ti 84 lcd screen replacement made in china

This website is using a security service to protect itself from online attacks. The action you just performed triggered the security solution. There are several actions that could trigger this block including submitting a certain word or phrase, a SQL command or malformed data.

I have a TI-84 Plus CE-T With fancy broaken LCD rainbow colors on the screen. I took it apart and after inspecting the LCD, I couldn’t find any part number markings that gave me any results on searx, aliexpress or ebay. I figured that this screen is probably using a somewhat standard protocol and pinout. Therefore I was wondering if I might be able to replace it with a simularely specced lcd from aliexpress with the same amount of pins on the connector. I ended up finding this part

MathPrint� mode for free download. The OS 2.53MP enables users to input and view math symbols and formulas, including stacked fractions, in their handhelds exactly as the equations appear in textbooks.

MathPrint� mode for free download. The OS 2.53MP enables users to input and view math symbols and formulas, including stacked fractions, in their handhelds exactly as the equations appear in textbooks.

Select a forumAnnouncements----------------Cemetech News and AnnouncementsCemetech LabsContestsGeneral Discussion----------------Technology & Calculator Open Topic-- Electronics/Hardware Development-- Calculator Hardware, Electronics, Robotics-- Calculator HelpYour Projects-- Do It YourselfGeneral Open Topic-- Politics & Rants-- Reviews-- Humour/JokesProgramming----------------Computer & Mobile-- [Comp] C and C++-- [Comp] Python, Java, and C#Calculator Programming-- TI-BASIC-- z80 & ez80 Assembly-- Casio Development & Programming-- HP Calculator News, Projects, and Support-- [TI] C and C++-- Axe & ICEWebsite Markup & Scripting1337 Programming TipsCemetech"s Projects----------------Doors CS and Doors CSESourceCoderMember Features-- TI-Freak8x-- Merthsoft-- MateoConLechuga-- JamesV-- PT_Other & Upcoming Projects-- The BASIC Elite-- mobileTunes-- CALCnet 2.2-- SuggestionsProgramming the TI-83 Plus/TI-84 PlusUsing the TI-83 Plus/TI-84 PlusWebsite-- Website Suggestions-- Website Bug ReportsCreativity----------------GraphicsMusicArtOther HobbiesPlay @ Cemetech----------------Minecraft-- Cemetech-MC 1.16-- Cemetech-MC 1.12-- Cemetech-MC 1.8FreeBuild & LEGO-- FreeBuild General-- Suggestions & Troubleshooting-- Content & Gallery-- Servers & ActivitiesUnreal TournamentOther Games

I am current considering use of this display but have not ordered it yet. I am a little hesitant due to the number of users reporting problems with these displays.

I am in the process of creating a PCB footprint for this display, but when applying the 3D model provided, the hole spacing of the display"s connector pads do not appear to be on a 0.100" grid. Is this an error in the model or are the display"s connector pins on a weird spacing? If so, what is the spacing? Any help would be much appreciated.

I would say use your best judgement. I have only looked at the reviews under the review page and they all seem fine; unfortunately, I don"t have time to comb through all +200 comments. That being said, you can check out our Technical Assistance page linked in the banner above and our returns policy. Another resource for you is the [forum]https://forum.sparkfun.com/], where you can check for what issues users had and if there was a resolution to the issue.

As shown in the hookup guide the through hole header connections should be about a .1" spacing (image of connection into a normal breadboard). I"ve never used Blender and there person who made the models is no longer around to ask... so I can"t speak to the accuracy of that file.

I had a project that needed some live data display, and looking for the cheapest low-power solution for our loggers lead me to the Nokia 5110 LCD. Once you get the backlight current under control, you can power the entire display from a digital pin, and if you use shiftout for soft SPI you can then get rid of the Reset and CS control lines. This brings the display down to any four wires you can spare on your build (incl. the power pin) and a ground line. This is much more manageable than what you see with the standard hookup guides if your mc is I/O limited like our pro-mini based loggers:

This LCD (I have the old-old kind) is absolutely my favorite. Yes, it has a board-to-glass connector that ranges from bad to abysmal, but it offers such a simple interface and so many pixels for so little money (obviously less if you buy only the panel.) Here are some clever things I"ve discovered:

Will fully operate on as little as 2.0V. That"s power (Vdd) and i/o. It can be driven at 2MHz at these speeds; in fact, the LCD will work at even lower voltages but the contrast fades quickly and your microcontroller will likely approach its lower voltage limit too.

The initialization sequence is magic. Here it is, in case you were, like me, frustrated over and over again with varying sequences that others claimed to work but they didn"t for you:

Reset/initialization can be picky - the datasheet says that the delay from power-up to reset mustn"t exceed 30ms. What I found to be the best is to set SCE low and reset high at system bootup, wait 5ms for voltages to stabilize, lower reset, delay by ~1uS (1 nop @ 8MHz will do), raise reset then send it the initialization sequence above.

The LCD will work with the chip-select pin (SCE) tied to ground. This means that if it"s the only device on the SPI bus, don"t bother framing the i/o with a chip-select pin. If the bus is shared, frame the entire transaction, not every individual byte you send to the LCD. Interestingly, the display also seems to work fine with a floating Vdd pin - it must draw sufficient power just from i/o via clamping diodes; not surprising when you consider how low-power it is.

The Vout pin: Looks like you don"t have to worry about it on this product, but the bare LCD will generate positive 6-9V on that pin. This wasn"t totally clear to me from reading the datasheet.

Magic indeed! I"m using a different approach for the RST (pull low as soon as possible, hold low for 100ms, then set high and continue) just as a matter of making sure I get a clean boot, but otherwise I end up using your init sequence verbatim. Saved me a ton of work since I"d been fighting with it for two hours.

(2) The second command byte (the 0xE0 shown above) is not arbitrary. It is 0x80, or"ed together with a 7-bit Vop value. I found my display to be somewhat sensitive to this value. At Vop=0xBF, my unit was initializing electronically, but had a blank display (or solid-black, I can"t remember now.) Anyway, I had to play with this value, and 0xB3 ended up working for me, so if you are initializing to this sequence and having trouble, try varying this parameter. The technical details of this parameter are explained in the datasheet, section 8.9, but really, you will just have to play with it.

Note the datasheet says reset can be low when VDD is applied, and then has to be at least a 100ns "low" pulse. So powering up VDD, then applying !SCE at 1 us and !Reset at 2 us meets that spec just fine. I have had perfectly consistent startup operation with this sequence.

(4) !SCE tied to ground - yes, you could, but I"ve had fine results framing each byte with the !SCE signal - there"s nothing wrong with that. Yes, framing a whole transaction is fine, too. Note the datasheet timing diagrams show that !SCE serves a reset function on the incoming serial shift register, so if you ever get a glitch in your serial transmission that gets the bit count off, respecting the !SCE timing (whether on byte- or transaction-basis) will automatically get the display controller"s serial receiver synchronized back up on the next byte or transaction.

(5) If you are using a PIC to run ths thing, and using the PIC"s USART or EUSART in a synchronous mode, be sure to note that the LCD controller expects the MSBit of each byte to be transmitted first on the serial line. The PIC 18F EUSART transmits the LSBit first. For now, I have lots of extra code space, so I"ve wasted a 256-byte section on a lookup table that reverses the bits in a byte. This way, I just write my initialization code normally, and I have a TransmitCommandByte() function that looks up every byte it sends so I don"t have to think about that.

As a separate issue from the command bytes, all my graphic data are already created with the proper bit orientation, so I don"t have to look them up to do the reversal.

It may be possible to compensate for the low voltage by tweaking the contrast or updating slowly, but in my testing about three years ago journeys beneath 2.0V led to the display whiting slowly out like the face of an inexperienced fighter pilot in an aerobatic manoeuvre.

Thank you! I"m not quite sure I do want an LCD yet, to be honest, I"m just considering the different options available. I"ll check out the Sharp component, thanks!

Advice for others: It took me quite a while to get this working on an ARM Cortex. Since there is no way to read from the LCD, it is very hard to know if SPI is working without doing everything perfectly. SO:

Use the contrast value that Kuy gives (0xE0) or darker - this will cause the display to be all-black so you can confirm it works. THEN you can adjust the value and focus on getting characters to write.

The correct fix is to delete the "\" (backslash) character in the comment for the 0x5C entry. The backslash is the "continue on next line" symbol, so the next line is not seen as part of the initializer, thus making everything following it offset by 1. Just change the "\" character to the word "backslash" and everything will be good:,{0x00, 0x7f, 0x41, 0x41, 0x00} // 5b [

The problem I had was solid black display screen. No matter the combination of bias and contrast values that I set. The unit wasn"t totally defective, because under a strong lamp light you could see the display trying to show the letters and pictures that are in the tutorial for Arduino that I got from SFE.

Finally, after a long time of trials, I fixed it--> The problem: the bezel wasn"t tight to the board, so the contacts of the display didn"t work well.

Conclusions: You get what you pay, don"t expect well manufactured products and tested components comming from China at half the price you pay here. And second, it is totally no good idea to start up a new development with new technologies with untrusty material. Maybe you can try some cheap components after you are sure that you are used to certain technology (a display, a sensor, whatever), but don"t try to break the barrier of new knowledge and tricky materials at the same time; because you won"t be able to know who is the faulty.

The problem with these displays, and it seems to be common to all the manufacturers, is that the PCB material is too thin. After a month or two, the board slowly bows away from the glass display panel under pressure from the conductive rubber connector strip. Pressing on the top centre of the metal part of the display makes it work again, but only temporarily.

If the LCD module is soldered to another board and the two top screws installed and tightened carefully to pull the bow out of the module it seems to prevent (or solve) the problem.

The only real fix for this display is to redesign the metal cover with a hold-down tag in the top centre, or make it with a thicker PCB that doesn"t bow under pressure from the connector. Until that happens I won"t be buying any more.

The Arduino code works fine for me, but I can"t make any thing work with Pic. I found this code at http://www.sunbizhosting.com/~spiral/ , but can"t figure out what I"m doing wrong.

I"m using voltage dividers to supply 3V in the inputs of the LCD, because of the Arduino works in 5V. LCD Vcc and LED are powered from the 3.3V output of the Arduino. The LCD only displayed something when I used: R1=470K,R2=820K. I have tried several values to obtain 3V, but the LCD showed nothing. I don"t understand that.

I"m interfacing this LCD with ATMEGA 32. Its been more than a week that I"ve been trying to get it right. All I get is the LED dimming effect. Here is my initialization code..CE=1;

Has anyone played around with minimum and maximum SPI clock rates with this display? I know the datasheet says up to 4 Mbit/s, but assuming this display is the only thing on the particular SPI bus, what"s a good minimum clock rate before you might start to notice delay in the display refresh? I mean TVs are 60 Hz - 120Hz.

I have a similar board made by mib-instruments and bought from ebay years ago. It has been my standard spi test tool because it"s so easy to work with. http://www.ebay.com/itm/Nokia-5110-LCD-84x84-dot-martix-backlight-PCB-RED-/320684678723 (specs http://i1119.photobucket.com/albums/k636/mib_instruments/diy/LCDC2A0SPEC.jpg)

I almost have it working satisfactorily but I find that the bottom 1/5th of the screen does not function correctly. Sometimes it has some random blocks that are black, most of the time it is blank. I am not sure what would cause this. Is it safe to assume it is a defect on this module?

These LCD"s need cleaning. I have an average failure rate of about 15-20% on delivery. The most common problem is that the contrast is too high, and there"s constant flickering / changing of contrast compared to the other 80% of them.

The solution is fairly simple, unclip the LCD from it"s board and clean the pads on the PCB with 99% IPA. Then remove the lcd back plate and contact bar. Sometimes the contact bar is stuck fairly well to the glass, peel off carefully. Clean the contacts on the LCD glass with IPA, if any residue from the contacts is left on, rub it off carefully with IPA / tissue.

I love this little display! I wanted to be able to create images for it but nothing I saw did exactly what I wanted. So I wrote a processing sketch that creates 84x48 squares on the screen and allows you to click to turn them on or off. Also has buttons to invert, move up/down/left/right, and flip horizontally/vertically. Then, it saves the hex data to a text file to copy to your code. You can also load an image (any size, any colors) and it will scale it, convert to b/w, then put it in the rest of the program so that you can alter the pixels or move it. It isn"t perfect for every occasion but I"ve found it useful and I hope others might too. It is heavily commented so it should be easy to figure things out and change them if you want something different. http://thewanderingengineer.com/2014/07/12/nokia-5110-screen-photo-to-bitmap-converter/

I"ve got 3 of these and wired them up as suggested in the hookup guide and downloaded the example code, and the only thing that seems to work is brightening and dimming the LED. I"ve started again from scratch at least twice, tried all 3 units, and verified all the connections match the hookup guide, and I still the ssame result; no graphics, no text, just a border. Any ideas as to what might be wrong?

Never mind, I had no problem running it at 5V directly from the Arduino with the contrast value bumped up to 45 or above. With 10K ohm resistors on the control lines as shown in the sample hookup, I get a blank screen. Also removing the backslash from the sample code got garbage characters, so I left it in.

Anyone taken these things apart yet? You know the flexible rectangular blocky thing that connects the contact pad on the board to the LCD itself? What are these called?

Got mine running last night and found two problems with the code, one of which was the backslash a couple of others have already noted. Second was that the LCDCharacter() writes two blank vertical lines, one before the character and a second after, when only one is needed. Without the extra blank you get at least one additional character on each line. I"ll probably also move the ASCII font table to PROGMEM space to save on RAM and then start to work on some big digits for a clock.

