Author Topic: Advantest/ADCMT R6581 VFD to TFT conversion  (Read 2531 times)

vf33184 and 1 Guest are viewing this topic.

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Advantest/ADCMT R6581 VFD to TFT conversion
« on: January 25, 2025, 12:50:01 pm »
Hi all,

I have moved my implementation of the VFD replacement on the R6581 to here as I felt it was getting a bit heavy for MickleT's original thread (see below).....my intention was not to take over his thread, but it was starting to head that way!
MickleT's Advantest R6581 VFD replacement thread:
https://www.eevblog.com/forum/repair/advantest-r6581-vfd-replacement/

The background to my TFT conversion you can find here:
https://www.eevblog.com/forum/repair/advantest-r6581-vfd-replacement/msg5732473/#msg5732473
https://www.eevblog.com/forum/repair/advantest-r6581-vfd-replacement/msg5743707/#msg5743707

In this thread I'm further documenting the mods and new functionality I am adding to my TFT conversion.

Again, my project could not have been done without the great work by MickleT. He not only reversed engineered much of the R6581 DMM but then came up with a great OLED conversion for failing VFD's.

##################################################################

Video:
Published on my TFT LCD effort:
https://youtu.be/rv6x6VsWyRo

Files:
These will be updated:
         

Test HEX files:
R6581_VS_Display_Test2.hex
Test writing to the TFT LCD without actual VFD SPI data but actions when a VFD SPI packet is received.
This initial version will count each time the TFT is updated on the MAIN display, and on the AUX display it will count each time data is captured from the R6581 SPI bus.
Any VFD SPI error codes are reported. TFT LCD refresh is set to 100ms. See R6581_VS_Display_Test2.hex attachment below.

R6581_VS_DisplayTestTFTonly1.hex
TFT counter only, VFD SPI is disabled.

Instructions:
https://www.ianjohnston.com/R6581_TFT/Instructions.pdf

Parts:
LCD & LT7680A-R (c/w flat flex & FPC conn) = https://www.buydisplay.com/bar-type-4-58-inch-320x960-ips-tft-lcd-display-spi-rgb-interface
The LCD (on it's own) is also available here = https://www.adafruit.com/product/5805
1 x 33uF 100V electrolytic capacitor (optional)
1 x 100uF 10V electrolytic capacitor
1 x 1uF polyester capacitor
2 x 10mm M3 brass standoffs
1 x STM32 Blue Pill (STM32F103C8T6) - 64k version.
1 x 5-way male SIL header
2 x 20-way arduino style female header
2 x Strips 2mm double sided tape
1 x Strip 6mm single sided foam

1 x LCD holder 3D .STL files for easier LCD mounting (no tape or foam required):
     See ZIP file below, courtesy vf33184 (Vladimir).

Screenshots:
Before (not my actual unit, mine was worse!):


After (with splash screen showing):


Disclaimer:
This work is experimental. Whilst I offer friendly support, you are on your own in terms of responsibility if things go wrong for you, or you don't like the result. You have been warned.

Ian.
« Last Edit: February 01, 2025, 10:14:02 am by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 
The following users thanked this post: Mickle T., faraday

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #1 on: January 25, 2025, 12:50:50 pm »
MOVED FROM HERE:
https://www.eevblog.com/forum/repair/advantest-r6581-vfd-replacement/msg5776769/#msg5776769

Hi all,

I have made some mods to my TFT implementation of the display.

MOD. 1 - 9th Jan '25
I have always wanted a 1 VDC range for the R6581. Now, we do get a 1000 mV range however I guess I am too used to my 3458A where it will display X.XXXXXXX VDC.
So, I added a new mode whereby if you are on the 1000mV Range and you hit the DCV button it will switch the display out to a new 1 V Range.
It gets its data from the 1000 mV Range and so you get an extra digit, one more than the 10 V Range for instance at max 8.5 digits.
See photo below.

The Blue Pill uses two additional inputs to interface to the DCV button, see photo below.
Two are required because the button is multiplexed and I need two inputs to somewhat read a button press.

GIT source = https://github.com/Ian-Johnston/R6581_VS_Display
GIT .HEX file = https://github.com/Ian-Johnston/R6581_VS_Display/tree/master/VisualGDB/Debug

Only upgrade to this version is you hook up the new inputs to the DCV switch.

New Pcb gerbers attached which include the new A11/A12 connections.

Any issues then let me know (there are a couple of very small things).

