RPN Scientific Firmware for the Spikenzielabs Calculator Kit

[Maximilian Hohmann]

Hello!

(09-28-2017 06:46 AM)Alejandro Paz(Germany) Wrote:  For like 14 € looks really good !

Just looked it up, the price actually is 10,59€ shipping included. One would have difficulties to buy the parts for that price. And I wouldn't know how to manufacture the laser-cut acrylic case.

This is the assembled (and working) calculator - next to a real calculator for size comparison. I didn't apply the labels and keycaps yet of course...

[pier4r]

Nice project!

Little OT.

(02-24-2016 08:48 PM)securd Wrote:  Hi All,

I have always had a fascination with early scientific calculators, and I'm impressed by how the engineers in the 1970's did so much with so little. ...

to an Arduino port of the BigNumber arbitrary precision library, coded up high precision Taylor series approximations for the transcendental functions, and voila!

I agree that the engineers in 1970 did a lot. I also see that nowadays we do a a lot through libraries done by others.

Surely making a wrapper of a given library is not trivial, but doing everything by oneself would be a ultra challenge (and likely one won't have a final library as polished as the existing ones. At least not so quickly).

[Maximilian Hohmann]

Hello!

Old thread ... new developments :-)

The Chinese kit manufacturers came up with yet another calculator kit which might be the basis for DIY projects. This one comes with a 2x16 characters inverse backlit dot-matrix LCD screen and 20 keys, mounted in a very smart laser-cut acrylic housing with slanted display. Power comes from 2 2032 Lithium cells (not included in my kit). Why this one needs two cells as opposed to the LED calculator above which only needs one I don't know.

I assembled mine in less than two hours, there is not much soldering to do but cutting out the key labels and fiddling them into the keys takes some time.

Apart from being a basic calculator (no operator precedence or other smart features installed) it comes with three pre-programmed applications which are selected with the "mode" key: Calculate resistor values form their color code, calculate resistors required to operate LEDs and a simple HEX<>DEC converter.

Unfortunately this time they did not use an Atmega 328 processor (esily programmed in the Arduino environment with lots of resources awailable) but a chip called "15W413AS" about which I could not find a lot of useful information yet.

These are the bulding instructions: http://www.diyleyuan.com/jc/L8Q.html and below are some pictures of my kit and the finished calculaor. Maybe someone finds this interesting... Ah yes, I paid 11,95 Euros shipping included. Buying the parts individually would have cost more I guess!

Regards
Max

[Luigi Vampa]

No Atmega, a bit of a shame. Nevertheless, thanks for sharing!

[John Keith]

Judging by the key labels, it looks like its original purpose had to do with resistor color codes.

[revwillie]

Nice. just bought two kits for $25 USD. Despite the lack of RPN, I could use a resistor color decoder and I love a cheap soldering project.

[jklsadf]

For anyone interested, that cheap diy calculator kit Maximilian posted about can be reprogrammed to use RPN, even though it doesn't use an atmel microcontroller. See https://github.com/jjj11x/stc_rpncalc

picture:


Qt GUI emulator showing functions:

[jpcuzzourt]

(10-07-2019 10:15 PM)jklsadf Wrote:  For anyone interested, that cheap diy calculator kit Maximilian posted about can be reprogrammed to use RPN, even though it doesn't use an atmel microcontroller. See https://github.com/jjj11x/stc_rpncalc

This is good news. So the micro is 8051 compatible, and has a serial flash bootloader.
These kits are available for about 11 bucks on aliExpress now.

If only it had a nice wide 'ENTER' key...

I may get one to play with.
EDIT: I've ordered one and read through the project at github. Because the included micro is fairly limited, I would want to consider a redesign to use an AVR instead.
But initially I'll just wait for it to arrive in a couple of weeks, build it, and try the RPN code that Jeff has developed already.
With the backlit LCD, this thing would be a battery hog. I'm interested in seeing how the keys feel. Jeff references some Omron replacements that are easier to operate.