I"m using this LCD for a large Arduino UNO project, but I"m running out of SRAM memory space. I was wondering if I used PROGMEM on the LCD ASCII array if that would help. If so, does anyone know what the right code for this would be? After looking through a lot of PROGMEM examples, I"m not advance enough to really grasp everything that"s going on. Any help you can give would be a great help. Thanks in advance!

I used one of these LCDs with an Arduino to display GPS information. I wrote a few functions that can display large numbers (28 px high) if anyone is interested, this lets me display speed, heading etc. A writeup of my project is here: http://mechinations.wordpress.com/2014/04/07/gps-sailing/

These are great displays. I ran into a problem using them with the nRF24L01+ radio transciever, which requires the use of the SPI bus. If one attaches both the radio and the display MOSI and SCK pins to pins 13 and 11 as instructed in the hookup guide, the SPI traffic of the other device (in this case the nRF24L01+ radio) will prevent the display from functioning. The easy solution is to move the Nokia 5110 MOSI and SCK pins to any other digital pin. This should be made clear in the hookup guide, where it says there is no choice but to use the hardware SPI pins for the display. I found out that is not true at all. I hope his helps others with the same problem. Despite the occasional bad display these carry much more information that the comparably prices 16 x 2 LCD and use fewer pins too boot. What a deal!

I was able to get this work only after using 5v, and lowering the contrast by int from 55 to 45 to remove the "black box" flickering behind the screen drawings. When I power with 3.3v, nothing - just the LED and dimming etc of the screen. I"d love to use 3.3v (and it seems based on responses here) that it should run only on 3.3v. Any ideas anyone?

It should run on 3.3V only, I found it very sensitive to the contrast value. 55 works for me, but 45 makes nothing show up. You might need to play with this value.

This is a great display for the money, certainly the best bang for the buck of you can live with B&W and lower res graphics. I have a lcd driver for Arduino I will post on http://www.marchdvd.com/5110 so take a look there it draws text aligned on pixels boundaries of 8 and draws lines and has invert video options.

I just started messing around with this LCD using a STM32F103 microcontroller running at 72MHz... it works great. The only problem I had, and I suspect others might have if they are using fast processors, is that you have to deliberately introduce the setup and hold time delays on the DC pin... if you don"t you will get spurious pixels written to the display. I used a delay of 10uS, although the spec says 100nS is fine.

I been trying to display 8-10 images as an animation. The individual images all work, but when I trey more than 8 images, nothing happens on the screen, and above 2 images, I get errors in the drawing of the images. Is there some kind of buffer or violent reset I can do so that the data doesnt get jumbled up?

I just spent the last couple hours struggling with this LCD because of something very stupid of me. I was using an atmega328p in AVR-GCC and using hardware SPI. Thinking i didn"t need MISO I hooked it to DC. The LCD worked absolutely fine until I tried to set the x and y position in the ram. It started acting weird every time I tried it. Finally I put dc to another pin and BAM NO PROBLEMS. Looking back I feel pretty stupid but hopefully this post will save someone else the same mistake. Other than that great LCD for my projects

I need to know where to drill the mounting holes in a board I"m designing, can someone get more precise measurements of the physical dimensions of this board?

I used the ASCII font given in the example code in one of my projects and ran across a mystifying bug. I was developing code on a PIC12F675 to drive an OLED through its SPI interface. (Yes there are enough pins and memory to do this!) I was using the HI_TECH C compiler and tried several different approaches and never could get the "]" character to display. After pulling what is left of my hair out for 2 days, I realized there was a "\" in the comment for the previous line of the ASCII font definition which caused the compiler to treat the array entry for ] as a comment! Just replace the "// 5c \" comment with "//5c backslash" and everything works! I don"t know if this effect is peculiar to the HI-TECH compiler only.

The Energia folks have an example program for this LCD and the TI Launchpad written using their Arduino style tooling. I"ve updated their example and added the ability to report back the temperature over a UART. It is a very simple hardware setup since both systems are 3.3v. http://joe.blog.freemansoft.com/2012/08/digital-thermometer-with-ti-lanchpad.html

1) There seems to be a serious issue related to contrast and/or connections between the display and the breakout board. When running "Black on White" text, the background gradually darkens to a black rectangle covering the etire display area but will revert to normal if the centre of the metal surround is squeezed towards the pcb, however, putting a sprig clip on it to keep it comoressed doesn"t seem to cure the problem, but just slows the raste at which it fails - waggle it and it comes good again.

2) I"m really struggling to find unformation on using this display with the Arduino. The example (pcdtest.pde) provided with the Adafruit libraries (Adafruit_PCD8544 and Adafruit_GFX) won"t even compile and the only library I have found that I can make any sense of using is the PCD8544 library from Google (http://code.google.com/p/pcd8544/downloads/detail?name=PCD8544-1.4.zip) and I can"t really uderstand how to do graphics with that.

I did find another example (did"t save it and can"t find it again) that worked with the Adafruit libraries as it was supplied (including graphics), but trying to change it in any way beyond changing the text of the "Hello World !" string (which actually shoed on screen as "Hffmmp Wpsme !", so obviously a coding problem there!!!) just locked everythig up.

I tried using the "LCDAssistant" package to create a logo from a graphic that I resized to a b&w jpg of 84x48 but every byte generated was 0x00 so that was not right. I tried fiddling with the settings (flying blind) but still got nowhere - does anybody know the settings for LCDAssistant and this display and has used it successfully?

Hi - for some reason (way above my "new to s/w" head) the code hates the \ character, even in comments. If you take out the backslash in the character definition section it works. My tweak is:

One of the things that I test regularly is a commercial item that features a 16x4 (HD44780) display. Currently I have a 20x4 on a flying lead that I plug in to determine if a display failure is down the lcd display or the main board.

Is there any way to get the 5110 Graphic Display to work with signals that were feeding to an HD44780? - if I could build that in then I would have a complete multi-testing set up in one box.

I"m very out of date hadrware-wise (I was out of electronics for 30 years) and I"m not a programmerby any stretch of the imagiation (even a relatively simple sketch takes hours eliminating the errors one by one).

I have the other test functions working ok by using the library and cobbling together code from projects I"ve found on the web (very inefficent coding when put together I"m sure, but it works) but I"d dearly love to add this extra functionality.

I am using arduino example and while i am getting proper images and text i also get some odd horizontal flickering. It looks like several horizantal lines across the screen on the background with image/text on foreground. I tried switching to only use digital pins on my arduino leonardo but i still see this behaviour. Any ideas?

Make sure your power is regulated correctly, and that all of your soldering points and connections are good. It may simply be a bad screen. If you still run into problems, shoot us an email at techsupport@sparkfun.com

Is there mechanical documentation for this (location of mounting holes, location of screen relative to board, etc)? Yes, I can measure it but I"m lazy...

