HP-41 modification: what is this (see photos)?

[Cococollector]

An HP-41CX fullnut I bought long ago shows a lengthy hole on the underside near one of the rubber feet.
It‘s (the calc‘s, not the foot‘s) speed is quite double of that of my other 41s!
Opening it, I find an extra PCB and additional wiring. CAT 2 doesn’t reveal any hidden ROMs.

What have I found here??? Any clues? A turbo modification?

Klaus

[Allen]

What do you get running EMDIR?

[KimH]

This might be a "Key Triggered" Turbo.

The C4 Capacitor is missing and the wires around the Inductor would indicate something like this.

The chips on the additional PCB (PCX) could be "Flip-Flops" and the additional thin wires from the PCB could be the ones determining the KEYs to press to go UP or DOWN.

Also the Beeper seems to have been replaced by something smaller, it would normally be taped onto the CPU and much larger in brass type metal.

So now... what is the key-code?

[ThomasM]

Yes, it is probably an early version of the 41 Turbo-2, a Keyboard Controlled Speedup made by PETSS Electronics in Belgium. You ca find a short description (only in german language, sorry) here:
https://calc.fjk.ch/library/wuw_cursor_88_1.pdf

[Cococollector]

@Allen: EMROOM reveals 753, but not in the expected way (see 1st image). Looks like SW mixed with ED. And I can edit the number to the right, <- deletes and replaces with _, 1-9 replaces _, R/S quits with either new number oder NONEXISTENT depending on unknown condition. ED shows same display except for an R instead of D. With EMDIR it shows the SW face (oh, uh, Apple watch has faces, 41 just a display), which I can only get rid of via ON. R/S does nothing in this state. Comparison:My halfnut shows D001, then X with proper number of registers, here 124.
Found strange behaviour in the CAT section: CAT 1 and 2 deliver properly, CAT 3-6 deliver… the same as CAT 2!
ON/ backspace Sequence leads to MEMORY LOST, but never shows the message itself. I only deduce this from other symptoms, i. e. resetted time, flags, display, size…
Conclusion for now:
additional X Memory confirmed, if unexpected size
I let it sit a while without energy to see whether something recovers.

Edit: MEMORY LOST shown again. Strange CAT behaviour remains.

@KimH:
Where does the C4 belong? On the 41’s board or on the new PCB? Do you have any schematics?
Any idea, what key sequence there could be? What was used in the past, if ever known? If there‘s a key sequence, what use could be the hole of?
I attached a close-up of the PCB area, the hole in the housing is situated over the end of the wire’s red insulation on the lower right.

[KimH]

(06-11-2022 09:07 AM)Cococollector Wrote:  @Allen: EMROOM reveals 753, but not in the expected way (see 1st image). Looks like SW mixed with ED. And I can edit the number to the right, <- deletes and replaces with _, 1-9 replaces _, R/S quits with either new number oder NONEXISTENT depending on unknown condition. ED shows same display except for an R instead of D. With EMDIR it shows the SW face (oh, uh, Apple watch has faces, 41 just a display), which I can only get rid of via ON. R/S does nothing in this state. Comparison:My halfnut shows D001, then X with proper number of registers, here 124.
Found strange behaviour in the CAT section: CAT 1 and 2 deliver properly, CAT 3-6 deliver… the same as CAT 2!
ON/ backspace Sequence leads to MEMORY LOST, but never shows the message itself. I only deduce this from other symptoms, i. e. resetted time, flags, display, size…
Conclusion for now:
additional X Memory confirmed, if unexpected size
I let it sit a while without energy to see whether something recovers.

@KimH:
Where does the C4 belong? On the 41’s board or on the new PCB? Do you have any schematics?
Any idea, what key sequence there could be? What was used in the past, if ever known? If there‘s a key sequence, what use could be the hole of?
I attached a close-up of the PCB area, the hole in the housing is situated over the end of the wire’s red insulation on the lower right.

All i Can sat is that the 130pF should be under the bundle of wires (between the 2 diodes) in your new picture. I replaced it with 33pf and that was close to 100% nothing on the Inductor

[KimH]

(06-11-2022 08:22 AM)ThomasM Wrote:  Yes, it is probably an early version of the 41 Turbo-2, a Keyboard Controlled Speedup made by PETSS Electronics in Belgium. You ca find a short description (only in german language, sorry) here:
https://calc.fjk.ch/library/wuw_cursor_88_1.pdf

So if I read this correctly the machine wakes up in turbo mode “*” and “-“ should speed down - for card reader and hp-il. Turning off and on resets to double speed

[Cococollector]

Thomas in his wisdom gave the right clue: it‘s the ETS turbo board made by PETSS in Belgium according to the aforementioned CURSOR article.
BEEP sounded in normal speed after pressing - (minus) and * simultaneously.
And yes: - and * set speed to normal, double speed after switching off and on again.

If it‘s an early version: is there a need for the capacitor maybe some variant?

Thanks a lot.

Klaus

[ThomasM]

Hello Klaus,
I do not own the version of the board you have, it looks a little different. So I think it's an earlier (or maybe later?) version.
I never installed it into my CX because I bought a W&W CY Turbo. So the Board is still in unused condition. I'll send you a copy of the Manual of the ETS Board (not a Manual, only 5 Pages) with the description and installation hints.
Hope this helps.

[Cococollector]

Hey great, thanks!

[Kiwi]

I think Petss even made a 2x turbo version.

[Cococollector]

After having been provided with a manual, I can confirm the key combination (* plus -) for normal speed. The faulty behaviour concerning X Mem functions remains.
I contacted Jo Vandale of PETSS in Belgium (interview with HP Solve on CES 20212 here) in hope of getting more insight.

Klaus

[Cococollector]

Though the hardware part seems to be solved, I post my further findings in this thread, not to open a new one:
I'm now sure, that commands having to do with X MEM lead to stop watch mode:

- ED: SW mode + 3-digit(!) difference display, sometimes alphanumerical, boxed star or other characters displayed (e. g. 00:00:00,0|D9*3, |=goose, *=boxed star). Registers always >600, which I deemed not possible.
- EMROOM: same as above, but with registers displayed (e. g. 00:00:00,0=D753)
- EMDIR: stop watch mode, no registers (00:00:00,00 ), leaving SW mode via SHIFT + <-, then X shows 1000 afterwards, no stack lift.
- CRFLAS: see EMDIR, X: 0,503
- CRFLD: see EMDIR, X: -0,513
- CLFL: see EMDIR: X: 0
- PURFL: see EMDIR, X: -1000
- SIZE?: see EMDIR: X: 0

Is there a connection between stop watch and extended memory (registers next to each other) and due to a maybe faulty RAM chip the content of one register gets swapped into one of the X MEM registers?
If not, that else could be the reason for this phenomenon?

Klaus

[Cococollector]

Update concerning ETS kit:
Jo from PETSS responded finally and is racking his brains.
Eagerly awaiting answers to my questions.