I am having a problem with a new GM328A (red pcb) kit.
In this topic and the author's manual repeatedly reiterated condition, which solves 99% of all problems with measurements and calibration, replacement of the weak Chinese +5V regulator to a
QUALITY (MCP1702, LP2950)!
In this topic and the author's manual repeatedly reiterated condition, which solves 99% of all problems with measurements and calibration, replacement of the weak Chinese +5V regulator to a QUALITY (MCP1702, LP2950)!
I had a similar issue when upgrading from 1.12k to 1.13k. The problem is in the 7550 and the capacitance of 10 microfarads, to the right of the crystal. Change the 7550 or match the indicated capacitance with a large ESR within> 3 ohms
I swapped the capacitor, the 7550 and the TL431 between boards and the problem stayed with the board. I then replaced them all with other parts and the problem again stayed with bad board and the good board still worked fine with parts from the bad board.
Any other ideas as to the source?
Any other ideas as to the source?
Compare the selftest run on your boards with the picture:
Any other ideas as to the source?
Besides the Atmega or the ZIF socket (it can be bypassed for tests), other user found a bad PCB with a tiny shortcircuit among traces, and we have found some shorted ceramic capacitors too (SMD and/or through-hole).
As stated in the original post, ATMEGA was swapped between boards and problem stays with board; the microcontrollers are healthy in the other board.
ZIF socket bypassed for SMD pad but same results. Show 0 ohm continuity between all pad and zif pins as well no cross shorting there.
All resistors measured in circuit with ATMEGA removed and show in spec and same between boards within ~1%.
Edit: I pulled every transistor (except the LED one ) and every capacitor and tested and swapped between boards. Still same problem that always stayed with the bad board.
Could not see any damaged/shorted traces anywhere. Metered for shorts in the microcontroller socket and ZIF and found nothing. I'll do a bit more digging then get config data. Thanks.
Could not see any damaged/shorted traces anywhere. Metered for shorts in the microcontroller socket and ZIF and found nothing.
Remove the SRV5-04 and 6V8 protection elements temporarily. Check the operation of the tester without them. If the problem persists, show the results of the T1-T7 hardware tests.
I've updated the comparison table for the most popular clone models. Several new models have been added as well as more information about the color palette of the displays. Fixed minor bugs and minor changes in description.
Hello Yuriy_K
I tried to test your "ST7735_RGB_en16MHz-Version" with a GM328A. The board is original with a 16MHz-Quartz. Fuses are (low=0xF7; high=0xD9; ext=0xFC). But the result is only a lighted LCD. Have you made any changes in the hardware, needing other fuses?
Best wishes
Have you made any changes in the hardware, needing other fuses?
Firmwares from page 277 are tested on the displays shown on the same page and earlier. If you have a different display, show its large image. Fuses in the picture.
Hello Yuriy_K,
I tested your firmware ST7735_RGB_en16MHz again with changed ext. fuse, - low fuse and high fuse are the same as before and in your posted picture. When I start the tester only the backlight of LCD switched on. Nothing others happens. As you can see: board is GM328A-kit and LCD is with ST7535 driver. I think, your vers. of firmware is based of K-H Kuebbeler's firmware. I wonder, when I flashed firmware 1.13k, all is ok. What could be the reason for this mistake? Curiously is, after starting the tester, led goes on - but by pressing the knob short or long - led and LCD staying on.
Best wishes
You yourself confused me. You are trying to run my firmware from page 291. This firmware is after last changes by Karl Heinz, I don't know how it works on old GM328A. Try the firmware from page 277 ST7735_RGB_en.zip it is 16MHz. She should definitely work.
Read the next steps on page 292.
Hello Yuriy_K
R8 and R9 are checked (32,9k and 3,29k) are testet, Q1 not desoldered and testet. Your firmware "ST7735_RGB_en_kit_16MHz" testet with fuses (l=f7; h=DC; ex=FC).
After starting such strips appeared in blue later white than red, finalla was displayed "not calibrated". When I put a m328 with firmware 1.12k or 1.13k - tester operates normaly. I have no explanation for it.
Have you or another one a solution
Have you or another one a solution
Send a Makefile from the firmware that works correctly. Show the image on the display in the normal connection of the display to the board.
Solder the same capacitors in parallel with the 10 uF electrolytic capacitors.
Hello Yuriy_K
R8 and R9 are checked (32,9k and 3,29k) are testet, Q1 not desoldered and testet. Your firmware "ST7735_RGB_en_kit_16MHz" testet with fuses (l=f7; h=DC; ex=FC).
After starting such strips appeared in blue later white than red, finalla was displayed "not calibrated". When I put a m328 with firmware 1.12k or 1.13k - tester operates normaly. I have no explanation for it.
Have you or another one a solution
Hi Rober!
Did you change the quartz from GM328? I know the original is 8MHz, the software version you want to use is compiled for 16MHz. Maybe that's where the problem lies.
Hello colleagues.
With some delay, for which I apologize, I am posting the Romanian language file, updated and functionally verified, for ver. 1.45m
@ Yuriy_K
the makefile is from 1.13k; output to lcd is the same as before. Sorry, which caps in parallel to which 10µF?
@ Dumidan
The quarz was changed a long time ago to 16 MHz.
Sorry, which caps in parallel to which 10µF?
In your diagram, there are two of them (indicated by red arrows) soldered to them in parallel. At the end of the previous page, I wrote that when installing version 1.13k, it is necessary to adjust the electronic components. it is necessary to replace the 7550 and select the capacitances of these capacitors. This is due to the large current at a frequency of 16 MHz (against 8 MHz) and a slightly higher load on the processor due to color switching commands.
Judging by the fact that "?" is displayed correctly, the software is installed correctly and it is only necessary to restore the normal operation of the TTester by hardware.
I compared the files, everything needs to be corrected "hardware".
The 7550 is replaced by mcp1702 long time ago also changed TL431 to LT1004; caps of 100n in parallel to both electrolyt caps are already installed during construction; 10µf cap near quarz changed by one with ESR > 3 ohm - I had only one with 10µF and low height. The result is displayed on cld the same as before.
Your situation is similar to the GM328A with the BGR gamut, where the display does not allow you to transfer the TTester to 16MHz. The attached picture shows an image with such a quartz. Try to choose the parameters of the stabilizer and blocking capacities yourself. Try replacing the 1702 with a more powerful LM1117 and increasing the blocking capacitances to 47uF. There is nothing more I can do to help.
Hello Yuriy_K
I had today more time and tested both versions ST7735_BGR from page 277 and got an output. But this is mirrored and stands on head. After test routine is indicaded a value of about 60 pF without any component in test socket(without adjust).
If your screen is horizontally and/or vertically flipped, and if you build your own firmware, you can define the screen orientation according to your needs. For instance, for the m-firmware:
//#define LCD_FLIP_X /* enable horizontal flip */
#define LCD_FLIP_Y /* enable vertical flip */
#define LCD_ROTATE /* switch X and Y (rotate by 90°) */
But this is mirrored and stands on head. After test routine is indicaded a value of about 60 pF without any component in test socket(without adjust).
Hello Rober
On the current page there is a picture with the results of selftesting. Compare with your results and show the difference. The RGB and BGR firmware differ only in the preset colors for the displayed elements. In all other respects they are the same. Decide which firmware works for you and send the picture to flip and mirror it no problem.
Hello Yuriy_K
it's a little bit difficult to make the selftest with mirrored presentation. I dont know, where in the program I can do changes to flip vertically and horizontally lcd. In firmware of madris it is in config.h , but in k-version it is not descripted and I don't have your version of firmware to do any changes in h- or c-files. After then, with a normal presentation I'll do the selftest
In order to change the screen orientation of a GM328: on the the m-firmware you use the config_328.h file, on the k-firmware you use the makefile of the mega328_GM328 folder. Then you "make" the firmware and test.
After then, with a normal presentation I'll do the selftest
See the firmware in your personal mail.