$20 ESR LCR Transistor NPN PNP Mosfet Meter

[ScottJ]

I purchased one of the chinese boards that says "Booster edition ver 2.2 2012/11/26 Transistor - L/C ESR Tester" on ebay.
Signature bits confirms it is ATMega168.

On this unit, the fuse bits were not locked, so I was able to read flash and eeprom info and existing fuse settings.

I flashed Markus/ComponentTester-106m/  that was compiled for m168 but the LCD display seems scrambled.

Has anybody else successfully flashed this device with the Markus fork of firmware? 

I read an older PDF description of this project that is very detailed and helpful.
It suggested the Chinese clone boards need a 0.1uF capacitor on the ATMega168 for power decoupling across Vcc/Gnd.
I'll probably try that next.

EDIT:  I forgot to program new fuse bits "make fuses; sudo avrdude ...."  -- so now it's working.
   It's in continuous mode now, but I'll fix that.


thanks,
Scott

[madires]

I read an older PDF description of this project that is very detailed and helpful.
It suggested the Chinese clone boards need a 0.1uF capacitor on the ATMega168 for power decoupling across Vcc/Gnd.
I'll probably try that next.

Karl-Heinz' documentaion is always up-to-date (http://www.mikrocontroller.net/svnbrowser/transistortester/Doku/trunk/pdftex/english/). In the German forum several chaps reported success with modifying (I should say improving :-) the Chinese testers and flashing new firmware.

EDIT:  I forgot to program new fuse bits "make fuses; sudo avrdude ...."  -- so now it's working.
   It's in continuous mode now, but I'll fix that.

Simply press the test button a tad longer when powering on.

PS: The new "m" version will have an improved UI and a new diode detection logic.

[ScottJ]

Thanks for the link to the current documentation.  I checked out the SVN repository, but didn't look under Doku for the current docs.
I discovered the continuous mode and internal menu from the documentation in your branch -- I really like it!

I'm using usbtiny from adafruit.com on ubuntu so I needed to change the makefile a bit to accomodate my environment.

I like Karl-Heinz's use of serial output using PC3 (usart pin?) for output.  Do you have plans to use this?

Also - my Chinese clone board has a 2v zener for voltage reference, but I can't find any place in your code where to read the PC4 (ADC4) voltage reference.
I measured my voltage as 1.955 volts, and wanted to specify somewhere in the code for precision.  but I can't find it.
Am I missing something?

My only problem now is during the self test or Adjustment modes, when I'm told "Create Short Circuit" and then connect all three test pins together, it does not proceed.
I see in the code that it loops waiting to confirm all 3 points are shorted, but in my case it does not proceed.
I confirmed resistance from TP1,2,3 back to input pins on TQFP32 ATMega168 - all traces are connected with expected resistance.
Any ideas?

I was next going to add a debug mode during Create Short Circuit to display shorted pins.
I'll just write to line 1 LCD something like "12" if pins 1,2 are shorted. eventually it should write "1231" to show all are shorted.

The serial output mode would be useful, to write debug messages to a terminal.

Thanks again for our contribution on this project, I'm really excited to play around with this.
If I make any significant changes, I'll post my code to my github account to share.

thanks,
Scott

[madires]

Thanks for the link to the current documentation.  I checked out the SVN repository, but didn't look under Doku for the current docs.
I discovered the continuous mode and internal menu from the documentation in your branch -- I really like it!

Thanks!

I like Karl-Heinz's use of serial output using PC3 (usart pin?) for output.  Do you have plans to use this?

No, but it's just a short and simple piece of code.

Also - my Chinese clone board has a 2v zener for voltage reference, but I can't find any place in your code where to read the PC4 (ADC4) voltage reference.
I measured my voltage as 1.955 volts, and wanted to specify somewhere in the code for precision.  but I can't find it.
Am I missing something?

No :-) The external reference is supported by Karl-Heinz' k-version. I decided to skip that feature because the external reference barely improves the accuracy for the 168 and 328. It would be much better to have a reference switchable between 1V and 5V for the ADC. But that would break the orginal design.

My only problem now is during the self test or Adjustment modes, when I'm told "Create Short Circuit" and then connect all three test pins together, it does not proceed.
I see in the code that it loops waiting to confirm all 3 points are shorted, but in my case it does not proceed.
I confirmed resistance from TP1,2,3 back to input pins on TQFP32 ATMega168 - all traces are connected with expected resistance.
Any ideas?

The short circuit test sets up a voltage divider for two probes and expects the voltage at both probes being Vcc/2 +/- 30mV. That's done for all 3 probe combinations. You could check if Vcc is stable. Which voltage regulator is used? If you measure resistors or capacitors are the values ok or somewhat off? If you run the PWM tool and check the output with a scope does the frequency match?