PS. For the coders, The Blue Pill doesn't have a hardware FP and the software FP was being a pain, i.e. I wanted to use atof but it just continued to return empty, so I wrote around it (ugghh!). There doesn't seem to be any downsides though. I’ll return to this soon as atof is a lot less code.

UPDATE - 9th Jan '25
Digital input A11/A12 logic tweaked.

UPDATE - 10th Jan '25
1 VDC mode is now quite a bit more seemless, once it is set then it will remain in 1 VDC mode even if overrange or the DMM is in AUTO mode.
GIT hex file updated.
Still to do: try to reverse engineer the keypad better to fully isolate that button.

UPDATE - 15th Jan '25
Vertical Back Porch for TFT set to 11 rather than 10 (10 is per spec!) due to TFT flickering issue when running the R6581T for 20hrs.
Possibly the a heat/stability issue with the LT7680A-R or the TFT itself. They do run pretty warm in there (I might look at a fan upgrade)!

UPDATE - 16th Jan '25
Horizontal Back Porch for TFT set to 30 rather than 20 (20 is per spec!) due to TFT still flickering issue when running the R6581T for 12hrs.
I have attached the original HEX file and this new one.

UPDATE - 19th Jan '25
Fix issue where random pixels might appear at right hand vertical edge.
See new HEX file.

UPDATE - 20th Jan '25
Display still flickering at times, found that TFT timings was incorrect, the TFT is 400x960 but with partial glass, i.e. -80. So, must be treated as 400x960 in software.
See new HEX file.

UPDATE - 23rd/24th Jan '25
Still chasing the flickering problem which seems to return randomly, or temperature related.
Porch and start timings - LCD_VBPD, LCD_VFPD, LCD_VSPW, LCD_HBPD, LCD_HFPD, LCD_HSPW
1 -  17, 15, 1, 50, 30, 10         Haven't tested fully
2 -  17, 15, 2, 50, 30, 10         Stable with freezer spray & heat gun, single vertical line at right
3 -  17, 15, 3, 50, 30, 10         Flickers when cold, no vertical line
4 -  17, 15, 4, 50, 30, 10         Flickers badly
5 -  17, 14, 1, 50, 30, 10         Flickers badly
6 -  17, 14, 2, 50, 30, 10         Stable with freezer and heat gun, has a single vertical line
7 -  17, 14, 3, 50, 30, 10         Stable with freezer and heat gun, has a single vertical line
8 -  17, 14, 4, 50, 30, 10         Stable (except once on startup) with freezer and heat gun, no vertical line - ORIGINAL
Option 6 seems to be the closest to the AdaFruit example but has a vertical line down the right hand side which I can't get rid of. It's the most stable.
Well, I can by adjusting other parameters but it makes the TFT unstable (flickers).
Wierdly, my other R6581T, and ADCMT unit, has no flicker or pixel issues at all.
5 HEX files below to choose from:
R6581_VS_Display_17_14_2_50_30_10.hex
R6581_VS_Display_17_14_3_50_30_10.hex
R6581_VS_Display_17_14_4_50_30_10.hex
R6581_VS_Display_17_14_2_50_30_10_45Hzrefresh
R6581_VS_Display_17_15_2_50_30_10_45Hzrefresh

Gerbers updated to V1.3 due to short across 3 tracks on the Pcb.

See further posts in this thread for more updates.


MOD. 2 - 10th Jan '25
When the LCD shows "DISPLAY OFF" i.e. when the user actions this via the menu, it now turns off the LCD completely (blank). Pressing the MENU button revives.
It reduces the backlight to minimum, and turns off the LCD via register 0x12 of the LT7680.
A user asked for this and I guess it's good for TFT long term saving especially for those running the DMM 24/7.

UPDATE - 12th Jan '25
Bug fix on the backlighting brightness control when implementing DISPLAY OFF.

See further posts in this thread for more updates.

Ian.
« Last Edit: January 28, 2025, 01:19:10 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 
The following users thanked this post: Mickle T.

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #2 on: January 25, 2025, 12:58:00 pm »
UPDATE - 25th Jan '25
New version in Post #1 available for download (see link to my GIT).

With the issue some folks have been experiencing with flickering on the TFT and the various HEX files I had produced to give users a choice to see which one was most stable for them I decided to bring them all together in one HEX file and give the user the ability to select at R6581 boot.

See Post #1 for the link to my GIT for the latest HEX file, or the download.

This functionality requires the DCV button mod to my Pcb (two small wires soldered between the BluePill and the R6581 DMM board. See previous post.