Thats a really good catch. The text editor doesn"t catch it, so I wonder what the issue is. It must be some sort of continue to next line character.I would have spent all night trying to figure it out. Thanks

I see that this item is Out of Stock (back-order allowed). I have a embedded system I have been documenting / publishing the design for over the last month, so I would like to know if Sparkfun Plans on keeping this part and if so for how long? Is there a part you would recommend for replacement if this one is going away?

Might I suggest you (SFE) source some of the Electronic Assembly"s LCD Dog-S series. I think they would be a step up from these at a reduced price. I don"t think that they website is up to date, but their part number is LED39x41-GR.

I made a little font generator for the Nokia 5110 in the processing programming language (processing.org). It allows you to convert any font and any character that you can display on the screen into a list of hex codes that can be directly used in an embedded system (I"m using msp430). Just type a character and the corresponding hex codes will be in your clipboard and you can copy them into your program. It starts with an example with the chinese character for 5. It should work on any system that can run processing (e.g. mac osx).

I am guessing that you are seeing the PCB bent away from the back of the display, up near the "top" of the display region, when it is viewed in normal operation, right?

This is because this is where the connections are on the back of the "display glass". These connections are carried down into the PCB via a small strip of that elastomeric connector material. This material works by being compressed between the two substrates (the glass and the PCB) to hold it in place and make the connections between the pads on each surface. As a result of being compressed, it puts a force "up" on the glass, and "down" onto the PCB. In this particular design, this has the result of bowing the PCB in the units I have purchased.

My thought, if I were up to me, would be to redesign the metal hold-down bracket, to have one more metal tab going down through an additional slot in the PCB, right there at the center of the top edge of the hold-down bracket. This would carry the majority of the reaction to the force exerted by the elastomeric strip, and should minimize board bowing.

I finally got around to running this LCD on my 3310 PCB. It is working fine with one minor problem. The SF 3310 display hides to first line of bytes for some reason and I had to offset everything to compensate. The 5110 doesn"t do this as behaves as expected. I haven"t heard anyone else report this so maybe my initialization code is different.

Using a 3V source, my LCD often worked OK using bias 0x14 like the other examples, but sometimes it would appear gray and faded. The fading would lessen if I touched the panel lightly with my hand for a few seconds, then let go, so maybe it"s a temperature-dependent thing?

After writing a program to let me interactively adjust Vop, temperature coefficient, and bias, and experimenting with different values, I found that bias 0x15 worked much, much better than 0x14. I"m also using Vop 0xD0 and temperature coeff 0x04, but those don"t seem to matter as much.

Ack! After two days of working nicely with 0x15 bias, I reset the board today, and the LCD appeared way over-dark. I changed the bias back to 0x14 and it looks perfect. What the heck?! I think there must be some temperature-sensing or temperature-dependence going on, so the same init values may produce good-looking results one day but not the next.

If you are having problems with the black pixels in images/warping PCB, use original Nokia 5110, I happened to have one at home and it works as it should, no bad connections degrading image quality.

Does anyone know whether this can be stripped of its backing so it can be used in transmission? I would love to use this as a modulator for a laser beam. Or if someone knows a similarly cheap transmission LCD that would be fine too.

Has anyone created an eagle footprint for this? I"ll be doing so when the order arrives. Basically just the outline with mounting holes and the 8 pin header.

Stuck. Blank LCD. Added 0x20, changed Vop to 0xB3. Guessing connections may be the issue? 3.3v for LED and VCC. GND to GND. Remainder connected to Arduino via voltage dividers. What am I doing wrong?

This is a great little lcd. When I first wired it up, the backlight was shorted (accidentally) against my 5v rail, so i got some magic smoke, and burnt to LEDs but it re-soldered the offending joints and it works very well now. Something to note: the refresh and write times are much, much slower if you use 5 volt logic. I stuck in a logic level converter and it ran at least 5x faster.

I did the same thing (see my comment above), but I still can"t seem to get anything to display. 3 of the 4 backlight LEDs still light up when I give them power, but I"m wondering what joints you re-soldered to get it to work again.

You can also use FastLCD to convert your bitmaps - google it. It outputs BASIC code, but you just search and replace &h to 0x and you"re grand. It has the added advantage of being an editor for touching up output.

I found that same 0xB3 value works just right for the two units I have. I wonder if some of the difficulties people have getting them going is from using "E0" or "BF" or some of the other values I"ve seen posted. When I first powered mine up, I got NOTHING on the screen, and I would have thought it was dead, or assumed mine had "bad connections", if I hadn"t known to play with that Vop value...

I recently obtained a virtually identical LCD from a Nokia 5160, and although its backlight LEDs are green, not white and conversely use different voltages, I had success hooking up the LEDs" Vcc pin to a PWM capable pin on the microcontroller, allowing me to control backlight intensity (I didn"t need a current limiting resistor for this either, but adding one will help reduce current drain on the controller).

Seems like the PCD8544 library does it"s own SPI bit managing and it really doesn"t like me using the SD library (also talks SPI) at the same time. I"ve made sure I"ve got all the SPI pins matching for both libraries (MISO, MOSI, Clock are the same and each device has it"s own Select), but it looks like the SD.begin() call just breaks the SPI bus for the 5110 and it becomes non-responsive. The LCD works just fine if I don"t initialize the SD library and the SD card works fine if I do initialize the SD card.

I"m pretty sure I tracked down the problem- the PCD8544 library uses software SPI while the SD library uses hardware SPI and I"m pretty sure the Arduino can"t do both over the same SPI clock/miso/mosi pins. Anyone know if this LCD will work with hardware SPI?

Hi! I"m having the same problem that you explained a lot of time ago, maybe you"ve forgotten it but, what where the changes you made to the PCD8544 library?

I accidentally (well, intentionally, but stupidly) plugged the LED backlight line directly into 5V when I first got this, and I saw (and smelled) the magic black smoke escape. I believe the black smoke only came from one of the backlight LEDs burning out, because I only notice one of them not working now. The other backlight LEDs still light up okay, but I also now seem to be having trouble getting the display to show anything. I"m just wondering if I could have messed anything else up. It seems like others have had their issues with this display, so I was hoping I maybe had something misconfigured. I"m fairly certain I have the pins assigned properly, but maybe I"ll tinker with the contrast. Any recommendations for setting the contrast value?

I"ve had issues with the LCD not showing anything intermittently. You got to make sure that all the connections are secure, and for the reset pulse, be sure to have a delay that"s 30-50 milliseconds long.

As much as I love SFE products and will continue to order from them, this is one product I would not recommend. The connection between the LCD unit itself and the carrier board is via those rubber polymer connectors. All the planets must line up properly for them to work. In this case, the carrier board was warped preventing the connection from working. You will find other such remarks in the comments area.

Out of curiosity - what indication is it giving you that it is powering up, if there is nothing on the display? There are no outputs brought out from the controller, so what feedback are you seeing that leads you to believe it is powering up?

I"ve followed the linked-to Arduino example and I get nothing on the display. Should it just work without any other components? The link mentions a possible cap on VOUT but there"s no such pin. Googling has suggested my Duemilanove"s digital pins will be @5V but I need 3.3V?

