m328 transistor tester wont powers on

[you]

hello,
i have this   transistor tester the M328 with atmega  328 chip i bought it from ebay as a kit , it worked succesfully for a about 2 years ,and then  it stops working  after  plugging 14v battery on it , stupid mistake from me.when i  tried to turn it on again it  wont  start by pressing the rotary button  i can see the  light on the  display for a half second after that nothing happen all i have is a black screen.
please i need help fixing it i have already changed the electrolyctic capacitors but still same  problem.
i ve joined 2 pictures of the m328  i took them from google just to show how my device   looks like. 
thanks in advance for any help.

[Specmaster]

Have you checked the transistors for shorts / open circuits, checked the other caps for shorts. There will be at least 1 may be 2 voltage regulators that look a normal BJT transistor, you'll need to check the part number and grab yourself a copy of the data sheet off the internet and then check and see what voltage they are providing on the outputs. Also you say that the LEDS flash on for a second and then stay off, have you checked to see if the device is drawing any power after the LEDs go off?

[AlcidePiR2]

I had the same problem.

The tester has been hanging in the shelf  for a while, and I have ordered another one in order to have a basis for comparison.

Your message make me look again at it.  No obvious part was looking bad, so after failing to copy the good working Amega328 (which is protected),  and veryfying that I had
about 5V on the VCC pin, I swap the chips and it worked.

So, at least in my case, the problem came from the ATMEGA328P chip that was corrupted.

I have a TL866a programmer, but up to now, I have no ben able to reprogram a new Atmega328P with the proper firmware to make it work again.
If some one can tell me  how to do it, I will appreciate.

[Specmaster]

You probably cannot reprogram the chip because if it has seen a massive over voltage it has been destroyed internally, hence why when you swapped it over for the good one from your new tester, it worked.

[madires]

Most likely the ATmega is toast. Based on the image your tester runs a k-firmware. Please see https://yadi.sk/d/yW8xa5NJgUo5z/M328Kit for the schematic and a precompiled k-firmware.

[AlcidePiR2]

You probably cannot reprogram the chip because if it has seen a massive over voltage it has been destroyed internally, hence why when you swapped it over for the good one from your new tester, it worked.

Thanks, but this is what I figured out. This is why I swap the chips to test.

[AlcidePiR2]

Most likely the ATmega is toast. Based on the image your tester runs a k-firmware. Please see https://yadi.sk/d/yW8xa5NJgUo5z/M328Kit for the schematic and a precompiled k-firmware.

Thanks a lot !  This will be very useful. I will  try it soon.  Indeed, I checked the  schematics and it looks like the one.

[AlcidePiR2]

@madires

I have flashed  the Atmega328P  with the Karl-Heinz-Kübbeler  8 Mhz  firmware, .hex  in code and .eep in data, in HEX.

But it did not work.

The remaining part of the hardware is fine, as if  I take  the IC from the new one, its works. I have changed nevertheless the regulator, but
the real problem was a nearly short in the old Atmega328.
And if I put the new programmed IC in the new device, it does not work.

I was  confused with the fuses settings, soi I did it both with  CKSEL3=0  ON of OFF : None worked.

Attached is the setting of the fuses I used


The quartz in the boards is 8 Mhz

Any additional hint ?

[indman]

AlcidePiR2
Do I see an ChipID FF FF FF in the shell? This should not be the case with a healthy ATMega328.

[AlcidePiR2]

AlcidePiR2
Do I see an ChipID FF FF FF in the shell? This should not be the case with a healthy ATMega328.

This is probably because I did not read the ID.
If I do it  :  I get 1E 95 0F

[indman]

Now 1E 95 0F is correct. Once again, completely erase the contents of the entire chip, then make sure that the chip id is correctly defined and re-write code hex and data intel hex (not bin). Next, carry out a complete verification of the contents of the chip memory(verify after bird).

[tooki]

I wonder why a 14V battery would kill it? These things all use a voltage regulator to produce 5V, which is what the circuit uses. Those regulators work just fine on 14V, albeit with higher dissipation. The only thing I can think of overvoltage killing would be the latching power circuit, as it’s the only part of the device that’s really exposed to the raw battery voltage.

[Specmaster]

Could it be because of the increased voltage to the regulators that there may be a momentary surge passed through while regulators are stabilising which is above the safe withstand of the chip? With the case of a 9V battery, this surge is within limits, but 14V not guaranteed? 

