TC1 lcr-meter transistor-tester fix

[indman]

It still won't work and the micro shows what's in the attached photo-

Once again I repeat to you, if you did not understand me before - your clone has a hardware problem and no other firmware will fix it! Use a multimeter to check the resistance of all measuring ports of the processor relative to + 5V and GND.

[LesioQ]

My tester TC-1 M-Tester died after a dozen of tests done (original factory graphical firmware).
The first problem I noticed was erratical behaviour - sometimes it booted OK but failed to test, sometimes it was hung-up on logo screen.
Sometimes the main screen display had pixel errors (image distorted).

Upon some testing I noticed it bahaves better if 5V supply is lowered somewhat. I ended up adding a Si diode right after (U3) 5V stabilizer with caps. Now Vcc is 4.2V.

It may be that communication with a LCD working on 3.3 Volt supply was causing problems.
Now it's testing OK, but fails to complete a Self-test - hangs at 7-16%.

Maybe someone can elaborate on that.

[madires]

The tester needs 5V to run properly. Possibly a power supply issue, e.g. a broken MLCC. Do you have a scope to check the output of the boost converter?

[LesioQ]

Sure I do. I was searching for some fluctuations on Vcc 5V, but it looks OK. HV is 39V, (after C13 & C14 replacement).
Vraw (5V stab. input) is about 9V.

[madires]

A dodgy display (ST7735)?

[mtcaus]

Solved the no power up issue. When the button is pressed, there is no display and Step-up/Boost IC at U5 is so hot. It could burn your finger. The fault is the Schottky barrier diode at D2. The diode with marking code SS18 (1A, 80V) was tested fine both in-circuit and out-of-circuit.  Replaced it with SR2100 (2A, 100V) and the unit is working again. Measured the peak voltage of 52V at the cathode side of SR2100 Schottky diode when the button is pressed. Since the voltage rating of 1206 SMD capacitor at C11 is only 50V.  Eventually C11 will give up.  I replaced it with an electrolytic capacitor Rubycon 4.7uF @50v.  I couldn't find a higher rated voltage of capacitor.  Ideally should be like 100v.  However, when I tested the Rubycon withstand voltage, it could go up to 100v.  No wonder Rubycon capacitors are good.  Anyway, fixed the unit by replaced two components at D2 and C11. See photo fixed.jpg.

From the transformer, when the button is pressed, there are two outputs, 9v (for powering the unit) and 37v(at least for Zener diode testing). See photo Bottom.jpg. Mine was like 2v on either path.  Obviously there was a short somewhere.  Through the process of elimination, I removed Schottky diode first.  I got the 9V back. At first I thought ATMEGA324PA was the fault, but it was tested fine when applying 5V directly. See photo Top1.jpg and Top2.jpg.  At first replaced D2 with SS24(2A, 40V) for testing. Obviously the voltage is not rated high enough. The diode was so hot and high pitch sound can be heard from the transformer when power up.  Since I don't have any other SMD Schottky diode with a voltage rated that high.  I used a through-hole Schottky diode SR2100 and there is enough space to install it. For the capacitor at C11, it will give up eventually if the peak voltage repeatedly go above the rated 50v 1206 SMD capacitor. This maybe the reason why people reported C11 shorted in the forum.

My LCR-TC1 is from the original creator.  The original has 3 Chinese characters (浩祺心) on the top of the circuit board. See photo Top1.jpg. The component layout may be slightly different than yours since there are so many clones out there.  However, the general layout is the same.

I have been troubleshooting this unit for a while.  Replaced numerous Step-up/Boost IC ALxxx at U5 with different numbers.  xxx are three numbers.  The numbers do not matter if anyone who cares to know.  Even bought STC15L104W chip at U4 and programmed it. By the way, I couldn't find the original U4 firmware. Luckily I didn't overwrite the original chip.  Obviously none of them worked until I replaced Schottky diode at D2. The time and money spent may not justify, but the feeling of successfully fixed the unit is worth it.  Hope this post will help someone out there.

[masster]

I bought today my TC1. As many other users before me, I calibrated the tester then started to check parts.
After maybe 8 parts measured, the tester turned off as usual by himself, but never came back on. It is dead. I didn't check any capacitor, only resistors, transistors, and a diode.
I thought the battery is low, although the tester showed 3.85V. I charged it on USB and after 10 minutes the LED turned from red to green. Yet still no luck. The tester is dead.
Full of confidence, I opened the case, hoping to change two capacitors and move on. But... surprise! Although on the outside, my TC1 looks identical to the rest of TC1's in this topic, on the inside there is another story. A totally different parts layout. See attached photo. I managed to read the codes on each IC and annotated the photo accordingly.
So my question is: how should I diagnose for the fault? Anyone happens to have the schematic for this version?
Many thanks in advance if you can help me.

[masster]

Update:
The output of 78L05 (VCC) is shorted (3.5 ohms). What is the best method to decide which component is shorted? Capacitors, input protections, IR receiver, MCU, ATMEGA itself.

[TurboTom]

If you have a thermal imager, attach a current limited lab supply to the shorted supply line (set to maybe 100mA) and observe what component heats up. This is also possible without such gear: if you use a sprayer (the type used for misting flowers) filled with isopropanol or ethanol and spray the board with a very thin layer of the liquid, you can usually observe where it evaporates first. Good luck!

[masster]

78L05 has a maximum output current of 100mA with proper heat dissipation.
How much current draws the device on 5V when measuring?

[madires]

I don't know the specific current draw for that model, but usually it's about 15mA on average for the 5V rail.

[mtcaus]

TC1 will very likely break down after continuous tests in a short period of time because it always generate high voltage for zener diode test every time.  The components along that path simply cannot take it. Your unit is the newer design, but it's basically the same.  Instead of one transformer with two outputs, your design simply uses two sets of inductors and step-up converters. By looking at the photo, you have a burnt resistor R24 circled in green. You should also check all the components circled in red. I saw your circuit online somewhere before.  Just search around and you can find out the value for your R24.

[masster]

@mtcaus

Very good catch on R24. It would help to know its value.

I can't explain myself why the resistor R24 failed. "High voltage" is not 6000 Volts or so, maybe 30 Volts at most, and R24 should be in the kiloohms range. Moreover, I didn't insert any part in the K pin (cathode for Zenner diodes)

I have checked the components marked in red. They are all OK. With a thermal camera, when I press the TEST button, the 78L05 regulator and the 4 capacitors (2 on the input side and 2 on the output side) glow like fire...

[indman]

masster,a malfunction in these clones with certain skills in repair and understanding of the principle of their work is found in a couple of minutes.  The schematic diagram of your clone will roughly correspond to the scheme on the LCR-T7H, only the numbering of the elements will be different. R24 will correspond to R13 on my circuit and will be 10Kohm.
If you have a heated 7805, then there is a high current consumption.We need to determine the cause.Remove the 7805 from the circuit and check its serviceability separately using an external power supply. Also, temporarily remove the U2 diode assembly(SRV05) and the 6V8A suppressor from the circuit. Then check with the multimeter whether there is a low resistance at the output of the 7805 (3.5 ohms)?

[Per Hansson]

By looking at the photo, you have a burnt resistor R24 circled in green. You should also check all the components circled in red. I saw your circuit online somewhere before.  Just search around and you can find out the value for your R24.

Isn't it just installed with the bottom side facing up?

[masster]

@indman
Thank you for your reply. I know you researched this tester heavily. The only thing that keeps me from doing a better job in diagnosing the issue is the microscopical size of the parts. I wanted to minimize the intervention on the board because I really hate it working with heavy magnifying glasses for electronics or with a microscope.
Anyway, I have desoldered  in turn the following parts: 78T05, the two filter capacitors on the 5V output, the input protections (6V8A and SRV05-4), and the IR receiver.
Shorted parts found: SRV05-4 and the IR receiver. Yes, I know. The latter being faulty was a total surprise.
I find that the quality of the parts used in these tester clones is deplorable (POS), hence the low price. In fact, the seller didn't offer any guarantee. Although after replacing the bad parts the tester will most probably work, it is very unreliable. Who knows what else will break? My advice? Don't waste the money on those China clones. Make one yourself with good quality parts, maybe in bigger size to avoid doing neurosurgery on the board, and definitely I would try finding a better input protection because right now if somebody is messing with the input pins by shorting them inadvertently or forgetting a cap is not discharged, he will look again(!) at a dead tester.

@Per Hansson
You are right. The resistor was soldered upside down. I measured it after desoldering. It has 10k indeed.

[indman]

Shorted parts found: SRV05-4 and the IR receiver. Yes, I know. The latter being faulty was a total surprise.
I find that the quality of the parts used in these tester clones is deplorable (POS), hence the low price.

Your problem does not in any way confirm the low quality of Chinese parts, although such a reality certainly exists.
A short circuit immediately in 2 parts allows us to conclude that there was some surge of increased voltage, which these parts took over. Put it back in place 7805 (if intact) and turn on the tester without SRV05-4 and the IR receiver. The device should turn on and work without them.
Very often the culprit of similar malfunctions is static electricity.

Who knows what else will break? My advice? Don't waste the money on those China clones. Make one yourself with good quality parts, maybe in bigger size to avoid doing neurosurgery on the board, and definitely I would try finding a better input protection because right now if somebody is messing with the input pins by shorting them inadvertently or forgetting a cap is not discharged, he will look again(!) at a dead tester.

Of course, you are right that a thing made with your own hands will bring much more benefit and experience than a purchased one!
But on the other hand, thousands of Chinese clones work well for many people and they don't bother looking for other options.

[masster]

@indman
I have soldered back in place just the 78L05 and 6V8A. The display came back to life, but instead of showing "No, bad part, or unknown part" message, it shows Cell 1.84V between pin 1 and 3.
I remind you that I didn't test any capacitor whatsoever from new. Moreover, when I worked on the board, I used an ESD wristband.

[madires]

Most likely the ATmega was damaged too.

[indman]

I remind you that I didn't test any capacitor whatsoever from new. Moreover, when I worked on the board, I used an ESD wristband.

It takes a very good reason to make several parts malfunction at once.Wouldn't you agree?
If you rule out static electricity and an undischarged capacitor, it will not be easy to find the cause of such an occurrence with your clone.
I do not exclude the possibility that somehow in an incredible way the voltage that is created to test the Zener diodes (25V) hit the input of the measurement ports of the processor and knocked out the protective assembly, IR and the ports themselves.In my practice there have been several cases where the part for testing was placed on the K contact on one side and on one of the 1-2-3 test contacts on the other side.

[masster]

1) What is the maximum voltage/current supported by the input ports of ATMega?
2) Is there a better solution to protect the input ports against charged capacitors?
3) Is there a better solution to protect the input ports against wrongfully connecting tested parts? Fool proof and full proof.

[indman]

1) What is the maximum voltage/current supported by the input ports of ATMega?

This information can be found in the datasheet of the AVR. MaxVoltage -5.5V, MaxCurrent for 1 port - 40mA,for all ports -200mA

2) Is there a better solution to protect the input ports against charged capacitors?

You can replicate the protection circuit using a relay with 3 contact groups as shown in the Karl-Heinz manual.
But as correctly noted by the authors of the project - no protection provides a 100% protection guarantee. The same goes for the answer to your question №3

[masster]

Most likely the ATmega was damaged too.

If the display says "Testing..." and after 5 seconds it shows "Cell, 1.40 V between pin 1 and 2" , is that definitive proof the ATMega is damaged?

[indman]

Yes, this is one of the characteristic signs that one or more measuring ports of the processor are damaged. To make sure of this definitively, you can check the resistance of each port relative to GND and VCC. Provided that the SRV05-6V8A protective assembly is removed and nothing else is connected to the ports in parallel.You can also believe whether the nominal values of the measuring resistors 680ohm and 470k are in order. If they are in good condition, and the port resistance is very different, then this indicates damage to the processor.

[masster]

I decided to remove the 6V8A and SRV05-4 to measure the resistance between each pin of ATMega and GND. See schematic below. Where there is no resistance value, the resistance is higher than 2 Mohm.
I find abnormal the resistance between VCC and GND, which is still low IMO. I have used an external power supply for VCC and I have measured a current consumption of 170 mA. That is way too high, over what 78L05 can deliver. Hopefully there is another faulty filtering capacitor on the VCC bus. I didn't remove them all. I still have hope that ATMega is not toasted.

Please have a look at my resistance measurements and tell me what else you find odd. I don't like the values on pins 34 to 37, corresponding to PA0 to PA3.

Right now, with external VCC supply and battery connected, I get the Cell of around 1.2V between 1 and 2. Sometimes I get a Cell between pin 1 and 3.