I have mine working now! The secret was oz-solutions"s linked-to observation about the second code example which does more initialisation. I suspect that the default contrast etc makes the screen blank.

I bought one of these that has the problem where you have to squeeze the top to make it work. Is there any way to fix it? If I take off the aluminium bezel am I in for a hard time?

Don"t do this. Each divider will be burning 20x the entire amount of current that the display needs to function, and the whole assembly will waste 100x the LCD"s needed power and many, many times more than even the atmega needs to run at full speed. This will kill battery life.

I got this little screen for my Netduino but it just won"t work! I"ve tried everything, and even worked on it with someone who had the same screen but it won"t work. We got it all right, but it just won"t work. If anyone here gets it to work with the netduino please let me know.

Hi, I just bought this wonderful LCD but I"m having huge huge problems connecting it..could anyone please point me in the right direction? Since there are pins that aren"t metioned in the code, for example the 6 - DNK(MOSI)...

It looks like the board just has 4 white LEDs in parallel and no current limiting resistors. To drive them safely from 5V you should probably use something along the lines of at least a 22 Ohm resistor which should limit the current to 80mA (20mA per LED).

Does anyone know the diode rating and package size, also does anyone know where to get the rubber ferroius connector behind the LCD mine is defective. Has anyone come into issues with the breadboard the LCD is connected to, a few aren"t working for me.

Yes, we have noticed that the PCB was bowing and as a result the LCD now only works when we press down on the metal strip at the top. I hope that only a small number of these LCDs have this problem. We"re expecting a shipment to arrive today, I will be running more tests.

I got two of these defective boards. I tested one with every mechanical thing I could think of to fix it, even soldered a paper clip as a spring to hold pressure from the back side. Nothing worked permanently. I ended up cracking the glass trying just to get the thing to work! I cracked it more out of frustration.

My second one had the exact same problem. I took the glass off and have applied superglue to the back of the glass and am clamping it (more gently this time) overnight to try to just get my $20 worth (at this point) out of the last one I have.

Edit: After leaving glue to dry overnight, LCD simply does not turn on anymore. All the connections are good, but absolutely nothing shows on the LCD now at all. Only the LEDs come on.

Did you get either of the LCDs to display anything, at any time? Is it possible that the connections were OK, but you were not initializing or driving them correctly? Or did they start to work at one point, and then fail at some later point?

When I originally tried to get mine working, I was seeing NOTHING on the display. Then I had to get my initialization sequence correct, and adjusted my Vop value (ultimately using a byte of 0xB3) before the screen would display anything visible at all.

Note that the backlight LED"s are soldered onto the breakout board, and have nothing to do with the circuitry of the controller and LCD. So just because the backlights are shining doesn"t tell you anything about the operability of the LCD itself.

I have two display boards that both work, but I do see the bowing along the "top" edge of the metal bracket. I haven"t taken one apart yet, but I assume this is the edge nearest the elastomeric strip, which is creating this bowing force.

It depends on the code that you are using to control the LCD. If you are using the Arduino example above, the pins are defined in the beginning of the code.

FWIW I have connected this LCD with a 5V power supply to a 5V Arduino board with no level conversion and it worked. Presumably this may reduce the lifetime of the LCD.

According to the Datasheet this is operating within the "Limiting Values" listed on page 17 of the driver IC. "Stresses above those listed under limiting values may cause permanent damage to the device."

I am attempting to use this with a Duemilanove (ATmega328). Up til now, I have been powering it with the 3.3V line, including the LED. The datasheet for the LDC claims: "VDDmax = 5 V if LCD supply voltage is internally generated (voltage generator enabled)." The logic levels should be kept from 2.7V to 3.3V. Since the Duemilanove uses 5V logic levels, I am using a simple voltage divider on the communication line with no issues.

It would be nice to have a real spec on the LEDs, though. After looking at BlackJester"s graph I went ahead and wired the LEDs to 3.3 with no current limiting resistor and all seems fine -- but I don"t really know if I"m limiting the life of the parts....

It says under the absolute maximum ratings section of that datasheet that VddMAX=5V (internally generated voltage) and VinMAX= Vdd+0.5. So I would think that 5V logic is ok as long as the power supply is at 5V also.

The Absolute Maximums section of a datasheet (also known as Limiting Values as in the Philips datasheet here) is only used to specify conditions that might cause permanent damage to a device, and should not be used to infer normal operating parameters.

Normal operating parameters are found in the DC Characteristics and AC Characteristics tables. Note that in DC Characteristics, the VDD2 is listed as min 2.7V and max 3.3V. This is the manufacturer"s specification for how to run the device to its specified performance.

If you run outside this range, performance is not specified, and you"re on your own. (Maybe it won"t work at all, will work inconsistently, or maybe it will work but other specs will be off - it may draw more current, timing may be off, etc...)

The maximum logic value of 3.3 volts made me cautious of driving the LCDs at the native 5 volts of my Teensy AVR. That said, running purely off 5 volts seems to do no harm to the LCD.

For those interested, I have taken a few measurements of the current draw of the LED backlight of my LCD. As I said earlier, powering the LED with 5V external has caused permanent damage to one, perhaps two of the four LEDs. So, use the following graph at your own risk.

I made some changes/additions to the example code linked above. The main addition is a function that makes it easy to turn individual pixels on/off.

Is there any more documentation available for the additions to the LCD? For example, the datasheet has no information (that I could find, at least) on the LED. Everything seems fine on 3.3V, but what is the current limit on the LED? (note: if it wasn"t for work, I would just mess around with it myself.)

If you want to wire up several up these to a single microcontroller, you might take advantage of my freshly GPL"d C++ driver library for PCD8544 devices. It"s templated, so you can avoid duplicating code all over the place. Here"s a picture of two PCD8544 screens running off of an ATmega328. (The screens are operating independently, even though they happen to be showing the same logo graphic in that picture.)

Here is a PicBasic Pro example for the 3310, which should be compatible with the 5110. http://www.picbasic.co.uk/forum/content.php?r=174-Using-Nokia-3310-LCD

Fantastic! It appears from your example link that this uses the same controller as the Nokia 3310 that I"ve already used in past projects. The only thing that made it so cumbersome was trying to connect to its fine pitch press on type connector. This gives me a great low cost display option that is easy to connect to.

If anyone doesn"t have experience with this LCD, take a peak at the Arduino example link above to see just how easy it is to use. If you use plain C on your AVRs, I have sample code on http://tinkerish.com.

