HP 15c Collector's Edition save/restore does not work

[fossy]

Hello

Could anybody get this to work?

I have a HP 15c_CE with new firmware (HP_IAR_USB 120ms.bin - checksum 0A0Ah - 2024-06-03) and the VoyagerSave.exe from https://hpcalcs.com/download.

As mentioned in https://hpcalcs.com/downloads/apps/Voyag...ctions.pdf the calculator should go into HID mode after pressing [ON]/[f]&[ENTER] - and leave it with a hit on [ON].

Even this is not true for me. The calculator keeps only in the HID mode while I keep holding [f]&[ENTER]. When I release [f]&[ENTER] the calculator is switched on in normal operation and HID mode is off.

I cannot imagine that the holding from [f]&[ENTER] is the "plan" to save and restore calculators memory.

When I hold the calculator in HID mode with my left hand and key in with my right hand on the PC

Code:

VoyagerSave.exe -s memory

I get a file memory with exactly 0 Bytes. Even this does not work.

Does this work (and if how) for anybody?

[Divasson]

Hi,

It works for me.

While the calculator is off, press [f]+[ENTER], hold them pressed, and then press and release [on] (nothing happens yet), THEN release [f]+[ENTER] and the calculator switches on in HID mode.

The back-up file has always 2048 bytes - no matter the memory size mode we're in.

If you do not get a permanent HID link then we have a problem...I'd reload the firmware and check that it is the same using the SAM-BA check button.

[ijabbott]

My symptoms are slightly different. In Windows 11, I get a HID device called "HID-compliant vendor-defined device". When I run `VoyagerSave.exe -s filename`, it seems to hang there. I can Ctrl-C back to the CMD prompt, but no file has been created.

(I re-flashed the firmware and verified it just to make sure it was running the correct firmware, but it made no difference.)

Then I tried it in a different USB port and it worked!

The non-working USB device has a path like the following, and is a USB 2.0 port:

Code:

computer
 + ACPI x64-based PC
    + Microsoft ACPI-Compliant System
       + PCI Express Root Complex
          + PCI-to-PCI Bridge
             + AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft)
                + USB Root Hub (USB 3.0)
                   + Generic USB Hub
                      + Generic USB Hub
                         + USB Input Device
                            + HID-compliant vendor-defined device

A working device in a USB 2.0 port has a path like the following:

Code:

computer
 + ACPI x64-based PC
    + Microsoft ACPI-Compliant System
       + PCI Express Root Complex
          + PCI-to-PCI Bridge
             + AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft)
                + USB Root Hub (USB 3.0)
                   + Generic USB Hub
                      + USB Input Device
                         + HID-compliant vendor-defined device

It has one fewer levels of USB hubs than the non-working device.

Here is another working device in a USB 3.0 port on the same XHCI controller:

Code:

computer
 + ACPI x64-based PC
    + Microsoft ACPI-Compliant System
       + PCI Express Root Complex
          + PCI-to-PCI Bridge
             + AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft)
                + USB Root Hub (USB 3.0)
                   + USB Input Device
                      + HID-compliant vendor-defined device

Here is another working device in a USB 3.0 port on a different XHCI controller:

Code:

computer
 + ACPI x64-based PC
    + Microsoft ACPI-Compliant System
       + PCI Express Root Complex
          + PCI-to-PCI Bridge
             + Renasas USB 3.0 eXtensible Host Controller - 1.0 (Microsoft)
                + USB Root Hub (USB 3.0)
                   + USB Input Device
                      + HID-compliant vendor-defined device

So it seems to be sensitive to the number of USB hubs in the chain, or at least the number of USB 2.0 hubs.

[fossy]

Thanks ijabbott!

I had to change the USB-Port. Now it work's.
For me it works with USB2 and USB3. It did not work with a long cable. But firmware flashing worked with the long cable.

[Johnh]

I got my cable from Jose, and worked through the instructions. No real problems getting the firmware updated, with the right checksum etc, 0A0Ah -, dated 2024-06-03.

Observations so far:

It took a bit of squinting at the POGO plug to notice that the two hooked claws are different widths, and that this is what determines the right way around.
Instructions are clear, though the process is not very intuitive, great care needed at each step to select and enter each code
After the update, key-bounce can still happen, on key release if you linger too long on a key press. But with a consistent click and release, it seems better than before.

Can't yet get the calculator to appear at a USB port to use VoyagerSave, which is why I came to this thread. But tomorrow I'll try again using the advice in the posts above.

Naturally, in working through the updates above, I've lost all entered programs. I have them saved as a file, and in an emulator, but having to key them in has been stopping me from developing more tools for my structural design work. But if I can get the save feature to work, I'll be more confident to do more and properly utilise this great old machine.

[AnnoyedOne]

(09-27-2024 01:18 PM)Johnh Wrote:  Can't yet get the calculator to appear at a USB port to use VoyagerSave...

