$20 LCR ESR Transistor checker project

[timelessbeing]

He doesn't tear it apaaat, so I can't tell if it's AVR based, but it's likely unrelated. It only does L, C, and ESR.

However, at 9:00, the guy pulls out a Frejek AVR-Transistortester clone.

[Seekonk]

I have no idea what Seekonk's issue with it was.

My board was indeed a T4 version.  First cap I tested was 130pf.  It was really close, but I knew the 18 ohm ESR had no relation to anything.  I just wondered when it would start reading correctly.   People just tend to believe anything that is digital.  I have issues when they display a parameter that they know is inaccurate or has more digits than can be justified.  As a stock meter it appears you can do a lot better for just a little more money.  It looks like more effort is going into the encoder versions if you want to update the purchase.

It takes far too much effort to slog through this thread.   The unmodified performance of the various versions was buried a long time ago. Seems like that a performance comparison would a good separate thread to make a sticky out of.  I went for cost since I just needed inductance measurements at my camp.  It looks like a lot of effort has gone into modifying these.  Just suggesting people do a reality check with their standard versions.

[timelessbeing]

Tutorial "How to use the Arduino as ISP Programmer (In-System Programmer)":
https://www.arduino.cc/en/Tutorial/ArduinoISP

The attached picture documenting my Arduino Duemilanove associated with the ATmega ISP Adapter.

I used official ArduinoISP sketch and this following commands:

For writing only Flash
avrdude -c avrisp -p m328p -P COM3 -b 19200 -e -U flash:w:"TransistorTester.hex":a -U eeprom:w:"TransistorTester.eep":a

For writing only Fuses (L=F7, H=D9, E=04)
avrdude -c avrisp -p m328p -P COM3 -b 19200 -U lfuse:w:0xF7:m -U hfuse:w:0xD9:m -U efuse:w:0x04:m

For writing Flash + Fuses (L=F7, H=D9, E=04)
avrdude -c avrisp -p m328p -P COM3 -b 19200 -e -U flash:w:"TransistorTester.hex":a -U eeprom:w:"TransistorTester.eep":a -U lfuse:w:0xF7:m -U hfuse:w:0xD9:m -U efuse:w:0x04:m

Note:
For proper operation it is important to use a baud rate of 19200 and set the correct number of the communication port for Arduino.

Tom, correct me if I'm wrong, but I think you have a mistake here. I believe when using the ArduinoISP sketch to do your programming, you should use "-c arduino" parameter for AVRDUDE. The "avrisp" should be used for the official Arduino brand programmers. Maybe by chance, this method works anyway. I don't know.

[timelessbeing]

@ Seekonk
The most important hardware to look for (in terms of accuracy) are precision regulators or voltage references, and precision resistors (0.1%) on the test points. To my knowledge, no Chinese clones have regulators/references with sub 1% precision. But it's easy enough to upgrade yourself. It seems the firmware is more important, and can be easily updated on any model. The other things (graphical display, rotary encoder, ZIF socket) are sort of luxuries, and it's up to the user if they feel it's worth it. In my opinion, they are, and cost very little.

[timelessbeing]

For writing only Fuses (L=F7, H=D9, E=04)
avrdude -c avrisp -p m328p -P COM3 -b 19200 -U lfuse:w:0xF7:m -U hfuse:w:0xD9:m -U efuse:w:0x04:m

Pardon me if I'm mistaken, but I think the efuse value you give is incorrect.
According to the Atmel datasheet (Table 29-12), 4.3V Brown Out Detection is:
BOD Level 2 = unprogrammed, BOD Level 1 = programmed, BOD Level 0 = programmed, or 100
Therefore, the entire extended fuse byte would be 1111 1100 or 0xFC

EDIT: I just tested it myself and even though it doesn't follow the datasheet, the MCU doesn't seem to care if the first 5 bits are 0s or 1s. In other words, 1111 1100 and 0000 0100 do the same thing.

[tom666]

Everything is fine. Because some numerical values refer to fuses containing undefined bits. Value FC set this bits to '1' and value 04 set this undefined bits to '0'. Depending on the target device these fuse bits will be read either as '0' or '1' (may result in the verification error).

[tom666]

Tom, correct me if I'm wrong, but I think you have a mistake here. I believe when using the ArduinoISP sketch to do your programming, you should use "-c arduino" parameter for AVRDUDE. The "avrisp" should be used for the official Arduino brand programmers.

Everything should be okay because Arduino as ISP is very similar to the STK500 1.x.

[speedy29]

@tom666
Programmer arrived, processor successfully written with avrdude program.
With extreme burn program, I had problems in the schivere Fuse.

Thank you... 

P.S. I should also repair a FISH8840, you put the backup without merged, those are standard for 328P?   
And 'possible to obtain the complete string to perform a backup with avrdude?
Grazie

[tom666]

@speedy29

Attached zip file contains the original firmware for Fish8840. It is of course possible to use the k-version software from the SVN repository (directory mega328_fish8840).