I was next going to add a debug mode during Create Short Circuit to display shorted pins.
I'll just write to line 1 LCD something like "12" if pins 1,2 are shorted. eventually it should write "1231" to show all are shorted.

Add also the output of the measured voltages to give you a hint.

The serial output mode would be useful, to write debug messages to a terminal.

Please have a look at Karl-Heinz' code and add the software UART :-)

[raresvintea]

A bought a ebay kit, and it's excelent! But i want to modify to measure ESR cap in circuit! I saw some esr-meters on net that uses 2 diodes, what is the best solution and the safest. I can eventualy use a swich when i need to take esr measurements in circuits. Please help me

[madires]

A bought a ebay kit, and it's excelent! But i want to modify to measure ESR cap in circuit! I saw some esr-meters on net that uses 2 diodes, what is the best solution and the safest. I can eventualy use a swich when i need to take esr measurements in circuits. Please help me

The in-circuit ESR meters use a low voltage which is below the typical silicon threshold. That requires an opamp to amplify the low voltage at the cap and also high test currents which the TransistorTester doesn't provide. So the answer is: sorry, it won't work.

[raresvintea]

Thanks! 

[MacAttak]

Usually isn't all that hard to pull the cap from the circuit for testing... though once you've gone that far they are usually cheap enough that you might as well just drop in a brand new one no matter how far gone the ESR measures just to be sure... meh...

[madires]

BTW, 1.08m is released:
- leackage current (helpful for germanium diodes and transistors)
- inductance measurement
- several minor improvements

http://www.mikrocontroller.net/svnbrowser/transistortester/Software/Markus/

[Winston]

Any suggestions for better component test sockets than the 0.1" female header used in the Chinese clones, sockets that will more easily allow insertion of a wider range of lead diameters?

[madires]

Any suggestions for better component test sockets than the 0.1" female header used in the Chinese clones, sockets that will more easily allow insertion of a wider range of lead diameters?

I've added 4mm banana sockets and use short test leads with crocodile clips or probing clips. For SMD stuff I use one of those SMD probing tweezers.

PS: 1.09m is released (fixed pin assignment bug for MOSFETs, added IGBT detection and hFE considers leak current now).

[Winston]

I just thought of using a 18-pin DIP ZIF socket, paralleling the connections between each 5-pin cluster in the socket to reproduce the 5-pin header pinout on the PCB.  The remaining 4-pin socket pins would be grouped in adjacent pairs as T1 and T3 and paralleled:

http://www.ebay.com/itm/2Pcs-New-Universal-18-Pin-DIP-Test-ZIF-IC-Socket-/170578063963?pt=LH_DefaultDomain_0&hash=item27b73eb25b

My clone is marked Ver. 2.4 2013/06/05 and has a zener diode in the D1 position.  In the docs, this is said:

"One error is the assembly of the part "D1", which should be a precision 2.5V voltage reference. Assembled is only a zener diode. This part should be removed. You can mount a LM4040AIZ2.5 or LT1004CZ-2.5 precision voltage reference at this place. A missing voltage reference is noticed by the software, so that you must not install the voltage reference."

I don't understand the meaning of the highlighted portion.  It says that a missing voltage reference is detected by the software, so don't install one.  That doesn't make sense.

[madires]

My clone is marked Ver. 2.4 2013/06/05 and has a zener diode in the D1 position.  In the docs, this is said:

"One error is the assembly of the part "D1", which should be a precision 2.5V voltage reference. Assembled is only a zener diode. This part should be removed. You can mount a LM4040AIZ2.5 or LT1004CZ-2.5 precision voltage reference at this place. A missing voltage reference is noticed by the software, so that you must not install the voltage reference."

I don't understand the meaning of the highlighted portion.  It says that a missing voltage reference is detected by the software, so don't install one.  That doesn't make sense.

The voltage reference is optional and if you put one in it will be detected by the software. But you should remove the cheap zener because it might cause problems if the software thinks that it's a voltage reference. Or the brief version: Cheap zener != voltage reference :-)

[edavid]

Has anyone found a suitable case?

Something like this:

http://www.ebay.com/itm/140957444342

[Winston]

Has anyone found a suitable case?

Something like this:

http://www.ebay.com/itm/140957444342

Wow!  Looks like someone else thought of the zif socket idea.

I plan to use this case:

http://www.digikey.com/product-search/en?WT.z_header=search_go&lang=en&site=us&keywords=sr251-ib-nd&x=-1057&y=-51

[Macbeth]

Any suggestions for better component test sockets than the 0.1" female header used in the Chinese clones, sockets that will more easily allow insertion of a wider range of lead diameters?

I've added 4mm banana sockets and use short test leads with crocodile clips or probing clips. For SMD stuff I use one of those SMD probing tweezers.

PS: 1.09m is released (fixed pin assignment bug for MOSFETs, added IGBT detection and hFE considers leak current now).

I was intrigued by this transistor tester and fancied making one myself, then found this chinese complete kit posted. It comes in a rather naff case, but with hook probe leads and a ZIF socket, seemed a bargain to me. I've got it, but no instructions. I shorted all the leads and reset it and it goes through an intruiging cycle of self tests then reports "Test End. Version 1.07k"

I notice all your firmware versions are 1.0xm - Can I flash your latest stuff?

The great thing is the chinese box shifter actually sent me 2 of these by accident and so I don't mind fecking one up. (Hasn't asked for it back, and we do not have the luxury of free or even reasonably cheap international return delivery in the UK, so seller can whistle for it even if he finally works out his stock is -1). I will end up giving it away for free, as in gimme P&P and someone gets it for free.

[Stonent]

Where did the rest of this thread go? I could have swore it was more than 3 pages.

[KD0CAC John]

Macbeth ,
What would P&P be , I have followed this thread from the beginning and waiting for a consensus and forgot about till you post
John

I thought this thread was longer also , or maybe another parallel  

[JoeO]

Where did the rest of this thread go? I could have swore it was more than 3 pages.

There are a few threads that discuss these boards.

[madires]

I was intrigued by this transistor tester and fancied making one myself, then found this chinese complete kit posted. It comes in a rather naff case, but with hook probe leads and a ZIF socket, seemed a bargain to me. I've got it, but no instructions. I shorted all the leads and reset it and it goes through an intruiging cycle of self tests then reports "Test End. Version 1.07k"

I notice all your firmware versions are 1.0xm - Can I flash your latest stuff?

Sure, compile the latest version and flash the ATmega. The current versions are:
- 1.09k  Karl-Heinz' official release
- 1.10k  under development (trunk)
- 1.10m Markus' (my ;-) firmware

All versions are available at http://www.mikrocontroller.net/svnbrowser/transistortester/Software/ (trunk = k-firmware under development, tags = releases of k-firmware, Markus = m-firmware). The main difference between the k and m firmwares is that the k-firmware includes support for ESR measurement and the m-firmware has a simple menu system and a PWM tool.

[madires]

Where did the rest of this thread go? I could have swore it was more than 3 pages.

This one?

[tdsnet]

Hi,
i have downloaded new software 1.10m but i can't find schematic with all hardware option for this release
Where i can find it?

Thanks a lots

[madires]

i have downloaded new software 1.10m but i can't find schematic with all hardware option for this release
Where i can find it?

Please see http://www.mikrocontroller.net/svnbrowser/transistortester/Doku/trunk/pdftex/english/, click "Download GNU tarball" and extract ttester.pdf. The m-firmware supports the standard schematic, 2.5V voltage reference, protection relay and external "high" voltage for zeners.

[NikWing]

though it says 1.10k in the pdf

the pdf is very complex. it's a transistor tester that can also be used to measure coils and capacitors?

@madires: I was on the hunt for a LC-Meter that also can measure crystals, ESR etc
on eBay I found different devices, some (the ones that look kinda neat) only have Chinese text on the LCD
so I found this thread and read that they copied it from the project you're working on

what I want to do: make a portable device with a color gLCD that also shows the symbols and info like this one:


right now I'm just collecting informations for my next arduino project (I plan to put an FT232, mega328 and everything else on a PCB)

since I'm an arduino beginner: would it be possible to get some help to implement the code for the gLCD?
do you see any problems using a gLCD? (as in connection problems, not enough free space on the mega328) etc
will a step up converter be troublesome for measurement accuracy?

I'm a bit  puzzled at this point hehe

[madires]

the pdf is very complex. it's a transistor tester that can also be used to measure coils and capacitors?

Yes, but inductance is quite limited and the range for capacitance is 5/30pF up to 20mF (based on the firmware version). The Transistor Tester isn't actually designed for that purpose, it's more a "let's see what we can achieve" feature.

@madires: I was on the hunt for a LC-Meter that also can measure crystals, ESR etc
on eBay I found different devices, some (the ones that look kinda neat) only have Chinese text on the LCD
so I found this thread and read that they copied it from the project you're working on

For a LCR meter I'd use a completely different measurement method.

what I want to do: make a portable device with a color gLCD that also shows the symbols and info like this one:


right now I'm just collecting informations for my next arduino project (I plan to put an FT232, mega328 and everything else on a PCB)

The display shouldn't be any problem but it could require more I/O pins. I don't know if it would match the standard Tester circuit or if some design modifications are needed.

since I'm an arduino beginner: would it be possible to get some help to implement the code for the gLCD?
do you see any problems using a gLCD? (as in connection problems, not enough free space on the mega328) etc
will a step up converter be troublesome for measurement accuracy?

The gLCD requires more code than a standard LCD module and you might need to store a font in the flash. There are some libs for such displays. The first step is to read the datasheet :-) A boost converter works fine (there's a HAM version of the Tester running with a lithium coin battery and a boost converter).