Clonix issue with built-in HEPAX

[Jurgen Keller]

A friend of mine gave me a 41CX with a built-in HEPAX module 1C. He wants to have Zenrom and Skwidbc in addition. I thought a Clonix would do the trick and burned one. However, whatever configuration I choose it does not work, i.e., the 41CX behaves erratically when turned on and only cools down after a reset. It seems the problem is not related to the ROM image I program into the Clonix. I tried different ones, same effect. Any idea what is going wrong?

[Sylvain Cote]

(11-10-2023 02:52 PM)Jurgen Keller Wrote:  A friend of mine gave me a 41CX with a built-in HEPAX module 1C. He wants to have Zenrom and Skwidbc in addition. I thought a Clonix would do the trick and burned one. However, whatever configuration I choose it does not work, i.e., the 41CX behaves erratically when turned on and only cools down after a reset. It seems the problem is not related to the ROM image I program into the Clonix. I tried different ones, same effect. Any idea what is going wrong?

Thanks to Richard, Robert and Diego, I have an original HEPAX 1C module and Clonix modules, I will check on my side tonight.
I am assuming that you assigned pages C to F for the Clonix ROM images.

[Jurgen Keller]

(11-10-2023 03:18 PM)Sylvain Cote Wrote:  

(11-10-2023 02:52 PM)Jurgen Keller Wrote:  A friend of mine gave me a 41CX with a built-in HEPAX module 1C. He wants to have Zenrom and Skwidbc in addition. I thought a Clonix would do the trick and burned one. However, whatever configuration I choose it does not work, i.e., the 41CX behaves erratically when turned on and only cools down after a reset. It seems the problem is not related to the ROM image I program into the Clonix. I tried different ones, same effect. Any idea what is going wrong?

Thanks to Richard, Robert and Diego, I have an original HEPAX 1C module and Clonix modules, I will check on my side tonight.
I am assuming that you assigned pages C to F for the Clonix ROM images.

Exactly, I tried those pages. Looking forward to your results, very appreciated!

[Sylvain Cote]

Hardware used:

• K149 PIC programmer
• Clonix & NoV's Configuration Utility v4.2
• HP-41CX Fullnut
• Adv. HEPAX ROM+QROM module
• ZenROM ROM module
• NoV-64d configured as a Clonix-D → HEPAX not loaded and QROM disabled
  

Test #1 : 41CX + HEPAX (port #1) + ZenROM (port #4)

• Reset HEPAX content
  • OFF
  • remove module from calculator
  • wait 10 seconds
  • insert module in calculator
  • ON
    
• CAT 2
  
  Code:
  

  -HEPAX 1D
-ZENROM 3B
• HEPDIR
  
  Code:
  

  H:DIR EMPTY 
2610.0000 // free registers
• Create a data file
  
  Code:
  

  "ABC"
50
HCRFLD
• HEPDIR
  
  Code:
  

              // display: ABC .. DA .. 50
2558.0000   // free registers
• MCED
  
  Code:
  

        P.6  P.7  P.8  P.9  P.A  P.B  P.C  P.D  P.E  P.F  comment
x000  007  000  00B  00C  00D  00E  000  000  000  005  ROM ID
xFE7            000  008  009  00A                      HEPAX prev page, 000 = none
xFE8            009  00A  00B  000                      HEPAX next page, 000 = none
  • HEPAX ROM is at page 6
  • HEPAX RAM are at page 8 to B
  • HEPAX RAM file system links: 8 <-> 9 <-> A <-> B
  • ZENROM ROM is at page F (ROM module)
    

Test #2 : 41CX + HEPAX (port #1) + Clonix-D (port #4)

• Clonix-D config
  • Flash block #1 : ZENROM in page F
  • Flash block #2 : ZENROM in page F
    
• Reset HEPAX content
  • OFF
  • remove module from calculator
  • wait 10 seconds
  • insert module in calculator
  • ON
    
• CAT 2
  
  Code:
  

  -HEPAX 1D
-ZENROM 3B
• HEPDIR
  
  Code:
  