Did you power-on the calculator with f+ENTER+ON to invoke USB mode? (This only works with firmware 0A0Ah 2024-06-03), This isn't necessary when updating the firmware.

A1

[Johnh]

(09-27-2024 01:27 PM)AnnoyedOne Wrote:  

(09-27-2024 01:18 PM)Johnh Wrote:  Can't yet get the calculator to appear at a USB port to use VoyagerSave...

Did you power-on the calculator with f+ENTER+ON to invoke USB mode? (This only works with firmware 0A0Ah 2024-06-03), This isn't necessary when updating the firmware.

A1

Thanks yes, that's all fine.

I think I have a battery problem related to my issues. I've had the calc one year, on the original batteries still running well. But, before starting the updates yesterday, I switched off and removed one battery to test it. Putting it back, all data had been lost, The other hadn't maintained the memory. This morning i tested both and found 2.7V and 3.3V. I'm suspecting one battery has never been connected? So I replaced the low one with a better one, at about 3.05 V

Now the calculator is showing itself as a 'USB input device' under 'Human Interface Devices' in Windows 10. No sign of anything called 'Voyager' as described in the instructions, either there on in Windows Explorer. Do we expect it to appear in file manager/win explorer in the same way as a USB stick or USB connected phone would?

Im assuming it should connect, provided the firmware and f/enter -on keys are pressed as instructed, without starting Voyagersave? Audible beeps are indicating that it is connecting and disconnecting as expected,

Anyway, Im backing off until I can get a new set of batteries.

Any insights will be appreciated.

[Johnh]

....further on this, in case this helps others:

Battery issues?

I got a new set of CR2032's, both measuring 3.34V

Calculator switches on fine

I load up some inputs and a small program, all is good. I switch off.

Now I remove one battery, being the one nearest the centre. On replacing it, all data has gone, like a full reset. The other battery has not maintained the charge.

I repeat with removal of the battery nearest the corner. Data is maintained, in fact I can switch on and run it on just the center battery.

Does this suggest anything?


VoyagerSave

Its going now, and below I attach a screen shot.

In win10, I open up a command prompt, which for me, says
C:\users\johnh> (johnh is my user name on this pc)

I chose to put the downloaded VoyagerSave.exe file in the root of my C drive. I guess I can put it anyplace I choose so long as I know where.

i connect the POGO cable, calc is off, i press and hold 'f' and 'enter', press and release 'on', release 'f' and 'enter', and I'm connected

There is no reference to 'Voyager' anywhere on my PC, the calc is there as a USB input device, in 'Device Manager' under 'Human Interface Devices', but without reference to what it is. No reference in Explorer etc. But it seems to work ok.

So my command to save the memory was
C:\Users\johnh>c:\VoyagerSave.exe -s hp12240828a
hp12240828a was my choice of a file name

This file was written, and was placed in my 'johnh' user directory

So I did that, with a program and some inputs in the calc . then cleared all data in the calc and checked it was empty

Then the same command, replacing -s with -r, read my data and program back in again, and it ran fine after switching calc off and back on as normal

So I can load up my routines again, and write more now.

A bit worried about the battery issue though!

[AnnoyedOne]

(09-28-2024 02:25 AM)Johnh Wrote:  Does this suggest anything?

Yes. The batteries are supposedly in parallel. Do you have a multimeter (on ohms, no batteries) that can be used to verify this? Could be a simple connection issue.

(09-28-2024 02:25 AM)Johnh Wrote:  VoyagerSave

Its going now, and below I attach a screen shot.

Good. Its not rocket science.

(09-28-2024 02:25 AM)Johnh Wrote:  No reference in Explorer etc.

To get an icon/Windows Explorer reference USB Mass Storage, MTP (Media Transfer Protocol), PTP (Picture Transfer Protocol) or something custom (driver required) would've had to be implemented on the calculator. My guess none would fit (112KB total of code space). But HID would.

A1

[Johnh]

(09-28-2024 12:31 PM)AnnoyedOne Wrote:  Yes. The batteries are supposedly in parallel. Do you have a multimeter (on ohms, no batteries) that can be used to verify this? Could be a simple connection issue.

Thanks for replying, I did the multimeter tests, batteries removed. From the side contacts that connect to the exposed upper '+' battery faces, there is a direct zero-Ohms connection.
But there is no measurable connection at all between the two 'spider' '-ve' contacts under the batteries. this is at any Ohms range from 200 to 2M, with meter leads either way around.

Curious! the first time Id every opened the battery compartment after a year of use was yesterday. If its intended as a simple fully-parallel connection between two batteries, then it isn't that in my case. But it occurs to me that sometimes when two batteries are sharing a load in parallel, it is good practice to feed each of them through a forwards diode, to block direct current flow from one battery to another in the case that one has more charge. That would come with a small loss of voltage to the circuit, but would have suppressed my meter test?

Does anyone else notice a similar issue with an HP15C CE? ie, can you remove either battery when switched off, and then replace it and your data is still there when you switch on? or as on mine, does it reset as soon as you remove the more central battery?

I'm just out or warranty with this machine, which I bought from Calculator World in Australia. (If I had to do a work-around fix myself, I think I could solder a fine insulated wire between the '-ve' contacts, fed up and over via the recesses used to lift the batteries, without opening up the back. A last resort though....)

VoyagerSave

(09-28-2024 12:31 PM)AnnoyedOne Wrote:  Good. Its not rocket science.

I agree, its Computer Science!

Making these cable connections and updates is different to achieving similar on other consumer devices, such as phones or audio interfaces etc. Sam-ba and the use of the command prompt is not familiar territory for most of us these days. Im happy to work through it, and Im really happy that we have these features. But from this and other threads, some have encountered difficulties, and sharing experiences here can help.


John

[AnnoyedOne]

(09-28-2024 09:25 PM)Johnh Wrote:  If I had to do a work-around fix myself, I think I could solder a fine insulated wire between the '-ve' contacts, fed up and over via the recesses used to lift the batteries, without opening up the back.

That's what I do. Taking off the back means removing rubber fetuses, screws, etc. Or use a strip of Aluminium foil. Even easier. Make a (paper) template first though.

(09-28-2024 09:25 PM)Johnh Wrote:  I agree, its Computer Science!

I'm a retired electronic engineer. I've soldered more wires than you can count on your HP-15C CE Computer Scientists don't do hardware.

As for "command prompts" I was using those 40+ years ago Before PC's.

A1

[DA74254]

following this thread, and new to the 15C CE.
Can Moravia or The Calculator Store confirm if the 15C CE runs on 3 volts with the batteries in parallel or if it is running on 6 volts?
My 35s has 2 CR2032 cells, but apparently run on 3 volts with the batteries in parallel.

[AnnoyedOne]

(09-29-2024 05:35 PM)DA74254 Wrote:  ...confirm if the 15C CE runs on 3 volts with the batteries in parallel or if it is running on 6 volts?

The former. 3V. Two cells in parallel.

https://eb9376.thecommerce.es/en/i/the-i...a-hp15c-ce

If they were in series removing one cell would result in power loss. Mine doesn't.

A1

[AnnoyedOne]

(09-28-2024 09:25 PM)Johnh Wrote:  But it occurs to me that sometimes when two batteries are sharing a load in parallel, it is good practice to feed each of them through a forwards diode, to block direct current flow from one battery to another in the case that one has more charge.

Often that is best -- yes. In the case of two CR2032 cells in parallel their relatively high internal resistance makes diodes etc unnecessary.

With other batteries/power sources diodes might be necessary due to their low source impedance.

A1

[Geoff Quickfall]

Too add to the save restore functionality>

My windows 10 machine:

Run command com as admin.

Cd to correct directory for voyagesave.

Configure 15CE with three finger press as directed.

Get a chirp indicating computer as connected

Cannot find any connection notice in device menu anywhere.

However, the voyagesave -r Or -s works perfectly.

————————————————————————

My window 11machine (brad new asus)

All of the above but no chirp time, no device manager info.

But it works.

Go figure.

[Johnh]

(09-29-2024 07:19 PM)Geoff Quickfall Wrote:  Too add to the save restore functionality>

My windows 10 machine:

Run command com as admin.

Cd to correct directory for voyagesave.

Configure 15CE with three finger press as directed.

Get a chirp indicating computer as connected

Cannot find any connection notice in device menu anywhere.

However, the voyagesave -r Or -s works perfectly.

————————————————————————

My window 11machine (brad new asus)

All of the above but no chirp time, no device manager info.

But it works.

Go figure.

Yes that exactly corresponds to what I found. A few corresponding tweaks to the instructions online would have saved some head scratching this weekend.

[Johnh]

Thanks fir the great insights into my battery problem, and so it's a direct parallel connection of the two batteries that is intended.

Opening up to get right inside needs the rubber feet removed. I kind of like them, they are solid and neat. How likely is removal and replacement of them going to risk a less than good result? eg damage, unevenness or a tendency to fall off?

I may pursue a work-around just within the battery compartment, as discussed above.

[rprosperi]

I've not removed feet from my 15c CE, so not absolutely clear on the type of adhesive used or if they would stick well when re-installed, even if very carefully handled, etc. but it's not a big deal if you try it and they won't stick well as replacements are available here, and not expensive:

https://www.ebay.com/itm/145927326004

Some may consider $1.56/ea expensive, and indeed it is for about $0.04 of neoprene, but the seller deserves some margin for all the effort and making it easy for me, and I've found them quite nice.

[Johnh]

Thanks for responses! Ive fixed my battery issue with a simple workaround, described here:

https://www.hpmuseum.org/forum/thread-22...#pid192561