$20 LCR ESR Transistor checker project

[madires]

Doesn't seem to be a genuine ATmega328, possibly an APT32F172K8T6.

[indman]

Anyone see this board?

Yeah, a lot of people have seen it. You bought a disposable device. There is no free firmware for this microcontroller!
Only from the manufacturer of these clones.

[golub2017]

Here is the better picture, chip is Atmega 328P, could I backtrace pins from mcu to points on board. For example MOSI in on pin 16 on MCU, can I measure with DM where is going?

[indman]

Here is the better picture, chip is Atmega 328P, could I backtrace pins from mcu to points on board. For example MOSI in on pin 16 on MCU, can I measure with DM where is going?

On your board is not the Atmega328P but the chip APT32F172K8T6! The Atmega328P labeling is a fake!

[Firesledge]

Hello,

My tester (an M328Kit+TFT) worked for several years but now displays a completely white page on power on and stays like this. Then the devices shuts down after about 10 seconds. It seems it’s a display-related issue, because if I put a LED in the socket, it flashes briefly multiple times a few seconds after power up, indicating a test of the component. Same behavior whatever the power supply (battery or adapter). So I have a few questions:
– How to make sure that the display module is faulty, and not the ports of the ATMega?
– What is the reference of the display module if I want to replace it, or how could I obtain this information?

Thanks, and I apology if I’m posting in the wrong forum, in this case please redirect me to a better location for troubleshooting.

[OLderDan]

Is there a workaround to measure low value inductors on the multi function tester-tc1 that show as a resistor? I have tried measuring the resonant frequency with a known value capacitor and calculating manually but I clearly ate some of the pi I was meant to be multiplying with and cannot seem to arrive at a sane value.
My device is stock and I am prepared to flash an alternative firmware if there is one that can help with low value inductors.

[oitar]

Hi guys, replaced the Atmega 324PA with a brand new 644PA(both 44 pin TQFP, full pin-compatible, right?). I cannot read the new controller's signature(AVRDDUDES). Is it because the new chips is expecting an external clock, or am I missing something else? I left everything as-is on the board when the AT324PA was working, and using the same interface as before.

[LeWidget]

Hallo. Just wondering if there's a known source for a LCR tester with a legit non-clone chip or are they all still a hit or miss?

[madires]

– How to make sure that the display module is faulty, and not the ports of the ATMega?

You could use a DSO or LA to check if the SPI bus signals are ok.

– What is the reference of the display module if I want to replace it, or how could I obtain this information?

It's an ST7735 based display.


BTW, welcome to the forum!

[madires]

Is there a workaround to measure low value inductors on the multi function tester-tc1 that show as a resistor? I have tried measuring the resonant frequency with a known value capacitor and calculating manually but I clearly ate some of the pi I was meant to be multiplying with and cannot seem to arrive at a sane value.
My device is stock and I am prepared to flash an alternative firmware if there is one that can help with low value inductors.

The measurement range of the normal inductance measurement starts around 10 µH. For lower values you can run the k-firmware with the SamplingADC method (if your tester has a genuine ATmega).

[Yuriy_K]

Is there a workaround to measure low value inductors

I don’t know the possibility of TC-1, but to measure small inductances you need to have a Ttester on which you can install k-firmware and 16 MHz quartz. Here are examples of such measurements on ATMega328...

[madires]

Hi guys, replaced the Atmega 324PA with a brand new 644PA(both 44 pin TQFP, full pin-compatible, right?). I cannot read the new controller's signature(AVRDDUDES). Is it because the new chips is expecting an external clock, or am I missing something else? I left everything as-is on the board when the AT324PA was working, and using the same interface as before.

Please try a lower ISP speed ( avrdude's -B option). The factory default fuse settings let the ATmega run with the internal RC oscillator and the 8:1 prescaler enabled, resulting in a 1 MHz clock.

[madires]

Hallo. Just wondering if there's a known source for a LCR tester with a legit non-clone chip or are they all still a hit or miss?

It's still a lottery. Your best bet is to look for a DIP ATmega.

[Firesledge]

You could use a DSO or LA to check if the SPI bus signals are ok.

Ah indeed, the DATA, RS and RST lines look fried while other signals are working. So I have to replace the ATMega.
Thanks for the advice!

[jm_araujo]

Couldn't resist the "T7" clone for less than 10€ on Aliexpress, the AT-AY I had has no case and requires 9V batteries, that are always flat when needed.
The "T7" has a nice case, internal Li-ion and charges via USB.  Didn't really trust the firmware it came with, and out came the screwdriver and soldering iron.

It had an 48 pin QFP IC in a blue PCB with no markings. After some searching, found "previous" versions of the "T7" had LGT8FX8P in 32 pin version. Found the datasheet and there is a version of the same IC in 48 pin, and all significant pins matched my PCB (power, cristal, programming).
Rather than trying to port the "transistortester" software to LGT8FX8P  that has some serious memory limitations for this project (no EEPROM) and some other differences, I compared the pinout to the  ATMEGA328P in QFP32 (which I had) and found I could make it work bending some pins and a couple of bodge wires (thankfully most of the extra pins in the 48pin package were not connected). I only had to cut two traces in the PCB (under the ATMEGA).
It is working fine with Markus 1.52 version at 16MHz

I don't recommend to anyone to go down the same path, but if someone wants some more information, feel free to ask.

[madires]

Nice job! In another thread gmc posted an image of a new TC-1, also with a QFP48L LGT8F328 (https://www.eevblog.com/forum/testgear/tc1-lcr-meter-transistor-tester-fix/msg5518339/#msg5518339). I guess we'll see more 48 pin LGT8F328s.

[Twiggy]

Hi

I got the hiland m644 with madires firmware 1.42m at the moment.

Now i'm learning that it's useful to measure capacitor ESR for repair purposes, and also that some ESR meters can do that while the cap is in circuit. Can I do that reliably enough for repair purposes with the m644 with a more recent firmware? I tried reading some random in circuit caps from a power supply but it wasn't really reliable (different component identified every time)

I figured I would ask before purchasing a mesr-100 or shannon tweezers.

cheers and thanks for the thread & firmware.

[Feliciano]

If you apply 5V or so pulses to a Capacitor in-circuit, you'll end measuring the Cap plus the other components directly connected.
On the other hand, if you apply less than 0.6V pulses, you have chances of measuring more the Capacitor than the PN junctions usually connected to it.
It's my understanding that our CTester uses less than that when selecting the ESR 1--3 function.

And regarding the improvements on different versions, you can consult the release notes on madires' github.

[Twiggy]

the kind of pulse it uses is not stated in the README i have, however it is stated that the ESR tool can measure capacitors in-circuit (in §3.1.8 for the latest readme on the repo).
The 0.1v measurement seems to be a strong selling point of lcr tweezers such as st42.


I guess I need to play a bit more with it on different test boards.

[Feliciano]

Maybe the readme of m-firmware is not detailed enough on that regard. Still, you can look for "C+ESR@TP1:3" on the k-firmware manual.

Friendly reminder: please make sure you discharge a Capacitor before attempting measurements.

[indman]

I got the hiland m644 with madires firmware 1.42m at the moment.
Can I do that reliably enough for repair purposes with the m644 with a more recent firmware? I tried reading some random in circuit caps from a power supply but it wasn't really reliable (different component identified every time)

I have expressed many times earlier in this thread on the issue of measuring C/ESR in-circuit - this device is not designed for this operation for many reasons. If you are satisfied with the reliability of the test 50/50 you can use as a very simple probe. In other cases it is much more reliable and correct to use specialized devices for in-circuit measurements! For example, ST42 is designed for this purpose.

[madires]

If you apply 5V or so pulses to a Capacitor in-circuit, you'll end measuring the Cap plus the other components directly connected.
On the other hand, if you apply less than 0.6V pulses, you have chances of measuring more the Capacitor than the PN junctions usually connected to it.
It's my understanding that our CTester uses less than that when selecting the ESR 1--3 function.

The difference is that the ESR function (menu) doesn't probe anything else, just capacitance and ESR. If someone repairs a lot of stuff it's better to get a dedicated ESR meter or LCR meter with a low voltage AC signal for in-circuit testing (has still some limits).

[Feliciano]

I think the in-circuit measurements can only be used as a preliminary test, and in case of doubt it's better to remove the DUT for a proper assesment, or indeed to have an specialized equipment for the task.

But what about the approach of the k-firmware?

C+ESR@TP1:3 The additional function C+ESR@TP1:3 selects a stand-alone capacity measurement with ESR (Equivalent Series Resistance) measurement at the test pins TP1 and TP3. Capacities from 2uF up to 50mF can be measured. Because the measurement voltage is only about 300mV, in most cases the capacitor can be measured in circuit without previous disassembling.

Does the m-firmware follows a different approach (i.e. voltage)? Because with 0.3V we would be measuring mostly the C and the possible R (physical or semiconductors' leakage) around it.

[madires]

No, it's the same - more or less. BTW, 0.3 V is the maximum voltage while checking large caps, i.e. the measurement stops when the cap reaches 0.3V (capacitance measurement). But the charging pulses can be theoretically up to Vcc for very small capacitances (pulses are Vcc plus a resistor to limit the current). ESR is a different story and uses very short pulses while also changing the polarity. The voltage depends on the capacitance and the ESR (higher capacitance -> lower voltage, higher ESR -> higher voltage). The resulting voltage should be quite low. However, a semiconductor in parallel might react to the pulses, as the voltage isn't regulated by the tester.

[Feliciano]

I see, so I think the wording of que quoted paragraph can lead to this misunderstanding. Maybe it could be rephrased like:
"Because the measurement stops when voltage reaches about 300mV, in most cases the capacitor can be measured in circuit without previous disassembling."
And I'm not sure whether replacing "most" with "many".

Just my two cents on the matter. I leave it to you guys.