[jklsadf]

The tactile switches it comes with are decent, with positive tactile feedback, they just require a lot of force to press. The atmega328 in the DIP-28 package (used to be used in the arduino) has the same number of pins, but a different pinout including power pins, so it wouldn't be a drop in replacement. There are other people who have done fully custom calculators with an AVR microcontroller, including the original kit calculator this thread was started for, ARC, and also SCARY.

SCARY fits an impressive amount of functionality into only 8KB of flash. I've considered using similar trig algorithms (which valentin albillo also used for an HP 12C implementation earlier, not sure if he actually came with the algorithm or not, but I don't know of any earlier examples) because of flash size, although I think the ones used in the HP 35 are potentially more accurate.

There are also people who have done much more advanced fully custom calculators with ARM microcontrollers. The diyleyuan kit is definitely the easiest to get though, and widely available.

[jklsadf]

With regards to the battery life, the original firmware turned off the calculator automatically to preserve battery life (I think within less than a minute after the last key press). As you noted, the backlit display draws a lot of power. The soft-latching power switch is quite effective in the off state, and draws almost no power when off. I measured it once, but don't remember the exact number off the top of my head.

I personally have the calculator plugged in to and powered from the USB-to-serial converter all the time, so it hasn't been an issue for me to have it on all the time. (The form-factor with the tilted screen really is ideal for desktop use, and makes the screen a lot easier to see than most "pocket" calculators with a flat screen.) I haven't really looked into trying to put the microcontroller into low power modes, since the backlight is probably one of the biggest power draws. I do turn off the backlight while doing calculations, but that's more of an indicator that a button was pressed and calculation is in progress. An auto power off would be a nice optional feature to have, although I personally like having it stay on all the time.

[tarantula]

(03-01-2016 10:19 AM)Maximilian Hohmann Wrote:  Regarding your projects I have a couple of questions and a suggestion:

- Does the housing of the Spikenzie calculator offer enough space for mounting the microcontroller in a socket? I ask because I have already "fried" a couple of them, mostly because loose wires on the breadboard got in contact with the wrong pin. And in-site-programming always comes with a certain risk of connecting the wires in the wrong order... De-soldering the controller from that calculator PCB would be beyond my capabilities.

- On one of the pictures I can see a MacBook pro, similar to what I am using. What ISP programmer and which software are you using to flash the controller in-site? And what settings do you use for bringing down the power consumption that low?

I got for myself a SpikenzieLabss Calculator Kit because I was inspired by the RPN capability (thanks to securd for sharing his code).
I bought it for $18 from it's nominal $45 price.
I didn't see it as a soldering project, and I didn't want to solder the ATmega328P on the PCB either. Instead I left it un-soldered. The chip comes with its pins flared out slightly (I skipped the SpikenzieLabs instruction step 11). I used a ruler to spring-press the pins while I was placing the chip pins into the pcb holes. At first, segment "a" was not lighting up. I bent outward the chip pin #15 and reinstalled the chip (un-soldered). It worked as advertised by SpikenzieLabs. I didn't want to solder the programming header either. I used a male 6-pin header on my FTDI programming cable and placed it in the PCB programming holes at an angle to make a contact. I programed the RPN firmware, tested it our (only 0.2uA when it is OFF) and assembled the plastics. I placed 0.5" wide Kapton tape on the LED seven segments to protect the black coating from being scratched. I was not pumped with the double sided tape on the keys. If I have an issue with the keys coming apart down the road, I will use cyanoacrylate glue.
This calculator is better than this:
http://files.righto.com/calculator/sincl...lator.html
I own that Sinclair piece of "rubbish" with a VFD display, which is a good reminder of the "Worse is Better" design principle.
https://www.jwz.org/doc/worse-is-better.html

[toml_12953]

(09-30-2018 10:47 PM)revwillie Wrote:  Nice. just bought two kits for $25 USD. Despite the lack of RPN, I could use a resistor color decoder and I love a cheap soldering project.

There is RPN firmware available for it.

https://github.com/jjj11x/stc_rpncalc