Hold the DCV button in whilst powering up the R6581 and you should get the following on the LCD.
Alternatively, hold DCV and hit the reset button on the Blue Pill.



The current loaded TFT timing will be displayed, and from there you can press the DCV button to scroll around 6 (currently) loaded TFT timing configurations.
When you see the one you want simply reset power to the R6581.
The new config is saved to Flash (EEprom emulation).

On normal boot the LCD will display as part of the splash screen the currently loaded configuration.


The currently available configs:
VBPD VFPD VSPW HBPD HFPD HSPW Refresh Rate COG (Adafruit/BuyDisplay)
--------------------------------------------------------------------------------------
17     14     2       50      30     10      60                AdaF
17     14     3       50      30     10      60                AdaF
17     14     4       50      30     10      60                AdaF
17     15     2       50      30     10      60                AdaF
17     15     2       50      30     10      45                AdaF
10     12     3       80      30     20      60                BuyD

Most of the above settings effect the LT7680A-R TFT driver IC, the last one is to select either Adafruits or BuyDisplays config data for the ST7701S COG IC.

Note 1:
Holding DCV on boot doesn't always work, you may need to retry up to 10 times before it will 'catch' (I will try to fix this if I can).

Note 2:
I'd like to have the ability to select individual settings for each parameter, but there's only so much I can do on a single DCV button.

Info:
Why is it so hard to get TFT LCD Porch timings and screen coverage correct?
Getting TFT LCD porch timings right is tricky because it requires balancing the physical panel's requirements, the controller's limitations, and the interdependence of resolution, refresh rate, and pixel clock.
Not to mention cheap TFT LCDs from China, lack of good manufacturers tech info and in this case I think these ones or the LT7680A-R may be slightly sensitive to temperature etc.

Ian.
« Last Edit: January 26, 2025, 04:04:01 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Offline jj131415

  • Newbie
  • Posts: 7
  • Country: cn
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #3 on: January 26, 2025, 07:52:09 am »
hi Ian,
I extended the temperature range and tested different FW for your reference:
1) 17_14_2_50_30_10_45Hz ---OK @ 12C~55C, with vertical line
2) 17_14_2_50_30_10_60Hz ---OK @ 12C~55C, with vertical line, I'm using this version now.
3) 17_14_3_50_30_10_60Hz ---OK @ 12C/55C, flickering around 30C, with vertical line
4) 17_14_4_50_30_10_60Hz ---OK @ 20~55C, flickering @ 12C, no vertical line
5) 17_15_2_50_30_10_45Hz ---OK @ 20~55C, flickering @ 12C, no vertical line

Thanks.
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #4 on: January 26, 2025, 12:12:35 pm »
hi Ian,
I extended the temperature range and tested different FW for your reference:
1) 17_14_2_50_30_10_45Hz ---OK @ 12C~55C, with vertical line
2) 17_14_2_50_30_10_60Hz ---OK @ 12C~55C, with vertical line, I'm using this version now.
3) 17_14_3_50_30_10_60Hz ---OK @ 12C/55C, flickering around 30C, with vertical line
4) 17_14_4_50_30_10_60Hz ---OK @ 20~55C, flickering @ 12C, no vertical line
5) 17_15_2_50_30_10_45Hz ---OK @ 20~55C, flickering @ 12C, no vertical line

Thanks.

Thanks, thats good info.

Using my new code, i'm currently switching between two profiles, the one from my pic above - 10, 12, 3, 80, 30, 20, 60, BuyD and the one you are on.

It's wierd, because my other R6581T (ADCMT branded) has never, ever flickered and is rock stable at all temperatures.

Ian.
« Last Edit: January 26, 2025, 01:02:47 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #5 on: January 26, 2025, 01:44:06 pm »
UPDATE - 26th Jan '25
New version in Post #1 available for download (see link to my GIT or directly download the new HEX file).

Power up with DCV button pressed to enter TFT Timing Adjust screen now much more reliable.

Ian.
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 
The following users thanked this post: faraday

Offline jj131415

  • Newbie
  • Posts: 7
  • Country: cn
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #6 on: January 26, 2025, 02:27:24 pm »
Hi Ian,
I have another R6581T, it's more new if compare with my R6581D. I will try it later.

Thanks.
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #7 on: January 26, 2025, 10:46:56 pm »
Hello,
I wanted to have better solution to hold new display in place so here is my design of LCD holder. It works well for me and was successfully tested also by Ian.
It can be easily 3D printed. There are 3 versions of STL file, details are described in ZIP file.


Vladimir
« Last Edit: January 28, 2025, 07:29:53 pm by vf33184 »
 
The following users thanked this post: IanJ, TheDefpom, bsw_m

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #8 on: January 27, 2025, 07:16:54 am »
Hello,
I wanted to have better solution to hold new display in place so here is my design of LCD holder. It works well for me and was successfully tested also by Ian.
It can be easily 3D printed. There are 3 versions of STL file, details are described in ZIP file.

Confirmed, no need for double sided tape any more. LCD is easier to fit, secure in place and the important 42mm LCD horizontal position is adhered to.

Much thanks.

Ian
« Last Edit: January 28, 2025, 04:43:08 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #9 on: January 30, 2025, 12:23:39 am »
I have been testing actual firmware version 25-01-26. All six display settings are causing flickering in my case. None has vertical white line issue but I have to note that during assembly I have painted white frame of display by permanent black pen what can be hiding this vertical line that was mentioned by jj131415 for some configurations.

Test setup: R6581T with FW A01, room temperature 22C, meter without enclosure, duration 24h (the behavior was not changing during this period)

I have also made couple of capturing videos. Very common is that at the moment of flicker many symbols on the display are replaced by ? symbol. Not sure if this can help with troubleshooting under what condition display shows ? instead of proper symbol e.g. digit. Most of these flickers are happening on digits, some on yellow text below digits and very seldom is flickering of green text above digits. Probably there is some relation with refresh rate of these symbols?

Also noticed that when I change setting from 1000mV to 1V, after power cycle it returns back to default 1000mV. I would rather prefer to remember last state before power cycle or even better have an option to change this setting permanently in boot menu. But I can live with this behavior for now because flickering issue is most important and we should focus in that one first.

Vladimir
 

Offline coromonadalix

  • Super Contributor
  • ***
  • Posts: 7109
  • Country: ca
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #10 on: January 30, 2025, 02:39:13 am »
can a test / specially crafted FW could test the refresh rate of the display    or use the same  dmm characters placement indicators layout to see where it could be slower or have some bad timing ??

could it be the bluepill mcu  having a hard time to keep up ?
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #11 on: January 30, 2025, 12:57:07 pm »
I have been testing actual firmware version 25-01-26. All six display settings are causing flickering in my case. None has vertical white line issue but I have to note that during assembly I have painted white frame of display by permanent black pen what can be hiding this vertical line that was mentioned by jj131415 for some configurations.

Test setup: R6581T with FW A01, room temperature 22C, meter without enclosure, duration 24h (the behavior was not changing during this period)

I have also made couple of capturing videos. Very common is that at the moment of flicker many symbols on the display are replaced by ? symbol. Not sure if this can help with troubleshooting under what condition display shows ? instead of proper symbol e.g. digit. Most of these flickers are happening on digits, some on yellow text below digits and very seldom is flickering of green text above digits. Probably there is some relation with refresh rate of these symbols?

Also noticed that when I change setting from 1000mV to 1V, after power cycle it returns back to default 1000mV. I would rather prefer to remember last state before power cycle or even better have an option to change this setting permanently in boot menu. But I can live with this behavior for now because flickering issue is most important and we should focus in that one first.

Vladimir

Interesting!
If you see ?? then it's either characters I haven't converted, or it's data corruption. In this case it's clearly data corruption. So I guess the incoming SPI data is somehow being corrupted.
I notice from your screenshot you are running GPIB at the time......I've tried hammering GPIB and see no issue here.
I notice also you have the INFO annunciator lit......whats that for?......I thought it was "unused" per the Japanese user manual.

On my units I never get ?? at the same moment as flickering, for me it's two different issues. Kind of leads me in this case to power supply dropping out on the LCD or BluePill. Might be worth adding further caps across teh +5VDC & +3.3VDC on my board.

I have just recently got saving to flash working so yes, I can make the 1VDC/1000mV modes more permanent for the user. I'll get to that soon.
*IDN? gives me ADVANTEST,R6581T,0,A01......not sure about my ADCMT branded unit.
With both, I get no flickering.......but like I have said it can appear for no reason, but mostly temperature issues.

I have emailed my contact at Levetop the manufacturers of the LT7680A-R TFT controller IC to see what they make about the flickering. Hoping they can experiment with one! They are normally very helpful to me so we'll see what happens.
Hoping they can give me some definitive settings for the LT7680A-R and also the S7701.

Ian.
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #12 on: January 30, 2025, 01:10:10 pm »
can a test / specially crafted FW could test the refresh rate of the display    or use the same  dmm characters placement indicators layout to see where it could be slower or have some bad timing ??

could it be the bluepill mcu  having a hard time to keep up ?

Possible, but you can see flickering even when pressing the MENU button and displaying static text.
Also, the Blue Pill just sends relatively slow SPI data to the LT7680A-R, and it takes care of writing to the TFT......so even if the SPI stops then the TFT will just remain with the last data on it.

TFT flickering is a known phenomenon. The frequencies involved (between LT7680A-R IC and the TFT) are up at 60MHz......and akin to controlling an old CRT in some respects.

Saying that, i'll create a version of firmware for SPI & TFT testing. Check back soon.

R6581_VS_Display_Test2.hex
The attached TEST hex file runs the TFT refresh rate slower, 100ms instead of 35ms.
On the MAIN display it will count each time the TFT is updated.
On the AUX display it will count each time data is captured from the R6581 SPI bus.

R6581_VS_DisplayTestTFTonly1.hex
TFT counter only, VFD SPI is disabled.

UPDATE:
See new HEX file, any VFD SPI error codes are now reported.

I'll add more 'tests' when I can in order to detect errors/corruption, but for now it's running the TFT refresh rate slower, that's about it so let me know.

Ian.
« Last Edit: January 30, 2025, 08:08:17 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #13 on: January 30, 2025, 07:27:11 pm »
Interesting!
If you see ?? then it's either characters I haven't converted, or it's data corruption. In this case it's clearly data corruption. So I guess the incoming SPI data is somehow being corrupted.
I notice from your screenshot you are running GPIB at the time......I've tried hammering GPIB and see no issue here.
I notice also you have the INFO annunciator lit......whats that for?......I thought it was "unused" per the Japanese user manual.

On my units I never get ?? at the same moment as flickering, for me it's two different issues. Kind of leads me in this case to power supply dropping out on the LCD or BluePill. Might be worth adding further caps across teh +5VDC & +3.3VDC on my board.

I have just recently got saving to flash working so yes, I can make the 1VDC/1000mV modes more permanent for the user. I'll get to that soon.
*IDN? gives me ADVANTEST,R6581T,0,A01......not sure about my ADCMT branded unit.
With both, I get no flickering.......but like I have said it can appear for no reason, but mostly temperature issues.

I have emailed my contact at Levetop the manufacturers of the LT7680A-R TFT controller IC to see what they make about the flickering. Hoping they can experiment with one! They are normally very helpful to me so we'll see what happens.
Hoping they can give me some definitive settings for the LT7680A-R and also the S7701.

Ian.

Hi Ian,
those symbols TLK, LTN and INFO from my picture are ghosts during flicker only, in a normal situation I have only AZERO and FRONT lit plus SMPL blinking. I have not been using GPIB yet except short test that it works, got ADVANTEST,R6581T,0,A01 response to *IDN? so the same as your.  I am just curious how come that e.g. INFO is lit during flicker stage when you never call it via SPI to be lit?

Regarding power supplies, I have measured power consumption of whole display including Blue Pill an it is just 1.1W so I doubt it can overload meter power supply if whole meter has 30W.  I will try to setup monitoring of incoming 5V to your board to see if there are any glitches.

UPDATE: I have been measuring 5V input to your board, there is less than 1mV fluctuation in voltage so I don't think there is an issue with power supply quality/stability.


Vladimir
« Last Edit: January 30, 2025, 08:34:54 pm by vf33184 »
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #14 on: January 30, 2025, 08:10:39 pm »
Interesting!
If you see ?? then it's either characters I haven't converted, or it's data corruption. In this case it's clearly data corruption. So I guess the incoming SPI data is somehow being corrupted.
I notice from your screenshot you are running GPIB at the time......I've tried hammering GPIB and see no issue here.
I notice also you have the INFO annunciator lit......whats that for?......I thought it was "unused" per the Japanese user manual.

On my units I never get ?? at the same moment as flickering, for me it's two different issues. Kind of leads me in this case to power supply dropping out on the LCD or BluePill. Might be worth adding further caps across teh +5VDC & +3.3VDC on my board.

I have just recently got saving to flash working so yes, I can make the 1VDC/1000mV modes more permanent for the user. I'll get to that soon.
*IDN? gives me ADVANTEST,R6581T,0,A01......not sure about my ADCMT branded unit.
With both, I get no flickering.......but like I have said it can appear for no reason, but mostly temperature issues.

I have emailed my contact at Levetop the manufacturers of the LT7680A-R TFT controller IC to see what they make about the flickering. Hoping they can experiment with one! They are normally very helpful to me so we'll see what happens.
Hoping they can give me some definitive settings for the LT7680A-R and also the S7701.

Ian.

Hi Ian,
those symbols TLK, LTN and INFO from my picture are ghosts during flicker only, in a normal situation I have only AZERO and FRONT lit plus SMPL blinking. I have not been using GPIB yet except short test that it works, got ADVANTEST,R6581T,0,A01 response to *IDN? so the same as your.  I am just curious how come that e.g. INFO is lit during flicker stage when you never call it via SPI to be lit?

Regarding power supplies, I have measured power consumption of whole display including Blue Pill an it is just 1.1W so I doubt it can overload meter power supply if whole meter has 30W.  I will try to setup monitoring of incoming 5V to your board to see if there are any glitches.


Vladimir

I've added a another test HEX file whioch doesn't use SPI at all, so give it a go......it should just count upwards for the main TFT line.
The other test2.hex file will be useful to you also, as it _should_ report any VFD SPI errors. I never get any SPI errors so not easy to test.

PS. Have you tried another Blue Pill board.....just incase?

Ian.
« Last Edit: January 30, 2025, 08:15:26 pm by IanJ »
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #15 on: January 30, 2025, 08:41:06 pm »
Interesting!
If you see ?? then it's either characters I haven't converted, or it's data corruption. In this case it's clearly data corruption. So I guess the incoming SPI data is somehow being corrupted.
I notice from your screenshot you are running GPIB at the time......I've tried hammering GPIB and see no issue here.
I notice also you have the INFO annunciator lit......whats that for?......I thought it was "unused" per the Japanese user manual.

On my units I never get ?? at the same moment as flickering, for me it's two different issues. Kind of leads me in this case to power supply dropping out on the LCD or BluePill. Might be worth adding further caps across teh +5VDC & +3.3VDC on my board.

I have just recently got saving to flash working so yes, I can make the 1VDC/1000mV modes more permanent for the user. I'll get to that soon.
*IDN? gives me ADVANTEST,R6581T,0,A01......not sure about my ADCMT branded unit.
With both, I get no flickering.......but like I have said it can appear for no reason, but mostly temperature issues.

I have emailed my contact at Levetop the manufacturers of the LT7680A-R TFT controller IC to see what they make about the flickering. Hoping they can experiment with one! They are normally very helpful to me so we'll see what happens.
Hoping they can give me some definitive settings for the LT7680A-R and also the S7701.

Ian.

Hi Ian,
those symbols TLK, LTN and INFO from my picture are ghosts during flicker only, in a normal situation I have only AZERO and FRONT lit plus SMPL blinking. I have not been using GPIB yet except short test that it works, got ADVANTEST,R6581T,0,A01 response to *IDN? so the same as your.  I am just curious how come that e.g. INFO is lit during flicker stage when you never call it via SPI to be lit?

Regarding power supplies, I have measured power consumption of whole display including Blue Pill an it is just 1.1W so I doubt it can overload meter power supply if whole meter has 30W.  I will try to setup monitoring of incoming 5V to your board to see if there are any glitches.


Vladimir

I've added a another test HEX file whioch doesn't use SPI at all, so give it a go......it should just count upwards for the main TFT line.
The other test2.hex file will be useful to you also, as it _should_ report any VFD SPI errors. I never get any SPI errors so not easy to test.

PS. Have you tried another Blue Pill board.....just incase?

Ian.

Hi Ian,
Here is my result using R6581_VS_Display_Test2.hex    It was running almost an hour. No errors reported, none flickering noticed. Should I run another HEX file too?

Don't have another BluePill now, will buy one.

Vladimir
« Last Edit: January 30, 2025, 08:56:23 pm by vf33184 »
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #16 on: January 30, 2025, 08:51:22 pm »
Hi Ian,
Here is my result using R6581_VS_Display_Test2.hex    It was running almost an hour. No errors reported, none flickering noticed. Should I run another HEX file too?

Don't have another BluePill now, will by one.

Vladimir

If you are saying you swap between the normal HEX file and TEST2 and you are getting issues with the normal HEX file and not with TEST2........then the fact I dropped the TFT update rate from 35ms to 100ms on TEST2 may be the fix!
I have attached a 100ms TFT refresh version of the normal HEX file.

Ian.
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #17 on: January 30, 2025, 10:44:23 pm »
Hi Ian,
Here is my result using R6581_VS_Display_Test2.hex    It was running almost an hour. No errors reported, none flickering noticed. Should I run another HEX file too?

Don't have another BluePill now, will by one.

Vladimir

If you are saying you swap between the normal HEX file and TEST2 and you are getting issues with the normal HEX file and not with TEST2........then the fact I dropped the TFT update rate from 35ms to 100ms on TEST2 may be the fix!
I have attached a 100ms TFT refresh version of the normal HEX file.

Ian.

Hi Ian,
unfortunately this 100ms version didn't fix the problem. Screen update rates is visibly slower, but still flickering, just less frequently. Again with ? symbols.

Vladimir
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #18 on: January 30, 2025, 10:47:26 pm »
Hi Ian,
unfortunately this 100ms version didn't fix the problem. Screen update rates is visibly slower, but still flickering, just less frequently. Again with ? symbols.

Vladimir

Does TEST2 display ?? At any time, and does it flicker?

Ian
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #19 on: January 30, 2025, 11:15:58 pm »
Hi Ian,
unfortunately this 100ms version didn't fix the problem. Screen update rates is visibly slower, but still flickering, just less frequently. Again with ? symbols.

Vladimir

Does TEST2 display ?? At any time, and does it flicker?

Ian

I didn't notice any ? symbol or flicker, have been testing both SPI enabled and SPI disabled firmware.

Vladimir
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #20 on: January 30, 2025, 11:21:02 pm »
I didn't notice any ? symbol or flicker, have been testing both SPI enabled and SPI disabled firmware.

Vladimir

So, the TFT SPI is ok, and for some reason then, you’re getting VFD SPI corruption. Perhaps try another BluePill just in case you have a dodgy I/O pin.
Try buying from a different supplier.

Ian
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #21 on: January 31, 2025, 11:16:55 pm »
I didn't notice any ? symbol or flicker, have been testing both SPI enabled and SPI disabled firmware.

Vladimir

So, the TFT SPI is ok, and for some reason then, you’re getting VFD SPI corruption. Perhaps try another BluePill just in case you have a dodgy I/O pin.
Try buying from a different supplier.

Ian

Hello Ian,

I have ordered new BluePill from different shop, should arrive early next week. I have also ordered another display and LCD driver board, it is somewhere on the way from China. When it arrives I will try to swap display or driver board too.

Regarding several SPI buses. Do I understand well that TFT SPI is a bus between display and display driver, and VFD SPI is between display driver and BluePill? If so, what exactly TEST2 HEX file is testing? I would like to understand it better to help me narrow down possible place of an issue I have. For now I am a bit shooting in the dark and trying to find the root cause.


Thank you.

Vladimir
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #22 on: February 01, 2025, 10:32:17 am »
Hello Ian,

I have ordered new BluePill from different shop, should arrive early next week. I have also ordered another display and LCD driver board, it is somewhere on the way from China. When it arrives I will try to swap display or driver board too.

Regarding several SPI buses. Do I understand well that TFT SPI is a bus between display and display driver, and VFD SPI is between display driver and BluePill? If so, what exactly TEST2 HEX file is testing? I would like to understand it better to help me narrow down possible place of an issue I have. For now I am a bit shooting in the dark and trying to find the root cause.

TEST2 is primarily for testing the LCD without actually capturing VFD SPI data. It uses the VFD SPI routine only to trigger writing to the LCD.
You should see the MAIN line and AUX line count upwards at different rates, the MAIN line should count slower.
Both should be very even counting, no long pausing or resetting to zero etc.
The very bottom line will report VFD SPI issues if they are received.

The flickering issue is not related to SPI data, it's an issue with the actual TFT drive from the LT7680 driver board and the TFT LCD.

If you are getting ?? on MAIN or AUX using TEST2 then that indicates a fundemental issue with the Blue Pill and likely nothing to do with the SPI data being captured.

If you get good counting on AUX and MAIN using TEST2 then the problem is with the captured VFD SPI data.

For info, there are actually 3 SPI interfaces being used on the Blue Pill:
SPI1 = Full duplex SPI MASTER to the driver board. This is used to set up and write to memory in the LT7680A-R driver IC which is then in turn passed to the TFT.
SPI2 = This is an SPI SLAVE used to capture data from the R6581 display board.
SPI(bitbang) = This is an SPI MASTER connected to the TFT itself (COG chip on glass) and is used once at startup to set up the actual TFT itself. Written once only.

Ian.
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 

Online vf33184

  • Contributor
  • Posts: 20
  • Country: cz
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #23 on: February 01, 2025, 05:24:42 pm »
TEST2 is primarily for testing the LCD without actually capturing VFD SPI data. It uses the VFD SPI routine only to trigger writing to the LCD.
You should see the MAIN line and AUX line count upwards at different rates, the MAIN line should count slower.
Both should be very even counting, no long pausing or resetting to zero etc.
The very bottom line will report VFD SPI issues if they are received.

The flickering issue is not related to SPI data, it's an issue with the actual TFT drive from the LT7680 driver board and the TFT LCD.

If you are getting ?? on MAIN or AUX using TEST2 then that indicates a fundemental issue with the Blue Pill and likely nothing to do with the SPI data being captured.

If you get good counting on AUX and MAIN using TEST2 then the problem is with the captured VFD SPI data.

For info, there are actually 3 SPI interfaces being used on the Blue Pill:
SPI1 = Full duplex SPI MASTER to the driver board. This is used to set up and write to memory in the LT7680A-R driver IC which is then in turn passed to the TFT.
SPI2 = This is an SPI SLAVE used to capture data from the R6581 display board.
SPI(bitbang) = This is an SPI MASTER connected to the TFT itself (COG chip on glass) and is used once at startup to set up the actual TFT itself. Written once only.

Ian.

Hello Ian,
thank you for very good description of different SPI buses and its function.

I have been running TEST2 HEX file again, this time more than 24 hours.  MAIN display counted 1 million and AUX display 6.5 million samples, VFD SPI report no errors (see picture). Counting is very smooth without any interruption, haven't noticed any glitches or flickers.

But with regular firmware some symbols are sometimes being replaced by "?" (shown on picture in my previous messages) just probably for single display refresh cycle, next cycle it's being corrected. Based on your description it would lead me to conclusion that:
a) no issue on communication between BluePill and LT7680A-R driver
b) no issue on communication between BluePill and TFT COG
c) issue with data being spied on SPI link between main meter CPU and Hitachi CPU on display board. If so, then I don't know why TEST2 firmware reports "VFD SPI - no errors". I don't know how do you evaluate issue on VFD SPI, if each data batch has some parity check or some other mechanism?

If there is corrupted data at VFD SPI input, BluePill replace unknown symbol by "?" and send it to TFT or BluePill sends unknown received character down to TFT and then TFT itself replace unknown character by "?" symbol?

Sorry for many questions, I am just trying to find out where is my problem. 

Thank you.

Vladimir
 

Offline IanJTopic starter

  • Supporter
  • ****
  • Posts: 1848
  • Country: scotland
  • Full time EE & Youtuber/Creator
    • IanJohnston.com
Re: Advantest/ADCMT R6581 VFD to TFT conversion
« Reply #24 on: February 01, 2025, 10:28:32 pm »
Hello Ian,
thank you for very good description of different SPI buses and its function.

I have been running TEST2 HEX file again, this time more than 24 hours.  MAIN display counted 1 million and AUX display 6.5 million samples, VFD SPI report no errors (see picture). Counting is very smooth without any interruption, haven't noticed any glitches or flickers.

But with regular firmware some symbols are sometimes being replaced by "?" (shown on picture in my previous messages) just probably for single display refresh cycle, next cycle it's being corrected. Based on your description it would lead me to conclusion that:
a) no issue on communication between BluePill and LT7680A-R driver
b) no issue on communication between BluePill and TFT COG
c) issue with data being spied on SPI link between main meter CPU and Hitachi CPU on display board. If so, then I don't know why TEST2 firmware reports "VFD SPI - no errors". I don't know how do you evaluate issue on VFD SPI, if each data batch has some parity check or some other mechanism?

If there is corrupted data at VFD SPI input, BluePill replace unknown symbol by "?" and send it to TFT or BluePill sends unknown received character down to TFT and then TFT itself replace unknown character by "?" symbol?

Sorry for many questions, I am just trying to find out where is my problem. 

Thank you.

Vladimir

You've got it right.....just to add that the SPI errors thats can be reported are probably rather limited. It's a built in mechanism I am am not 100% sure of all the failure modes it can report on.
Can you post a photo of the termination of the 3 wires down onto the R6581 display board.

Ian.
Ian Johnston - Original designer of the PDVS2mini || Author of WinGPIB
Website: www.ianjohnston.com
YouTube: www.youtube.com/user/IanScottJohnston, Odysee: https://odysee.com/@IanScottJohnston, Twitter(X): https://twitter.com/IanSJohnston, Github: https://github.com/Ian-Johnston?tab=repositories
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf