$20 LCR ESR Transistor checker project

[anwfeb]

yes this is just a clone, and i just want to try to fix if it's possible..
testing the test button in cold conditions (when the clone hasn't been turned on at all in a few hours) the results are accurate, but when the clone has been turned on in a few minutes the test results change, even testing 0.22 ohms will get a value of 0.26 ohms that's a big difference in clone conditions has been turned on in a few minutes..
so I have to wait for the clone condition to be turned off in a few hours then the first test of the start button is usually an accurate measurement value of 0.22 ohm the results are very accurate.. different from when the clone has turned on the second and then third test even in the next few minutes the measurement value starts to change..

[indman]

What type of resistor are you using? Try to measure another highly stable type of resistance. Preferably without internal inductance, that is, non-wire.

[anwfeb]

I use this resistor, but when I try to jumper pins directly pins 1 and 3 only with solder wire the results also changed for example 0.01 - 0.03 - 0.04

[indman]

Once again, try to measure another type of precision resistor. And you can also try to solder high-quality 100nF filter capacitors directly to the power pins of the processor. Also, 431 sources in the presence of a high-quality LDO + 5V can worsen the result.

[anwfeb]

Once again, try to measure another type of precision resistor.
I have tried with other resistors that are more precise and the results are the same, even when I connect the jumper pin for selftest only on pins 1 and 3, the result can't be exactly 0.00 ohms, it always changes 0.01 - 0.03 - 0.04

And you can also try to solder high-quality 100nF filter capacitors directly to the power pins of the processor.
does that mean I solder directly into pins 5, 17, 27, 38 with 100nf high quality, is it enough with 1pcs 100nf on one of the vcc pins 5v to GND, or 4pcs 100nf for pins 5, 17, 27, 38 to GND

[indman]

I understand that you are currently using k-firmware? Try m-firmware and compare the results.

I solder directly into pins 5, 17, 27, 38 with 100nf high quality, is it enough with 1pcs 100nf on one of the vcc pins 5v to GND, or 4pcs 100nf for pins 5, 17, 27, 38 to GND

To check the effect, 2 capacitances between 5-6 and 17-18 pins are enough

[anwfeb]

To check the effect, 2 capacitances between 5-6 and 17-18 pins are enough
okay i will try

do you have the latest (m) version? i already have your previous file(m) from yandex disk

[indman]

Use the working version from the Yandex disk. I have not yet collected a more recent version for this clone.

[anwfeb]

okay
i have tested LCR-TC1 1.43m, test results with monitor RCL values ​​always change 0.25, 0.24, 0.22, 0.26, 0.28, 0.21, 0.29 to 0.31 i have calibrate (selftest and adjustment) and results in monitor RCL same as version(k) results are variable

[Feliciano]

So now for the table of clones we have at least 4 "T7" variants:

• The one with a 324, transformer, and an STC
• The one with a 644
• The one with a 324, with a non-STC power controller
• The one with a 328(?), without transformer and power controller

Attached to the (new/old) T7 images, the circuit doesn't look the same as the previous ones with a step-up transformer.
firmware v3.12k?

Is there another way to differentiate this from the others aside the "v3.12k" reference?
BTW: you can look at the atmegas's date code to guess whether this is a newer or older version.

[indman]

Feliciano,all the most interesting is yet to come, since the TC1 clone also has a modification based on the ATMega328 in a DIP package 

[Yuriy_K]

i have tested LCR-TC1 1.43m, test results with monitor RCL values ​​always change 0.25, 0.24, 0.22, 0.26, 0.28, 0.21, 0.29 to 0.31 i have calibrate (selftest and adjustment) and results in monitor RCL same as version(k) results are variable

Nobody canceled the change in readings when the microprocessor warmed up. If you use the TTester for single measurements, then Selftest must be performed on a "cold" TTester. If you need to take multiple measurements, then the Selftest results should be saved after 4-5 cyclic repetitions. Saving the results of the "warmed up" TTester. Check and compare measurements after such a method.

[anwfeb]

hi Yuriy_K
after I noticed when the clone was in "cold" conditions with "warm" conditions the difference result in measurements still continued to occur, but now I prefer selftest when the clone has been used "warm". the measurement changes are more stable than when "cold". I still use the firmware color that you made, this is specifically for my clone because i have replaced the screen with a different offset, thank you very much for your help making the firmware..

if you don't mind I want to try with the rotary encoder version, can you make a firmware for me? your firmware works fine for my clone..

[Feliciano]

Feliciano,all the most interesting is yet to come, since the TC1 clone also has a modification based on the ATMega328 in a DIP package 

The bright side is it will be easier to repair.

[anwfeb]

Use the working version from the Yandex disk. I have not yet collected a more recent version for this clone.

Hi.. indman
did you change to the source file? i tried compiling your makefile with change source files CFLAGS += -DWITH_ROTARY_SWITCH=2 but it doesn't work for my clone TC1.

[indman]

anwfeb,yes,I explained everything in a text file ReadmeModEn.txt

Several variants of the modified m-firmware 1.45m for clone GM328A(AY-AT)! You can read all the details ReadmeModEn.txt file in the attached archive!

[anwfeb]

yes.. i have read it and i have adjusted the encoder PIN for my clone. Thank you for the information

[indman]

Several variants of the modified m-firmware 1.45m for clone GM328A(AY-AT)!

Sorry, I have missed a very important nuance when creating the firmware. For the correct work of the Mod1 firmware a little hardware modification of the display module and connection of Buzzer to the PD5 processor port is needed. Without these modifications the firmware will not work. So I added another basic Mod configuration in which the "Continuity" function disabled. This firmware variant, as well as the Mod2 variant, will work on your device without any hardware modifications. I've updated the archive with firmware and description!

[de_light]

Probably been asked before, probably by me

What does it mean when testing a cap esr when it reads 0.00ohms?surely that can't be right.

[indman]

Madires already answered that question of yours here before
https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg3665140/#msg3665140

[madires]

For a nice Easter weekend:
v1.46m
- Added option to disable text based pinout for 3-pin semiconductors when component symbols are enabled (UI_NO_TEXTPINOUT, suggested by bffargo@EEVblog).
- Pin designators for MOSFET's intrinsic diode, IGBT's flyback diode, BJT's flyback diode, BJT's base-emitter resistor and Schottky-clamped BJT's base-collector diode are now also color coded.
- Fixed output of symbol pinout for symmetrical JFETs (shows now 'x' for drain and source).
- Reworked management and output of pin designators for 3-pin semiconductors.
- Updated Romanian texts (thanks to Dumidan@EEVblog).
- Updated font font_8x12t_win1251_vfp.h (thanks to indman@EEVblog).
- Revised component symbols. Removed cycles and added pin designators (suggested by indman@EEVblog). Old symbols can be still used (set SYMBOLS_<size>_OLD_<format>.h in config_<MCU>.h).
- Display probe/testpin IDs with reversed colors (UI_PROBE_REVERSED, suggested by Feliciano@EEVblog).
- Added numbers 1-3 with reversed color to several fonts (thanks to Feliciano@EEVblog).
- Updated Russian texts (thanks to indman@EEVblog).
- Updated Spanisch texts (thanks to pepe10000@EEVblog).
- Updated Romanian texts (thanks to Dumidan@EEVblog).
- Updated Polish texts #2 (thanks to Jacon@EEVblog).

Please download at:
- https://github.com/madires/Transistortester-Warehouse/tree/master/Firmware/m-firmware
- https://github.com/kubi48/TransistorTester-source/tree/master/Markus

[pepe10000]

Hello.

I have tried the new version, but I have noticed something strange.

If I use the English language and the default font FONT_10X16_HF that I have always used, the letters on the screen come out wrong:



I have tried to use the Spanish language with the same font FONT_10X16_HF and the result is similar, they cannot be read.

But, if I use the font FONT_10X16_ISO8859_2_HF, the letters come out fine, both in English:



Like in Spanish:



The problem is that this font FONT_10X16_ISO8859_2_HF takes up more space and forces me to disable some features so that the firmware does not exceed the available space.

This is normal?.

Would it be possible to go back to using the standard font FONT_10X16_HF which takes up less memory?

Thank you and greetings.

[madires]

The characters are shifted by one. Let's check the font... Found it. Please edit font_10x16_hf.h and insert after the bitmap for '(':

Code: [Select]

  0x00,0x00,0x06,0x00,0x18,0x00,0x30,0x00,0x20,0x00,0x40,0x00,0x40,0x00,0x40,0x00,
  0x40,0x00,0x40,0x00,0x40,0x00,0x20,0x00,0x30,0x00,0x18,0x00,0x06,0x00,0x00,0x00,   /* 0x19 ) */

Somehow I managed to delete those two lines. 

Edit: The next character  (0x1a '*') is also affected. Please replace the bitmap data with:

Code: [Select]

  0x00,0x00,0x10,0x00,0x10,0x00,0xD6,0x00,0x6C,0x00,0x28,0x00,0x38,0x00,0x28,0x00,
  0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,   /* 0x1a * */


[pepe10000]

Hello.

Perfect, adding these two lines the screen can be read correctly again.

Tomorrow I will recompile everything again so I can activate the features that I had to remove for it to pick up the firmware so I can get the most out of it.

Thank you very much for everything and greetings.

[indman]

Madires,I updated the Russian manual in pdf format and also slightly corrected the Russian translation of var_russian.h