The Museum of HP Calculators

HP Forum Archive 14

[ Return to Index | Top of Index ]

HP-41 Clonix Adapter Clarification (hope so ;-)
Message #1 Posted by Diego Diaz on 23 Dec 2003, 1:12 p.m.

Hi all,

I've received some mails regarding what seems to be an obscure point in the Clonix ad.

Just to try putting things as clearer as possible here is an abstract of the Clonix functionality and (re)programming.

Clonix's processor is a PIC (a sort of multipurpose microcontroller) with FLASH ROM. This ROM can be re-programmed at user's will.

A serial "Programmer" uses the standard RS-232 serial port of your PC to program the PIC's ROM. It has a DIL socket, thus an "Adapter" (a tiny PCB board) is needed to lead every needed pin from the DIL socket into the Clonix module's edge connector.

(PC serial port) -> (Programmer) -> (Adapter) -> (Clonix)

-> means "connects to"

All the related info is detailed in the Clonix documentation (see appendix C), including a Programmer's schematic (public domain) and the Adapter PCB layout.

Hope this helps to clarify any misunderstood.

Thanks a lot and Season's Greetings to all of you.

Diego.

      
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #2 Posted by Giuseppe Marullo on 23 Dec 2003, 1:19 p.m.,
in response to message #1 by Diego Diaz

Diego, could you please tell me why is not possible to use a serial bootloader and save the pic programmer hassle?

I tell you so because I do have 2! Epic Programmers (ICSP on parallel port) but I seldom use them, much more easy to go with the bootloader and a simple serial cable.

Giuseppe Marullo

            
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #3 Posted by Diego Diaz on 23 Dec 2003, 5:51 p.m.,
in response to message #2 by Giuseppe Marullo

Hi Giuseppe,

The main reason for not including the, otherwise useful, bootloader into the Clonix code is that (please correct me if I'm wrong) it needs a proper interface chip (max232 or the alike) to handle RS-232 signals level, and there is no physical room for that chip inside the tiny module shell. Furthermore the RS-232 interface would need an extra connector (I can't figure out how to share both functionality in the module's edge connector) thus the bootloader implementation should become even more difficult. This and the low price of serial PIC programmers makes my decision on the subject.

Anyhow, these are the reasons I found when I look at the PIC Data Sheet and the Bootloader application note (AN851), and, as stated above, I may be wrong, should you have further info please mail me and I'll gladly study the possibities for further developements.

For those interested on the subject please take a look at:

http://www.users.bigpond.com/vk3jaj/p18f252/

You seem to have some expertise in PIC handling, so your help will be greatly appreciated.

Best wishes and Merry Christmas for you and yours.

Diego.

                  
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #4 Posted by Dave on 27 Dec 2003, 2:11 p.m.,
in response to message #3 by Diego Diaz

It is possible to do a simple serial/TTL conversion with a resistor and a diode. Check This page on piclist.com for several parsimonious TTL<->serial techniques. Also, single RS232-receivers are available very inexpensively and take up very little board space, such as this maxim chip. The reciever is probably technically a better solution, and will work in more circumstances. I imagine that an adapter could be made for a serial cable; perhaps some board space could be reclaimed by reducing the edge connector to a smaller connector, if fewer pins are used for serial programming.

**To delete this message, use "delete" **

                  
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #5 Posted by Giuseppe Marullo on 27 Dec 2003, 7:48 p.m.,
in response to message #3 by Diego Diaz

Diego, first of all Merry Xmas(late) and a Happy New Year to you and yours too.

I am not aware about the space problem inside the module, I was thinking about 4 tiny holes (.1" spaced) somewhere on the module to interface the pic. I need to build one before I could try a solution.

Don't use appnotes alone to evaluate bootloaders, usually there are better ones around, so their limitations vary widely. AFAIK the 18f comes with a little booloader already onboard, but I didn't use 18f yet. Now, it seems the time is right...

This way, you could avoid others adapters, and a special rs232 cable will handle all the functions. module plus cable, period.

The cable has 4 pins (gnd, +5V, Serin and Serout), with a max233 IC inside the DB9 connector.

It is a common solution on little robotic boards, usually the board provide the +5v for the converter, this time is better to provide it stealing from a USB port or a PS/2.

Yes, there is a way to steal some power from the serial port( thanks Dave), or to even avoid the max233 but there are drawbacks also. One of the most affordable design is used inside a Stamp2 (www.parallaxinc.com) but I don't have the schematic right now.

Diego, I have seen your documentation, and I will try to build one, but actually I don't have the pic handy. Here they are not easy to find in small qty.

I have 2 HP41C Memory modules that are going to be Clonixed. Will I ever need two of them at the same time?

Now "the" question:

On a 41CX, which is the best configuration now (Clonix era)?

HPIL interface + 1 Doubled EXT MEM + 1 Clonix + 1 Card reader

Which roms would you stuff inside the Clonix?

