Anyone want hp 41 to 41 infra-red data transfer?


Hi all.

I am developing an infra-red module for the 41 for engineers.
Many have expressed their desire for simple low cost data
transfer to the PC.

But would anyone want, need or use a system where two 41 machines
could transfer data, programs or key assignments from calculator to calculator?

It strikes me this could just be a useful add-on for firms with several machines used by a number of colleagues



A more useful module might be a replacement of the HP-IL module that does not exchange data with IL devices, but has a direct RS/232 or USB or parallel interface.
I do know that there are such gadgets for the IL bus, and I already use the 82166 converter to exchange data with the PC for quite some time, but HP-IL is nowadays exotic. When we are talking about adding hardware modules to the HP-41 - which I consider a great idea - then they should be universally usable and available, which is except occasional auctions at eBay, not really the case for the IL converter, for instance.

For bidirectional infra-red transfer (a sender-only module is available already, the IR-Printer module) with a PC you would need some standard like IRDA, which is strictly a serial protocol, so replacing the IR xmit/rcv diodes with RS232 drivers would be quite easy.



Interfacing to PC is possible by using the IL-Module and the HP-IL/PC interface card. For this application some nice software solutions exist, the leading tool is EMU41. By using EMU41 it is also possible to transfer data and programms between two HP-41 handhelds.

Also directly data exchange between two HP41 is possible by using two HP82166 IL-Converters. The powerfully HP82166 IL-Converter (nice to use for any hardware interfacing application) is available from some sources without problems :-)

If I know correctly, there exist infared interfacing solutions from HP-41 to PC which uses the Inrared Printer Module and an infared eye with special software. This application is undirectional, that means only data / programm upload to PC is possible.

People who like to use the HP-41 also prefer the belonging HP-IL system, because with this the fourty one mutates to an powerfully computer system...

With best regards from Germany - Christoph Klug


You misunderstood me (probably intentionally). I do use IL converters for interfacing purposes already.

The problem with such solutions is that they add a lot of stuff dangling from the handy HP-41 box. This is similar to the problem of the MAC-mini which is by itself a nice and small system, but if you need any additional component, you'll end up with a desktop full of additional boxes and cables and wall warts to powerup the whole stuff. When the original poster came up with his idea of an IR module, I immediately thought about a bidirectional version of the IR printer module - namely same in size and selfpowered by the HP41 box. The IL converter needs an external supply, hence a wall wart (at least two converters fit onto the busprint and only need a single supply.

Besides, I still consider IL as a proprietary, outdated solution although it was at its time technically interesting. With the documentation from the Kane/Ushijima book, one could rebuild the HPIL state machines in a CPLD, I suppose, and recreate IL converters, but the serious question (besides hobbyism and nostalgia) is why? More interesting might be nowadays an entire replacement of the CPU board in a HP41 box with some powerful CPU that emulates the whole machine at much higher speed.

However, this might be considered as heresy in a collector's forum...



Dear Holger,

from the hardware standpoint both, the HP-41 box and the HP-IL system are nostalgic solutions, today we have much better replacement systems. But most of us really like the fourty one system (including HP-IL) not only some exotic collectors :-)

From the technical standpoint your idea about replacing the internal HP-41 box hardware by a fast speed CPU solution (including RAM-Box features and some other advanced features) sounds really good. Than we use the identical operation mode with keyboard and display and handy size, but get much more calculating power...

Yes - the HP-41 is a vintage calculator or computer system - but also the device mixture of nostalgic HP-IL equipment plus modern add ons like I/O-Board, IL2000 or the MLDL2000 from Meindert or Clonix / NoVRAM from Diego make possible new and powerfully applications.

Therefor I think anything is welcome to HP-41 world : Using the old well known system parts same as new any hardware extension and future system solution. Please design all this new parts hardware compatible to the existing system - only than we are able to create fantastic HP-41 applications :-)

With best regards from Germany - Christoph Klug


Hey Holger!

I agree with you totally, for what it's worth.

I would love a fast hp41! If possible, I'd settle for something which was 16 times faster. Monte D's project if it gets to be reality, will be really great. I think it's an amazing effort.
He is aiming for 50 x normal speed.

I am even considering doing something down the track a bit,
either at 16 or 64 x normal speed. But what bothers me is the ROM copyright situation, since there is not a lot of point in keeping
the serial roms... so one would have to COPY the rom.

The other thing is, is that such a practical idea? We'll have to see what Monte comes up with. I am a little concernied about power consumption in these add-ons / replacements, in general.



Hi Chris,

HP-IL is a really nice old system. Howver, it is bulky and expensive. Surely now some of the mechanical stuff must be getting unreliable. And though this next factor is not usually a problem, it's also slow.

I am thinking of a fast compact, reliable two way system which
is reasonably priced, although it won't be "cheap". Something that can be used in-field which "emulates" ;-) the quality level of the
Hewlett Packard equipment itself. And which NEVER crashes the calculator.

If I was going to control instrumentation, I'd go for the modern equivalent of HP-IB. Why muck around with HP-IL at all, except for fun?



And though this next factor is not usually a problem, it's also slow.

The HP-41 was only able to operate the HPIL at about one-thirtieth of HPIL's maximum speed; so the bottleneck was the 41, not HPIL.

And which NEVER crashes the calculator.

HPIL never did. (I know you weren't saying it did-- I just wanted to clarify.) in 19 years of using my 41 daily, much of it with HPIL, the 41 only crashed a couple of times, and that was due to my own errors in synthetic programming. I have not had a MEMORY LOST condition either in over 15 years.

If I was going to control instrumentation, I'd go for the modern equivalent of HP-IB. Why muck around with HP-IL at all, except for fun?

Although not as fast as HPIB (IEEE-488), HPIL had some advantages over IEEE-488. These include auto-addressing and no address conflicts, the ability for the controller to find various pieces of equipment regardless of their position in the loop, interfacing to more devices at a time, longer connections possible, less-expensive and lighter cables, and the pulse transformers that allowed the grounds of two consecutive devices in the loop to be up to 1000 volts apart. In the late 1980's when my 41cx was mostly used to control a rack of IEEE-488 equipment, it was nice that IEEE-488 (HPIB) and HPIL were so much alike, because the interface converter was in most ways completely transparent, making it look to the 41 like the IEEE-488 equipment was actually HPIL equipment on the loop. The 41 insisted on being the controller on the HPIL, and did not have HPIL interrupt support; but these were limitations of the HP-41, not HPIL. The HP-71 does not have these limitations. When I've connected the 41 and 71 together, the 41 had to be the controller. For those who have more than one 41, it would be nice to be able to connect them together to transfer files and programs directly.

IR has its place, just not in interfacing to multiple pieces of equipment at once. To add a comment to someone else's about RS-232, Maxim's MAX3100 UART has an IrDA mode.


Hi Garth.

Thanks for the feedback, very pertinent and much appreciated.

I take your points on HP-IL. Yeah, they would be great features...
I hadn't actually considered all those great "selling points"... ;-)

If there was a large market for HP-IL I would be stirred to do something. But I guess most of the non mechanical hardware is still running fine. As for HP-IL on the PC, who cares? Maybe I am a bit closed minded on that one.

I would have really liked to have HP-IL when it came out. I can think of a lot of situations where it would have been rather useful, even in my own limited hobbyist electronics work at the time. Certainly I would never have used it at work, since my jobs were repairs, not testing and design. And far from high-tech.
"We're not in Santa Clara (anymore) Toto!". My working life was mainly a basic drudge involving a multimeter, a soldering iron and
a screwdriver (!), but "it's all good" as the young people say now.

I am seriously considering doing some development work towards an analogue voltage data logger in a plug-in ROMPACK. Any thoughts?