Afghanistan - AFGAlbania - ALBAlgeria - DZAAmerican Samoa - ASMAndorra - ANDAngola - AGOAnguilla - AIAAntigua and Barbuda - ATGArgentina - ARGArmenia - ARMAruba - ABWAustralia - AUSAustria - AUTAzerbaijan Republic - AZEBahamas - BHSBahrain - BHRBangladesh - BGDBarbados - BRBBelarus - BLRBelgium - BELBelize - BLZBenin - BENBermuda - BMUBhutan - BTNBolivia - BOLBosnia and Herzegovina - BIHBotswana - BWABrazil - BRABritish Virgin Islands - VGBBrunei Darussalam - BRNBulgaria - BGRBurkina Faso - BFABurma - MMRBurundi - BDICambodia - KHMCameroon - CMRCanada - CANCape Verde Islands - CPVCayman Islands - CYMCentral African Republic - CAFChad - TCDChile - CHLChina - CHNColombia - COLComoros - COMCongo, Democratic Republic of the - CODCongo, Republic of the - COGCook Islands - COKCosta Rica - CRICote d Ivoire (Ivory Coast) - CIVCroatia, Republic of - HRVCyprus - CYPCzech Republic - CZEDenmark - DNKDjibouti - DJIDominica - DMADominican Republic - DOMEcuador - ECUEgypt - EGYEl Salvador - SLVEquatorial Guinea - GNQEritrea - ERIEstonia - ESTEthiopia - ETHFalkland Islands (Islas Malvinas) - FLKFiji - FJIFinland - FINFrance - FRAFrench Guiana - GUFFrench Polynesia - PYFGabon Republic - GABGambia - GMBGeorgia - GEOGermany - DEUGhana - GHAGibraltar - GIBGreece - GRCGreenland - GRLGrenada - GRDGuadeloupe - GLPGuam - GUMGuatemala - GTMGuernsey - GGGuinea - GINGuinea-Bissau - GNBGuyana - GUYHaiti - HTIHonduras - HNDHong Kong - HKGHungary - HUNIceland - ISLIndia - INDIndonesia - IDNIreland - IRLIsrael - ISRItaly - ITAJamaica - JAMJapan - JPNJersey - JEJordan - JORKazakhstan - KAZKenya - KENKiribati - KIRKorea, South - KORKuwait - KWTKyrgyzstan - KGZLaos - LAOLatvia - LVALebanon - LBNLiechtenstein - LIELithuania - LTULuxembourg - LUXMacau - MACMacedonia - MKDMadagascar - MDGMalawi - MWIMalaysia - MYSMaldives - MDVMali - MLIMalta - MLTMarshall Islands - MHLMartinique - MTQMauritania - MRTMauritius - MUSMayotte - MYTMexico - MEXMicronesia - FSMMoldova - MDAMonaco - MCOMongolia - MNGMontenegro - MNEMontserrat - MSRMorocco - MARMozambique - MOZNamibia - NAMNauru - NRUNepal - NPLNetherlands - NLDNetherlands Antilles - ANTNew Caledonia - NCLNew Zealand - NZLNicaragua - NICNiger - NERNigeria - NGANiue - NIUNorway - NOROman - OMNPakistan - PAKPalau - PLWPanama - PANPapua New Guinea - PNGParaguay - PRYPeru - PERPhilippines - PHLPoland - POLPortugal - PRTPuerto Rico - PRIQatar - QATReunion - REURomania - ROURussian Federation - RUSRwanda - RWASaint Helena - SHNSaint Kitts-Nevis - KNASaint Lucia - LCASaint Pierre and Miquelon - SPMSaint Vincent and the Grenadines - VCTSan Marino - SMRSaudi Arabia - SAUSenegal - SENSerbia - SRBSeychelles - SYCSierra Leone - SLESingapore - SGPSlovakia - SVKSlovenia - SVNSolomon Islands - SLBSomalia - SOMSouth Africa - ZAFSpain - ESPSri Lanka - LKASuriname - SURSwaziland - SWZSweden - SWESwitzerland - CHETaiwan - TWNTajikistan - TJKTanzania - TZAThailand - THATogo - TGOTonga - TONTrinidad and Tobago - TTOTunisia - TUNTurkey - TURTurkmenistan - TKMTurks and Caicos Islands - TCATuvalu - TUVUganda - UGAUkraine - UKRUnited Arab Emirates - AREUnited Kingdom - GBRUnited States - USAUruguay - URUUzbekistan - UZBVanuatu - VUTVatican City State - VATVenezuela - VENVietnam - VNMVirgin Islands (U.S) - VIRWallis and Futuna - WLFWestern Sahara - ESHWestern Samoa - WSMYemen - YEMZambia - ZMBZimbabwe - ZWE

My Sharp EL-811 was a generous gift in June 2020 from a gentleman in Wisconsin who purchased the calculator in 1972 for around $400 while he was at the University of Wisconsin, Milwaukee. $400 in 1972 dollars has the buying power of about $2,700 in 2021! Five years later my Sharp EL-5805 scientific calculator below with LCD display cost about $40 or about $180 in 2021 dollars. This demonstrates how rapidly calculator technology increased, and calculator costs decreased, in the 1970s. All of this paved the way for personal computers and eventually smart phones. My Sharp EL-811 is in good cosmetic and working condition. It comes with the original case with a piece of paper in it with handwritten calculus formulas. The calculator originally came with a rechargeable battery pack Model EL-84 with six 1.2 volt NiCad cells. These were corroded and no longer functioned after nearly 50 years. I instead wired in a battery pack with five 1.5 volt single use batteries. It works fine with this battery pack. Looking at an internal view you can see the two Rockwell integrated circuits, 10572PA-7152 and 10573PA-7150. You can also see a variety of other individual components. This photograph compares that with the single printed board of a modern Casio fx-300ES Plus scientific calculator which I recommended to my middle school math students. The modern Casio has a retail price of about $13. Without the cover, the Casio weighs 89 grams compared to 529 grams for the Sharp EL-811. The label at the top of the calculator gives a power range for the Sharp EL-811 of 1.3 to 1.9 Watts. That compares with the 1973 Sharp EL-805 below with an LED display of .02 Watts.

Datamath (go to "calculators related to TI") states the price was $110, nearly $500 in 2005 dollars! Sharp LCD Calculators and Vintage Calculators have excellent information about the history of Sharp LCD calculators. Wikipedia has information about LCD technology. Purchased on eBay on 11-27-05 for $26 with $6.50 shipping. Excellent working condition. (The display is fine. The photo is just fuzzy.) Excellent cosmetic condition except for light scratches on the back and a name and social security number lightly scratched onto the back. Soft case and instruction manual. Flip up cover over the display like the EL-808 below. Much smaller that the EL-808, with dimensions of approximately 8cm x 12cm x 2cm.

Casio Memory 8R, (1975 Japan) horizontal design, purchased on eBay on 5-22-05 for $2 ($3.50 shipping). Keys and logic in good working condition. Power works sometimes with the AC adapter but the wire at the adapter is loose. The adapter can"t be opened. The calculator takes two AA batteries. The connection with the + end of the battery is a problem. It works intermittently by placing aluminum foil at the contact and the two silver areas above it. See lower left of board in interior view 1 revealed when you remove back plastic cover. Folding that board up reveals interior view 2, showing the main circuit board connected by 12 wires to the board with the keys. Interior view 3 shows the main circuit board. Find comprehensive information at Vintage Technology. I purchased a second one on eBay in very good cosmetic and working condition on 3-2-06 for $2.50 plus $5 shipping.

