$20 LCR ESR Transistor checker project

[indman]

0.1% SMD's are only a dollar each, I wonder if I could have soldered wires to them and used them instead.

Maybe I'm wrong, 0.1% resistors are good, but for this project it is quite enough to select measuring resistors from the 1% set with a multimeter. This selection is then easily controlled using the built-in hardware tests T1-T6. It's just important that the pairs of resistors are close in resistance. I didn't notice much difference in readings compared to 0.1% precision resistors. Do not forget that this is just a SIMPLE TESTER, in which there are still a lot of factors that do not allow equating it to metrological instruments.

[.RC.]

I was just being guided by others on ways to improve on the chinese kits  https://github.com/blurpy/transistor-tester/blob/master/resources/AY-AT-J1.3.png

Is this the way to check zeners on this model.    It was a 12V Zener.   40V being input, 28V showing, so 12V Zener?

I only enabled  HW Zener

[madires]

Nice! Connect the Zener diode and the external power supply in parallel to the tester's input, while limiting the current to a few mA (typically 3 mA).

[oitar]

OK, I connected PD0 of my mega324pa(which is a configuration pin, pulled up to 5V, not used in m-firmware) to the test button, and can now use the test button properly. I did the Test, and Adjustment(now with the 1.49m) number of times, completed successfully, and saved in Profile1.
However the issue persists: 1. Zener diodes over 10 V are measured perfectly, below 10 V(or thereabout) are shown as ~1/4th of their voltage.
2.Resistors are shown in a very strange way: I put a brand new 10 kOhm(which measures 4247 Ohm after Adj) multiturn pot, and starting from 0 slowly go up. I get this in Ohms: 0.91, 70.5, 535.6, 960.2, 1224, 1610, 2104, 2424, 2601, 11.59k, 12.72k, 14.07k, 15.23k, 16.60k, 18.20k, 2418, 2569, 2854, 3043, 3329, 3615, 3862, 4187, 4322(at the end).
I.e. going linnearly, albeit low, then for a while overshooting by double, then back to low, and finnishing at around half the actual value.

Please be aware that this is NOT A FIRMWARE ISSUE, since it had a similar behaviour with the original Chinese firmware.
I would like an opinion whether this is a hardware defect(why does it complete the Adjustment then?), or a wrong port pin specified somewhere(checked number of times, schematic is very similar to the T7 in the Clones memo)?
I have checked the resistance of the 3 probes, and the readings are practically the same to GND and VCC for all 3. Also the resistance of the 3 680 Ohms, and 470 kOhms are within 0.1% of each other.

[madires]

Could you please repeat the Zener test (with a Zener <10V and another one >10V) and measure Vcc (5V) with a DMM? Is Vcc stable?

[elecdonia]

Is this the way to check zeners on this model.    It was a 12V Zener.   40V being input, 28V showing, so 12V Zener?

Another method is this:

1) Connect a 10k ohm resistor instead of the zener shown in your photo. Temporarily set the zener diode aside (not connected). Otherwise keep same connections to your 40V power supply. After replacing zener with 10k resistor then the Transistor Tester should display the DC voltage provided by your power supply (~40V).

2) Next, connect the zener to be tested directly across the 2 “voltage test” terminals of the Transistor Tester, with zener cathode (banded end) to + terminal of voltage test connector.

3) Transistor Tester should then display the actual zener voltage (12V). You won’t need to subtract the voltage displayed by the Transistor Tester from the power supply voltage to obtain the zener voltage.

The 10k series resistor limits the maximum zener test current to <4mA (when external power supply voltage is 40V).

Most  “factory built” Transistor Testers with self-contained zener test feature use a 10k series resistor to limit the test current. Although the Transistor Tester technical reference manuals provide a schematic for an optional “constant current source” for zener testing, I don’t think this extra complexity is necessary. The 10k series resistor all by itself is adequate.

[elecdonia]

Call me a sucker for punishment, but after recently failing with my purchase of the LCR-TC2 from this vendor (I wanted an ATmega324 but got the LGT8F328P), I've taken a gamble on the LCR-TC1 model from the same vendor. https://www.aliexpress.com/item/1005004717377243.html
What gives me hope?

LCR-T7 / LCR-TC1:color screen / M644 chip

I succeeded at getting a unit containing a genuine Atmel ATmega324 in Jan. 2023. I purchased it from an eBay vendor who also offered several different Transistor Tester models, quite similar to the Aliexpress listing in your post.

The item I selected was priced at $23 USD vs. $12-15 USD for all other models offered by the same vendor. The eBay vendor described it as “LCR-TC2 improved” (I expected to pay ~$23 USD because genuine Atmel ATmega324 as bare IC chip sells for >$8 USD).

Curiously my mega324 unit also displays V2.3E for its firmware. I have confirmed it contains a genuine ATmega324 (44 pins, not 32). I’m nearly ready to install OSHW Transistor Tester firmware into it. I will post my results when finished upgrading.

Please let us know what arrives for you!

[Obelix2007]

Is this the way to check zeners on this model.    It was a 12V Zener.   40V being input, 28V showing, so 12V Zener?

I only enabled  HW Zener

... connecting your z-diode is wrong for a measurement without calculation:

[elecdonia]

Is this the way to check zeners on this model.    It was a 12V Zener.   40V being input, 28V showing, so 12V Zener?I only enabled  HW Zener

... connecting your z-diode is wrong for a measurement without calculation:

Is that a 20k ohm series resistor in your photo?

[Obelix2007]

... yes, it is - this prevents the BZX83C47 from heating up. With higher current, the zener voltage increases and the distance to the measuring voltage (50 volts) becomes too small to measure accurately - with this arrangement, the measuring current is approx. 250 µA!

Regards Horst

[.RC.]

Here are some pricey ones  https://www.aliexpress.com/item/1005003670951373.html

I want upgrade 3.  "After sales upgrade，New skin shrimps will be given away and accidentally burned within one year，"   

I think they mean a different firmware and a one year warranty.

I guess this is the same basic model without the accidentally burnt shrimp skins. https://www.aliexpress.com/item/1005003680113751.html

All claim 48V zener capability.

[elecdonia]

Here are some pricey ones  https://www.aliexpress.com/item/1005003670951373.htmlI want upgrade 3.  "After sales upgrade，New skin shrimps will be given away and accidentally burned within one year，"   

Interesting that this vendor posts photos of the PC boards. Most do not. Also all 3 of their PC board photos show MCU chips with 44 pins. To the best of my knowledge there aren’t (yet) any cloned/faked Atmel MCU chips with 44 pins. Popular “fake” Atmel MCU chips all have 32 pins. Yet another “tell” is the presence of j4, a 6-pin Atmel ISP programming header in the traditional 2x3 pin format. Transistor Tester PC boards with clone/fake MCU never have this 2x3 pin programming header. Some do have what appears to be a programming header, but never in the 2x3 pin style.

Chinese-> English is often entertaining. For example:

https://www.ebay.com/itm/324243157482

Product parameters:
Product Type: Electronic Tube Amplifier
Number of channels: two channels
Output rent resistance: 4-8 euros
Power supply: AC12V
Product net weight: 170g
Product size: 81*81*43MM

[oitar]

Could you please repeat the Zener test (with a Zener <10V and another one >10V) and measure Vcc (5V) with a DMM? Is Vcc stable?

Sorry about the mess on the desk! See attached measurement of an 8.2 V Zener with the dmm over the diode.One with the dmm over the VCC.
Another with a 16 V Zener. And one with empty test terminals.
I can put a trace scope on the VCC if needed, but it looks pretty steady.

[indman]

However the issue persists: 1. Zener diodes over 10 V are measured perfectly, below 10 V(or thereabout) are shown as ~1/4th of their voltage.
Please be aware that this is NOT A FIRMWARE ISSUE, since it had a similar behaviour with the original Chinese firmware.

In this case, show us the compiled firmware and the configuration files you used.
It will also be nice if you show the results of hardware tests T1-T6 from the Selftest item.

[oitar]

However the issue persists: 1. Zener diodes over 10 V are measured perfectly, below 10 V(or thereabout) are shown as ~1/4th of their voltage.
Please be aware that this is NOT A FIRMWARE ISSUE, since it had a similar behaviour with the original Chinese firmware.

In this case, show us the compiled firmware and the configuration files you used.
It will also be nice if you show the results of hardware tests T1-T6 from the Selftest item.

See attached makefile, config.h, config_644.h, .hex file, and .eep file in one archieve.
Second archieve has screenshots of each stage of Self Test, as well as the Show Values screen.
The T4 stage did not show any values, but flickered very quickly, before asking to "remove short".
Didn't want to upload a video. Hope this is what you needed?

[indman]

Second archieve has screenshots of each stage of Self Test, as well as the Show Values screen.

Well, a hardware problem was immediately discovered in the T2-T3 tests. In the photo, I highlighted with a red question the values ​​that indicate a problem with the measuring resistors or controller ports. The values ​​in these tests should always be of the same order and close in value. Now you must remove all the extra wires that you use for the ZIF panel and also power the tester not from an external source, but from a 3.7V battery. Then repeat tests T1-T6 and show us the results.
Show us also a good quality photo of the reverse side of your clone where is the microcontroller and parts?

[indman]

I completely forgot one more thing. Such poor readings in the T2-T3 tests can be caused by a faulty SRV05-4 protective assembly. Remove it for a while and repeat the hardware tests again.

[madires]

Sorry about the mess on the desk! See attached measurement of an 8.2 V Zener with the dmm over the diode.One with the dmm over the VCC.
Another with a 16 V Zener. And one with empty test terminals.
I can put a trace scope on the VCC if needed, but it looks pretty steady.

So Vcc is stable and the problem is somewhere else. Please follow indman's suggestions next.

[oitar]

I completely forgot one more thing. Such poor readings in the T2-T3 tests can be caused by a faulty SRV05-4 protective assembly. Remove it for a while and repeat the hardware tests again.

Thanks, I will try removing the OV protection.
With the battery back, and Self Test performed(all T readings are basically the same), the zener/resistor issue is still present.

[oitar]

Thanks, I will try removing the OV protection.
With the battery back, and Self Test performed(all T readings are basically the same), the zener/resistor issue is still present.
[/quote]

No luck with the DZ1 chip removed, I'm afraid! Re-did the tests, similar results.

[indman]

No luck with the DZ1 chip removed, I'm afraid! Re-did the tests, similar results.

Have the test results not changed at all, exactly the same as before?
One more question - why did you change the port for the Test button from PD2 to PD0?

[vk3em]

ha ha, well I now have gone from zero testers a month ago to two and another currently on a plane.  Then I bought six resistors for $2.57 each and six for $1.60 each to "fix up the kits" Along with a couple of precision voltage regulators and a couple of precision voltage references.

Wow, that's expensive! A local online shop sells 0.1% TH resistors (Yageo) in single quantity for around EUR 0.25 - 0.30 (680 and 470k are 0.27).

E96 resistor values are more common in 0.1% than E12 where 0.1% values are non standard. Look for 681 Ohms (instead of 680) and 475 kOhms (instead of 470k).

Luke

[oitar]

Have the test results not changed at all, exactly the same as before?
One more question - why did you change the port for the Test button from PD2 to PD0?
[/quote]
They look pretty identical...

I could never get the Test Button to operate properly via the U3(as in my board) chip. I re-programmed it as stated, but could not get it to go low on pin8(i.e. PD2 on the MCU). So using it before only woke-up the tester, but could not enter the menu, so only usable way was to program it as "continuous".
I know replacing it with the 2 transistors will be better(also gaining a port), that's coming next, but I don't think this affects the measurements.

Tomorrow I will check the battery port if it can measure low voltage, since it is adjacent to the zener port. And the zener port is indeed failing in [low] voltage measurement. As to the weird resistor results- no idea what to do yet.

[indman]

They look pretty identical...
I could never get the Test Button to operate properly via the U3(as in my board) chip. I re-programmed it as stated, but could not get it to go low on pin8(i.e. PD2 on the MCU).

These strange and incorrect test data quite convincingly hint at a hardware problem with your tester. It is very likely that your controller is damaged or defective. Try the 2-transistor control option. If you have already checked the 680Ω and 470kΩ resistors and they are OK to I have no more ideas how to fix this problem.

[Obelix2007]

... no idea what to do yet.

... my suggestion is to swap the "small" Atmega324 for a 644 or 1284!!

Greeting Horst