I'm an HP-IL enthusiast, so I'd be using it whether I had a practical application or not. But as of now, HP-IL is the best way I have to move data back and forth between my HP-41s and the rest of the world, what I like to call "the 21st Century" in this context. I realize that you are planning to do a better solution for the 41, and that the MLDL2000 will enable USB for a similar, but not identical purpose. But neither of those solutions are here right now, or, more to the point, were six months ago when I started feeding my old HP equipment habit.

But there are several things I do with HP-IL that no HP-41 based solution will do. Those are a) all the stuff I do with an HP-71B, and b) all the stuff I do with an HP-75. Enthusiasm for the HP-41 doesn't often translate to an equal enthusiasm for the 70 series. That means there are fewer options for a 70 series user today. So HP-IL remains the best solution for transferring data to and from those systems. I may be interested in purchasing any soluion for two way HP-41 data transfer you come up with, but it won't replace my need for HP-IL, let alone my enthusiasm. 8)

But a discussion of HP-IL's merits and demerits is interesting. It obviously failed, or was killed in the marketplace. HP announced the death of HP-IL at the same time they withdrew the HP-71B. Some HP-IL peripherals remained on the price list until the HP-41C reached end-of-life. But HP-IL as a commercial solution for data collection died with the 71B. HP noted then that they had basically no replacement to serve the market niche HP-IL was aimed at. That was low cost, battery operated test equipment. I think that partly this reflected the lowered cost of other interfaces, RS-232 in particular, and the systems that hosted them, such as the IBM PC and its clones. But partly this may have reflected lessened importance of battery powered data collection generally. I don't know if such systems were actually needed less, or if they just didnt grow as fast as other segments. Either way it would have lead to less market share.

Surveying is one place where battery power remained important, and where the HP-41C therefore remained popular. But that soon enough transitioned to specialized systems, and also the combination RS-232 and the HP-48. Today we see the HP-48GX holding on to importance in the surveying world, partly because of excellent custom hardware and software systems to interface to "total stations." I'm not a surveyor, so I don't know for sure why they remain so popular. But I can guess that they are relatively low cost, and low power solutions. That's the niche that HP-IL was designed to fill, oddly enough.


But a discussion of HP-IL's merits and demerits is interesting. It obviously failed, or was killed in the marketplace. HP announced the death of HP-IL at the same time they withdrew the HP-71B.

I say the "failure" was that of HP's marketing and management, not the product. The 71 especially was made for technical professionals; yet not one in ten even knew it existed. Of those who did, not one in ten had any idea how much power was there (which was of course partly due to the user groups and third-party suppliers). I can't imagine that HP recovered their development cost on the 71, but that was not the fault of the 71 or its designers. It probably made Susan Wechsler, Nathan Meyers, James Dickie, Thomas Lindberg, Stanley Blascow, James Donnelly, Laurence Grodd, Charles Patton, Robert Miller, and the rest of the design team just sick to hear their creation was being canned!


That must have been particularly tough when measured against the success of the HP-41C. HP really hit the "sweet spot" with that machine. I don't know if it's second in unit sales to the 12C (which surely must hold the top spot) or if some other machine holds that distinction. But if you combine longevity on the price list with enthusiasm for the machine then, and even more remarkably, now, you have got to hand the 4C the crown as HP's most successful calculator ever.

Still, I'm having more fun with the 70 series. The 71B is cool, but the 75C is more of a challenge, since there's less materual extant for it now.


Couldn't agree more!



I think I've gained a small insight into this topic over the last couple of days. The HP-IL card for the 80 series is rare. I finally obtained two in one lot. Plugging on in to my HP-87, I tried to get it to talk to my other machines and peripherals. It's probably a stretch to say that the HP-85 implementation of HP-IL is nearly as primitive as the 41C's, but I tend toward that point of view. I'll cover the technical details elsewhere, but lack of a full-function HP-IL implementation in the HP-85 series may have contributed to the failure of HP-IL as a product.


