$20 LCR ESR Transistor checker project

[KD0CAC John]

I wander if its a low volatage issue ?
I leave mine in a box with the battery disconnected and have no issues .
So maybe keeping track of batteries voltage and not using if below ? what ever ?

[madires]

Enter the menu and adjust the contrast. From the documentation (k-firmware):

Contrast This function is available for display controllers, which can adjust the contrast level with
software. The value can be decreased by a very short key press or left turn with the rotary
encoder. A longer key press (>0.4s) or a right turn of the rotary encoder will increase the value.
The function will be finished and the selected value will be saved nonvolatile in the EEprom
memory by a very long key press (>1.3s).

[richfiles]

Has anyone tried this (assumed to be) recent Chinese revision? They are calling it a 2016 EZM328 (GM328R) transistor tester.
http://www.ebay.com/itm/2016-EZM328-GM328R-transistor-tester-ESR-frequency-meter-square-wave-genera-/272165236396

It costs a little bit more, but the unit has a socketed DIP ATmega328, A rotary encoder/test button, graphic display, headers for frequency and flying test leads, the standard ZIF socket we've seen on most of these units, and claims to have a precision reference source. The listing says it's using "the 2016 version of the latest V1.12K software".

Has anyone purchased one of these units yet, and if so, does it really have an actual voltage reference, or just a zener? Is it worth the few bucks premium over other testers? I do like the fact that it uses through hole components (good for modding), and the new screen configuration makes it no bigger than many of the SMD versions. How hard is modding these Chinese units to support things like the input protection feature I've seen elsewhere on this thread?

I'm strongly considering purchasing one.


I'm still skimming things (81 pages is a LOT ), but I'm genuinely curious if anyone has come up with a PC board that unifies all the options that are possible. Input protections, higher zener testing, support for ATmega 644 or 1284, etc? I saw a unit with built in protection quite a few pages back, but it used the text LCD, and I just like the graphic LCD too much!

These devices remind me of a more advanced version of the transistor tester on my old DMM. It was a Radio Shack 22-174 and it had a three terminal transistor tester that did not require you to know the part or pinout, You just inserted the part any way, pressed the test button, and it told you if it was PNP or NPN, told you the pin configuration, and gave an hfe measurement. Obviously, not as advanced as this unit, as it only did those two types of transistors, but this was also made way back in the 1990s.

[madires]

Has anyone tried this (assumed to be) recent Chinese revision? They are calling it a 2016 EZM328 (GM328R) transistor tester.
http://www.ebay.com/itm/2016-EZM328-GM328R-transistor-tester-ESR-frequency-meter-square-wave-genera-/272165236396

Looks like a nice clone, but the ISP header is missing, as usual. I think it would be more easy to add the relay based protection than the SRV05-4 (SMD). Since the clone got a 2.5V reference you just need a resistor, NPN, a tiny relay (e.g. Fujitsu FTR-B3 4.5V), freewheeling diode and some wire.

I'm still skimming things (81 pages is a LOT ), but I'm genuinely curious if anyone has come up with a PC board that unifies all the options that are possible. Input protections, higher zener testing, support for ATmega 644 or 1284, etc? I saw a unit with built in protection quite a few pages back, but it used the text LCD, and I just like the graphic LCD too much!

There's no such board yet. I'm working on a modular dev board with ATmega644 and most (all?) options. If someone likes to help, it should be simple to fork a graphic LCD board and a character LCD board for users. Seems I should move from Eagle Freeware to Kicad to overcome the board size limits.

[jleg]

Has anyone tried this (assumed to be) recent Chinese revision? They are calling it a 2016 EZM328 (GM328R) transistor tester.
http://www.ebay.com/itm/2016-EZM328-GM328R-transistor-tester-ESR-frequency-meter-square-wave-genera-/272165236396

It costs a little bit more, but the unit has a socketed DIP ATmega328, A rotary encoder/test button, graphic display, headers for frequency and flying test leads, the standard ZIF socket we've seen on most of these units, and claims to have a precision reference source. The listing says it's using "the 2016 version of the latest V1.12K software".

if you don't mind a bit of soldering, i'd go for this one:
http://www.ebay.com/itm/M328-LCD-12864-DIY-Transistor-Tester-Kit-Diode-Triode-Capacitance-LCR-ESR-Meter-/161913228003?hash=item25b2c7cae3:g:pvwAAOSwf-VWZ~zL

This one does have a VRef, and 0.1% Rs - and flashing latest firmware is also possible, if you have some socket with ISP headers for the Atmega...