[madires]

Theoretically the M328Kit should survive a power supply voltage of 14V. But we don't know if the LDO is a genuine 7550 or something much cheaper from Shenzhen.

[AlcidePiR2]

Theoretically the M328Kit should survive a power supply voltage of 14V. But we don't know if the LDO is a genuine 7550 or something much cheaper from Shenzhen.

I my case,  it has always be powered by  a  9V battery.  I dont remember what could be the reason for frying the Atmega328, as it occurred two years ago.

In any case, I have replaced  the 7550 by a MCP1702 but I am still with my problem of flashing the    IC.

[AlcidePiR2]

Now 1E 95 0F is correct. Once again, completely erase the contents of the entire chip, then make sure that the chip id is correctly defined and re-write code hex and data intel hex (not bin). Next, carry out a complete verification of the contents of the chip memory(verify after bird).

I did all that. It does not work.

@madires
By the way,  on the working device, it  the version is said to be 1.12k  .  It seems that the one you send me is 1.13k.  Does this matters ?

[indman]

AlcidePiR2,
then the problem is not in the firmware. What happens when you click the Test button?
If you insert a regular Light-emitting diode into the test terminals, it flashes after pressing the Test button?
https://yadi.sk/d/yW8xa5NJgUo5z/ - this is my repository with firmware for different clones and they are tested on a lot of the same clones.

[AlcidePiR2]

AlcidePiR2,
then the problem is not in the firmware. What happens when you click the Test button?
If you insert a regular Light-emitting diode into the test terminals, it flashes after pressing the Test button?

I believe that it is the firmware for  the following reason :
I have two  similar  testers : A and B.

A is brand new and works.
B does not work.

If I take the IC from A and put it in B, then B works.
If I take the IC from B and put it in A, then A does not work.

When I start B,  the LCD lits on, and stays like that. Blank white.
The led lits very little. Much less than on  A.

[indman]

Show photos of your clones. I do not believe in miracles.

[AlcidePiR2]

In fact, the led lights. 

Here is the picture of both testers. A  ( the new one ) is the black SMD.  B ( non working IC) is  the red through hole one.
I removed the screen  for better view. ( There is no effect by permutting the screens).

[indman]

AlcidePiR2,
Well, that's all and it became clear, you have other clones than in the photo posted by: you 
Your clones are called AY-AT ColorKit or M328Kit+TFT(GM328A)
The firmware for this clone is here https://yadi.sk/d/yW8xa5NJgUo5z/

[AlcidePiR2]

AlcidePiR2,
Well, that's all and it became clear, you have other clones than in the photo posted by: you 
Your clones are called AY-AT ColorKit or M328Kit+TFT(GM328A)
The firmware for this clone is here https://yadi.sk/d/yW8xa5NJgUo5z/

Thanks, yes,it seems that there was a misunderstanding.
Now I see that the schematics is nearly the same, except  probably for the screen, and for some resistor values.
I should have checked better.

As soon as It is dowloaded, I will test the proper firmware.  Hopefully it will work.

[indman]

Here is the main topic on various clones, where they are actively discussed! There, I recently suggested testing the latest m-firmware 1.39m from a respected madires!
https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/msg3063792/#msg3063792

[AlcidePiR2]

Thanks a lot Indman and Madires.

Yes there was a misunderstanding, abut now it is working wiht the 1.13 k  firmware.

I had look to this more deeply, but severals years ago. The  thread on  tester was not so large and the clones not so numerous.
I had not touch to my  programmer for a while,  so was not that confident in its use.  I am quite pleased to see that it  works now, and that things are logical,
with good reasons for not working.

@You : most probably you have the same problem. Your IC is  burned. Check  if you draw a lot of power  from  the battery.  In my case,
even before measuring the intensity,  it was reflected by the fact that the VCC  decreased to 4.5 V ( instead of 5).

[Doctorandus_P]

With a bit of searching on github you'll find a master project with about 20 different LCD variants of this project.

But how much time do you want to spend on this?
If the uC has seen a massive overvoltage and died because of that, then the controller on the LCD is very likely also burned out. The rest of the whole circuit is just a few passives and a very nice soft power on/off switch, but even that is a few cheap transistors, which means the whole thing is beyond economical repair.

So, if the power on/off circuit works properly and your uC has 5V and still does not work, it's just game over.

I've bought a bunch of these things. They are cheap and easy development boards for any project that needs a M328 and a LCD.