Dear Don,

your idea about creating a analogue voltage data logger plug in rom pack sounds really nice. Helpfully for this system would be a resolution of 12bit or more, also a multiplexer for more than one measurement channel would be nice.

For HP-41 was the CMT-300 digital voltmeter box available, which was designed from Corvallis Micro Technologie. Furthermore exist the HP3468 digital voltmeter which includes the HP-IL interface.

From CMT also the additional CMT-200 digital I/O box was available for HP41. This includes a 8bit digital Input / output ports. In combination with the voltmeter box you get a compete data logger system for HP41. Furthermore from CMT was available the fantastic RAM-Disk (with HP-IL). This includes 128KByte memory up to 512KByte memory plus RS232 interface option.

May be you are able to add some of this features to your analogue voltage data logger plug in rom. That means : Additional digital I/O lines for realising control applications plus RAM memory for storing data.

My own battery powered I/O-Board and the IL2000 System includes some analogue voltage input and output channels (with 8bit and 12bit resolution) plus some digital 16bit input / output ports for control applications.

With best wishes from Germany - Christoph Klug


Hi Chris.

Thanks (!) for the info...

You may want to check out my last post in the "Who moved my cheese?", thread Namir started...



If I was going to control instrumentation, I'd go for the modern equivalent of HP-IB. Why muck around with HP-IL at all, except for fun?

Interestingly, when you read Kane&Ushijima's book on HP-IL, there is very much in common between HP-IB and HP-IL, so one could in principle call HP-IL something like a serial version of HP-IB (surely there are differences in topology, in electrical parameters and the device identfication mechanims, but the logic behind both is rather similar).

Concerning control instrumenation, it should be also asked what the benefit of a 8 bit bus technology is nowadays. Serial communication is not slow by itself - see the PCI-Express connection lines in a modern PC. HP-IL took into consideration that the controller itself, i.e. here the HP-41, is not fast enough to feed large amounts of data to the ring. A CPU with a 300kHz clock, and effectively a 1/6kHz cycle time it power-saving, but definitely not fast (for normal calculations, you wouldn't need more anyway, BTW).

When the question were to design a new communication system to replace HP-IB, I'd rather think about some multi-master serial USB type than the conventional parallel HP-IB handshaking. Powerful, ad also fault-tolerant protocols are now easily written in software; HP-IB is a child of the 80s; it could be rather easily "coded" in TTL.



Multi-drop (RS485?) might be really nice.

The can bus is similar (I think...).



RS-485 multidrop might be nice, but it is not really multi-master. CAN is a totally different bus and requires a specific controller (some PIC's have it nuilt inside), and software may be somewhat complicated.

For easy interfacing with peripherals SPI or I2C could be useful, since there are ADCs, DAC, memories and many other interface IC's that can immediately connect to it. This is also realtively easy to implement in a microcontroller by bitbanging a port (if the controller does not already have I2C or SPI.

On your remarks about current consumption (further down in this thread): any microcontroller will consumer more power than the HP41 if you let it run at several MHz. I considered a microcontroller in my MLDL2000, but decided against it due to the high power consumption. When running, the MLDL2000 consumes about 6-7 mA, on top of the 5 mA that the '41 uses, and it used no extra power if the NUT is not running, apart from some uA's to keep the SRAM alive.

I would strongly advise a (small) CPLD to create an external interface, which could be powered from something outside when it is connected. My MLDL2000 uses USB power when that is connected.



Hi Meindert!

Yes, I have been thinking SPI or I2C. As you say there
is some nice hardware which uses it.

My memory is not too good about RS485...

I agree completely about your comments on USB.
I think I will get into CPLD, it's very nice.
I did some tinkering with Lattice stuff about ten years ago.



Dear Don,

supporting I2C-Bus would be a nice idea. Much hardware boards and chip solutions exist, which are controlled by I2C.

Tony Duell from HPCC created a I2C Bus interface system for the HP48 :-)))

With best regards - Christoph Klug


On your remarks about current consumption (further down in this thread): any microcontroller will consumer more power than the HP41 if you let it run at several MHz.

The mentioned MSP430 is listed with typically 500uA/MHz in active mode, and about 17-22uA in (32kHz) power-down mode (datasheet of msp430f1611, 48K Flash + 10K RAM, SPI, I2C, ADC).



Thanks for the info Holger! I may try to get a sample and
port the clonix firmware to the different chip. A one chip soluion to all my different projects would very nice.



It is probably a two-chip solution; you'd better have level shifters to interface to the HP bus, and maybe a tiny voltage regulator (Z diodes already consume to many mAmps). The MSP is 3.3V, however TTL tolerant at its I/O pins, but the HP41 is not even TTL compatible.


The MSP looks promising, but you will need to buffer the inputs (HP41 I/O is about 6.5V) and drive the outputs (HP41 is CMOS so you will need to reach the full 6.5V to keep power consumption low).

This is in general a problem to connect state-of-the art technology with good old TTL or CMOS hardware.



Hi Meindert. Thanks for your thoughts. I agree completely...

Can't be bothered with trying to pack in even more electronics...(!)

What a shame...

(Hand soldering one chip one resistor and a crystal was trying enough ;-)

However, I will try to find a nice CMOS microcontroller which will cope with 6 volts. I think that's why Diego had a 47 ohm resistor in the Vcc rail, to drop a little voltage. Never got around to asking him. Diego? Are you there? (Hi!) The thing is, when
the coconut is sleeping (display on) the 41 internal psu is generating 6.2 volts on the rail AND the PIC chip is not drawing any current so it is seeing the full 6.2 volts on it's VCC pin, so I do wonder a bit about the long term reliability of the basic Clonix circuit. I considered using a diode to drop 0.6 volts...


Edited: 23 Sept 2005, 8:28 a.m.


I didn't look at ANY specs at all on the chip...

I am not interested, as there is no space in a rompack for level shifters at all (!)

What a shame though...

Thanks guys!



This is all possible with the MLDL2000, it has USB. By replacing the USB print it can be used to build an interface to almost everything, that would be a simple hardware expansion by implementing the UART in the CPLD and add the RS232 level shifters in a very simple piece of hardware.
The design is completely open and when I have finished the current production run all design files and sources will be published.
In addition, the MLDL2000 will hold ALL currently known ROM images.



True, but for me MLDL2000 has the wrong form factor - it blocks the cardreader :-) In a cardreader case, you could even put a replacement for the entire HP41 CPU. This is why I talked about a replacement of the IL module.



I don't feel the need to transfer data or program between 2 41's. It's much more useful to have something that can transfer data and program between a 41 and a PC.


Hi Chan. Thank you for your feedback. I appreciate it.

What do you feel is the best? USB or universal RS232?
What operating system would you be running on the PC side?
What would you want the PC program to look like?
e.g. is text only okay or do you a Windows or X graphical environment?

What functions do you need in the system? Just program and data upload and download, or a complete machine memory dump? Would you be interested in a program execution TRACE? And are you interested in using it help write MCODE or just in the daily ("normal") use
of the calculator, e.g. data backup?



What do you feel is the best? USB or universal RS232?

USB is becoming far more common in consumer electronics that get operated as appliances by people who have absolutely no concern for what goes on inside the boxes, but RS-232 is much more friendly to anyone who would want to make their own equipment. I think the maker of MLDL2000 said that making a USB controller gets a lot more complicated and is impractical for the 41.

RS-232 requires a lot more power, but there have been some hand-held terminals made that had no batteries and operated entirely on power scavanged from the RS-232 lines coming from the equipment the terminal was connected to. Of course the power has to come from at least one end, so you couldn't connect two 41's together to transfer data if both ends were expecting the other to supply the power.

