Announcement: OpenRPN Project


Greetings all,

I posted a week or so ago proposing the ground-up development of a new calculator. Current news is that I'm hammering out details finding a host and will have a forum up and running within the next couple of weeks. Anyone with any interest is encouraged to contribute. Thanks in advance to all, I look forward to working as a community to design the perfect calc.


As I said before, count on me!

As a side note, I just quoted a local LCD display maker, to build a prototype display is about US$900 (machinery to build more than one)... ouch... but it is at least possible...

Best regards,



I've already thought of a solution for that, actually. If we use a dot-matrix LCD, the lowest row can be covered with a white silk screening leaving the desired characters clear so that "deg" "rad" "grad" etc. will appear. That way we can get away with a $10-20 LCD instead (at most).

PCBs can be made using toner transfer sheets, eyelets for through-holes, double sided alignment with some clever fixturing. Surface mount components can be applied using solder paste and a toaster oven.

I've found a source for cast aluminum cases, and am getting in touch with some resin modellers about making keytops. This should keep total costs under $100. As a side benefit, it will be waterproof and fire resistant.


The 16x2 or 20x2 LCD is fine, and cheap. I was only thinking in a more "elegant" solution in the future, and this custom-building isn't too expensive if you can reuse it for hundreds of displays. Just an idea... I really like 7 or 14 segment displays ;) They can do matrix displays too, with controller (COG also), very thin.

Fire and water proof is awesome! I like the idea of metal case with solid plastic inside as some (old) Sharp or Casio are, the feel is pretty good and the circuit is well protected. And with screws, NOT melted plastic pins!!

As we won't(?) profit with this project, there isn't the need to sell a new calculator every time the old one is damaged, if the old one doesn't get damaged at all... IF we can build a very good hardware.

I was thinking on to build the PCBs, they are very good and cheap.

For the keys... do you have any solution? Here are some silicon-rubber makers that can do a good job, but is silicon-rubber... It won't last long enough and doesn't have THE click! Double molding isn't necessary if we can "engrave" the signs on the keys with laser and fill in the traces with epoxy paint. Even if the paint goes off, is only a matter of repainting and cleaning.

I'm "polishing" some ideas to discuss later.

Best regards,



It certainly seems the best way to make the keys themselves in epoxy resin in a silicone mold. Very high detail, low cost, durable, and avaiable in many colors. The only hard part about "the click" is finding a good source for dome-sheets. It will need to be stainless steel or kevlar like HP has used before. I've found a few sources for these before, but I'll have to see if they can sell small enough runs.

All in all, things sound good so far. I really want these to be easily serviceable if they need it, but design them to be indestructible in the first place.

Once the forum is up we should be able to better organize a general design path and layout all of the important milestones.

Best regards,


What about those lettle switches that are used on so many things today--like control buttons on printers, on RC toys etc. They have a very powerful CLICK. ?




Absolutely. Tactile dome sheets are still readily available. One of my favorites is actually the keypad on my cell phone! So it's really a matter of finding a mfg who is willing to produce limited runs at a reasonable cost, or sell us surplus. It may even be worth getting in touch with kinpo as we may be able to source them "from the horse's mouth."

In all honesty, the most challenging parts to make will be the keyboard assembly and the lcd. Like I said, I've found a good way around the keytops... so we're down to just the dome sheets.


More good news, I've found a major supplier of stainless steel domes used in tactile keyboards. In the process of getting a quote right now. If anyone is interested, here's a link:

I've also gotten in touch with Randy Sloyer about some keytops to use as a basis for the molds (just a bit of reverse engineering). I'll be making some modifications, the main one being a rubber membrane will be fused to the edges of each key during the casting process. This can be taken out to the edges to function as a gasket, providing a watertight seal... Coffee spills, pink lemonade, dirt, dust, mold, no problem: put it under the sink for a few minutes!

Keep the ideas coming! So much of the logistics in making the case, keyboard, etc. are solved that we'll need to choose all of the necessary, processors, controllers, and I/O stuff soon.


A few configurations of off-the-shelf snap-dome keypads are available from Pac-Tec.

Since these are only rectangular arrays with fairly large spacing, they would not be ideal for a final design, but might be good for a prototype, or to scavenge some domes.


I also like the idea of using off-the-shelf tact switches: these would be easy to replace, come in a variety of tactile forces, and aren't all that expensive.


ALPS manufactures tactile switches for games, phones , toys, etc.

There are many types, travel distance, force, etc.
Some examples:

I was able to order a few samples from the local (Brasil) ALPS subsidiary.



I am in for this project, just don't make everything as cheap as possible. We are going for the perfect calculator, remember. Pick a fast processor with plenty of memory space. Most important for me is to offer I/O expansion and connection to a PC (USB???)



You can now buy quite cheap SMD single chip USB drivers. Use a mini USB connector.

Will the ROM be flash? Easily upgradeable. Use the USB port.


LOL, I think the possibly the best thing about a project like this in the HP community is that everyone is pretty unanimous about what they want. I've been finding the absolute best/most durable components I can. Like I said, I alreday decided to make the case waterproof and fire resistant!!!

Everyone wants tactile keys and a big enter key, plenty of flash rom space and low current requirements.

The one thing I fully anticipate is something that all open-development projects encounter (and I want to make this flexible enough to prevent a major schizm) is that sub-projects will result. The three major ones I can already see are just form factors: horizontal (voyager), upright (pioneer), and graphing. If we keep the components as modular as possible, we can certainly satisfy everyone involved.

I personally think this may need to go as far as creating a piece of software to allow each individual to build a custom rom by selecting key locations, and it generating a printable overlay. We certainly have the resources, the time, and the committed individuals to make this happen.

Keep the ideas coming!


If it simplifies things, I'd be happy with something that looks and functions like a lightning fast 41cx (but with a true two line display). I can count the number of times I've used the graphing function on my 48GX on one hand.

Take care.



I too rarely use the grpahing capabilities of my calculators. More often, I use the table feature, when analyzing graphs. It allows you to really "see" what's going on easier than with the picture itself. Perhaps a simple "table" app could be added in.

I believe the best way to go about designing this calculator is to have a single (or a few) calculator bodies, and then, via different roms and keyboard maps, make 2 or 3 different models (maybe like the 16c, 15c, and 12c were: all the same body, but different applications). I don't know too much about electronics, but if an uprgadable ROM is included, then it should be fairly simple to make the basic, fast, scientific calculator, the financial calculator, and the ultra-powerful uber scientific better than graphing calculator.

Just my thoughts



I agree with Wayne, the high-powered graphing capabilities of the so-called graphing calculators do not do much for me. I would like to lobby for a "super" 42S display. Normal display would be the whole 4 level stack, and Last X. If soft-labels will be used for the top row of keys, then display of the soft labels should be included simultaneously with display of the above registers, i.e., the soft labels do not reduce the number of stack levels that are displayed. If I recall correctly, the 42S display is 16 X 131 pixels for a two line display. So based on the above concept related to the 42S display, it would take a 48 X 131 pixel display to do the above, not counting any pixels used for annunciators. This pixel grid could then also be used for the rudimentary graphics capabilities of the 42S display.

Of course the above assumes that a 4 level stack will be used. I can see that some might want a 5, 6 or ? level stack. Stack height might be user definable feature. Some might even want an RPL-style “infinite” stack, although I am not a fan.



Excellent suggestions. My only concerns with a four or five line display are:

1) Fragility of a large screen.
2) Size of the calculator if it includes a large screen.
3) Cost of large screen display.

I also really like the 42S display. Perhaps compromise by adding one line to it (for a total of 3) to allow showing soft-key labels while still displaying x and y registers?

Take care.


PS-I do like the idea of an aluminum or steel body.


This Message was deleted. This empty message preserves the threading when a post with followup(s) is deleted.


Although not worth a huge amount of extra effort, I think a 7 segment (or dot matrix) LED option would be nice.

In my limited experience, LEDs usually have a much longer practical life than LCDs; my junk box has a significant quantity of LCDs that have gone black etc.

(And then there are nixie tubes for that "retro" feel :-)

Perhaps the display unit can be a separate module. This might help divide the effort too.


I have a lot of LCDs, some up to 20 years of age, and they're all in perfect condition. The calculator packaging should protect the LCD from being poked or twisted, but only common sense will keep it from sitting in the sun for long periods of time or getting the deadly high heat/humidity combination. The circuit design must make sure the LCD is not fed with DC. One of my LCDs is in a cheap (under $2) little clock I had on the car's dashboard for 12 years, from 1985 to 1997. The car was always parked outside and got awfully hot in the sun, but no humidity at the same time and little UV. There are different grades of LCDs too though. The widest specified operating temperature range I have is -20 deg C to +70, but my testing found it still somewhat useful from -25 to +85. At -25, it took a looooong time for the dots to turn off or on. At -27, it turned clear so you couldn't see anything. At +85, it was virtually negative. It all returned to normal when it got back into the specified operating range. I ran the temperature down to about -50 and up to nearly +100 for perhaps 15 minutes, and did not damage it. LCDs are sunlight-readable, take very little battery power, and make less electrical noise than something like a LED matrix or vacuum-fluorescent, which should make it easier to get the FCC certification for the product.

Edited: 15 Apr 2004, 2:51 a.m.


Like I've said before, I think it's essential to uphold a rigid standard that each major component be modular. In many open development projects, other projects with almost identical goals will break off and do their thing... There's no point in being divided when we can commit to pooling our talents.

That said, displays seem to be generating the greatest controversy as of yet. The two that I find most attractive are VFD's and LCD's. It's a real tradeoff of power consumption vs. reliability and visibility. I'm intrigued by the possibility of using lithium cell batteries as a power supply, as it may improve the unit's ability to handle a VFD.

I'd love to get some opinions (aside from single/double/graphing size, etc. don't worry, as I'm certain this project will grow to encompass plenty of options).


Even a small VFD module can take a couple hundred milliamps, meaning that operating from small batteries is pretty much out of the question. It still won't be sunlight-readable either. The good news is that there are standard modules in almost any kind of display. You can get modules with the same mounting holes and interface requirements from many manufacturers, with the drive circuitry built in.


