$20 LCR ESR Transistor checker project

[madires]

One more question, please. Is this the right repo? https://github.com/Mikrocontroller-net/transistortester
There hasn't been a commit there in one year

That's outdated (migration of mikrocontroller.net's SVN). Please use https://github.com/kubi48/TransistorTester-source!

[madires]

I received an LCR-TC2 from Aliexpress that I ordered in December 2022. The board appears to be updated. The boot screen says firmware is 3.1E and chip U3 seems to have been removed. No options for changing the power off delay (which is annoying short).

Haven't seen any schematics for that version yet, but without the additional control MCU it should be easier to upgrade to an OSHW firmware. And you could customize the user interface settings to your liking.

[dazz]

One more question, please. Is this the right repo? https://github.com/Mikrocontroller-net/transistortester
There hasn't been a commit there in one year

That's outdated (migration of mikrocontroller.net's SVN). Please use https://github.com/kubi48/TransistorTester-source!

Thanks. I was having issues with the calibration process, so I will use that instead 

[elecdonia]

…I was having issues with the calibration process, so I will use that instead.

With Transistor Testers that were previously in use, I go into the menu and select “Self Test” which works its way through several different screens, first with “probes shorted,” then with “probes isolated,” and finally with a film capacitor connected to pins 1-3. I use a 1.0uF capacitor for this. Don’t rush it by pressing the button to move on to the next screen. It will advance all by itself after each section finishes. I had to practice this several times before I was patient enough to wait and let it proceed at its own pace.

In particular it appears that capacitor ESR results are more accurate after performing this full menu-driven “self test” with the 1.0uF film capacitor connected to 1-3 when the display message calls for it.

I’ve also used the faster self-test with probes shorted before pressing power button. But this “fast self-test” by itself didn’t permanently shut off the “not calibrated” message. It was only after the longer self-test (from the menu) that the “not calibrated” message disappeared permanently.

I discovered this “not calibrated” issue with an unmodified “GM328 color-kit” tester running the original Chinese factory firmware it came with. This firmware displays “1.12k” but I suspect the Chinese modified it. Unfortunately its ATmega328P is locked.

Among several transistor testers in my workshop I use this one the most. I’ve had only one issue with it: I accidentally toasted its SRV05-4 overvoltage protector with a charged capacitor. The tester worked well again after removing the failed SRV05-4 except it began showing the “not calibrated” message. I observed no damage to the ATmega328P or any other parts. But I did need to run the full menu-driven self test after that incident in order to clear the “not calibrated” condition.

[elecdonia]

I received an LCR-TC2 from Aliexpress that I ordered in December 2022. The board appears to be updated. The boot screen says firmware is 3.1E and chip U3 seems to have been removed. No options for changing the power off delay (which is annoying short).

Haven't seen any schematics for that version yet, but without the additional control MCU it should be easier to upgrade to an OSHW firmware. And you could customize the user interface settings to your liking.

I’ve been trying to get my hands on one of these for 2 months. The first unit I received, labeled LCR-TC1, contained APT32F172K8T6 MCU. The 2nd unit, also labeled LCR-TC1, contained LGT8F328P MCU. I’m now awaiting unit #3, said to be model LCR-TC2.

My goal is to acquire a unit containing Atmel ATmega324 which I will then upgrade to ATmega644 along with adding the paddle switch on the side for menu access.

[elecdonia]

Note: 1. For all fuses “1” means unprogrammed while “0” means programmed.

With it programmed with 0xFF it sets all bits to "1" meaning unprogrammed, this means that the oscillator has a limited frequency range of 8 to 16MHz and startup mode: "slowly rising power" this is a low power operation mode for the crystal oscillator.
With it programmed with 0xF7 instead the CKSEL3 bit is progammed and this means that the crystal oscillator has a frequency range of 0.4 to 20MHz with the same starup mode as before, this is a higher power mode for the crystal oscillator.

I have an unmodified GM328A+ Transistor Tester, factory original v1.12k firmware, 8MHz, which fails to initialize its ST7565 monochrome LCD at startup when the room temperature is cold (10-15 C). But it starts up perfectly if I warm up the ATmega328 by holding my thumb over it for 20 seconds. This tester always works perfectly at normal room temperature. I plan to check the fuse bits to learn which oscillator settings are in it. Perhaps “low-power oscillator” mode starts up too slowly at very cold room temperature?

I own several Transistor Testers. The others work well at low temperatures. I’ll post what I find when I check this GM328A+ unit. I also intend to update it to current v1.13k.

[Feliciano]

IIRC "low power" xristal oscillator runs at 3.3V, and the "normal" or "full swing"are 5V. The ctesters I've seen have used normal oscillators, so I set the fuses accordingly.

[elecdonia]

Hello. Got 2 LCR Testers today from AliXpress..  New LCR-T4 and LCR-T7 Color. Both have new firmware version 3.12K 
Does anybody know where to find this firmware.. or is it a hoax..  mean Chinese Variants.. ? tnx for info JP

Would it be possible for forum members who recently purchased LCR- series Transistor Testers to post photos of their PC boards?

I’m asking because the Chinese manufacturers are now using 3 different MCU chips in current production LCR series Transistor Testers. Only one of these MCU chips is compatible with existing open-source -k or -m software. The other two MCU chips can use only proprietary software created by their Chinese manufacturers. That software cannot be updated or replaced by open-source Transistor Tester software.

At this time we can only determine which MCU is inside an LCR series Transistor tester by viewing a photo of the PC board. The model number, which could be LCR-TC1, LCR-TC2, LCR-T4, or LCR-T7, reveals nothing about what is inside. Often the vendors don’t even know what they are selling. They don’t post photos of the PC boards either.

Forum user indman maintains a table of clone versions of Transistor Tester hardware here:
     https://yadi.sk/d/yW8xa5NJgUo5z
The file name is TableClonesEn.pdf (or TableClonesRu.pdf)
However the large number of different Transistor Testers recently manufactured in China is on the verge of becoming a tsunami of randomness. Well-known model names such as LCR-TC1 or LCR-T4 are being applied to several entirely different hardware/software combinations. Some are incapable of running open-source -k and -m software. But the vendors don’t disclose this.

I recommend we should all pitch in and assist with separating the good and bad from the ugly. Photos of the PC boards will be a big help,

[MrSqueaky]

I received an LCR-TC2 from Aliexpress that I ordered in December 2022. The board appears to be updated. The boot screen says firmware is 3.1E and chip U3 seems to have been removed. No options for changing the power off delay (which is annoying short).

Haven't seen any schematics for that version yet, but without the additional control MCU it should be easier to upgrade to an OSHW firmware. And you could customize the user interface settings to your liking.

I've made a schematic the for LCR-TC2 T7-PLUS v1.2 I received. I also traced the tracks on the PCB in case I've made a mistake somewhere in the schematic. Ordered from this seller https://www.aliexpress.com/item/1005004717377243.html . I soldered on the battery connector. Hope this helps.

[madires]

IIRC "low power" xristal oscillator runs at 3.3V, and the "normal" or "full swing"are 5V. The ctesters I've seen have used normal oscillators, so I set the fuses accordingly.

low power oscillator:
- reduced voltage swing (no voltage given)
- can't drive other clock inputs
- more susceptible to noise
- low power consumption

full swing oscillator:
- rail-to-rail swing
- can drive other clock inputs
- less susceptible to noise
- power consumption higher than for low power oscillator

[madires]

[I've made a schematic the for LCR-TC2 T7-PLUS v1.2 I received. I also traced the tracks on the PCB in case I've made a mistake somewhere in the schematic.

Matches nearly the pin assignment of the TC1 (just without the additional control MCU). PD1 and PD2 are swapped (push button and power control) and the display pinout could be:

Code: [Select]

/RES   PB4
D/C    PB5
SCL    PB7
SDA    PB6


[indman]

PD1 and PD2 are swapped (push button and power control) and the display pinout could be:

Madires, for this clone variant your m-firmware as well as k-firmware(if ATMega644) is suitable.You only need to swap PD1 and PD2 in places. This has already been tested.

[madires]

Great! I'll add that one to the list of clone settings (Clones file).

[indman]

Madires,I am afraid that your Clones file may increase in size in an incredible way, given the number of different modifications that the Chinese industry produces!

[elecdonia]

Madires,I am afraid that your Clones file may increase in size in an incredible way, given the number of different modifications that the Chinese industry produces!

Wow! It will require much time and effort to categorize so many different versions.

          At first glance I suspect all 6 PC boards in these photos have either
          Atmel ATmega324 or ATmega644 MCU for the following reasons:
1) All MCU seen in these photos have 44 pins.
2) Also all of these PC boards have a place to mount the 6-pin Atmel ISP programming header. This 6-pin ISP interface is unique to Atmel products. To the best of my knowledge it isn’t used by any other MCU manufacturer.

          In contrast:
1) All non-Atmel MCU I’ve observed in Transistor Testers have only 32 pins.
2) In addition the non-Atmel Transistor Testers do not have the Atmel 6-pin ISP header. Instead they have other types of programming interface, such as SWD (single wire debug) or CH340X USB IC

[indman]

elecdonia, I wanted to show exactly the varieties of seemingly the same topology of a printed circuit board with ATMEL MCU. To this gallery, you can add a variant with an ATMEL MCU in a DIP case.As for the variants with non-ATMEL MCU, they will make the same large gallery. 

[madires]

Madires,I am afraid that your Clones file may increase in size in an incredible way, given the number of different modifications that the Chinese industry produces!

Indeed! However, one has to do what one has to do.

[elecdonia]

Details about MCU chips found in recent production Transistor Testers:

Atmel ATmega324 has been used in Transistor Testers for several years, but recently had a big price increase ($5-15 USD) and became harder to get. This MCU has 44 pins (11 pins on each of the 4 sides). ATmega324 is supported by existing -m (Markus) Transistor Tester software. However, ATmega324 software is not available in -k (Karl-Heinz) version.

Atmel ATmega644 has same 44-pin package as ATmega324 and can be used with same PC board layout. ATmega644 has 64k flash memory compared to 32k for ATmega324. Therefore ATmega644 accommodates larger software with additional testing/measurement capability, higher resolution display fonts, and multiple user-selected languages. ATmega644 is supported by both -k and -m Transistor Tester software versions.

LGT8F328P - First used by Transistor Tester manufacturers in 2021. Price: < $1 USD. 32-pin package (8 pins on each of the 4 sides). This is same package as Atmel ATmega328. It is said to be “similar” to Atmel ATmega328. However its pinout is different and it has several additional features. For example the LGT8F328P can operate at 32MHz. The LGT8F328P has been ported into the Arduino IDE. Clone versions of Arduino Pro and Nano boards containing LGT8F328P are available. But no open-source Transistor Tester software for LGT8F328P has been officially released yet. However, one Transistor Tester enthusiast has a GitHub project to adapt the -m software to LGT8F328P:
     https://github.com/DurandA/transistor-tester-lgt328p

APT32F172K8T6 - First used by Transistor Tester manufacturers in 2022. Also < $1 USD. 32-pin package (8 pins on each of the 4 sides). Pinout, internal architecture, and instruction set are totally different than Atmel ATmega328. Only similarity is 32-pin package. Very few development tools are available for compiling and debugging APT32F172K8T6 software. Existing open-source Transistor Tester software does not support this MCU. Warning: Unfortunately some Transistor Tester manufacturers are deliberately mis-labeling the APT32F172K8T6 as Atmel ATmega328P.
 
Conclusion:
ATmega324 and ATmega644 are the only 44-pin MCU chips found in Transistor Testers.
Therefore a quick look at the PC board will distinguish these MCU chips from the other two.

[indman]

Atmel ATmega324 has been used in Transistor Testers for several years, but recently had a big price increase ($5-15 USD) and became harder to get. This MCU is in a 4-sided TQFP package with 44 pins, 11 pins on each of the 4 sides. This MCU is fully compatible with existing -k and -m Transistor Tester software.

No.There is an m324p MCU in the k-firmware makefile, but the firmware compilation ends with an error.That is,in fact, ATmega324 is not yet supported by Karl-Heinz.

[madires]

Could you please post the error message? I remember that my avr-gcc package came with a broken iom324p.h.

[indman]

madires,Karl-Heinz has not been answering my questions lately. Check out its current documentation - ATMega324 is not listed in the list of supported MCUs. I provide a screenshot of the error

[elecdonia]

No.There is an m324p MCU in the k-firmware makefile, but the firmware compilation ends with an error.That is,in fact, ATmega324 is not yet supported by Karl-Heinz.

Thank you for correcting me. I have just confirmed this:
          https://github.com/kubi48/TransistorTester-source
contains compiled ready-to-load -k firmware for ATmega644 only. But nothing for ATmega324. I will edit my post.

[indman]

Is there any functional -k software for ATmega644 at this time?

I didn't quite understand your question?For ATmega644, there have been working k-firmware for LCR-TC1(T7),Hiland644,LCD2004 for a long time

[madires]

I provide a screenshot of the error

The MCU type is missing. Edit setup.mk, copy the section for m328p, change 'ifeq ($(PARTNO),m328p)' to 'ifeq ($(PARTNO),m324p)', and 'MCU = atmega328p' to 'MCU = atmega324p'.

[elecdonia]

Is there any functional -k software for ATmega644 at this time?

I didn't quite understand your question?For ATmega644, there have been working k-firmware for LCR-TC1(T7),Hiland644,LCD2004 for a long time

Karl-Heinz firmware for LCR-TC1 with ATmega644 MCU is in Yandex repository:
     https://disk.yandex.ru/d/yW8xa5NJgUo5z/LCR-%D0%A2%D0%A11/Firmware/Karl-Heinz%20K%C3%BCbbeler/English/16Mhz

Yandex repository also mentions this recently modified Karl-Heinz ATmega644 firmware for LCR-TC1, -TC2 -T7:
     https://disk.yandex.ru/d/yW8xa5NJgUo5z/LCR-TC2/Firmware/Karl-Heinz%20Kubeller/English/FirmwareModNew
Additional details in forum post #8001:
      https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg4578967/?topicseen#msg4578967

Please confirm. I am hoping to obtain LCR-TC2 containing ATmega644 soon. Or if I get one with ATmega324 I will replace MCU with ATmega644. Then I will install this firmware described in forum post #8001