I myself do have the FSI164A HPIL-to-dual-RS232 interface converter; but I don't think there's any way to transfer a program over it unless you "print" it to the interface converter. Added funcions like this would be welcome.


What operating system would you be running on the PC side?

Linux or OpenBSD

What would you want the PC program to look like? e.g. is text only okay or do you a Windows or X graphical environment?

Text only is not only okay, it is prefered since it is scriptable.

What functions do you need in the system? Just program and data upload and download, or a complete machine memory dump? Would you be interested in a program execution TRACE? And are you interested in using it help write MCODE or just in the daily ("normal") use of the calculator, e.g. data backup?

All of the above :)


I had you in mind, Geir... ;-)



It does block the cardreader, but who needs a cardreader when you have the MLDL2000 ;)
You are right, even better would be the complete replacement of the NUT's guts, then there is need for a communication device the size of a module.



Hi Meindert.

Wouldn't the process of backing up your mag cards to your
mldl system be very laborious and maybe even wear out the 41 connectors (plugging and unplugging...)?

What if you have a power failure or corruption of the memory in your mldl2000? All that work doing the transfer down the drain...



Backing up your mag cards is a one-time exercise. Just load as much as possible cards in the '41 or in Extended Memory and then transfer to a ROM image (in the MLDL2000) with existing utilities. This can then be saved on the PC as a main backup.
When programmed in Flash the programs will be safe for a long long time, and can be reloaded form the PC backup whenever you want.
Some software could be written to make the process much easier and that should not be a very difficult task.



Hi Meindert. That project of yours is very nice. You certainly put a lot of work into it.

I am considering doing a stand-alone USB port for the 41.



Hi Holger.

I am also planning some sort of wired connection to a PC for the 41.
But that is a rather different project. I am hoping that will be
a straight RS232C cable which you can simply plug in to the 41 to upload and download either the complete machines memory, or just the user data and / or user program area. That is actually pretty involved, as is anything which connects to the 41.

The difference I see with the projects I am doing is that they must be 100% reliable. That is, they MUST NOT crash the calculator and make it lose it's memory. In other words, full commercial grade peripherals, noy hobbyist experimental equipment. It's a whole different kettle of fish...

Thanks very much for your feedback. It's appreciated.


Possibly Related Threads…
Thread Author Replies Views Last Post
  HP-41(CL): The easiest way to transfer FOCAL programs from a Linux PC to the HP-41 Geir Isene 13 5,609 12-05-2013, 02:40 AM
Last Post: Hans Brueggemann
  HP-41 module covers and side port cover Nate Martin 1 1,517 12-02-2013, 01:51 AM
Last Post: Giancarlo
  HP 41 Advanced Programming Tips Michael Fehlhammer 0 1,158 11-28-2013, 06:11 PM
Last Post: Michael Fehlhammer
  [HP-41][HP-71B][HP-75C/D][HP-IL] Found a mother-lode of programs on rdj 2 1,668 11-26-2013, 05:31 PM
Last Post: rdj
  Sheet data importer for HP Prime Marek Russ 4 1,827 11-15-2013, 04:55 AM
Last Post: debrouxl
  HP-41 Clonix&NoV's SW Update. (For the non-Primer's guys out there... :-) Diego Diaz 21 5,653 11-13-2013, 09:00 AM
Last Post: Ángel Martin
  Another HP 35 Red Dot on eBay Michael de Estrada 5 2,384 11-09-2013, 03:07 PM
Last Post: aj04062
  HP-41 MCODE: The Last Function - at last! Ángel Martin 0 1,031 11-08-2013, 05:11 AM
Last Post: Ángel Martin
  hp-41 Cards Marcel Samek 8 3,122 11-07-2013, 11:13 PM
Last Post: Garth Wilson
  Non-Prime question alert: Hp-41 and synthetic instructions Marcel Samek 11 3,993 11-04-2013, 09:31 PM
Last Post: sjthomas

Forum Jump: