Author Topic: $20 LCR ESR Transistor checker project  (Read 4232059 times)

JohnKr and 12 Guests are viewing this topic.

Offline Messtechniker

  • Frequent Contributor
  • **
  • Posts: 840
  • Country: de
  • Old analog audio hand - No voodoo.
Re: $20 LCR ESR Transistor checker project
« Reply #9900 on: January 19, 2025, 09:43:42 am »
Note that you are using a tester and not a precision instrument. What you are seeing is probably a temperature dependence.
Agilent 34465A, Siglent SDG 2042X, Hameg HMO1022, R&S HMC 8043, Peaktech 2025A, Voltcraft VC 940, M-Audio Audiophile 192, R&S Psophometer UPGR, 3 Transistor Testers, DL4JAL Transistor Curve Tracer, UT622E LCR meter, UT216C AC/DC Clamp Meter
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9901 on: January 19, 2025, 10:35:08 am »
The inconsistency is much larger if I wait for it to shut down, then measure again.
If I make many measurement without letting it to auto-off, the repeatability is OK.

It looks like it would do some sort of self-calibration at each power on, probably when it also measures its own battery.  :-//

Is that expected, to self-calibrate itself at each power-on?

Offline indman

  • Super Contributor
  • ***
  • Posts: 1214
  • Country: by
Re: $20 LCR ESR Transistor checker project
« Reply #9902 on: January 19, 2025, 10:41:56 am »
Is that expected, to self-calibrate itself at each power-on?
No,that's not entirely true! As has been said many times in this topic, the stability of measurement results primarily depends on the stability of the +5V voltage that powers the microcontroller! The quality of Chinese components also plays a big role. So for now, don't expect much measurement stability from your tester.
« Last Edit: January 19, 2025, 11:39:34 am by indman »
 
The following users thanked this post: RoGeorge

Online iMo

  • Super Contributor
  • ***
  • Posts: 5640
  • Country: va
Re: $20 LCR ESR Transistor checker project
« Reply #9903 on: January 19, 2025, 10:58:37 am »
..Is this "normal" for the version I have?
What to do to get better repeatability?

The transistors have -2mV/C TC of Vbe, they heat up when in the socket (they are powered), or cool down with an air draft..
The voltage regulator there might have 300ppm/C (or more) easily..
The voltage of your battery fluctuates (thus the temperature and Vout of the 5V regulator)..
It also picks up EMI from the outside, as well us the mcus's ADC is 10bit only and usually noisy (like +/-2 LSB easily)..
Etc..
So taking that into account your measurements above are almost perfect..  :D

PS: for example at say 22C ambient the Vbe of a transistor could be say 550mV, when you keep the transistor in your fingers for couple of seconds (your fingertips are around 31C) you may get 532mV..
And in still air after insertion into the socket it will slowly rise (as its temperature goes down), and while in socket powered the Vbe wants to go down again as its junction heats up..
« Last Edit: January 19, 2025, 11:15:39 am by iMo »
Readers discretion is advised..
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 8321
  • Country: de
  • A qualified hobbyist ;)
Re: $20 LCR ESR Transistor checker project
« Reply #9904 on: January 19, 2025, 11:16:29 am »
It looks like it would do some sort of self-calibration at each power on, probably when it also measures its own battery.  :-//

Your T4 has a TL431 which is used to figure out the voltage of the 5V rail. This also impacts the voltage measurement of the internal bandgap reference. Both self-adjustments are repeated for each probing cycle. With a cheap 5V regulator and a TL431 as 2.5V reference you can't expect much. BTW, all this is explained in the OSHW documentation.
 
The following users thanked this post: RoGeorge

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9905 on: January 19, 2025, 11:25:08 am »
Mine seems identical with this one here:
https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg709934/#msg709934

I've just looked on the PCB, it has a TL431, but never dug the details before.  To my surprise, I see now that the TL431 is not used as a Vref, but tied to another pin.  ???