  H:DIR EMPTY 
2610.0000 // free registers
• Create a data file
  
  Code:
  

  "ABC"
50
HCRFLD
• HEPDIR
  
  Code:
  

              // display: ABC .. DA .. 50
2558.0000   // free registers
• MCED
  
  Code:
  

        P.6  P.7  P.8  P.9  P.A  P.B  P.C  P.D  P.E  P.F  comment
x000  007  000  00B  00C  00D  00E  000  000  000  005  ROM ID
xFE7            000  008  009  00A                      HEPAX prev page, 000 = none
xFE8            009  00A  00B  000                      HEPAX next page, 000 = none
  • HEPAX ROM is at page 6
  • HEPAX RAM are at page 8 to B
  • HEPAX RAM file system links: 8 <-> 9 <-> A <-> B
  • ZENROM ROM is at page F (Clonix-D module)
    

Test #3 : 41CX + HEPAX (port #1) + Clonix-D (port #4)

• Clonix-D config
  • Flash block #1 : ZENROM in page F ; SKWIDBC in page E
  • Flash block #2 : ZENROM in page F ; SKWIDBC in page E
    
• Reset HEPAX content
  • OFF
  • remove module from calculator
  • wait 10 seconds
  • insert module in calculator
  • ON
    
• CAT 2
  
  Code:
  

  -HEPAX 1D
-SKWIDBC
-ZENROM 3B
• HEPDIR
  
  Code:
  

  H:DIR EMPTY 
2610.0000 // free registers
• Create a data file
  
  Code:
  

  "ABC"
50
HCRFLD
• HEPDIR
  
  Code:
  

              // display: ABC .. DA .. 50
2558.0000   // free registers
• MCED
  
  Code:
  

        P.6  P.7  P.8  P.9  P.A  P.B  P.C  P.D  P.E  P.F  comment
x000  007  000  00B  00C  00D  00E  000  000  008  005  ROM ID
xFE7            000  008  009  00A                      HEPAX prev page, 000 = none
xFE8            009  00A  00B  000                      HEPAX next page, 000 = none
  • HEPAX ROM is at page 6
  • HEPAX RAM are at page 8 to B
  • HEPAX RAM file system links: 8 <-> 9 <-> A <-> B
  • SKWIDBC ROM is at page E (Clonix-D module)
  • ZENROM ROM is at page F (Clonix-D module)
    

Test Results

All the above tests were successful with my specific hardware setup.

[Jurgen Keller]

Sylvain, thank you much for your efforts and the detailed report. Good to see that - in principal - it works.

I've noticed a few things: the built-in HEPAX is version 1C, not a 1D. I don't know the differences between those two versions and if that could be a source for the problems.

Second thing is that I have a CLONIX-41, not a CLONIX-D. But again, I don't know if this makes any difference.

As I also have a NoV-32, I tried that with another 41CX, and I burned the HEPAX, Zenrom and SkwidBC into it. Works like a charm.

Third thing I noticed is that the file related functions did not work properly on the 41CX with built-in HEPAX. HEPDIR gave strange outputs and I could not create a file at all. I checked the links in the RAM pages and they were wrong, however, I could not edit them. The reason was that some of the RAM pages were write-protected. I unprotected all RAM pages and cleared them using CLRAM. In the end, the file related functions I tried, like HEPDIR, creating a file, etc., did work properly.

But ... when I insert the CLONIX-41, the calculator goes wild. Hmm, weird, I have no clue what's wrong with it.

[Jurgen Keller]

Some good news: I used HEXEDIT to fill 8FE0-8FFF with zeros and did the same for pages 9, A, B. Surprisingly, it works now!

Using MCED (available now and running fine) I verified the contents of the RAM page links and it's a little bit unexpected:

Code:


      P.6  P.7  P.8  P.9  P.A  P.B  P.C  P.D  P.E  P.F  comment
xFE7            00B  008  009  000                      HEPAX prev page, 000 = none
xFE8            009  00A  000  008                      HEPAX next page, 000 = none

HEPAX RAM file system links: B <-> 8 <-> 9 <-> A

I have no idea why that happened. Anyway, I changed it to the default and it still works fine (as expected).

