Replace HP-67 head

[kubla]

I purchased an HP-67 off ebay. When I received it, the card reader wasn't working. After repairing the drive wheel, it worked mechanically, but generated an error message when reading or writing cards. I replaced the capacitors that are known to cause reader problems. It still gave error messages. I measured the head continuity; one side read open, so the head was bad. I removed the head and sent it to JRF Magnetic Sciences. John didn't have an exact match, so he sent me the best match he had. I installed the head and the calculator is now working perfectly. I have attached before and after pictures. I apologize for the poor workmanship, but I had to use leftovers from my parts bins, so I didn't even have the correct wiring: wrong colours and low-temp insulation. I may order the correct wiring and redo everything to clean up the workmanship, but I'm so happy to have it working, I may put that off for a while.

Before

After

Edited: 10 May 2010, 2:13 p.m.

[Vieira, Luiz C. (Brazil)]

Hi;

Congratulations! Great job, indeed. Matching wires and soldering after-work is easier than what you've done already.

The worst of all should be the gap alignment, but it seems you've done it well. Based on the 'after' picture you´d have no chance to correct it if misaligned, so I'd like to ask if you used any optical device other than magnifiers to check the gap alignment.

Cheers.

Luiz (Brazil)

[kubla]

I assumed the head gap was perpendicular to the long axis of the head face and eyeballed with a straight edge to put the head face in line with the grooves in the mounting plate. The OEM head looks a little like the one on the left in the following pic; the new head looks a little like the one on the right, so long-axis alignment was easier on the new head.

Of course, if the gap and head face weren't right, I would have been hooped. If I had to do it again, I'd use a jeweler's loupe to eyeball the gap and head face.

I shimmed the head into place using small pieces of aluminum MIG wire, flattened to the correct thickness for a light interference fit. Using a small machinist square, I aligned the head face with the top of the groves in the mounting plate. That provides enough clearance so the card just clears the head. Two of the little white balls then squish the card against the head as the card is pulled through. I tacked the head with dabs of epoxy applied to the original mounting holes in the plate. I then epoxied the rear of the head to hold the shims and head in place. If I did it over, I'd probably just tack the head so I could adjust it if I had to. Once tacked, I'd also snip off the tops of the shims so they don't interfere with the rear case.

Edited: 12 May 2010, 10:16 a.m. after one or more responses were posted

[Joel Setton (France)]

Looks like you did a super job with the new head, I'm impressed!

Of course the real alignment test would be to try exchanging magnetic cards with another HP-65. Interchange compatibility is the "acid test" for both head-gap alignment and card speed.

Joel Setton

[kubla]

I did that with my other HP-67 and HP-41. The fixed 67 can read cards written by the other 67, the other 67 can read cards written by the fixed 67, and the HP-41 can read cards written by either 67. I wish I had some preprogrammed cards to see how it works with them.

Edited: 10 May 2010, 3:30 p.m.

[Joel Setton (France)]

Great! The project does qualify as a complete success. Once again, congratulations!!!

[Katie Wasserman]

I don't recall reading about any other head replacements on the HP-67/97 before, that's great to know that compatible parts exist! Was the problem an open coil? I had that problem on an HP-9810 calculator but a fellow collector had a spare so I was able to successfully replace that with an original (used) part.

Thanks for posting your repair story and congratulations.

-Katie

[kubla]

Yes, it was an open coil.

[Joel Setton (France)]

It's nice to know that a spare part is available! It might be a good idea to post the exact part-number of this replacement head, for future reference.

Joel Setton

[kubla]

Part number is D188380 from JRF Magnetic Sciences.