Fuse settings fot M328P:
lfuse:0xf7 hfuse:0xd9 efuse:0x04 (0xfc)

To backup the firmware, you can use the following command:

Code: [Select]

avrdude -c usbasp -p m328p -B 0.5 -U flash:r:"TransistorTester.hex":i -U eeprom:r:"TransistorTester.eep":iHowever, if the contents of the MCU is locked, so you're out of luck.

[speedy29]

Thank you.
Up I had already downloaded from your old 3d, and I asked you about merged, thanks.
As for ilo backup, they are aware of the possibility to read lock.
Thank you.

[mauroh]

The only limitation I can see is the time it will take the test, but this is not an issue for my needs.
Is there something else that will require a hardware change?

Just the 30 odd Ampere test current The tester can deliver 7mA (safely).

I think I need to better detail my initial intention...
We don't need to push 30A on the capacitor to measure C and ESR
It will be fine to use 2 or 3 amps to perform the measure safely in few minutes.
I made some tests with a constant current PSU (set at 2V and 2A) and an oscilloscope, and for the purpose of my project I can verify my 350F Ultracaps easily and precisely.

Just to prove the concept of using my M12864 to automate the measure, the idea was to just have a specific firmware function to control a constant current source ,a dummy load (for the discharge) and read the values of C and ESR with the M12864.
Of course I will need to add external hardware to provide the extra current/sink, and modify the firmware to extend the measure time, but theoretically should work.

In this days I made the last mod on the case, adding a slot for the ICSP and another to easily connect to the UART, to the Frequency counter and to the PWM/square wave output (see picture)  .

Mauro

[speedy29]

@ tom666
question, but the backups you've attached it to the device with graphic display, like the one I am attaching?
Thank you.

[tom666]

@speedy29
Yes, it's backup firmware from Fish8840 tester with the graphic display.

[speedy29]

Tom, as they say in Italian, "if you're not, have to invent".   

Thank you

[tom666]

Sorry, I don't understand this Italian proverb

[bitseeker]

Perhaps it's the equivalent of the old saying, "Necessity is the mother of invention."

[Gyro]

Or more likely...'If you didn't exist, it would be necessary to invent you!' (it's a compliment  )

[bitseeker]

Ah, I can see that now. It's a great compliment indeed.

[speedy29]

Or more likely...'If you didn't exist, it would be necessary to invent you!' (it's a compliment  )

 
Yes,  It's a great compliment indeed.
Thank you

[avrank]

Please help me,
I use a yellow board t3 t4, and compile file M, but the display is still not perfect as in the picture. What's wrong ? Thankyou.

[Gandalf_Sr]

I have an older EZM Electronic Studio version bought from eBay a couple of years ago (see picture below).  It uses a Mega328 DIP IC, seems not to have a ICSP connector, and after calibration, reports software vn1.07k

I'm looking at doing any upgrades that may be worthwhile but should I just go buy one of the newer ones for $20?

I believe I can upgrade the firmware but which version is recommended and what improvements does it have?
With no ICSP, do I take the Mega328 out of its socket and program like that?
What components are worth upgrading?

Thanks

[bianchifan]

What components are worth upgrading?

0.1% Resistors onboard but no precision voltage ref, perhaps you may add one.
If there is enough space you can switch the tactile for an encoder.
If you like to test small capacities and/or inductives with actual K-Firmware you need to change the crystal for 16MHz too.
Please refer to the actual ducuments on github/mikrocontroller.net.

[madires]

Please help me,
I use a yellow board t3 t4, and compile file M, but the display is still not perfect as in the picture. What's wrong ? Thankyou.

I can read "PWM" and "ESR". The other lines seem to be the same gibberish. That could mean that the items got the same broken pointer. When you enter the PWM tool, do you get a nice list of the frequencies or more gibberish?

[indman]

Hi,frends! Karl-Heinz Kübbeler and Pieter-Tjerk de Boer added possibility of check and testing of crystal and ceramic resonators in new revision(actual 638th) of an insertion! I tested all resonators which I have and was very pleased with this new function.Definition of resonators with a frequency from 1 MHz to 13 MHz is possible!
But for its work it is necessary to establish in tester hardware crystal with a frequency of 16 MHz. It is obligatory! And also it is necessary to add in the makefile the line CFLAGS + = - DWITH_XTAL and change line OP_MHZ = 16

[avrank]

Please help me,
I use a yellow board t3 t4, and compile file M, but the display is still not perfect as in the picture. What's wrong ? Thankyou.

I can read "PWM" and "ESR". The other lines seem to be the same gibberish. That could mean that the items got the same broken pointer. When you enter the PWM tool, do you get a nice list of the frequencies or more gibberish?

When enter the PWM, text display is good, as shown below mr...