[richfiles]

I got no problem with soldering, but man do i hate inverted LCDs! 

[macboy]

Has anyone tried this (assumed to be) recent Chinese revision? They are calling it a 2016 EZM328 (GM328R) transistor tester.
http://www.ebay.com/itm/2016-EZM328-GM328R-transistor-tester-ESR-frequency-meter-square-wave-genera-/272165236396

It costs a little bit more, but the unit has a socketed DIP ATmega328, A rotary encoder/test button, graphic display, headers for frequency and flying test leads, the standard ZIF socket we've seen on most of these units, and claims to have a precision reference source. The listing says it's using "the 2016 version of the latest V1.12K software".

if you don't mind a bit of soldering, i'd go for this one:
http://www.ebay.com/itm/M328-LCD-12864-DIY-Transistor-Tester-Kit-Diode-Triode-Capacitance-LCR-ESR-Meter-/161913228003?hash=item25b2c7cae3:g:pvwAAOSwf-VWZ~zL

This one does have a VRef, and 0.1% Rs - and flashing latest firmware is also possible, if you have some socket with ISP headers for the Atmega...

I bought that kit from a different seller and finally assembled it last night. Worked on the first go. This is an excellent value, and is fully featured: graphic LCD, rotary encoder, ZIF testing socket, socketed DIP ATMega328 chip (easy to re-program or replace), TL431 VREF, etc.  The PCB is high quality, with good sized component holes (not too tight or loose) and pads, nice red soldermask, and with all component valued silkscreened on - I didn't use any reference other than the silkscreen to fully assemble the kit.  I didn't mind the work of soldering it together, that was half the reason I bought it, just to spend some quality time with my soldering iron.

[stj]

it would be nice if they included a programming header - or atleast the pads for it!
you dont want to have to pull the chip every time there is a software update.

[casinada]

Sometimes it is handy to have a few processors around labeled with the different firmware versions, in that case you can go back to the one it was working right

[hapless]

These devices remind me of a more advanced version of the transistor tester on my old DMM. It was a Radio Shack 22-174 and it had a three terminal transistor tester that did not require you to know the part or pinout, You just inserted the part any way, pressed the test button, and it told you if it was PNP or NPN, told you the pin configuration, and gave an hfe measurement. Obviously, not as advanced as this unit, as it only did those two types of transistors, but this was also made way back in the 1990s.

I still like it better than any other DMM despite it being the most inaccurate (even the "free" [not-so-free after shipping and taxes] Harbor Freight meter is a lot more accurate). Couldn't find another DMM with automatic pinout detection. Attaching a picture of it measuring a 2N2222 so people would see what we're talking about. The problem is that it doesn't always properly identity leads, which causes it to show the wrong type and HFE. This is what initially prompted me to look for an alternative, and how I found the device that this thread is about.


I'm curious, with the same hardware available in a case, why do so many people insist on buying it without the case? The price difference is usually not that great.

[richfiles]

Shame Radio Shack hasn't sold that model for years, nor came up with an updated model, nor continued to be a viable hobbyist business...

I hate that I popped mine. 
I was measuring current through some LED strips and it got really late... I immediately knew I screwed up when I did it... I was tired, and had the wires & knob in the wrong places.

One of (among many) of the problems with that meter, is it only does NPN and PNP type transistors. Doesn't handle FET of any kind. These Atmel based testers are a next generation of test versatility. Given the low cost of these component testers, I'm genuinely surprised no company has tried to do the automatic component identifier/tester as a next gen meter selling gimmick. Personally, I WOULD buy a cheap crappy DMM just to have a proper ASIC based tester around for that and nothing else. Besides, if one of those testers got packed into a multimeter, Dave would have to shit on it for being a multimeter with a transistor tester! 

 As to why people buy the bare boards vs the case? I dunno. I looked at the cases, and I honestly did not like a single one's configuration. I literally thought "I could do better, and it probably won't feel like cheap and plasticky Chinese crap" With a home brew case, you also have room to add on extras, like input protection, rechargeable batteries, etc.

[SirKnowsALot]

I know I'm late to the party (typical for me), but I've read every post on this thread to get caught up.  This is an amazing community of very helpful people such as Markus, Tomas, and many others.

I bought this tester eBay auction: #http://www.ebay.com/itm/221961865013 and found it does a poor job at capacitance/inductance measurement, which is what brought me to this forum.  Based on what I read, I believe this to be a LCR-T4 12864LCD unit with unversioned Chinese firmware, although it does not match precisely any tester photo I've seen on this thread.  I hope to reflash with M- or K-firmware and solve my issues, and I think mega328_st7565_kit is the right branch, correct?

I've attached a photo of the back which shows the ISP connector I added.  One notable difference I found was that on my unit, the ISP arrangement was CORRECT.  I just plugged my USBISP with 10-to-6-pin adapter straight into my header and was able to pull the EEPROM, code, and fuse info from it.  At least I think.  I've included it here also, as it may help someone.  (Note MTester.bin->MTester.hex and MTester.eep->EEPROM.hex to make forum happy). edit: I guess I converted the code from Intel hex to binary at some point

If I rebuild code, how do I know that I have the build flags properly set?  Although this looks like a standard design, can I irreversibly damage it with improper options selected?  Could I reflash with my code+eeprom data to get back to where I started?  I thought I read somewhere that code pulled from the tester is not usable due to locking.  Since I did get something, would it not be valid?

Since I'm willing to forego the new, cool functions (rotary, func-gen, etc.), I hope this can be a smooth experience. 

Please forgive any newbie mistakes I've made.  Any of your expert advice is welcome.   

[madires]

I bought this tester eBay auction: #http://www.ebay.com/itm/221961865013 and found it does a poor job at capacitance/inductance measurement, which is what brought me to this forum.  Based on what I read, I believe this to be a LCR-T4 12864LCD unit with unversioned Chinese firmware, although it does not match precisely any tester photo I've seen on this thread.  I hope to reflash with M- or K-firmware and solve my issues, and I think mega328_st7565_kit is the right branch, correct?

Seems to be a slightly modified PCB of the T4 clone. For the k-firmware the best bet would be mega328_T3_T4_st7565 and for the m-firmware there's a display profile in the config.h file. The firmware you've downloaded from the clone looks fine. It got a lot of known strings So I'd guess, you could re-flash it.

[Rescator]

Hello,

I bought this tester Aliexpress :

http://fr.aliexpress.com/item/Hot-ESR-T4-Mega328-Digital-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-LCR/32579382733.html.

I am very disappointed because it does not recognize field effect transistors (tested: 2N3819 and MPF102) These are seen as bipolar NPN with a very high gain (between 100 000 and 510 000!)

The low value inductors (2.2 nH) appear as resistors (maybe normal ?)

Do you have the same phenomena ?

What can I do ? change firmware ?

thanks for your advices

Gerard

[madires]

I am very disappointed because it does not recognize field effect transistors (tested: 2N3819 and MPF102) These are seen as bipolar NPN with a very high gain (between 100 000 and 510 000!)

The low value inductors (2.2 nH) appear as resistors (maybe normal ?)

A 2N3819 shouldn't be any problem, but the standard inductance measurement starts at about 10µH. Unfortunately that clone runs a modified firmware, so I can't say anything about it. You could try the firmware mentioned in the post above.

[amspire]

The low value inductors (2.2 nH) appear as resistors (maybe normal ?)

2.2nH has an impedance of 0.000014 ohms at 1kHz. Yes I think not recognizing it as an inductor is very normal.

A 25mm wire loop (single turn) has a 10nH inductance. Every time you test anything, you make a loop. If it could see inductances that low, it would be seeing every resistor as an inductor.

[hapless]

[...]
I am very disappointed because it does not recognize field effect transistors (tested: 2N3819 and MPF102) These are seen as bipolar NPN with a very high gain (between 100 000 and 510 000!)

The low value inductors (2.2 nH) appear as resistors (maybe normal ?)

