Spice Low Power

[jjohnson873]

(10-04-2019 11:26 AM)Maximilian Hohmann Wrote:  Hello!

And regarding the original topic: Great as usual! Unfortunately my non-working "spices" have also some mechanical damage to the casing (due to the cheap brittle plastic that HP used for this series) and/or lots of corrosion damage due to barttery leakage which even includes the displays. So no suitable donor for an upgrade unfortunately.

Regards
Max

I echo this problem with battery corrosion that typically eats the copper in the flex cable used for charging as well as the battery terminals. I repaired an HP-34C last year and used some spring metal for a new battery contact and a three wire ribbon cable to reconnect the charging pc board back to the calculator electronics. Only after I was finished did I see the hairline crack in the case. While the very fine crack didn't affect the operation of the HP-34C, it was visible if you looked very hard in a bright light. ~ Jim J. ~

[Dave Frederickson]

(10-08-2019 12:36 AM)jjohnson873 Wrote:  I repaired an HP-34C last year and used some spring metal for a new battery contact and a three wire ribbon cable to reconnect the charging pc board back to the calculator electronics.

My 34C needs a new battery contact, too. Where did you get the spring metal?

Dave

[PANAMATIK]

(10-08-2019 12:36 AM)jjohnson873 Wrote:  I repaired an HP-34C last year and used some spring metal for a new battery contact and a three wire ribbon cable to reconnect the charging pc board back to the calculator electronics.

Could you show some pictures of the new battery contacts? It could inspire me for making a ribbon cable replacement.

Bernhard

[BobVA]

I just got a 34C as a possible LP conversion candidate. Battery contacts and ribbon cable were eaten away, but I spliced some wire-wrap wire to the last good bit of the ribbon, cleaned the power switch and was happy to see the calculator worked. For a day.

It then developed erratic operation, which seemed vary with pressure on the case, so I disassembled the "sandwich" (what a ghastly bit of engineering that is!), cleaned all the chips/contacts and reassembled it, but it didn't work any better and continued to degrade. Now it just lights a couple of segments.

I'm not sure if I damaged something, or if my tinkering was unrelated and perhaps it's a slowly failing power supply component that's the issue, but I was planning to convert it to an LP anyway, so...to my question: While I had it apart I noticed distinct impressions (grooves) in the printed circuit board under the IC legs, particularly under the leads of the large IC.

Is this normal? Or does it indicate that perhaps the printed circuit board is damaged (hence not being a good candidate for repair with the LP circuit)?

[jjohnson873]

Bernhart...I'll dig up the pics on the ribbon cable and post it.
As far as the clamshell construction....it was a poor choice IMHO, for HP to use this construction. Before you reassemble the IC's into the clamshell use DEOXIT GOLD to coat the pads of the pc board where the ICs leads contact. One of the challenges is the coplanarity of the IC leads. If one lead gets "lifted" or isn't coplanar with the rest of the leads, you'll have a lead that isn't in contact with the pc board and then your HP-34 won't be electrically functional.

[PANAMATIK]

(10-10-2019 03:12 AM)jjohnson873 Wrote:  Bernhart...I'll dig up the pics on the ribbon cable and post it.
As far as the clamshell construction....it was a poor choice IMHO, for HP to use this construction. Before you reassemble the IC's into the clamshell use DEOXIT GOLD to coat the pads of the pc board where the ICs leads contact. One of the challenges is the coplanarity of the IC leads. If one lead gets "lifted" or isn't coplanar with the rest of the leads, you'll have a lead that isn't in contact with the pc board and then your HP-34 won't be electrically functional.

I totally agree with the above. The grooves are the logical consequence of the plastic deformation of the board material with time and reason for bad contacts. Also oxidation is a problem. sometimes I had a black color coating (sulfid) on the pads. I used a drill with metal brush to clean the contacts, but very careful to not destroy the contacts.

The SPICE LP prototype boards are ordered and will arrive soon. Then I can check the contact behaviour. In my test calculator I have already put some solder to the pads for fixing the prototype wires, which also eliminates the grooves. I think this could possibly also a solution for repairing solderless boards without the LP.

Bernhard

[BobVA]

(10-10-2019 03:12 AM)jjohnson873 Wrote:  ...Before you reassemble the IC's into the clamshell use DEOXIT GOLD to coat the pads of the pc board where the ICs leads contact...If one lead gets "lifted" or isn't coplanar with the rest of the leads, you'll have a lead that isn't in contact with the pc board and then your HP-34 won't be electrically functional.

Thank you!

I did apply some DeOxit to the pads, but I'm not so sure about the second issue. I'll set IC's on something flat and see if I can spot any of the leads out of plane.

Unless it's extremely difficult I'll be very motivated to solder in the LP board, just to put an end to this nonsense. But I'll happily let Bernhard lead the way on testing that approach :-)

[PANAMATIK]

(10-10-2019 06:42 PM)BobVA Wrote:  Unless it's extremely difficult I'll be very motivated to solder in the LP board, just to put an end to this nonsense. But I'll happily let Bernhard lead the way on testing that approach :-)

I think I will recommend to solder the new circuit to the flexible board pads anyway, even before my first tests. Why should my Spice LP circuit have better solderless contacts than the original HP chips? It is a design flaw, which is not fixed, unless all components are soldered. The LP could only partially reduce problems, because it doesn't use the ROM and RAM connections, thus bad contacts will not result in catastrophic failure, but in missing LED segments or non reacting keyboards rows, which would be bad enough.

If the LP circuit for some reason (but hopefully not to expect) has to be removed later, it can be unsoldered relatively easy pin by pin, because the flexible pads can be depressed under each pin, while removing tin with a copper braid.

Bernhard

[PANAMATIK]

The First Spice Low Power board is running!

HP-34 SLP and original having the same display output.

For my testing and writing the SPICE LP software I couldnt use wobbly contacts, therefore I soldered the SLP and LED display to the flex board preliminary. All components are working; Real Time Clock, Flash memory, LED display and keyboard are doing their job, the software is complete, no hardware issues. I'm very happy.

I do recommend to solder the display anyway when you use the Spice LP. The solderless contacts are responsible for missing digits and LED segments of many of these calculators, which sometimes could be temporarily repaired when you hit the calculator on the table. It is easy to solder the display to the flex board, but solder only for one second to avoid melting the plastic. Also tin the display contacts before, they do often have an oxide layer.

Soldered SPICE LED display

The problem, which remains is the mechanical contact to the 40 pads. Unfortunately the pads of the flex board are outside of the circuit border, or do only touch the edges. There is no way for solderless inserting the SLP. I will try to solder the board for the first time in the next days. Second issue is that you have to remove some of the plastic from the board cover, which is no problem, but I will try to improve this with the next layout.

Fitting into the board space.

All SPICE models are implemented HP-31, HP-32, HP-33C, HP-34C, HP-37, HP-38C, but it is impossible to include also the classic or woodstock models, because the SPICE models have only 10 digits, instead of 12 (Woodstock) and 15 (Classic). But who wants to run a HP-25 on a HP-34C hardware anyway?

I received only two prototype boards. Fortunately both are OK. Now I have to test soldering the circuit to the flexboard pads and placing it inside the calculator case, then I will continue with an improved layout.

Is there any standard HP-34 program collection, which I could insert to the SLP? Or do you have an idea how to get good solderless contacts?

Bernhard

[BobVA]

Excellent news! That would be a great Christmas present for the community!

If you do develop a solderless installation process, I'd still like to have the option to solder it in, just to increase the reliability.

[Harald]

Very impressive Bernhard!
I must go through the box of spice parts and see if there is something in there that could be resurected

Cheers,
Harald

[PANAMATIK]

(11-08-2019 05:38 PM)BobVA Wrote:  Excellent news! That would be a great Christmas present for the community!

If you do develop a solderless installation process, I'd still like to have the option to solder it in, just to increase the reliability.

Glad to hear that soldering will not only be accepted but the preferred option. This will be the way to go. Indeed the HP engineers must have had bad times when they discovered that their product had a major design flaw with the reliability of the contacts.

Also I do not understand why they made a LED display with 11 digits, but the left most digit contains only a minus bar. Did this significantly reduce manufacturing cost? Or is the slightly longer Minus segment significant. As a result the Minus of the exponent has not the same length as the mantissa.

Nevertheless the SPICE models are fine calculators. They look more modern than the Woodstocks, and they are not brown.

Bernhard

[rprosperi]

(11-08-2019 10:34 AM)PANAMATIK Wrote:  The First Spice Low Power board is running!