Any consideration to a 28S/28C like 4-line display ?




"If we use a dot-matrix LCD, the lowest row can be covered with a white silk screening leaving the desired characters clear so that "deg" "rad" "grad" etc. will appear"

Why not a configurable HEADER area?



That's probably the perfect solution. I'll start looking into it. Thanks for the contribution!


I cast my vote too for this project. I used to me a member in PPC in the 80s and we did the PPC-ROM which showed the power and talents of a community likes ours. I also learned by attending the various PPC conferences that when folks like ourselves meet in person, the rate of communications goes through the roof!!

Here is what I feel we need to get organized a bit:

1. Have a more permanent forum to discuss this subject further. Maybe Dave Hicks can create a new area on his server. This way we stay on this site.

2. Have a group of voluneteers who are core organizers.

3. Plan for a conference so that all those who can travel and get together can meet in person.

4. Have fun doing this project!



I'll look forward to it. Sounds like it is going to be a blast! :)


I have no knowledge worth sharing except to say that you can count me in to purchase a calculator that this community develops. It might be worth having someone start collecting names/email addresses of those of us who would be willing to purchase one of these future calculators when they're finished (might help justify paying to develop necessary parts knowing you had 1000 future customers right from the start). I would pay a few hundred dollars to buy a one of these RPN calculators developed for reasons other than making a quick buck.

I'd pay even more if you can get NCEES Approval for use on the PE Exam :)


You raise an interesting point, when the design reaches completion it may be worth seeing how many units we can move. Since development costs will be sunk and non-existant, the final price would reflect actual production cost... at most with a very small markup with profits going directly to a fund for further development (prototyping does cost money after all).

One thing that is absolutely essential to me is that all of the information needed to build a fully-functioning unit be freely available.

I'll get in touch with my university, they have rapid prototyping, cnc, laser sintering, etc. as well as injection molding equipment. Perhaps I can pull some strings to help lower production costs further (if we choose to take that step).

Again, thanks to all for your ideas and contributions!


Warren makes an interesting point:

"I'd pay even more if you can get NCEES Approval for use on the PE Exam :) "

Should our new calculator be modularized so that you can remove whatever it takes to make it NCEES approved? Or, will there be a low-end model without all the bells & whistles (a la the '32S or 33S)? This might be a significant market share.


As I don't live in the USA, what is the 'NCEES' after all? How frequently do you need to execute its 'tests'? For what purpose?

I think that this project will be totally modular, but how do the guys at NCEES check what (software) modules is installed on the calculator? We can build a version with segment LCD display, but it is also capable of showing alpha characters.

OR they provide a SD/MMC card with the modules allowed for use on the test...

Best regards,



PLEASE forget about the NCEES business. Let us design a machine that is unencumbered by *artificial* limitations.

Also, I doubt that a beauraucracy such as the NCEES would ever "approve" a non-mass produced electronics item!

Also, count me in!



Edited: 16 Apr 2004, 11:02 a.m.


Agreed, even if we were to *try* for approval (this is purely hypothetical) we would never get it. Think about it... fully flashable rom, open source code, user configured keys?!? One unit will invariably differ from the next.

I will say this, however. If someone is able to get it approved, I'm all for it. More power to you.


I can't offer much from a technical standpoint, however, I think that this is a very worthwhile project. I like the open discussion of ideas. I currently am impatiently awaiting the arival of my HP33s having passed on my HP32sII to a fellow co-worker. I would also agree to support this endeavor and would like to be kept up-to-date on it's progress. I would like to be notified if a seperate forum is utilized. Thanks in advance for your hard work.

Possibly Related Threads…
Thread Author Replies Views Last Post
  FRAM71 Project aj04062 1 1,222 11-25-2013, 01:59 PM
Last Post: Hans Brueggemann
  [HP-PRIME] QPI project CompSystems 0 891 10-09-2013, 02:51 PM
Last Post: CompSystems
  OpenRPN Matt Agajanian 3 1,514 09-09-2013, 12:42 AM
Last Post: Paul Dale
  RPN-1200 Project Benoit Maag 0 843 08-25-2013, 01:05 AM
Last Post: Benoit Maag
  What do you think of this project??? Pierre 1 1,174 11-21-2012, 06:58 PM
Last Post: Eddie W. Shore
  TI 2550III project, help me I can help you. watchdoctor 2 1,427 10-19-2012, 10:13 PM
Last Post: Joerg Woerner
  neat project for the HP48 Han 5 1,923 09-25-2012, 04:58 PM
Last Post: Han
  What do you think of this project??? Namir 10 2,785 09-16-2012, 03:56 PM
Last Post: Eddie W. Shore
  WP-34s: Project donation page W. Bruce Maguire II 5 1,851 04-26-2012, 07:01 PM
Last Post: Geoff Quickfall
  USB-41 Module, project status update. Diego Diaz 8 2,636 04-06-2012, 10:17 AM
Last Post: Diego Diaz

Forum Jump: