$20 LCR ESR Transistor checker project

[LinuxHata]

the red pcb one with color screen.
known as "GM328"

This one: https://iduino.co.in/product/gm328-transistor-tester-signal-generator/

[Feliciano]

1. Is there a modification of firmware which just not power offs on each run and you do not need to check all this vbat and other things on each measurement?
2. Is there a larger and brighter display, which can be hooked to the color model? (I don't mind soldering an adapter, if there's no direct pin to pin compatible screen)

1. For m-firmware, you can disable ui_autohold to set continous mode. For k-firmware you can disable dpower_off (longer explanations on the config files)
2. The LCD controllers are limited. Probably you could find one compatible with the OSHW firmware that suit your likings, but for "larger and brighter" you should tell what we are comparing against to

[LinuxHata]

Thanks, I'm not familiar with these firmware versions, where I can read more about it?

Regarding the display, controller model is known, I guess, all I need is to find display with same controller and with same number of pixels XY, but larger physically and more brighter.

[LinuxHata]

I found this:

https://github.com/markus-seidl/component-tester?tab=readme-ov-file

There is mention of various LCD controllers, and I already have ST7920 display which is mentioned there which is big, bright and high contrast

But for firmware download it refers to already dead webpage.

[madires]

The mikrocontroller.net's SVN was shut down. Please see https://github.com/kubi48/TransistorTester-source for the current k-firmware (try the compiled firmware for mega328_color_kit) or https://github.com/madires/Transistortester-Warehouse/tree/master/Firmware/m-firmware for the m-firmware (just source code).

[Feliciano]

I'm not familiar with these firmware versions, where I can read more about it?

On the madires' github, under documentation, you can find the complete manuals of m-firmware and k-firmware (asuming your clone is compatible with those), or if you want the short versions, me and indman made some FAQs.

[Yuriy_K]

The use of color displays limits the functionality of Ttesters. Chinese manufacturers have monochrome ST7565 with a black background. I assembled a test bench to check the working one, on which I got the maximum number of measurable capabilities. I added an additional 3-volt battery to check IGBT transistors, added a microcircuit to check quartz and ceramic resonators. I gave the diagram earlier. Here are some examples of measurements and display settings.

[madires]

Yes, with a limited flash size there's a trade-off between functionality and fancy UI.

[Cosmic_Starlight]

Hello and good day/afternoon/evening to you all.

so ive bought a DIY kit of the clone of this LCR tester i believe is what is known as a AY/AT one (red solder mask)
and unfortunately it started with the wrong foot for starters it was missing 6 resistors in total and one capacitor (labeled 22)

anyway i managed to source the replacements parts locally and matched them to within 1-3% tolerance (i am fully aware you are supposed to use 0.1% ones but those arent sold here) and what i care about now is to get it working at least i can always upgrade them down the line

so i put the replacements in and unfortunately all the device does is light up the screen solid white and turns off (not even the power led lights up)
if i hold the encoder pressed the screen stays lit up solid white but still no power led and as soon as i release it it turns off

i would like to know if i could potentially diagnose whats wrong with it, all i have is a dmm, i could potentially buy something to flash the atmega again if its a case of the wrong firmware being flashed (as long as the flasher isnt too expensive keep in mind this is south america so everything is atleast 2x 3x the price down here)

i would love to be able to diagnose it and find whats wrong with it but i dont know where to even start, before i even attempted to power it on i made sure with the multimeter in continuity mode that it wouldnt beep between any two pads on any transistor

(about the attached image, never mind the missing components the replacements are already soldered in)

any help where to start checking stuff with a dmm is greatly appreciated

[madires]

Welcome to the forum! When you press the encoder's push button the ATmega's PD6 pin should go high (5 V) to switch T1. If there's no high signal at PD6 the ATmega could be unprogrammed. In case PD6 goes high check the circuit around T1 for any issues. Also check the polarity of the LED.

[snapper]

Only for those would 0.1% make sense.
Are the two 22 picofarads for the quartz correct?

[Cosmic_Starlight]

Welcome to the forum! When you press the encoder's push button the ATmega's PD6 pin should go high (5 V) to switch T1. If there's no high signal at PD6 the ATmega could be unprogrammed. In case PD6 goes high check the circuit around T1 for any issues. Also check the polarity of the LED.

thank you for the warm welcome!

i will check that hopefully my multimeter is fast enough to read it, if not would holding the encoder pressed suffice ?

Only for those would 0.1% make sense.
Are the two 22 picofarads for the quartz correct?

yes one of them is what i believe is a tantalum it has a number 22 on it i assume its picofarads, the other missing one that i had to source locally its also labeled 22 so assume its 22 picofarads too but this one isnt a tantalum but its one of those round circular ones that look like a lentil should work the same right ?

[snapper]

Keramik-Kondensator, 22 pF
Tantalum capacitors are usually in the µF range and are out of place here !

[Cosmic_Starlight]

Since I am still a beginner I don't know what they are that's what the guy at the store told me when I showed him the extras that came with my kit and asked for one like them but that said "22" so maybe he was wrong ?

The ones that came with it are the exact same ones you see on YouTube videos of the diy kit

I've attached a pic of one of the extras that came with the kit and one that looks pretty similar if not the same as the one I used as a replacement

[Cosmic_Starlight]

Don't know what happened, I was about to start measuring pd6 as suggested removed the screen and plugged the battery in and as soon as I did it the led turned on, unplugged it added screen back in same deal turns on automatically and screen is solid white doesnt turn off or anything, atmega removed same behavior also I can very faintly hear and sometimes at random something that sounds kind of like when you're frying something on oil but is very faint and seemingly at random

I've attached a photo of its back where the leads were soldered

[snapper]

https://www.google.com/search?q=Marking+of+ceramic+capacitors

[madires]

Pin PD6 drives an NPN which switches the LED and the PNP used as power switch. Since the LED is lit as soon as power is provided, despite the ATmega being removed, the NPN could have a short.

[Feliciano]

Things to check:
A shorted Capacitor (it happened to me once) or a transistor or the protection and/or MCU. Excesive current consumption. A component placed in the wrong position and/or orientation. And maybe a bad solder (specially if you hear faint sparks).

For reference on where to test with a DMM, I attach an schematic of a similar unit.

[snapper]

if so, then please provide the correct circuit diagram !!!

[Feliciano]

I said "similar" as reference.

Is there need to be aggresive?

[snapper]

This has nothing to do with aggression, but why should you make troubleshooting unnecessarily difficult for a beginner by using an unsuitable circuit diagram?

[banane]

Hello everyone  ,

I have purchased an LCR-T7 which I can also find this device in the table : https://github.com/madires/Transistortester-Warehouse/blob/master/Documentation/English/Clone-Comparison-Chart.pdf
I have attached a few pictures. There is no label on the MCU, probably rubbed off / etched away. 

I don't see a 6pin J3 connector here, i.e. I have to make do with the Chinese firmware which is pre-installed, right ?

Thank you and best regards
Christian

[Cosmic_Starlight]

Things to check:
A shorted Capacitor (it happened to me once) or a transistor or the protection and/or MCU. Excesive current consumption. A component placed in the wrong position and/or orientation. And maybe a bad solder (specially if you hear faint sparks).

For reference on where to test with a DMM, I attach an schematic of a similar unit.

Pin PD6 drives an NPN which switches the LED and the PNP used as power switch. Since the LED is lit as soon as power is provided, despite the ATmega being removed, the NPN could have a short.

Sorry for the delay In replying didn't sleep too well last night and didn't feel too well waking up

Am gonna try to be as detailed as possible within my limited knowledge

So the first thing I did was to clean the top of the PCB with 70% alcohol and a toothbrush since I saw that some of the soldering paste I used had managed to seep into the other side and dried it with a hair dryer

Then I went and reflowed all the transistors again, and now the issue of the led lighting up as soon as battery was hooked up was gone

Then I checked the power draw with the dmm in series with the battery and he connector and saw 0 on power consumption

so I decided to press the encoder to see what it would do and surprisingly this time the led turned on and it stayed on (this was without atmega or screen in)

Put the atmega and screen in, and it it turned on sort of. The led lit up and stayed on and the screen was solid white but most importantly it stayed on, pressing the encoder the led flashed off, holding it down turned the led off but he unit wouldnt shut off the only was was to remove the battery

So I decided to reflow all the solder again with the iron, no improvements but either the weird fizzling/cracking noise came back or it was never gone.

Power draw with battery in bit turned off is now 0.10-0.12 mA (no atmega or screen in) turned on with atmega in but no screen is about 0.46 mA

And about the dp6 pin on the atmega (am gonna explain how I measured it don't know if it was the wrong way or not) I put the negative lead on the battery negative pad and the positive lead on the dp6 pin then I turned it on and I saw the voltage rise to around to a little over 3 volts  then when it was on settle around 1.6-1.7 volts
(Keep in mind me multimeter update freq is around 3 times a second so maybe it isn't fast enough to see 6?

I've attached a few close ups of it, don't think any component is in the wrong orientation but you never know

I would attach more but the 8mb total limit isn't letting me

[Cosmic_Starlight]

man this thing almost makes me want to pull my hair out, so after i decided to desolder one of the S9012 (T3) to check it manually with multimeter and it appeared to be good when i soldered it back in and decided to test it again i was back to square one (led turns on as soon as battery is connected)

so i just said screw this and desoldered also the other S9012(T1) along with S9014 (T2) and 7550(IC2) didnt desolder IC1 because it was getting pretty late and i was getting tired

anyways since i had both T1 and T3 out i checked them with a multimeter and while initially both initially both appeared fine (reverse polarity didnt work and the value shown on the dmm on diode mode was the in the range of around 680 on both)

but one of them if i put the negative lead on the collector (its pin 3 right ? the rightmost one when the flat part is towards you) and the positive on the emiter it will show continuity or it was the other way around (positive on collector and negative on emiter) this was T1

and T3 also seemed fine but i if i put the positive on base and negative on colector it read fine but if i quickly moved the positive lead from base to emiter it briefly shoed continuity for like half or a quarter of a second  (is this correct or isnt that how transistors are supposed to work ? apply voltage to base, stuff comes out the emitter)

would love to be able to test both IC1 and IC2 but ive no idea how i think T2 tested fine

[madires]

I have attached a few pictures. There is no label on the MCU, probably rubbed off / etched away. 

I don't see a 6pin J3 connector here, i.e. I have to make do with the Chinese firmware which is pre-installed, right ?

Yes, the MCU is most likely an LGT8F328.