Do you have the same phenomena ?

What can I do ? change firmware ?

thanks for your advices

Gerard

Nobody knows exactly what's in the Chinese firmware (other than the the people who actually release it, and they don't seem to be too fond of sharing). For this reason alone I think it is a good idea to reflash with an "official" firmware.

Older firmware versions detected all inductors as resistors. Even older firmware could not read inductance value, either. This is normal. Flashing with a newer version should fix it.

You may want to try to read in and make a copy of the firmware that the tester came with first (not always possible, but worth a try). It's always good to have a copy.

[...]
If I rebuild code, how do I know that I have the build flags properly set?  Although this looks like a standard design, can I irreversibly damage it with improper options selected?  Could I reflash with my code+eeprom data to get back to where I started?  I thought I read somewhere that code pulled from the tester is not usable due to locking.  Since I did get something, would it not be valid?
[...]

Read the manual. It includes build options. Try to change what makes sense changing for your particular tester based on the description of the things that can be changed. If something doesn't work as desired, undo, rebuild, reflash.

The code you read should be valid. The problem with locking is actually being able to read it.

It would be difficult to do any damage just by flashing, but make sure your fuses don't deny SPI access and don't remap the RST pin. It is also not impossible to exceed the number of times you can write to the programmable memory, but that should not be a problem in most cases. Reflashing these devices is a pretty safe operation overall.

A 25mm wire loop (single turn) has a 10nH inductance.

It does?

[amspire]

A 25mm wire loop (single turn) has a 10nH inductance.

It does?

Yes. To get a 2.2nH inductance, that is a single turn with a 10mm diameter. Basically at that level, all wiring looks like an inductor. It is all way below the possible range of the $20 LCR circuit.

[Rescator]

Hello, thank you for all your answers.

I also have a big doubt on the firmware version installed.
When starting the tester, the version is not specified on the splash screen.

I do not know what firmware and which version to choose to replace

I study the problem before any change !

Gerard

[madires]

BTW, the optional Sampling ADC method in the current k-firmware is able to measure down to about 300nH (8MHz crystal). With a 16 MHz crystal you could push the limit further.

[SirKnowsALot]

Seems to be a slightly modified PCB of the T4 clone. For the k-firmware the best bet would be mega328_T3_T4_st7565 and for the m-firmware there's a display profile in the config.h file. The firmware you've downloaded from the clone looks fine. It got a lot of known strings So I'd guess, you could re-flash it.

I tried reflashing with 1.12k_r453 prebuilt by Tomas (https://www.mikrocontroller.net/topic/248078?goto=4120737#4105488) which turned the backlight on but nothing more.  I was able to flash the code I originally extracted back and at least it works again.  When I get some more time I will look more closely at the makefile and build my own version.  It is good to know that if you can get the original code out of the tester you can use it for disaster recovery.

[stj]

I tried reflashing with 1.12k_r453 prebuilt by Tomas (https://www.mikrocontroller.net/topic/248078?goto=4120737#4105488) which turned the backlight on but nothing more.  I was able to flash the code I originally extracted back and at least it works again.

display contrast is a mess in the makefiles,

try changing the makefile to read

CFLAGS += -DVOLUME_VALUE=50

that's the optimal setting for the GM328A tester, and it uses the same display as the T4

[enut11]

Hi All. I have a similar device that has proven to be very useful and well as accurate. I powered mine from 2 cell LiPo pack as it draws about 30mA when testing.
Unfortunately, I was testing some caps for a project when I zapped it with a 25v electro that had been sitting in the drawer for weeks, fully charged. Now my tester keeps rebooting. So sad.
I found it so useful that I ordered another one from HK at just under $20AU.
http://www.ebay.com.au/itm/322055635297?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
enut11

[hapless]

I like EZM's versions. U3 there may be able to protect the device next time you do that. But the old device may still be salvageable if you can replace the MCU.

[JacquesBBB]

In mine, I put the MCU on a zif socket. This is more easy to change the firmware, and was very useful to just  change the MCU in 10 seconds when I zapped mine.
https://www.eevblog.com/forum/beginners/how-to-blow-a-esr-meter/msg679338/#msg679338