Looks like a strange design choice.  Why using the 78L05 Vcc as Vref, then comparing the ADC count against a TL431?  Why not using TL431 directly, to generate a 5V ADC Vref?  What am I missing?
« Last Edit: January 19, 2025, 11:28:14 am by RoGeorge »
 

Offline indman

  • Super Contributor
  • ***
  • Posts: 1214
  • Country: by
Re: $20 LCR ESR Transistor checker project
« Reply #9906 on: January 19, 2025, 11:36:11 am »
Why using the 78L05 Vcc as Vref, then comparing the ADC count against a TL431?
Because 78L05 is usually a low quality stabilizer. Install a high quality stabilizer instead and the need to use an external Vref will disappear completely! ;)
 
The following users thanked this post: RoGeorge

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9907 on: January 19, 2025, 11:44:15 am »
BTW, all this is explained in the OSHW documentation.

Searched for 10 minutes and couldn't figure out which one do you refer to.  In which place are the docs please, or what term to search for?

Offline madires

  • Super Contributor
  • ***
  • Posts: 8321
  • Country: de
  • A qualified hobbyist ;)
Re: $20 LCR ESR Transistor checker project
« Reply #9908 on: January 19, 2025, 12:15:11 pm »
Looks like a strange design choice.  Why using the 78L05 Vcc as Vref, then comparing the ADC count against a TL431?  Why not using TL431 directly, to generate a 5V ADC Vref?  What am I missing?
IIRC, the external 2.5V reference was added later on and therefore had to be compatible with the original design (and firmware). Also, the tester switches between reference voltages (Vcc and internal bandgap) on demand. A fixed reference voltage at AREF would break that. And with an ATmega328 we don't have unused pins to switch external references, which in turn would make the circuit more complex (we want to keep the basic design simple). Another reason is that this design supports the measurement of a Vcc higher than 5V (the 5V regulator could output 5.07V, for example). Using the TL431 to create a 5V reference voltage for the ADC would limit measurements to that reference voltage.
 
The following users thanked this post: RoGeorge

Offline madires

  • Super Contributor
  • ***
  • Posts: 8321
  • Country: de
  • A qualified hobbyist ;)
Re: $20 LCR ESR Transistor checker project
« Reply #9909 on: January 19, 2025, 12:17:45 pm »
BTW, all this is explained in the OSHW documentation.

Searched for 10 minutes and couldn't figure out which one do you refer to.  In which place are the docs please, or what term to search for?

Please see https://github.com/madires/Transistortester-Warehouse/tree/master/Documentation and start with Karl-Heinz' ttester-1.13k.pdf.
 
The following users thanked this post: RoGeorge

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9910 on: January 21, 2025, 04:49:45 pm »
That was a lot of reading.  Took an entire afternoon to read all that.
And together with other things, took me 3 afternoons to clarify which is what.  ;D

But I'm grateful that documentation exists, and the project as well, thank you all, and thanks for support, too.

- first day did the reading
Not yet sure if mine is an RLC-T4 or a T3, but the schematic posted above matches mine, except one resistor, which I guess is a typo.  R12 on the PCB measures 100k, not 10k.  The value printed on the PCB silkscreen is also 100k, so I guess the R12/10k in the picture is a typo.  Funny thing, in some PCB versions (like mine) the programming header is mirrored.

The software docs (English PDF) is probably translated from German, and sometimes not very clear, but the English version of the README text file (the one gziped together with the sources) was crystal clear, and up to date.

- second day had problems with my PICkit 2, and instead of using another programmer, spent the time making the pikit2 programmer to work again.

- then another day spent fooling around with all kind of software options and settings.  I didn't know there are so many hardware and software options for this project.  It can do a lot more than the default.  :-+

A nice surprise was the PC connectivity, which I want to try.  But how come that there is no ADC logger implemented already?  A 3 channel ADC logger (for the 3 probes) would be a nice feature.  Or is it already there but I've missed it?



To summarize for newcomers like me:
- there are repositories for sources and another for documentation
- the project supports many models of MCUs and many types of schematics
- there are two flavors of firmware, 'k' and 'm', 'm' is for MCUs with at least 32k (MCU at least ATmega32)
- unless you are a developer, all you need is to download the last .gz version from https://github.com/kubi48/TransistorTester-source/tree/master/Markus (ComponentTester-1.54m.tgz)
    - unpack and cd into it
    - read the ./README
    - customized the 'Makefile', 'config.h' and 'config_328.h' according to your tester's hardware
    - connect the AVR programmer and type
            make upload

And now, fool around with all the new features for the days to come.  :D

Offline indman

  • Super Contributor
  • ***
  • Posts: 1214
  • Country: by
Re: $20 LCR ESR Transistor checker project
« Reply #9911 on: January 21, 2025, 05:11:49 pm »
Not yet sure if mine is an RLC-T4 or a T3, but the schematic posted above matches mine, except one resistor, which I guess is a typo.  R12 on the PCB measures 100k, not 10k. 
This schematic contains a lot of errors, so I don't recommend using it. I have attached a more correct version below
« Last Edit: January 21, 2025, 05:22:01 pm by indman »
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 8321
  • Country: de
  • A qualified hobbyist ;)
Re: $20 LCR ESR Transistor checker project
« Reply #9912 on: January 21, 2025, 06:45:34 pm »
But how come that there is no ADC logger implemented already?  A 3 channel ADC logger (for the 3 probes) would be a nice feature.  Or is it already there but I've missed it?

No. There are tons of data loggers much better than what an ATmega328 could do.
 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9913 on: January 22, 2025, 08:02:20 am »
I guess I've spent too much time with the project, because now I'm starting to see all kind of flaws.  ;D

Being open source I'm tempted to change too many details, which might break backwards compatibility with older hardware.  Another thing, I'm tempted to add all kind of upgrades, which will probably sink a lot of time for me (not a programmer), upgrades that will probably not be very useful, or might not fit in the memory without disabling something else.

How do you guys resist to cramming too many features?



- For my hardware version, one thing has to be addressed first is the 2 x 1N4148 diodes in series with the display (which is a 3.3V LCD powered at 4.3V, because the voltage drop of the two series diodes is too small at the quiescent current of the LCD).  When the contrast is set above 9, the LCD contrast is already showing signs of over voltage, and above 15 is unreadable, while the expected contrast range is from 0 to 63.

So either a resistor in parallel with the lcd (to increase the current through the series diodes) or maybe try to replace them with a Zener.


- Another high benefit and easy to do is to replace the 8MHz quartz with a 20MHz one.  Would like it to measure and display faster, but I would expect some loss in precision at higher ADC speed.  Not sure by how much, or if that would be visible at all in the final result.


- Maybe a worth trying HW change would be to increase the current through TL431 (reduce R15/2k2 to 1k or less).  The current now is (5V-2.5V)/2.2k = 1.1mA, out of which some current is needed for the Ref input of the TL431.  1mA is right at the lower edge of the TL431 specs.

TL431 still works at 1mA, but I remember some application notes where the recommended current for peak performance was 10mA.  That would be a lot of current for a 9V primary battery, but I am using a much bigger rechargeable NiMH pack, so I don't care about battery life.

(by the way, the measuring inconsistencies I was seeing before the firmware upgrade are no longer happening with the new firmware - didn't fully tested it, but the values seem rock solid now, with no HW changes yet)

Again, not sure if a higher current for TL431 will have any visible effect.  Are there any known results already (either positive or negative) about increasing the TL431 current?

Offline indman

  • Super Contributor
  • ***
  • Posts: 1214
  • Country: by
Re: $20 LCR ESR Transistor checker project
« Reply #9914 on: January 22, 2025, 08:21:49 am »
- For my hardware version, one thing has to be addressed first is the 2 x 1N4148 diodes in series with the display (which is a 3.3V LCD powered at 4.3V, because the voltage drop of the two series diodes is too small at the quiescent current of the LCD). 
The best option would be to install a separate 3.3V LDO to power the display instead of 2 1N4148 diodes. This is done in more competent clone designs.

- Another high benefit and easy to do is to replace the 8MHz quartz with a 20MHz one.  Would like it to measure and display faster, but I would expect some loss in precision at higher ADC speed.  Not sure by how much, or if that would be visible at all in the final result.
The design and circuitry of your clone is simplified and far from optimal, so I do not recommend changing the clock frequency above 16 MHz. Many owners of similar clones note unstable operation at higher frequencies. Better pay attention to those improvements to this clone, which I marked in red on my diagram. This will bring much more benefits than replacing quartz.

- Maybe a worth trying HW change would be to increase the current through TL431 (reduce R15/2k2 to 1k or less).  The current now is (5V-2.5V)/2.2k = 1.1mA, out of which some current is needed for the Ref input of the TL431.  1mA is right at the lower edge of the TL431 specs.
As I wrote above, it is better for you to avoid using the TL431 altogether. The best option is to install a high-quality LDO MCP1702-5.0 instead of 78L05

 

Online RoGeorge

  • Super Contributor
  • ***
  • Posts: 7070
  • Country: ro
Re: $20 LCR ESR Transistor checker project
« Reply #9915 on: January 22, 2025, 10:38:19 am »
I've noticed the changes in red, only that I'm not convinced they are necessary.  Here's what my doubts are about:

- the additional 0.1uF at the input of the 71L05 is not needed, particularly when the power source comes from a battery.  Adding a 0.1uF is a thumb rule, but the existing C1/1uF MLCC already has very low ESR, another 0.1uF won't help.

- 10uF instead of 1uF (the cange of C1/C3) might get the LCO to behave even worst, particularly the C3.  I've looked yesterday in the datasheet of a few 78L05 manufacturers, they recommend smaller C1/C3 values, and they don't use a second 0.1uF neither at input, nor at the output.

The thumb rule of adding 0.1uF decoupling does not apply to voltage regulators, in general.  In fact, for some other regulator models (not 78L05), adding a 0.1uF MLCC might turn them into oscillators, because MLCC capacitors have very low ESR.

- same for the C2/0.1uF at the output (existing in the unmodified version), not needed.  C2 should have been placed near the ATmega Vcc pin on the PCB, and unfortunately it was placed near the 78L05 instead.  I guess things got mixed during the PCB design.  This comment is not about the changes in red, but about the misplacing of the C2 on the PCB design.

- because of the above PCB design pitfall, the C* you added should be beneficial, I'll add those too, just that I didn't mentioned them as critical in my previous post

- the R* of 2k7 seem too big.  I guess the reasoning of adding them was to limit the maximum current into the LCD pins, but the R* combined with the input capacitance of the LCD pin will make a low pass filter (will damage the edges of the digital signals).  For custom builds with a longer ribbon cable to the LCD, the cable capacitance will add to the parasitic C of the LCD inputs, too.  Maybe the 2k7 R* + the C_LCD_input + C_wires might be too much for faster SPI bit-bang at 20MHz sometimes (only a guess, I didn't test it).

Found it, https://www.displayfuture.com/Display/datasheet/controller/ST7565.pdf the LCD pin capacity for mine is 8pF (page 59 of 72).  Corner frequency will be about 7MHz, but to pass nice square edges through it, the bit-bang will gave to be about 10 times slower.  I would expect problems if the equivalent bit-bang SPI clock goes higher than 500kHz.

Anyway, even without any series resistor, the current from an ATmega DO at 5V to a 3.3V+0.6V LCD input clamp diode probably won't be more than a mA or so but I didn't measure.  A few mA will be safe to handle by any input clamp diodes the LCD might have on its inputs.  My guess is, the real reason for no voltage level adapter is because the LCD inputs are in fact 5V tolerant (but at a first glance can't see 5V tolerant inputs mention in the datasheet, so it might not be).

- about the second C*/0.1uF at the AVCC (by the way, it's a good idea to always number or name any components on a schematic, so to easily refer to them later).  As opposed to the 0.1uF near Vcc (where it is added to keep the fast current edges inside the MCU), the 0.1uF is to keep out any incoming spikes, so it will be more effective with a series ferrite bead, or a shunt L, or even a small resistor between the Vcc and the 0.1uF+AVCC, to form a low pass filter.

On the VCC pin, the MCU don't care much about outside noises, so the 0.1uF role is to minimize the escape of the internal RF produced spikes into outside, into the rest of the PCB.  It keeps the inside MCU noises inside the MCU.

On the AVCC pin, the role of the 0.1uF is to keep the outside noises outside of the MCU, so to not influence the ADC and the other blocks sensitive to analog noise.  In fact, the datasheet recommends a low pass filter, not just a decoupling 0.1uF.  A ferrite bead, or a small R in series between the 5V and the 0.1uF at the AVCC should help even more.



Of course, these are generic comments, and I didn't try them in practice.  They might not apply much when the manufacturer sources its BOM based on lowest price alone.  So, if you observe better performance with the changes in red, definitely keep them.
« Last Edit: January 22, 2025, 10:51:08 am by RoGeorge »
 

Offline indman

  • Super Contributor
  • ***
  • Posts: 1214
  • Country: by
Re: $20 LCR ESR Transistor checker project
« Reply #9916 on: January 22, 2025, 10:59:39 am »
They might not apply much when the manufacturer sources its BOM based on lowest price alone.  So, if you observe better performance with the changes in red, definitely keep them.
You have a lot of unnecessary text and theoretical assumptions that you haven't tested. And I have many years of experience in using these devices and I have not just highlighted in red color changes that really correct some of the shortcomings of the circuitry. And they are naturally proven in practice by many people. Whether or not to apply them is up to you. ;)
 

Offline Yuriy_K

  • Regular Contributor
  • *
  • Posts: 188
  • Country: ru
Re: $20 LCR ESR Transistor checker project
« Reply #9917 on: January 23, 2025, 08:46:17 am »
I will add for greater clarity. The shown capacitors will be measured on different TTesters assembled according to the T4 circuit shown above. For comparison, the readings of the ESR NJ100S at a frequency of 30 kHz are closest to the readings of the k-firmware. Each of the TTesters was adjusted by the shown capacitors and additional inductance in AVcc and programmatically - FAKTOR_ESR . And let the gentlemen "theoreticians" try to get similar results...
« Last Edit: January 23, 2025, 08:50:33 am by Yuriy_K »
 

Offline mentholflash

  • Contributor
  • Posts: 38
  • Country: au
Re: $20 LCR ESR Transistor checker project
« Reply #9918 on: February 03, 2025, 06:20:44 am »
Hi all
Just wandering if this could be used to create a ttester?
it was $9
which is the reason for the question
Thanks
 

Offline hapless

  • Regular Contributor
  • *
  • Posts: 204
  • Country: us
Re: $20 LCR ESR Transistor checker project
« Reply #9919 on: February 04, 2025, 01:55:10 am »
What do you mean by"conclusions"? Broken out I/O points?

Hi all
Just wandering if this could be used to create a ttester?
it was $9
which is the reason for the question
Thanks

Looks like a simple enough breakout board for the chip. Without a schematic for that thing, I think that removing R2 is going to make it work as the heart of the tester. But there may be something else on the board that needs to be modified, especially if there are more components on the reverse..
 

Offline snapper

  • Contributor
  • Posts: 46
  • Country: de
Re: $20 LCR ESR Transistor checker project
« Reply #9920 on: February 04, 2025, 03:57:17 am »
« Last Edit: February 04, 2025, 07:36:33 am by snapper »
 

Offline hapless

  • Regular Contributor
  • *
  • Posts: 204
  • Country: us
Re: $20 LCR ESR Transistor checker project
« Reply #9921 on: February 04, 2025, 04:44:32 am »
Thanks for that. Guess it also needs AREF (pin 20?) to be decoupled from VCC and have a cap on it. Someone more eagle-eyed than I may find something else, but I don't see anything...
 

Offline masterpogo

  • Newbie
  • Posts: 2
  • Country: pl
Re: $20 LCR ESR Transistor checker project
« Reply #9922 on: February 10, 2025, 01:08:13 am »
Does anyone have the firmware from the post:
Reply #7879 on: September 22, 2022 and could share it? I don't have 10 posts yet to ask the user in PM and there is only such an option. This is New modification English k-firmware for GM328A(AY-AT) based on the current 899th revision 1.13K.Quartz 8 or 16 MHz.Font 8x15.
THX.
 

Offline CaptainBucko

  • Regular Contributor
  • *
  • Posts: 53
  • Country: au
Re: $20 LCR ESR Transistor checker project
« Reply #9923 on: Yesterday at 12:41:17 am »
Does anyone have the firmware from the post:
You may be better off installing the required software to compile it yourself.
Install WSL here with Ubuntu 22.04: https://learn.microsoft.com/en-us/windows/wsl/install
Install avr-gcc and downgrade: https://github.com/carl1961/Transistor-Tester-AVR-Compiled-files-On-Windows
Then you have total flexibility for all your needs.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf