Datron 1071 repair

[feedback.loop]

I take it the character displays aren't preferable? You could take your current approach and use small 1" OLED modules to emulate 1-2 digits from the Panaplex.

I don't see how to draw 9 large digits plus symbols on a two-row character display.
Regarding those .96" OLED on eBay, I did consider them, but they are either white or blue and won't span the window on the meter vertically (not by much though), so I preferred TFT. However, if I give up the color, or put a red or orange filter in front of the white ones, they might be a good option.

[z01z]

There are bigger OLEDs available, like 1", 1.1" and 1.3", though these are all in white.
There's also a vertical 0.96" display that might be a bit bigger than the horizontal one.
I got the 1.3" one from them (with the matching connector) and though it needs some extra circuitry to drive it, they do work, the datasheet has the needed schematics.

Another approach could be a custom panel with SMD orange LEDs laid out as the original segments, with a diffuse layer in front and a mask in between, that would separate the segments from one another, like that filament display posted here previously. This way you could emulate the middle dots as well, and with appropriately cut out masks, all the special symbols.
One could even etch a mask from a thin one sided PCB, though I'm not sure how the green tinge of the PCB with the orange light would look, probably terrible. Just a random idea .

[Le_Bassiste]

Remember this HP 6632B OLED display replacement project?

Just a thought. 

I don't see any details there. Do you have the model of the display used in that project?

They used this 16x2 panel: http://www.ebay.com/itm/IIC-I2C-1602-16x2-OLED-Module-Display-For-Arduino-PIC-AVR-ARM-/161162845021

But if you wanna go big, try this extra large 20x2 yellow OLED panel: http://www.mouser.com/ProductDetail/ELECTRONIC-ASSEMBLY/EA-W202-XDLG

EDIT: Wow, this 96mm wide 20x2 panel also looks nice: http://www.ebay.com/itm/IIC-I2C-2002-20x2-OLED-Module-Display-With-Arduino-Example-/191835417963

These larger graphic OLEDs also exist, but I don't know where you could get them: http://www.winstar.com.tw/products/oled-module/graphic-oled-display/256x64-oled.html

The first and the second are not graphic. The third is graphic, but won't fit into the meter and seems to cost about $150.

i got mine from www.texim-europe.com for approx. EUR 40 per piece.
quality is not too shabby, so i gave it a test drive with one of my K2001. they "fit like a glove" into the display frame of the DMM, with no modifications to the frame at all:

[nidlaX]

i got mine from www.texim-europe.com for approx. EUR 40 per piece.
quality is not too shabby, so i gave it a test drive with one of my K2001:

Oh my, I need to find a US dealer for these...

[Le_Bassiste]

rest assured, it is by no means as crisp and bright as the original VFD. 

[texaspyro]

I was thinking about doing a replacement front panel board for the HP-53132A counter...  the only way to get a new display for these counters is to send HP $1600+ for a new board... they won't sell the display separately.    Crystalfontz sells a 192x32 graphic LCD for a reasonable price.  They also have smaller displays.

[feedback.loop]

Reviving this topic.
I was asked to share the code (yes, after 2 years!), so here it is. But it is not the same code I showed in the video. At some point I decided to play with this project a little more. I shifted some more complexity to hardware to make things more reliable and repeatable across different meters. No particular reason, just wanted to clean things up a bit.
I cleaned up the interrupt signal by using 74LS74 flip-flop: phase2 clock to pin 3 (CLK) and the address selector to pin 2 (DATA), pins 1 and 4 are pulled up. The output from the flip-flop (pin 5) is used to interrupt the MCU and also to latch both address and data. I added one more 74LS374 to latch the address as well (only 4 lines are used). With this configuration the rising edge from the flip-flop happens when both the address and the data are stable, and the glitch is avoided entirely. Now there is no rush at all to read the address and the data in software. This drastically simplified the software. The interrupt routine was a crazy mess in assembly and very time sensitive. Now it is 3 lines in C, no naked interrupt trickery is needed. Enjoy.

[coromonadalix]

Thank you very much

[Johnny10]

So you changed the hardware and software for the Display LEDs you used a few years back?

What did Le_Bassiste reply have to do with the Datron project?
I do like that display of his also.

[feedback.loop]

So you changed the hardware and software for the Display LEDs you used a few years back?

What did Le_Bassiste reply have to do with the Datron project?
I do like that display of his also.

I am not sure I understand your point.
I was asked in a private message to share the code. I thought that reviving the topic to share with everybody would be a good idea.
I did make some changes since I published the videos, so I described the changes.
The displays I used are cheap TFT displays (not LEDs). I don't quite like them: not very bright and viewing angle is limited.
The comments by Le_Bassiste have to do with some other project using an OLED display.

[Johnny10]

OK,

Sorry I got lost in that thread.
Thought maybe you were saying you reworked your displays something similar to Le-Bassiste.

Thanks for clarifying.
I have a few broken Datron 4000 displays also.

[branadic]

In this case you want to follow this thread:

https://www.eevblog.com/forum/repair/datron-4000a-getting-at-test-points/msg2289120/#msg2289120

-branadic-

[feedback.loop]

I finally decided to try another approach for display replacement in this good old Datron 1071 meter. Design decisions are discussed in detail, and the code for ATmega164PA is shown. The new display is very bright with a very wide viewing angle, but it could use a better diffusing overlay.

[branadic]

Well done 
You can use a white paper (80g) between spacer and printed foil / photo film as a diffusor. This worked at least for me and this is what I've shown in my pictures. Sorry if I forgot to mention this.
Very early prototype using white spacer had printed paper instead of the printed photo film, fixed with Kapton tape to the board, which worked to, but I skipped that approach later.

-branadic-

[seebeck]

I saw a FLUKE 343A DC voltage calibrator, the spec is 0.003%...

[feedback.loop]

A sheet of paper was added between the spacer and the overlay film, a few more things were tried as well. It seems that a thicker spacer should help, but I don't have the means to make one at the moment.

[The Soulman]

A sheet of paper was added between the spacer and the overlay film, a few more things were tried as well. It seems that a thicker spacer should help, but I don't have the means to make one at the moment.

Use the transparent plastic of a cd case as a spacer?

Looks good already btw. 

[HighVoltage]

What happens, if you put a window like from the Aglient 34401A in front of your LED display?
It should defuse and focus at the same time.