$20 LCR ESR Transistor checker project

[indman]

As elecdonia recently mentioned, there are at least 3 kinds of MCU for this series of ctester. If the pinout is valid for ATmega, you could swap the damaged MCU, but you would also need to replace the power controller MCU with a circuit previously described in this thread, and program the new ATmega with the m-firmware.

This is not the case, you are slightly mistaken here.I do not see any difficulties and obstacles in order to bring this clone back to life and normal operation. LGT8F328 is 99% similar to ATmega328 in terms of the location of contacts. Therefore, here is the procedure that should be done:
1. Remove the old controller from the board.
2.Before installing ATmega328, carefully compare those contacts that are designed for normal ATMEL operation and activate them.This applies to pins 3,6,18,21
3.Compile a new firmware for ATmega328.This is perhaps the most difficult point for beginners, but also realizable.

[indman]

Madires,as you know, now it is impossible to say with 100% certainty that the clones of the LCR-TC series can be upgraded, as there may be "surprises" from Chinese friends inside the package.

Here is 1 more version of the LCR-T4 "surprise" from Chinese friends on LGT8F328. However, it can also easily be redesigned to work with genuine ATMEL. It remains only to purchase a few spare ATmega328 from the same Chinese friends!

[Feliciano]

2.Before installing ATmega328, carefully compare those contacts that are designed for normal ATMEL operation and activate them.This applies to contacts 3,18,21

indman, thanks for bringing this to my attention, but I think you forgot the pin 6.

[indman]

but I think you forgot the pin 6.

Yes, thanks you for noticing.I have apdate my post above. 

[cincin]

I read a half-dozen pages back but couldn't find a clear answer: Where can one reliably purchase one of the better one of these nowadays?

From what I read back in this thread, the good ones are not sold anymore, or are hard to find old-stock, and there's a million lesser clones being sold as the good ones everywhere. Is that correct?

So there are no known reliable sources for decent ones? I'm probably not looking to tinker with it much (may not need to be flashable if it's already useful), just to use it for part ID and tester.

Cheers!

[elecdonia]

I read a half-dozen pages back but couldn't find a clear answer: Where can one reliably purchase one of the better one of these nowadays?

I recently purchased this LCR-TC2 unit:
     https://www.ebay.com/itm/LCR-TC2-Transistor-Tester-Diode-Triode-Capacitance-Meter-V2-3E-with-Tester-Cable/295395097022/

After inspecting it internally and using it for several weeks I confirmed it contains the desirable ATmega324P MCU. Its performance and accuracy are very good. The zener test feature works properly without overheating the zener diode being tested. The internal rechargeable lithium cell provides plenty of hours of use. In fact I haven’t needed to charge it yet.

Notes:
1) This unit has a dark blue front panel with manufacturer’s logo at top left (2 Chinese characters).
2) Model number is “LCR-TC2”
3) Version is “V2.3E”
4) The ebay listing contains a color image of a detailed user manual provided by the manufacturer. My unit arrived with a printed user manual identical to what was shown in the ebay listing.

I should mention that several months ago I acquired two other LCR- units which contained alternative (non-Atmel) MCU. Both had issues and faults. In my opinion they were barely functional. Those units had light gray front panels without any manufacturer’s logo. The ebay descriptions did not show the color image of the manufacturer’s user guide.

Conclusion: I do not recommend purchasing any LCR- unit other than this LCR-TC2 V2.3E unit with the dark blue front panel which carries the manufacturer’s Chinese logo.

[cincin]

Thanks! Unfortunately, that seller doesn't ship to Canada. But I will try to find that specific unit somewhere else. If anyone has a reputable lead that ships to Canada, please do tell.
Cheers!

[elecdonia]

Thanks! Unfortunately, that seller doesn't ship to Canada. But I will try to find that specific unit somewhere else. If anyone has a reputable lead that ships to Canada, please do tell.Cheers!

The ebay vendor I ordered my LCR-TC2 V2.3 unit from ships to USA locations from a warehouse inside the USA. That might be why they don't ship to Canada.

I know of several other Chinese ebay vendors who offer the exact same LCR-TC2 V2.3E unit. I've ordered from ebay vendors listed below. They have nearly always shipped what I expected. If there was any issue they immediately issued refunds.

     alice1101983 (557,445) 99.7%
     worldchips (181,380) 99.6%
     modulefans (109,038) 99.4%
     diybox (73,854) 99.4%

Another thing I noticed is most ebay listings for this version of the Transistor Tester include all of the following words in the listing title:

           high precision LCR-TC2 V2.3E

Therefore I recommend using the above words for searching on ebay

[MrSqueaky]

I read a half-dozen pages back but couldn't find a clear answer: Where can one reliably purchase one of the better one of these nowadays?

I bought an LCR-TC2 back in December from Aliexpress https://www.aliexpress.com/item/1005004717377243.html.

It contained an ATmega324 and has worked well so far https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg4635052/#msg4635052

[ws786873]

I bought an AY-AT clone and bought the resistor and IC needed for the upgrade. But when I wanted to burn the M firmware, I found that I could not find the pre-compiled firmware that could be burned in the Windows system... It seems that it needs to be compiled and burned on the Linux system... Can anyone provide the pre-compiled M firmware? I used 16MHZ crystal oscillator.

[py-bb]

Any recommendations on getting a long and happy life out of the replacements? I apologise you clearly know a lot about the model and I regret treating mine as screwed (I used it for heatgun practice)

Replace diode D2 and step-up IC U7. Add a 2K2 series resistor between the junction of D2/C14 and the K pin of the ZIF socket. This requires cutting the PC board foil between these two points. Doing this will make the zener testing circuit unbreakable.

It’s a pity the manufacturer neglected to include the 2K2 series resistor.

Hi there, I've ordered 2 replacements as unfortunately being without this bit of tat sucks. Before I do the mods any recommendations? I have some diodes I need to test - should I be really careful in which way around they go?

I appreciate your advice on this as you clearly know a lot about these devices. Would you like to hear back if I do the mods?

Thanks

[GraemeG]

Assuming you have that particular version of the tester, my advice is DO NOT test anything on the zener test pins (K - A) until you have made the mods. You might get away with it if it is a higher voltage zener and you are fast but there is a good chance you will burn D2 and possibly damage the DC-DC boost IC.

I didn't have any SMD resistors of the right value but it is not too difficult to use a normal through-hole resistor. Cut the track near to the K pin and carefully scrape the copper clean on the track leading to D2. With the resistor lying flat on the board near the end of the ZIF sock pins, solder one lead of the resistor to the K pin (easy) then carefully adjust the other lead to align with the cleaned track. It shouldn't be too difficult to solder the lead to the track if you have a reasonably fine soldering iron tip.

[Feliciano]

I bought an AY-AT clone and bought the resistor and IC needed for the upgrade. But when I wanted to burn the M firmware, I found that I could not find the pre-compiled firmware that could be burned in the Windows system... It seems that it needs to be compiled and burned on the Linux system...

If you prefer the m-firmware, you won't find it pre-compiled online, as there are many features one person may want to enable and other person don't. It's not so difficult to compile it yourself, you may want to read the "Clones" file for instructions on what to modify for your particular unit.
Regarding your other question, it's not mandatory to use linux, there are Windows tools for the task, let's say AVRdudess and others. You will also need the cable for programing it, and the toolchains.
For more infiormation, you can search old posts of this thread.

[Fuzzy Star]

<offtopic>

...I have some diodes I need to test - should I be really careful in which way around they go?

What do you want to measure? Zener voltage? Forward voltage? There are simple ways to measure diodes (Google it). You don't need to risk your tester.
</offtopic>

[elecdonia]

Assuming you have that particular version of the tester, my advice is DO NOT test anything on the zener test pins (K - A) until you have made the mods. You might get away with it if it is a higher voltage zener and you are fast but there is a good chance you will burn D2 and possibly damage the DC-DC boost IC.

There is one item which can be safely tested across the zener test pins (K-A) on every version of the Transistor Tester, modified or not:

        This item is a 1k ohm resistor

No matter what condition the Transistor Tester is in, the absolute maximum current which will flow through this 1k ohm resistor will be safe. It will be no higher than 30mA.

In fact the “zener voltage” displayed by the Transistor Tester (for the 1k resistor) represents the mA flowing from K to A.

     This is just Ohms Law: The voltage drop across a 1k ohm resistor is exactly 1V per mA.

Transistor Testers with the original OSHW zener test circuit will display 2-3V for “zener voltage” with 1k resistor connected to K-A. The original OSHW zener test circuit is a 30V DC source feeding the “K” pin through an internal 10k ohm series resistor. The “A” pin is grounded. This internal 10k series resistor therefore limits the absolute maximum zener test current to 30V/10k = 3mA.

In contrast, unmodified (but working) LCR- units with APT32F172K8T6 MCU and the poorly designed U7 circuit will display 5-8V “zener voltage” with the 1k resistor connected to K-A. This is because DC-DC converter U7 actually does function as a constant current source (until it gets damaged).

An LCR-TC1 where the current-limiting portion of U7 has been damaged, but where the rest of the circuit still works, will apply about 30V across the 1k resistor. In this case the tester may display “unknown or faulty component”. However, a multimeter connected in parallel with the 1k resistor will read 25-30V.

Finally, if D2 and/or U7 failed catastrophically, then the tester may report a zener voltage of either 0V or 4V.

[elecdonia]

Hi there, I've ordered 2 replacements as unfortunately being without this bit of tat sucks. Before I do the mods any recommendations? I have some diodes I need to test - should I be really careful in which way around they go?

I have a Transistor Tester just like yours where the current limiting feature of U7 failed, but fortunately I avoided roasting D2. Therefore my unit is still generates 30V, but without any current limiting. I went ahead and installed the 2k2 series resistor between the junction of D2/C14 and the K test socket pin. This limits maximum short-circuit current to about 15mA.

IMHO putting in the 2k2 series resistor alone is good enough to prevent anything from smoking.

I continue to recommend replacing U7 too. The 2k2 series resistor protects U7. Even if a direct short is placed from K to A, the 2k2 series resistor keeps U7 safe and happy. After installing a new U7, testing with the 1k resistor should display a “zener voltage” of 5-6V. This indicates U7 is properly limiting the zener test current.

[MrSqueaky]

I've discovered a problem with the reference voltage on pin PA3 for my stock LCR-TC2. I was looking at the config.h file of the m-firmware and saw an option for defining the reference voltage so I decided to measure it on my tester but found it was only 45mV.

The TL431 works fine out of circuit so I cut the trace going to PA3 and now the reference voltage is 2.495V. So something is wrong with PA3 pin or it is not being configured properly by the firmware. The resistance between PA3 and ground is ~26 ohms when powered and open circuit when not powered. The voltage drop between 5V and PA3 is 5V so it must be being pulled low.

I'll try to compile and upload the m-firmware and see if the problem still persists.

[madires]

Have you enabled HW_DISCHARGE_RELAY by chance? That would ground the TP_REF pin (in your case PA3) between probing cycles.

[MrSqueaky]

Have you enabled HW_DISCHARGE_RELAY by chance? That would ground the TP_REF pin (in your case PA3) between probing cycles.

The reference voltage problem was with the original Chinese firmware (3.1E). I compiled the 1.48m firmware and uploaded it. Now when I press the test button I get a white screen that turns off after 30s.

The LCD is a 14 pin Z180SN009 (same as TC1 I think). I used the display configuration for the TC1 (ST7735). According to my multimeter LCD_RES stays low the whole time which might explain the white screen.

[madires]

The 'original Chinese' firmware is a modified k-firmware. And apparently they didn't disable the support for the discharging relay. When LCD_RES stays low then there could be something wrong with the configuration, e.g. same pin also used for some other function.

[indman]

I've discovered a problem with the reference voltage on pin PA3 for my stock LCR-TC2. I was looking at the config.h file of the m-firmware and saw an option for defining the reference voltage so I decided to measure it on my tester but found it was only 45mV.

This is not a problem, but a feature of the tester that you need to know and take into account. In this circuit, the PA3 pin of the MCU is periodically connected to ground (by software), so the reference voltage of 2.495V appears only when the measurement cycle starts for a very short time.

[MrSqueaky]

I've discovered a problem with the reference voltage on pin PA3 for my stock LCR-TC2. I was looking at the config.h file of the m-firmware and saw an option for defining the reference voltage so I decided to measure it on my tester but found it was only 45mV.

This is not a problem, but a feature of the tester that you need to know and take into account. In this circuit, the PA3 pin of the MCU is periodically connected to ground (by software), so the reference voltage of 2.495V appears only when the measurement cycle starts for a very short time.

I also have a round GM328A(BGR)2 tester and the reference voltage stays on for a couple of seconds after pressing the test button. I'm pretty sure it wasn't with the LCR-TC2. I overwrote the Chinese firmware so I can't test it anymore to double check

[indman]

I also have a round GM328A(BGR)2 tester and the reference voltage stays on for a couple of seconds after pressing the test button. I'm pretty sure it wasn't with the LCR-TC2. I overwrote the Chinese firmware so I can't test it anymore to double check

If you have a copy of k-firmware installed on your GM328A, then I'm sure the same behavior with Vref was also on your LCR-TC2 with Chinese firmware,which, as madires rightly pointed out, is a modified k-firmware.

[MrSqueaky]

I finally got the v1.48m firmware working on my LCR-TC2 with hardware SPI.  The pin assignment for the LCD conflicts with the hardware SPI. Tracks need to be rearranged to make a permanent solution. Bit banging works without any need for hardware modification but the output seemed a bit slow. Also, I had to half the hardware SPI speed otherwise I just got a white screen (possibly due to the jumper wires)

Code: [Select]

/* hardware SPI in config_644.h: */
#define SPI_HARDWARE
#define SPI_PORT         PORTB     /* port data register */
#define SPI_DDR          DDRB      /* port data direction register */
#define SPI_PIN          PINB      /* port input pins register */
#define SPI_SCK          PB7       /* pin for SCK */
#define SPI_MOSI         PB5       /* pin for MOSI */
#define SPI_MISO         PB6       /* pin for MISO */
#define SPI_SS           PB4       /* pin for /SS */

/* ST7735 using hardware SPI  in config_644.h: */
#define LCD_PORT         SPI_PORT       /* port data register */
#define LCD_DDR          SPI_DDR        /* port data direction register */
#define LCD_RES          PB0            /* port pin used for /RESX (optional) */
#define LCD_CS           SPI_SS         /* port pin used for /CSX (optional) */
#define LCD_DC           PB1            /* port pin used for D/CX */
#define LCD_SCL          SPI_SCK        /* port pin used for SCL */
#define LCD_SDA          SPI_MOSI       /* port pin used for SDA */

/* set SPI clock rate for hardware SPI display in ST7735.c: */
// SPI.ClockRate = SPI_CLOCK_2X;    /* set clock rate flags */
SPI.ClockRate = 0;    /* set clock rate flags - SPI2X = 0, SPR1 = 0, SPR0 = 0, SCK = fosc/4 */

/* set power management pins in config_644.h: */
#define POWER_PORT       PORTD     /* port data register */
#define POWER_DDR        DDRD      /* port data direction register */
#define POWER_CTRL       PD1       /* controls power (1: on / 0: off) */

#define BUTTON_PORT      PORTD     /* port data register */
#define BUTTON_DDR       DDRD      /* port data direction register */
#define BUTTON_PIN       PIND      /* port input pins register */
#define TEST_BUTTON      PD2       /* test/start push button (low active) */

/* set probe colors to match decal in color.h */
#define COLOR_PROBE_1         COLOR_RED
#define COLOR_PROBE_2         COLOR_YELLOW
#define COLOR_PROBE_3         COLOR_GREEN




[indman]

I finally got the v1.48m firmware working on my LCR-TC2 with hardware SPI. 

In my experience, for a 1.8" color display at 16-20MHz, there is no particular need for a hardware SPI.
The difference in speed compared to software SPI is not noticeable. For displays with a diagonal of 2.2 inches and above, hardware SPI really gives an increase in the speed of updating the picture, but still not as big as we would like.