than shown in the service manual I draw the schematics of that layout to show which part is where. I attach to my answer here.

During the weekend I now changed C4 and C6 as the card reader didn't read. What should I say: success!
I measured the old tantalum caps after unsoldering. They still seem to have good values. 22µF nearly accurate, 47µF gives 52.5µF. Nevertheless there obviously was a problem with them.

I also added some comments to the schematics as IMHO there are some errors in the service manual. Table 4-6: see agenda in PDF file. Figure 4-10: Q2 should be named Q3 and Q3 as Q2.

Andi

[Jlouis]

Thanks Andi,

It will be very important to me when I fix mine.

Cheers

[AndiGer]

Flogging a dead horse, I know ... old thread.

These days had another HP-97 on the repair desk. Didn't read any card but today I tried and it wrote the card. Successfully read in another machine.

So I found this thread and did what I had done before:
Unsoldered C4 and changed to another one with good values. C6 was ok as is the other one I still have. So resoldered original C6. The cards are read now :-).

I checked Vss and Vgg. IIRC Vgg before changing C4 was -11.25V, now is -11.50V. Maybe that was the culprit? Design voltage according to the service manual is -12.00V.
Vss before and after change of C4 = +6.08V (design +6.25V).

[teenix]

(11-23-2021 02:58 PM)AndiGer Wrote:  Flogging a dead horse, I know ... old thread.

These days had another HP-97 on the repair desk. Didn't read any card but today I tried and it wrote the card. Successfully read in another machine.

So I found this thread and did what I had done before:
Unsoldered C4 and changed to another one with good values. C6 was ok as is the other one I still have. So resoldered original C6. The cards are read now :-).

I checked Vss and Vgg. IIRC Vgg before changing C4 was -11.25V, now is -11.50V. Maybe that was the culprit? Design voltage according to the service manual is -12.00V.
Vss before and after change of C4 = +6.08V (design +6.25V).

It is still good information :-)

This would make sense as C4 and C6 filter the Vgg and Vcc power rails from the switching power circuit. The sense chip is powered from the Vcc rail, so if it is overly noisy maybe the amplifier and level detector circuits won't work properly giving card errors.

From what I've seen, the sense chip is also finnicky about the supply voltage being close to 6V.

Maybe for HP-65's and HP-67's with card reader problems, it might be worth changing all the capacitors in their power supplies - 4 used in 65 and 3 for 67.

cheers

Tony

[albertofenini]

Definitely worth a try !!!
I’m thinking about the HPs I have tried
to repair last weekend
I need to check if I have some new capacitors and I’ll give it a try

Take care !!!

[AndiGer]

(11-23-2021 10:18 PM)teenix Wrote:  This would make sense as C4 and C6 filter the Vgg and Vcc power rails from the switching power circuit. The sense chip is powered from the Vcc rail, so if it is overly noisy maybe the amplifier and level detector circuits won't work properly giving card errors.

I had a second 97 on the bench for quite a long time that didn't read and write although I changed the mentioned caps to new electrolytic ones. After Tony explained the task of these caps I hooked both calculators to my oscilloscope and found ripple on Vss. Put in another one of the same charge and now ripple is gone - and cards are read.
(Now after some reads the coupler/damper is gone but that's another story.)

So I'm glad I posted here these days.

[albertofenini]

Hi everyone
Which capacitors have the same function on the
HP67 cpu board ?
Thanks for help !!

[AndiGer]

Alberto,
HP-67 and HP-97 both have the same CRC chip 1820-1751.
According to the HP-97 service manual I could locate Vss at pin 2 of that chip, Vgg at pin 15. That were the points I measured with the scope @50µs and saw the ripple of Vss.
If you find ripple as Vss I think you can track down which one is the appropriate cap for smoothing Vss as a direct connection to the mentioned pin 2 of the chip, opposite side is ground.
Maybe this link to Kees's page helps as well as it explains schematically how the HP-97 power supply works:
https://www.keesvandersanden.nl/calculat...supply.php
IMHO the HP-67 could be similar. The power supply for the HP-97 shown there is also similar to that of the shown classics HP-35 and HP-45. And as we learned the HP-67 is more like a woodstock. So I guess it is a 22µF cap as well like C5 in the shown HP-25 power supply. With that guess I think it is the red cap with blue foot left to the torroid.

Be sure you have a replacement cap with low ESR. The one I re-removed had more than 5 ohms as had the original C4 tantalum cap (with more than 3.5 ohms read failed). The working replacement has around 1 ohm. I used a (cheap) LCR-T4 to test the caps.

[teenix]

(11-25-2021 01:17 PM)AndiGer Wrote:  Flogging a dead horse, I know ... old thread.
- - -
So I'm glad I posted here these days.

More clues for the puzzles.

Keep flogging that horse :-)

cheers

Tony

[Francois Lanciault]

I wrote the second post on this thread way back. For the record, changing the capacitors did not solve my problem.

Hoping the Teenix new board will make it work!

François