Giuseppe Marullo

                        
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #6 Posted by Raymond Del Tondo on 27 Dec 2003, 8:15 p.m.,
in response to message #5 by Giuseppe Marullo

Hi,

sorry for jumping into your discussion with Diego;-)

I just wanted to make my comment on the 'optimal' ROM config for the Clonix. My favourite config is as follows: David Assembler and Mainframe Labels ROM on pages 8 and 9, CCD Module on pages 10 and 11, and Extended IL ROM on page 15 .

Regards,

and a happy New Year to you and all others here.

Raymond

                              
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #7 Posted by Diego Diaz on 28 Dec 2003, 6:08 p.m.,
in response to message #6 by Raymond Del Tondo

Hi all,

With your permission, I'm going to make a turn in this thread, thus I'm going to make a question to you:

Why are you so interested in using a BootLoader?

What will the HP-41 community win by using a BL instead of the PIC programmer?.

In your own words:

"The cable has 4 pins (gnd, +5V, Serin and Serout), with a max233 IC inside the DB9 connector"

Well, should you admit the use of "active cables" then take alook at this:

http://www.olimex.com/dev/index.html

A nice PIC programmer inside the DB-9 shell, and it needs no extra power source!! :-)))

I can buid an adapter to plug at the PIC's end of this cable, basically the same wiring (5 pins) as for the standard programmer.

No extra connector needed. No risk of BL code corruption (a serious one). And cheaper indeed!!

Please take a look at a recent post to Dave's question:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/forum.cgi?read=49491

Apart from those HW & SW inconveniences, the BL code introduces a considerable delay in the PIC start-up time, I have not evaluated the effect of such delay in the overall Clonix behavior but it may potentially be harmfull.

Again, those are part of the inconveniences (to my mind) that makes my decision of choosing the PIC programmer approach and reject the BL option.

Your last question ("the" question), is quite more difficult indeed (thanks for your hint Raymond, I know it's a great config! BTW you're always wellcomed in my threads ;-). It all depends on the main use you need your 41 for. I've built over a dozen Clonix's for very different users and noone has asked for the same configuration. My best advice is to take a deep look into your needs, and choose wisely. Anyhow, the best thing of the so-called "Clonix-era" is that you no longer need to get stucked to the physical modules, should any configuration reveals as inapropriate for the intended use, just erase it and choose a better one!! Nice and clean!! ;-))

My best wishes for you and yours in the New Year.

Diego.

                                    
Re: HP-41 Clonix Adapter Clarification (hope so ;-)
Message #8 Posted by Giuseppe Marullo on 28 Dec 2003, 9:31 p.m.,
in response to message #7 by Diego Diaz

Wow, this little thingy is going to heat up our winter. You are always welcome to ask me(us) questions, as Raymond and all the others, obviously.

I would like to build one Clonix module before speaking too much about it and I have to say that is a very smart solution as is. That said:

>Why are you so interested in using a BootLoader?

It is a "method" like others with advantages and dis-advantages. The advantage I see here is that I always imagined that the module would have been a self contained unit. I don't like the idea to have a socket, a separate programmer (another!, cables, box ugh) etc. etc. I was thinking that the module connector is enough strange by itself.

I don't like the idea to depend on voltages generated from the serial port (laptops are terrible), so I will gladly steal power elsewhere. The olimex design is interesting, but I always got problems with the not-powered serial things in the long run.

I must admit that the serial programmer you are using is optimized and low cost but you are still tied to run where the pic programmer software is supported, dos for example. HP community could gain that they could use a wider range of equipment to program the clonix.

A serial bootloader can be fed from an Apple, dos, linux, PDAs.., whatever is able to feed byte on the serial port is able to reprogram the clonix. Think about reprogramming it from your Palm...

I don't know how often it is supposed to be reprogrammed, maybe this is not a big problem at all.

Anyway if there is the availability of a connector, you could basically program the module without any pod, and this is could be a reasonable "hack" of your clonix module.

>Apart from those HW & SW inconveniences, the BL code >introduces a considerable delay in the PIC start-up time, >I have not evaluated the effect of such delay in the >overall Clonix behavior but it may potentially be harmfull. This is an easy one, just place a magnet near the module when you want the bootloader activated and use a tiny hall sensor inside. If there is no magnet, the clonix runs normally. Or test the power supply when it comes from the other connector using a diode.

>Your last question ("the" question), is quite more >difficult indeed (thanks for your hint Raymond, I know >it's a great config! BTW you're always wellcomed in my I don't know much about the calculator, so I can't answer at the moment.

I have HPIL adapter, 2 X-Mem (I hope to fit them into one), a card reader so I could end with: 1 - X-MEM(double) 3 - CLONIX 2 - HPIL adapt 4 - CardReader / CLONIX2

Surely I need HPIL stuff, so EXT I/O and maybe DEVIL ROM should be present, but other than this, who knows. Raymond gave a great config, but where to grab rom images is another story.

I was thinking that a second CLONIX could be used to directly get the wild I/O I am researching... there are several I/O free on it, isn't right? Did you keep the SPI free? I am playing with a SecureDigital right now...

Giuseppe Marullo giuseppe.marullo@iname.com


[ Return to Index | Top of Index ]

Go back to the main exhibit hall