$20 LCR ESR Transistor checker project

[kelchm]

Yes you can use Arduino to write the Arduino bootloader or the ay-at fw on the new chips. Search Arduino as isp

I ended up being lazy and just got a USBasp. 

Has anyone else had issues with running a 20MHz crystal? I experienced really bad display corruption (missing/inforrect characters, random dots) and general flakiness after swapping the crystal and uploading the new firmware. I then swapped in a 16MHz crystal which seems to work fine.

I'm using 1.26M.

Yes I thought it might have been from the long leads I soldered on for flashing but haven't had time to downgrade back to 16Mhz.

Sent from my SAMSUNG-SM-G900A using Tapatalk

I tried a couple different MCUs, no change. I think it's an issue with the LCD.

[perieanuo]

Yes you can use Arduino to write the Arduino bootloader or the ay-at fw on the new chips. Search Arduino as isp

I ended up being lazy and just got a USBasp. 

Has anyone else had issues with running a 20MHz crystal? I experienced really bad display corruption (missing/inforrect characters, random dots) and general flakiness after swapping the crystal and uploading the new firmware. I then swapped in a 16MHz crystal which seems to work fine.

I'm using 1.26M.

Nope. It's the crystal or maybe incorrect capacitors for him if you compiled correctly. I use 20M for longtime

Envoyé de mon HUAWEI NXT-L29 en utilisant Tapatalk

[madires]

Has anyone else had issues with running a 20MHz crystal? I experienced really bad display corruption (missing/inforrect characters, random dots) and general flakiness after swapping the crystal and uploading the new firmware. I then swapped in a 16MHz crystal which seems to work fine.

I'm using 1.26M.

Some compatibles of LCD controllers don't support the maximum clock rate of the genuine controller. In this case you can slow down the SPI write function by adding a few wdt_reset().

[kelchm]

Some compatibles of LCD controllers don't support the maximum clock rate of the genuine controller. In this case you can slow down the SPI write function by adding a few wdt_reset().

Yeah, this is probably the issue. Slowing down the SPI write would defeat the purpose of running the 20MHz crystal though (faster display updates).

I wish there was a way to have a better idea of just what you were buying on eBay/Aliexpress/etc.

[jewelie]

Hi folks 

I think this is probably the best place to put this post, apologies if not as I'm not a seasoned poster here?

After reading these posts I recently got the through-hole GM328 version of the transistor tester (red board, through hole components so I can easily hack later, with the 160x128 display and 7735 driver, and also with terminals for voltage measurement and frequency generator signal output.)  By default it's running the 1.12k firmware, so I've not bothered hacking it yet.  On this occasion I cheated and bought it ready assembled so that if it didn't work they couldn't blame me and my soldering. 

One caveat I've had playing with it is that the external voltage measurement option seems low: 8.84V reads as 8.6V, 2.5V reads as 2.42V (so Vext is ~3% off.)

I know that's still really quite a small tolerance for measurement but it seems odd to be that off given that-

• Manually measured Vcc at pin 7 of the ATmega328 is 5.01V  VS  Board identifies it as 5.02V, so basically dead on (so Vcc accuracy measured by board is only ~0.2% off)
• Manually measured Vref going into the chip at pin 27/PC4 from the TL431 is 2.492V, so again pretty accurate (so the board's Vref accuracy is only ~0.3% off)
• Manually measured voltage from my own 2.500V Vref going through the 10:1 resistor divider to chip at pin 26/PC3 is 0.250V (so the 10:1 divider etc is bang on!)

So, any ideas why the Vext voltage measurements would be low by ~3% when everything else is so accurate?


Julie
x

[madires]

Intriguing! Could you please also check higher voltages >= 12V?

[hussamaldean]

Hi,
I got the LCR T5 one and the tester is so slow that the resistance mmeasurement takes like 12 second for 1K value,  the capacitance is 10 time less than the label value, and I can't self-test it. when it reaches 14%, the device shuts down
any solution ?

[jewelie]

Intriguing! Could you please also check higher voltages >= 12V?

Will do, when I've charged up my batteries.

However, there's something else odd about it anyway - although it says 1.12k it has IR Encoder etc so it must be a slightly hacked default firmware.

[madires]

I got the LCR T5 one and the tester is so slow that the resistance mmeasurement takes like 12 second for 1K value,  the capacitance is 10 time less than the label value, and I can't self-test it. when it reaches 14%, the device shuts down
any solution ?

My guess is that the ATmega is running at 1MHz. Please check the fuse settings (external quartz crystal and 1:1 clock divider).

[madires]

However, there's something else odd about it anyway - although it says 1.12k it has IR Encoder etc so it must be a slightly hacked default firmware.

Do you mean IR receiver/decoder or rotary encoder?

[jewelie]

However, there's something else odd about it anyway - although it says 1.12k it has IR Encoder etc so it must be a slightly hacked default firmware.

Do you mean IR receiver/decoder or rotary encoder?

Good point.  It has a hardware rotary encoder.  The firmware has options for connecting IR Decoder, IR Receiver, DHT11 Temp & Humidity sensor, etc.  I thought the proper 1.12k firmware didn't have that.  It sounds like all the Banggood ones ship with the same firmware.

[hussamaldean]

I got the LCR T5 one and the tester is so slow that the resistance mmeasurement takes like 12 second for 1K value,  the capacitance is 10 time less than the label value, and I can't self-test it. when it reaches 14%, the device shuts down
any solution ?

My guess is that the ATmega is running at 1MHz. Please check the fuse settings (external quartz crystal and 1:1 clock divider).

the crystal is 8MHz and how can I check the fuses ?

[imidis]

I like AVR Dudess (http://blog.zakkemble.co.uk/avrdudess-a-gui-for-avrdude/), use either an arduino as ISP or a USBASP or similar to read the fuses on the chip. (Either solder wires to the pins on the board or put it on a breadboard if you have the supplies. 

Then AVR fuse calc  (http://www.engbedded.com/fusecalc/) to see what they are set to.

Sorry nix the breadboard as it looks like LCR t5 is surface mount, but it does look like it has connections for the ICSP header.

[hussamaldean]

I like AVR Dudess (http://blog.zakkemble.co.uk/avrdudess-a-gui-for-avrdude/), use either an arduino as ISP or a USBASP or similar to read the fuses on the chip. (Either solder wires to the pins on the board or put it on a breadboard if you have the supplies. 

Then AVR fuse calc  (http://www.engbedded.com/fusecalc/) to see what they are set to.

Sorry nix the breadboard as it looks like LCR t5 is surface mount, but it does look like it has connections for the ICSP header.

when I launch it, it says libusb0 is missing !!!!

[imidis]

Sorry, been a while. I believe zadig will install the proper libusb files

http://zadig.akeo.ie/

scroll to libusb-win32

install wcid driver

[hussamaldean]

Sorry, been a while. I believe zadig will install the proper libusb files

http://zadig.akeo.ie/

scroll to libusb-win32

install wcid driver

I managed to upload the software but I ended up with burnt UNO
and the tester doesn't work anymore

[rddube]

Hello everyone,

It's been a while since I haven't logged in to this thread. I own a transistor tester that I bought on BangGood a while ago and it has version 1.12K on it.

Is that the latest version, or is there a more recent version? I am wanting to test small capacitors 5-10pf and I wonder if there is a newer version of the firmware. If so, could you point me in the right direction for the hex and eep files?

A photo is attached of my tester.

[imidis]

I managed to upload the software but I ended up with burnt UNO
and the tester doesn't work anymore

Like literally burnt? Hook up the wires wrong?  Anything happen when power applied to either?

[hussamaldean]

I managed to upload the software but I ended up with burnt UNO
and the tester doesn't work anymore

Like literally burnt? Hook up the wires wrong?  Anything happen when power applied to either?

nop, it was the bootloader not responding and I managed to fix it
I knew now what wrong with the old one

[jewelie]

Intriguing! Could you please also check higher voltages >= 12V?

Done.

8.84V reads as 8.6V, 2.5V reads as 2.42V and 28.4V reads as 27.8V.

As I say, it seems odd given the accuracy of the Vcc voltage it measures, the 10:1 resister divider for the Vext input and the internal Vref.  Resistance measurements seem about the same amount off too.

I've another ATMega328p coming on to which I'll put the real 1.13k firmware and see what we get.  With luck I'll be able to binary-chop my way into working out where the loss of accuracy is coming from.

Regards,
Julie

[madires]

I suspected the internal bandgap reference (1.1V). But since the higher voltages are off by the same percentage it's something else. The firmware simply measures the voltage and multiplies it by the factor of the voltage divider which can be changed by editing the resistor values in the source code.

[rddube]

Hello everyone,

It's been a while since I haven't logged in to this thread. I own a transistor tester that I bought on BangGood a while ago and it has version 1.12K on it.

Is that the latest version, or is there a more recent version? I am wanting to test small capacitors 5-10pf and I wonder if there is a newer version of the firmware. If so, could you point me in the right direction for the hex and eep files?

A photo is attached of my tester.

Anyone?

[madires]

Usually you'll find the links to the firmware repo somewhere in the last ten pages Running the k-firmware you could put a cap with maybe 47pF in parallel or use the SamplingADC. The m-firmware starts measuring at 5pF.

[rddube]

Hello Madires,

Thank you for your reply. In fact, it's been a while that I wanted to try the M-firmware. Can you point me in the right direction so that I get the good version for my tester which is a 2 line tester?

I made an ISP port for it so, is it possible to get just the hex and eep files without having to compile it?

Many thanks again for your help!

[madires]

Sorry, I don't provide compiled firmware because there are too many different hardware versions. The links for the latest firmware versions:
- k-firmware: https://github.com/svn2github/transistortester/tree/master/Software/trunk
- m-firmware: https://github.com/svn2github/transistortester/tree/master/Software/Markus