Great news Bernhard, thanks for the update! Well you've got great reason to be pleased, your initial board design worked right from the start, that says a lot about your design skills and close attention to detail.

I agree that soldering these silly contact connections makes sense. 'Original condition' is always nice when it works reliably, but these early Spice machines (especially when aged) are too flaky. Having a stable and reliable 34C will be a welcome improvement.

As for supported models, the full Spice/Spike lineup is fine IMHO; just that line alone represents a nice variety of features and models to select among.

As for program collections, I have these 34C solutions books which I'm happy to send you (each book has ~10 sample programs):

34C Applications
34C Mathematics Applications
34C Statistics Applications
34C Surveying Applications
34C Student Engineering Applications

Wishing you best of luck as you resolve the final soldering and packaging details.

Looks like SLP will be another winner!!

[PANAMATIK]

(11-08-2019 09:58 PM)rprosperi Wrote:  Great news Bernhard, thanks for the update! Well you've got great reason to be pleased, your initial board design worked right from the start, that says a lot about your design skills and close attention to detail.

Was easier this time, because the WLP hardware and firmware paved the way. But I'm pleased though, could have gone wrong so many things.

(11-08-2019 09:58 PM)rprosperi Wrote:  As for program collections, I have these 34C solutions books which I'm happy to send you (each book has ~10 sample programs):

This would be great. Can you make scans of the relevant pages and send to my e-mail?

(11-08-2019 09:58 PM)rprosperi Wrote:  Looks like SLP will be another winner!!

Hope so. Thanks to calculator guys like you

[PANAMATIK]

SPICE Low Power First Light!

The first converted HP-34E SLP calculator is running in an HP-38E from batteries. It fits!

HP-38E converted to Multi SPICE calculator HP-34 SLP including HP-31E/32E/33C/34C/37E/38C

The SLP circuit is soldered to the flex board with isolated copper wires. There was no way to get reliable contacts in non soldered mode. But this will be an achievement anyway. The wires will be already attached to the SLP circuit and cut to the right length when you get it, and it allows to adjust the curcuit to the right place easily.



After soldering you just fold the circuit to its place. Also the display is soldered to its pads, no missing segments anymore.

The SLP circuit inside the inner case.

Still I did not attach the IR printing diode and the beeper, but this should be easy.

Now I have to prepare the manual chapter for a complete description of the replacement procedure. It was easier than I thought. But still not as easy as just replacing the solderless chip, this was not possible.

Bernhard

[Gene]

Excellent!

[albertofenini]

Excellent work, I have a couple of these to repair ...
Looking forward to fix them now ....

[rprosperi]

(11-19-2019 06:34 PM)PANAMATIK Wrote:  Now I have to prepare the manual chapter for a complete description of the replacement procedure. It was easier than I thought. But still not as easy as just replacing the solderless chip, this was not possible.

Great news, thanks for the progress report and photos Bernhard, very cool!

Do you expect to be able to fit a USB connector in the cramped Spice chassis?

Advice for readers new to the Forum or these topics: Don't mistake easy for Bernhard to mean actually easy. These things are relative.

[PANAMATIK]

(11-19-2019 11:01 PM)rprosperi Wrote:  Do you expect to be able to fit a USB connector in the cramped Spice chassis?

Space for the USB/RS232 converter inside seems OK, but where to place Micro USB connector? The batteries consume all space in the front. I have to try the charger slot.

(11-19-2019 11:01 PM)rprosperi Wrote:  Advice for readers new to the Forum or these topics: Don't mistake easy for Bernhard to mean actually easy. These things are relative.

Haha! But the most difficult part of the whole surgery - as you all know - is and remains to get the calculator opened.

Bernhard

[PANAMATIK]

SPICE Battery Adapter is under way.

Many SPICE calculators have the flexible printed circuit board corroded and need a replacement.



The good one and the bad one. I tried to repair the contacts with wires, but it doesn't look nice.

Therefore I made a layout for a replacement SPICE battery adapter and I'm waiting for the manufacturers offer. I hope producing the flexible foil is not too expensive. Also I hope the new battery contacts, which I want to use have 1mm more space for the todays AA batteries. The todays AA batteries are a small amount too long and cause to break the chassis when inserted.





The outline of the battery adapter PCB was difficult to draw and I hope they will not charge extra for this complicated outline.

I want to know how many boards would be needed to have an idea whether my costs can be paid off.

Bernhard