$20 LCR ESR Transistor checker project

[vindoline]

Madires, Thank you very much for the reply! It looks like my inductor measurement are right in spec then! The capacitor measurements I've made seem pretty accurate, certainly in the ballpark for the nominal value. The next thing I'll try are some transistors, FETs, etc.

[quantumvolt]

An inductor with infinite inductance is a resistor.

Since you have (absolute values) V/I = 2*PI*f*L wouldn't you say that - for frequency f greater than zero - an infinite L is an infinite effective impedance V/I - that is to say an open circuit (albeit with a 90 deg phase shift). For steady state DC, any inductor - infinitesimal or infinite - would (as I read it) behave as a zero effective impedance - a short.

[paulie]

The k-firmware goes down to 30pF and the m-firmware down to 5pF. With the current hardware design and the implied measurement methods low value measurements aren't really feasable. The oscillator method (frequency counter) or the AC generator method (U and I) are suited much better for that purpose. That's something for a Tester+ ;-)

yes, ive had good luck with "subtract 30pf" method but also working with uhf etc circuits it would be nice to get down in the range of 1 or 2 pf. the mid price meters do that w/o too much trouble so i guess low value inductors are the most desirable feature for me atm. since theres frequency counter function then maybe a simple inverter with switchable caps would do the trick. then its only a matter of some simple conversions. looks like its time to dig into the source.

[dannyf]

For steady state DC, any inductor - infinitesimal or infinite - would (as I read it) behave as a zero effective impedance - a short.

And what's left is the serial resistance of an inductor.

So an inductor with infinite inductance will block any ac signal and let pass it the dc portion of the incoming signal.

Like a capacitor, in the opposite way.

[radioFlash]

I got one of these board from eBay for $20. It's amazingly useful--I primarily use it to identify transistors and verify the pinout.  I've run some tests on it for various values resistors, capacitors, inductors and voltage drop across some diodes. Attached are the results compared to a Fluke 187 multimeter, Brymen BM869 multimeter, and DE5000 LCR meter. The highlighted green columns are the "reference" columns that other values are being compared to.

* My board is marked as v 2.4 and is stock / as delivered with the Atmega168. It is running the 1.05k software.
* Transistortester resistance readings are within about 1% from 100 ohms - 1 Mohm. Low values have higher error. It doesn't look like it reads < .7 Ohms.
* Capacitance readings were generally close to the multimeter readings, but the DE5000 generally read much lower than the other meters
* Inductance couldn't be measured for the .8 - 1.6 uH inductors --it shows .7 Ohm resistance. Values for 10-
* Diode voltage drops read higher than the multimeters, but the Transistortester was able to produce a value while the multimeters could not

Testing notes:
* Capacitor readings were done at 1kHz on the DE5000 up to 100 uF and 100Hz for 2200 and 6800 uf Capacitors
* Red Led had a measured drop of 1.77V with 1mA, 2V with 20mA
* White led had a measured drop of 2.69V with 1mA and 3.27V with 20mA

I plan on updating the firmware when I get a programmer and possibly doing some of the hardware mods as previously suggested.

[madires]

* My board is marked as v 2.4 and is stock / as delivered. It is running the 1.05k software.
* Transistortester resistance readings are within about 1% from 100 ohms - 1 Mohm. Low values have higher error. It doesn't look like it reads < .7 Ohms.

Please update to 1.10k (1.05k is about 15 months old). That version measures low value resistors down to 0.01 Ohms.

[casinada]

I have revision 2.2 of that board and comes with the 8MHz crystal. Too bad they removed it from newer revisions. Mine is bricked after I tried to update the firmware.  I tried using the MiniPro TL866A and the USBAsp but fails. I don't know if the processor is cooked or something with the fuses. I have some Atmega328P-AU to replace the Atmega168. My USBAsp has the warning: cannot set sck period. please check for usbasp firmware update problem so I'm trying to breadboard another USBAsp to program the original one. I feel that for every problem I fix I get two new ones

[gtroc71]

I have revision 2.2 of that board and comes with the 8MHz crystal. Too bad they removed it from newer revisions. Mine is bricked after I tried to update the firmware.  I tried using the MiniPro TL866A and the USBAsp but fails. I don't know if the processor is cooked or something with the fuses. I have some Atmega328P-AU to replace the Atmega168. My USBAsp has the warning: cannot set sck period. please check for usbasp firmware update problem so I'm trying to breadboard another USBAsp to program the original one. I feel that for every problem I fix I get two new ones

Hey mate I had the same sck period problem but the info at http://www.rogerclark.net/?p=702 worked for me if you have an arduino handy.
Best of luck!

[casinada]

No Arduino here, that's why I have to create a basic Arduino or a basic USBAsp to fix mine.
http://www.fischl.de/usbasp/
http://blog.apexelectrix.com/how-to-update-usbasp-firmwares/
I have seen the solutions using the Arduino.
Now is breadboard time.
Thank you

[paulie]

that warning should be ignored. it has no effect on operation. its amazing how many have bricked their boards AND usbasp programmer thinking they had to update firmware. "if it aint broke dont fix it", even if you think it is because of that stupid message.

[all_repair]

* My board is marked as v 2.4 and is stock / as delivered. It is running the 1.05k software.
* Transistortester resistance readings are within about 1% from 100 ohms - 1 Mohm. Low values have higher error. It doesn't look like it reads < .7 Ohms.

Please update to 1.10k (1.05k is about 15 months old). That version measures low value resistors down to 0.01 Ohms.

Is there a place to submit features request?  I have a french made transistor tester, but the software has not been updated.  They do have a function to test opto-coupler.   On powerup, it shall detect if the opto-adapter is attached and so shall "mode" opto coupler testing.  Giving the CTR value accordingly.  Also is it possile to test MOSFET or IGBT driver chip, and those that are opto-isolated driver?

[madires]

Is there a place to submit features request?  I have a french made transistor tester, but the software has not been updated.  They do have a function to test opto-coupler.   On powerup, it shall detect if the opto-adapter is attached and so shall "mode" opto coupler testing.  Giving the CTR value accordingly.  Also is it possile to test MOSFET or IGBT driver chip, and those that are opto-isolated driver?

We're limited to three probe leads and a maximum voltage of 5V. For the autodetection of an opto-coupler we would need 4 probes. The MOSFET/IGBT drivers create voltages around 10V which are too high for the ATmega. At the moment I don't see any way to add the features above. If you got an idea please let me know.

[Hydrawerk]

I put my AVR tester into a metal housing and I am getting wrong measured values of capacitors. What to do with that? Maybe a plastic housing would be better?

[madires]

I put my AVR tester into a metal housing and I am getting wrong measured values of capacitors. What to do with that? Maybe a plastic housing would be better?

Just for values in the pF range or also for µF? How much are the values off? Have you connected the metal case to the tester's ground or is the tester completely isolated?

[Hydrawerk]

Just for values in the pF range or also for µF? How much are the values off?

I am getting some 50µF to 300µF readouts with test wires not connected to anything or sometimes when connected to a 300nF capacitor.

Have you connected the metal case to the tester's ground or is the tester completely isolated?

Yes, the metal case is connected to the tester's ground.
Well, this is strange. The fault appears only when simple non-switched PSU is used. When I use an industrial switched PSU, the problem never appears.
Semiconductor measurement is always correct.

[madires]

Yes, the metal case is connected to the tester's ground.
Well, this is strange. The fault appears only when simple non-switched PSU is used. When I use an industrial switched PSU, the problem never appears.

Could you please remove the connection between the metal case and ground and check again.

[Hydrawerk]

I will try it later...

[intrax]

Hi guys,

I'm currently looking at : http://pages.ebay.com/link/?nav=item.view&id=171313859915
Is it any good and is the software upgradeable?

[paulie]

as mentioned the ones with graphic lcd will never be up to date with latest firmware. and dont really show more info. i also have a feeling it will be more than $11 before bidding ends. there are many out there easily impressed by bling.

[intrax]

as mentioned the ones with graphic lcd will never be up to date with latest firmware. and dont really show more info. i also have a feeling it will be more than $11 before bidding ends. there are many out there easily impressed by bling.

Thanks your reply. Maybe it's better to wait for free-electron's board (whenever that may appear?) or look for an atmega328 with a character display as has been stated earlier in this thread.
Thanks.

[paulie]

personally i think its wise to stick with something compatible with original hardware. extra stuff dont hurt but its nice to take advantage of any innovatios cooked up by m and k. ie the recently added frequency counter which has just proven invaluable to me for measuring ultra low value caps and inductors (i did have to add an mb506 and 4040).

and if you are diy type dont forget you can throw together a quickie for less than $1 in parts:

[ecat]

Moving away from the $1 range, I have a couple of MCP3424s (4 ch, 18bit ADCs + Vref) sitting here looking for a home. Do you think there would there be any benefit swapping one in place of the internal ADCs and changing the code to accommodate of course 

MCP3424 http://ww1.microchip.com/downloads/en/DeviceDoc/22088b.pdf

[madires]

Moving away from the $1 range, I have a couple of MCP3424s (4 ch, 18bit ADCs + Vref) sitting here looking for a home. Do you think there would there be any benefit swapping one in place of the internal ADCs and changing the code to accommodate of course 

For some measurements an 18 bit resolution would be nice but the MCP3424 has some issues. It's really slow at 18 bits and can only use the internal 2.048V voltage reference. Therefore it would be only usable for high resolution measurments if the voltage is below 2V and if there are no timing constraints. The tester uses also the ATmega's analog comparator feature (taking the ADC pins as inputs for the comparator) and the ESR measurement performs exact timing of the internal ADC's S&H. You could add the external ADC in parallel and use it for more precise low voltage measurements, e.g. for resistors.

[ecat]

Moving away from the $1 range, I have a couple of MCP3424s (4 ch, 18bit ADCs + Vref) sitting here looking for a home. Do you think there would there be any benefit swapping one in place of the internal ADCs and changing the code to accommodate of course 

For some measurements an 18 bit resolution would be nice but the MCP3424 has some issues. It's really slow at 18 bits and can only use the internal 2.048V voltage reference. Therefore it would be only usable for high resolution measurments if the voltage is below 2V and if there are no timing constraints. The tester uses also the ATmega's analog comparator feature (taking the ADC pins as inputs for the comparator) and the ESR measurement performs exact timing of the internal ADC's S&H. You could add the external ADC in parallel and use it for more precise low voltage measurements, e.g. for resistors.

Thanks for the info. Sounds like more trouble than it's worth with the MCP3424.

[Hydrawerk]

Yes, the metal case is connected to the tester's ground.
Well, this is strange. The fault appears only when simple non-switched PSU is used. When I use an industrial switched PSU, the problem never appears.

Could you please remove the connection between the metal case and ground and check again.

So guys, there is no problem with the metal housing. The odd readouts are caused by some interference with the 230V power supply. The problem appears less if you use a switched power supply. The AVR tester's metal case should be connected to negative pole and you can put your hand on it to ground yourself, then it will be fine.