Casio Personal I, (1976 Japan) 8 digit flourcent display. In good working condition with either included AC adapter or two AA batteries. Purchased at La Mesa garage sale on October 8, 2005 for $2. More at Vintage Technology.

Hewlett Packard 35, (1972-1975) this historic model was the first scientific pocket calculator and signaled HP as an innovator in calculator design. It sold for $395, over $1,800 in 2006 dollars. Like many HP calculators the HP 35 uses Reverse Polish Notation (RPN). With RPN you first enter the number and then tell the calculator what to do with it. According to an HP article on RPN, RPN has several advantages over typical algebraic entry including showing intermediate answers and using fewer keystrokes. HP says it"s easy, and now even has a virtual RPN calculator at the above link to try it out. Made in USA. Mine appears to be the third of four versions of the HP 35. Over 300,000 were sold, vastly exceeding expectations. The HP 35 signaled the death of the slide rule. The TI 50 eventually competed with it with a significantly lower price but without the RPN preferred by many. Mine was purchased for $45 with $5 shipping in February 2006 on eBay. That"s expensive, but HP calcualtors, both used and new, tend to be expensive. The description said it works, but it did not come with batteries or a charger to test it. I had an HP charger (82026A) from a garage sale ($1) that I thought would work, but it has a slightly different two hole connection while the HP 35 calculator has three pins. After some experimenting I got it to work with three 1.2 volt NI-MH rechargeable batteries taped together with Aluminum foil to join negative and positive terminals and speaker wire to hook the battery terminials to the calculator terminals (upper one went to negative battery terminal). This closely approximates the 3.5 volts of the original battery pack. (I did not do this for long and I don"t know if this can cause injury or damage, so don"t necessarily do what I did!) With this arrangement you, of course, must charge the batteries outside the calculator in an appropriate battery charger. There is excellent information on the HP 35 at several sites including HP site, Museum of HP Calculators, Museum of HP Calculators Simulation, Wikipedia, Vintage Calculators, Making New Battery Packs for the HP 35 Calculator, Educalc.net (exploded view), HP Calculator History - Trascendental Functions, Datamath.org - Internal View HP 35.

Hewlett Packard 67, (1976-1982) The HP 67 is a programable scientific calculator that was introduced around the height of the calculator boom of the 1970s and at the beginning of the personal computer revolution. (The Altair 8088 computer was introduced in 1975 and the Apple 1 computer was introduced in July 1976.) The similar HP 97 was a desktop version with an integrated printer. The HP 67 sold for $450 in 1976, or about $2075 in 2020 dollars. That"s a lot of money but having that computing power in your pocket was well worth it to engineers and scientists. The HP 67 was the successor to the HP 65 introduced in 1974 with a suggested retail price of $795, or about $4,400 in 2020 dollars! The HP 65 was the first hand-held programmable calculator. At the time, Apple co-founder, Steve Wozniak, was a scientific calculator designer at Hewlett Packard and sold his HP 65 in 1976 for $500 to help finance the start of Apple Computers. In 2013 Wozniak stated, "The next month HP was going to introduce the HP-67, a better calculator, and my employee price would be $370." www.bizjournals.com. Wozniak had offered his Apple 1 design to Hewlett Packard five different times without success. HP made a programable electronic desktop calculator, the HP 9100A, in 1968 that could be considered an early personal computer. For marketing purposes, they called it a desktop calculator, however. Bill Hewlett said, "If we had called it a computer, it would have been rejected by our customer"s computer gurus because it didn"t look like an IBM. We, therefore, decided to call it a calculator and all such nonsense disappeared." History of the 9100A desktop calculator, 1968. The 40-pound machine was priced at $4,900, or around $37,000 in 2020 dollars! The 9100A logic circuit was built without integrated circuits. Bill Hewlett around 1970 challenged the HP 9100A developers to make a "shirt pocket" version. "I want it to be a tenth of the volume, ten times as fast and cost a tenth as much." (The HP 35, Consumer Electronics, An Origin Story, http://codex99.com/design/the-hp35.html.) The result was the HP 35 scientific calculator, followed by the HP 65 and HP 67 programable scientific calculators. The HP 9100A was the first "calculator" to use Reverse Polish Notation. https://www.calculator.org/articles/Reverse_Polish_Notation.html.

The serial number for my HP 67 starts 1703 which indicates it was manufactured in early 1977. https://www.hpmuseum.org/collect.htm#numbers. My HP 67 was made in the United States, perhaps at the HP Corvallis, Oregon facility, which is located on the Willamette River and near Oregon State University. The Corvallis Gazette-Times on August 8, 1974 indicated HP planned to build a calculator and research facility in Corvallis. Moving day was in September 1976. "Report from Corvallis: Now that the dust has settled . . .," Measure magazine (August 1978). See "How Hewlett-Packard layoffs were avoided by founders Bill and Dave," Tom"s OSU, October 15, 2016. Prior to the Corvallis facility, calculators were made at the Advanced Products Department in Cupertino. See vintagecalculators.com ("At first the HP-35 was made only in the USA (at the Advanced Products Department in Cupertino - the calculator division moved subsequently to Corvallis"). HP still markets calculators today. The manufacturing and much design are done overseas, however. For example, the HP 35s programable scientific calculator, which was introduced in 2007 to commemorate the 35th anniversary of the introduction of the HP 35, was designed in Taiwan and made in China. https://www.embedded.com/tear-down-scientific-calculator-boils-design-down-to-two-ics/.

My HP 67 was a generous gift from a gentleman in Vancouver, Washington in June 2020. He used it as an engineering student at Oregon State University in the 1970s. It is in great cosmetic condition. It comes with the charger and several programming cards. It turns on but gives an error message. I will have to research more the possiblity of getting it repaired. If I get it working I"ll see if I can learn to program it.

HP 12c, financial calculator using Reverse Polish Notation (RPN). First introduced in 1981, the HP 12c is still sold today! According to ajc.com/business (Atlantic Journal-Constitution), HP claims it is the oldest consumer electronic device still made. The new on-line price at hp.com is $69.99. With RPN, you add 2 + 3 with the following keystrokes: 2 Enter 3 + . This takes getting use to. hp.com has considerable information availble including features and specifications, and the 211 pagemanual. See also Museum of HP Calculators. Purchased at a garage sale 9-10-05 for $.25. Good cosmetic and excellent working condition. No case or manual. USA stamped on the back with serial number 2902A32194. The 12c has been made in other countries but apparently the USA ones are the most valued. (See Museum of HP Calculators). If anyone knows the approximate date of manufacture of my calculator, please e-mail me.

