Skip to main content ↑ ↓
Topic: Dangerous component/part tester (Read 301790 times) previous topic - next topic

Re: Dangerous component/part tester

Reply #345 –
[quote author="sa5bke"]I have built the circuit with free PCB from Markus_B, thanks again! I am having some problem, so I wonder if I could get a hint on what to check. The problem is that the LCD is full of 'garbage' when I push the test button. See attached image.

avrdude -c usbtiny -p m8 -U lfuse:w:0xc1:m -U hfuse:w:0xd9:m
avrdude -c usbtiny -p m8 -U flash:w:Transistortester.hex
avrdude -c usbtiny -p m8 -U eeprom:w:Transistortester.eep

- I have replaced the contrast resistors with a trimmer that is mounted below the PCB.[/quote]
I would have said that you did not flash the eeprom file. As you did that correctly I'd try flashing the files again. Maybe you get the files from an alternate source, like the zip file poster earlier in this thread.

What does it show when you measure a BJT transistor ? You should see some numbers for the Hfe, these numbers do not depend on the EEPROM contents and should show up fine.

If the contrast is not good, then you don't see anything. If you see these characters, then this means the contrast is usable.
Markus

Re: Dangerous component/part tester

Reply #346 –
What firmware are you using? The firmware I am using (I dont know about markus's) is for the Atmega328p; you may need to re-build the firmware for an atmega8 if you are using my firmware. That would forsure cause an issue; for example with the atmega328p firmware it is using PORTC which is at address 0x08(0x28) so its flipping those registers. So if you burnt that to a mega8; you are actually flipping your UCSRA register... because PORTC is at address 0x15(0x35) now. That would forsure cause you some issues... but then again I would be surprised the power on circuit would work; but I didnt check every address so perhaps the port/ddr that handles that is at the same address or what have you.

I do know I got a similure issue when I didnt flash the eeprom file with it; it looks like you did in the paste you shared but double check its correct one and is not corrupt or something similar. You might also want to put the eeprom and hex burning commands on the same command line; thats just nit picking but whatever. Also only burn the fuses once; dont do it for every upload (again nit picking; bit of a pet peeve)

EDIT:
Quote
When I power the circuit from the programmer via ISP port it first says something about battery low (because it is not connected) and then into 'Timeout!' so the LCD is actually working.

Wait, can you elaborate on this?

By power do you mean just +5v on the 5v ISP rails? or do you mean the USBTINY completely connected too the ICSP lines?

Re: Dangerous component/part tester

Reply #347 –
From the posted picture (Look a the unusual characters, and I bet the underscore is flashing) it looks like the LCD was initialized improperly.
I've made my one with the Atmega8 also(PCB credit: MickM). I had used the files from original archive. There's only small bug in capacitor value measurement there, but rest everything is fine. I attached those again.
Still learning
-Arup

Re: Dangerous component/part tester

Reply #348 –
[quote author="BrentBXR"]
By power do you mean just +5v on the 5v ISP rails? or do you mean the USBTINY completely connected too the ICSP lines?[/quote]

I just connect the USBTINY to the 6 pin programming port and the circuit seems to come to live (naturally as it will get power directly from the USBTINY and bypass the 'test button circuit').

[quote author="arupbsk"]From the posted picture (Look a the unusual characters, and I bet the underscore is flashing) it looks like the LCD was initialized improperly.
I've made my one with the Atmega8 also(PCB credit: MickM). I had used the files from original archive. There's only small bug in capacitor value measurement there, but rest everything is fine. I attached those again.[/quote]

I was using the original files first downloaded from a webpage where it built a custom archive of what my choices were. This morning I tried Arupbsk files and now I ran the avrdude in one line

Code: [Select]
avrdude -c usbtiny -p m8 -U flash:w:TransistorTestNew.hex -U eeprom:w:TransistorTestNew_English.eep

but unfortunetly the same issue when I power it up with battery and push the button. One new thing noted though: now it did not say anything about Battery when I again reattached the programmer, only a '0' (if I remember correctly as it was only a short moment on the screen) and then it went to 'Timeout!' and stayed there.

Thanks for all the the suggestions. I only had a short moment this morning before work to try Arupbsk files. I will try Markus suggestion later tonight or other ideas you might have. I forget to check if the underscore was flashing.

BTW, I did also test the LCD with the bus pirate yesterday and it seems ok.

Eric

Re: Dangerous component/part tester

Reply #349 –
Hi all,

I finally managed to get enough time and assemble the component tester. Special thanks to the PCB Markus gave me for free!. I all has gone smoothly, checking out the R9/R10 resistors swap and so on. But when I complete it and power it on, nothing shows on the LCD! Luckily I suspect it has something to do with the contrast. So I checked with the tester and I measured 2.7V at LCD pin 3 (Vcon).

That was wrong, since I read in the forums that the correct voltage for Vcon should be 1V aprox. I then installed a 10K pot between Vcon and GND and find the correct value at 2.7Kohm more or less. Since I dind't have such resistor available, I did put a 2k2 resistor at R8, removing R12.

So here I have a begginners question. I know that each LCD comes with different contrast voltage needs, but what I've found in other schematics involving the HD44780 LCD is that there is usually a single resistor between Vcon and GND. Why in this particular schematic are two 4k7 resistors between Vcc and GND?

Finally there are a couple of pictures of my finished project:


Re: Dangerous component/part tester

Reply #350 –
The original used a trimpot, so this board is a little different. I'm guessing the two places for a resistor is to be able to parallel a couple of resistors in order to get a value which best suits the contrast of the particular LCD.

Re: Dangerous component/part tester

Reply #351 –
Oops, I might have read your question incorrectly sorry, I was incorrectly thinking they were in parallel, but sounds like you're saying there is one to Vcc and one to 0V. If that's the case it's probably just the particular LCD needs to be biased to a particular voltage and so he's using a voltage divider (which is what the trimpot would do). Some LCD's may have an internal circuitry that only requires a pull down to set the voltage, though I haven't seen this myself. Most of the circuits I've seen just use a trimpot.

Re: Dangerous component/part tester

Reply #352 –
I actually think I should have provided space/flexibility for a trimmpot or two resistors. I don't remember how my LCD does it, but I think I could connect the contrast pin to ground.

There is always room for improvement, maybe for a Rev 2 !
Markus

Re: Dangerous component/part tester

Reply #353 –
Nice.
Still learning
-Arup

Re: Dangerous component/part tester

Reply #354 –
I still cant get mine to work off of the battery; but I never did replace the transistors; tho seems an easy task. I would just clip the heads off the transistors; heat the bottom of each three legs and pull with some pliers. then wick it up and replace; then I dont have to take off the LCD to get to the bottom...

but I like running it off 5v because it doesnt die and I can read the results and test things better. The jumper code I added (PB6&Ground I think) disables the battery checking; and the source knows its in 5v mode works really well. When you startup in 5v mode the welcome screen even lets you know by showing 'BAT' or 'PWR' on the top right.

I have not made any changes in a while (check the SVN change log; its been a while)

Re: Dangerous component/part tester

Reply #355 –
This is a fantastic little project here - I've always wanted something that does just this to identify unknown parts.

I took the eagle files from the code.google.com site, and toner etched by own (double sided) board the other day, and have a working component tester! I used a mega168 that I had lying around. I plan on tweaking the code a little bit (could do with a tidyup) and putting it in a proper case.

[attachment=1]

[attachment=0]

Re: Dangerous component/part tester

Reply #359 –
I built my own version of this handy tool last weekend. I replaced the linear regulator with a small boost-converter. Now the circuit can be powered with two AA batteries.
The PCB is sandwiched between the battery holder and LCD, so to provide a handy device without need for a case. I had a lot of fun designing and building the circuit, so thank you DP for highlighting this project.