On the one hand, I'm happy, but on the other hand I'm also wondering why this beast behaved that badly before. Usually, I want to know the root cause.

A big thanks to Sylvain whose detailed test report helped me a lot.

[Diego Diaz]

Hi Jürgen, Sylvain,

Sorry to get so late to the party.

As usual, Sylvain tests and result presentation are impeccable.

Some doubts about the built-in HEPAX behaviour arises:

Since real HEPAX use volatile SRAM, it should be erased every time the battery is removed from the HP-41CX. So there will be no need to CLRAM its contents.

Once the power is restored, the initialization program will allocate HEPAX code (usually into page #6 if no HP-IL or 82143A printer is present) and will build the HEPAX File System in the available memory assigning the usual XROMs B, C, D & E to pages 8, 9, A & B respectively.

@Jürgen, did you checked this without Clonix-(41/D) installed? This could help to confirm if the built-in HEPAX is behaving properly. BTW, the type of Clonix used (-D, -41 or -41d) will make no difference; or NoV-64(d) configured as Clonix-D for that matter...

@Sylvain, just a configuration hint, when using just one of the ROM banks in a Clonix-D (or NoV-64d) you can just mark the "Join Blocks" box to get the ROMs available regardless the port you plug the module in. This saves the need of programming both blocks with the same images.

Apart from that, I've also checked the proposed configuration (almost):
- ADV HEPAX but v.1D not 1C (real module) -> Port 1
- Clonix-41d: SKWIDBC pg. #C, ZENROM3B pg. #D -> Port 3

And found no problem at all.

Also intrigued by the unusual File System links described by Jürgen on his last post.

Please comment if you find the cause.

All the best.

Diego.

[Jurgen Keller]

(11-18-2023 06:30 AM)Diego Diaz Wrote:  Once the power is restored, the initialization program will allocate HEPAX code (usually into page #6 if no HP-IL or 82143A printer is present) and will build the HEPAX File System in the available memory assigning the usual XROMs B, C, D & E to pages 8, 9, A & B respectively.

@Jürgen, did you checked this without Clonix-(41/D) installed? This could help to confirm if the built-in HEPAX is behaving properly. BTW, the type of Clonix used (-D, -41 or -41d) will make no difference; or NoV-64(d) configured as Clonix-D for that matter...

Apart from that, I've also checked the proposed configuration (almost):
- ADV HEPAX but v.1D not 1C (real module) -> Port 1
- Clonix-41d: SKWIDBC pg. #C, ZENROM3B pg. #D -> Port 3

And found no problem at all.

Also intrigued by the unusual File System links described by Jürgen on his last post.

Yes, I checked that and the built-in Hepax seems to behave normally. The only thing I noticed was the unusual order of the Hepax RAM pages. However, I did not investigate that further.

Good to know that the Clonix type does not matter. That was not clear to me.

Thank you for double checking that the configuration with Zenrom and SkwidBC works on your system, too. As I stated earlier it now also works on the HP-41 with the built-in Hepax.

As this special 41CX is not mine and I already returned it to the owner, I unfortunately cannot do any further testing. However, the owner reported that when attaching a card reader, the problem pops up again and random characters appear in the display.

An unresolved question is also if the Clonix (configured with Zenrom at page E and SkwidBC at page F) has to be inserted into port 3 or can be inserted into any other port. I missed to test that (stupid me). I know there are some restrictions documented in the Hepax manual. It's not clear to me if the Hepax mapping done at startup can get confused by some configuration/port usage, or if Clonix and Hepax may interfere somehow in a bad way?

[Artur - Brasil]

Hi, Diego

I have the same problem with my NovRAM.
I "installed" one special module in some port, just for one function: get the current X-File name.
So, after some days, my CX became crazy.

I didn't have time to revert this situation.
Best regards from Brazil
Artur