Hewlett Packard 15C (1982-1989) (Large Image) Advanced programmable scientific calculator using Reverse Polish Notation. It shares a similar design with the HP 12C financial calculator above. Both are part of the HP 10C series which included the HP 10C basic scientific calculator, HP 11C mid-range scientific calculator, HP 12C financial calculator, HP 15C advanced scientific calculator and the HP 16C computer programmer calculator. (HP 10C Series - Wikipedia.) The HP 15C was made from 1982 to 1989. It originally cost $135. That"s about $315 in 2011 dollars - the price of a discount laptop computer. (For example, on the day I bought my HP 15C my local Radio Shack had a laptop computer on sale for $300.) While the HP 15C hasn"t been made for over 20 years, it still has a following. Bring Back the HP 15C seeks to have HP resume production. 14,847 people in 175 countries have signed a petition stating they would buy 83,363 HP 15c calculators. (Curiously, that"s an average of 5.6 calculators per person.) That site states the HP 15C is the best scientific calculator ever made by anyone for everyday use. Reasons: It uses Reverse Polish Notation (RPN) which is more efficient and natural, it was the smallest RPN calculator made by HP, the keyboard feel is very close to perfect, the landscape keyboard layout where you press keys with your two thumbs is better than a portrait layout where you press keys with one index finger, it has great battery life, the owner"s manual is top notch, and it is the coolest looking calculator ever made! HP 10c Series - Wikipedia states it sells used for between $200 and $400. On the day I bought mine (July 23, 2011), one sold on eBay for $419! The range on eBay for about the past 15 days was from $92 to $600 with 25 of them sold. I bought mine at a garage sale in the San Carlos area of San Diego for $1.00! It is in good working and cosmetic condition. It was pretty dusty with some yellowing of the covering over the screen. It works fine, however. It did not come with the cover or manual. There are computer simulators at Hewlett Packard 15C - A Simulator for Windows, Linux and Mac OSXand athp15c.com. As explained at hpmuseum.org the HP 15C supports a lot of advanced math including complex numbers, matrix math and many things beyond my comprehension! It uses three 1.5 volt LR44, S76 or similar button cell batteries which are readily available today. UPDATE: I was looking at amazon.com on 9-16-11 and noticed that HP now does sell a new edition of the HP 15c for $99. It"s the same as the old one but up to 100X faster. Pretty cool - an electronic product being reintroduced after 22 years! The petition referred to above apparently worked! Congratulations.

Canon Pocketronic, truly historic, the Pocketronic was the first pocket calculator. (For big pockets or hands - it"s about 8" x 4" x 2".) It is a direct product of Texas Instrument"s "Cal-Tech" project. The Cal-Tech (i.e. calculator technology) project set out in 1965 to use integrated circuits to build a calculator that could fit in one"s hand. The project was completed in 1967 with several working prototypes. Texas Instruments sought out a manufacturer and Canon, noted for its cameras, was interested to increase it business machine business. The result was the Canon Pocketronic released in Japan in April 1970. (I was in 7th grade.) It is very similar to the Texas Instruments prototypes including having a horizontal paper printout. It has no LED, LCD or other display - just the printout. It is powered by 13 rechargeable Ni-Cad batteries. It originally sold for $395, over $1,950 in today"s dollars!

The Cal-Tech program is discussed at the datamath site. On the left menu, click "History" and then "Datamath story". The datamath site also has a good article on the Canon Pocketronic which it describes as "the most important calculator in the history of Texas Instruments." On the main menu, go to "Calculators related to Texas Instruments," then to "Canon," then to the "Pocketronic." The datamath site also has several additional pictures. Vintage Calculators also has an excellent article on the Pocketronic and the Cal Tech project. See also Old Calculator Museum. Instructions.

My Pocketronic was purchased on e-Bay on May 26, 2005 for $42, with $12 shipping and insurance. It is in good cosmetic condition, but does not have the charger and therefore I do not know if it works. I assume the Ni-Cad batteries no longer function. Serial no. 406747. It has an "Export Inspected by Canon, Tokyo" tag on the front.

Canon Canola L121 (circa 1971) (large image) www.datamath.org states the first desktop calculator using Large Scale Integrated (LSI) circuits. According to Wikipedia an integrated circuit is a miniaturized electronic circuit etched onto a semiconductor material such as silicon. A large scale integrated circuit has tens of thousands transistors per chip. (See also webopedia.) Describing the L121 as a breakthrough in technology, www.datamath.org has photos and descriptions of the four main integrated circuits in the L121. The L121"s display used Nixie tubes which are sort of like vacuum tubes with 10 layers inside, each representing one of the numerals 0-9. (See Wikipedia.) After this time, Nixie tubes were rapidly replaced by orange Panaplex displays by Burroughs which appear to be like flat Nixie tubes, Light Emitting Diodes (LED), Vacuum-Fluorescent-Displays (VFD), and Liquid Crystal Displays (LCD). Each used progressively less power allowing more pocket sized devices. (See Datamath"s Display Technology of TI Calculators).

The Old Calculator Web Museum has a detailed discussion of the L121 including how the calculator sparked the interest of that Web site"s author as a youngster in an eventual career in computer science. I myself acquired an L121 because it reminded me of the electronic desktop calculator that was demonstrated by a speaker in my 8th grade math class. As indicated before, I was amazed at what the machine could do. I have no recollection of what brand or model of calculator it was, but it would have been the same vintage as an L121. John Wolf"s Web Museum has an excellent display of early Canon desktop calculators including the L121. Be Calc has some excellent close-up photos the L121 components. It lists a date of 1970, a year earlier than other sites. Classroom Tech shows a somewhat similar Monroe 620 made by Canon. I purchased my L121 on eBay on 6-16-06 for $9.99 plus $16.25 shipping from Oklahoma. It is in good working and operating condition with a cover and power cord.

Canon Palmtronic LE-10 (January 1972) Canon"s first LED display pocket calculator, following the Bowmar 901B introduced September 1971 by only a few months. Datamath.org states the original price was $259 or over $1,250 in 2006 dollars! It is a very solid machine with a very clear LED screen and ten digits. It runs on either a NiCad battery pack or four AA batteries which fit into a removable holder. It has an analog battery meter. To charge the NiCad Battery pack you fit the calculator on a cradle apparently attaching it with a screw that fits in a screw mount on the calculator. The screw mount happens to be the same size as a tripod mount; hence, it"s the only calculator I know that mounts to a tripod!

I purchased this on eBay on 6-13-06 for $15.75 with $5 shipping - a great deal! It is in near new cosmetic condition and operates perfectly. It includes the original (box), manual and cover, all in excellent condition. It also comes with a battery holder for AA batteries as well as a NiCad battery pack. Both the holder and the separate NiCad pack look new with absolutely no corrosion. It did not include the cradle for charging and AC operation, however. Datamath.org has some excellent photos of the charging craddle and internal views of the calculator. While made in Japan, the calculator uses Texas Instruments chips and display modules.

According to Datamath there were actually four versions