$20 LCR ESR Transistor checker project

[madires]

I would add them to the symbol (after removing the circle there's free space for them). So it would be pin designators as part of the symbol  plus the pin numbers around the symbol.

[indman]

I like that option! Although we are gradually coming to the same optimal variant of symbols representation, which is used in k-firmware

[madires]

Nothing wrong with that!

[lotec25]

The first step is to make the TC1 compatible with the OSHW firmwares. You can update the firmware of the management MCU U4 (https://github.com/atar-axis/tc1-u4) or replace it with a simple circuit (https://github.com/madires/Transistortester-Warehouse/blob/master/Hardware/TC1-Mod.pdf).

Ok i have my LCR-TC1 update to new firmware. Can someone please point me to what is needed to upgrade the hardware? The Dial Encoder etc?

Thanks

[madires]

Please see section 2.2.5 of https://github.com/kubi48/TransistorTester-documentation/blob/main/pdftex/english/ttester.pdf for adding a rotary encoder.

[Feliciano]

I would add them to the symbol (after removing the circle there's free space for them). So it would be pin designators as part of the symbol  plus the pin numbers around the symbol.

If we take a "clear enough" NPN symbol, we should distinguish B,C,E without need for more clues. Anyways the most important thing is to know which number is which termination, like 123=BCE, or 1=B, 2=C, 3=E, or something like that.

On the other hand, if we remove the icon circle, we could increase the figure_size within the same symbol_size, therefore perhaps it would be easier to identify BCE that way, or indeed the letters can be added to the symbols itself.

So, if I get correctly what you mean, as a picture worth thousand words, with the attached you can imagine some options we may have.

(PS: BTW I realize the arrow of the NPN should be a tad down-right on the currently used symbols).

[indman]

Feliciano, I've shown before how large symbols can look without a circle. Adding the B-C-E designations to them is not difficult.

[carrascoso]

Hi all.
I intend to build the "RingTester" circuit.
I want to put a female connector on the GM328 Color Kit itself, and a male connector on the "Ring tester" circuit.
In order to be able to put and remove the circuit according to my needs. And simply connect the component to be measured in DUT.
My question is:
Where the "4-Counter" and "2-Pulse IN" terminals should be connected on the GM328.
And in this case of a detachable circuit, whether or not I have to put the 680ohm resistor R1.

And if I'm not mistaken the best option for the GM328 Kit Color is to use the test leads as hardware is limited.

I ask for a little explanation and guidance.

Thanks
Happy New Year to everyone   

[madires]

The GM328A has a dedicated counter input already (PD4). I haven't tested using one of the probes as pulse output (RINGTESTER_OUT), but I think you can simply use one of the Rl pins of R_PORT. So for probe #1 you would set RINGTESTER_OUT to PB0 (same as R_RL_1), RINGTESTER_PORT to PORTB and RINGTESTER_DDR to DDRB. And you wouldn't need the additional 680 Ohms resistor.

[madires]

On the other hand, if we remove the icon circle, we could increase the figure_size within the same symbol_size, therefore perhaps it would be easier to identify BCE that way, or indeed the letters can be added to the symbols itself.

That's another reasonable idea! Let's see what other users think about a revamp of the symbols.

[Yuriy_K]

That's another reasonable idea! Let's see what other users think about a revamp of the symbols.

On my ILI9341 it turns out like this:

[TERRA Operative]

The latest OSHW firmwares: k-firmware 1.13k or m-firmware 1.45m

Ok, I'm completely lost with all this stuff. I think I'll just leave well enough alone, not really worth it to me to spend hours figuring it all out just for some cheap ebay trinket.
Thanks anyway.

[madires]

You could try the ready-to-burn k-firmware for the T4. For older models use https://github.com/kubi48/TransistorTester-source/tree/master/trunk/mega328_T3_T4_st7565, and for newer ones https://github.com/kubi48/TransistorTester-source/tree/master/trunk/mega328_T4_v2_st7565.

[TERRA Operative]

Thanks, that does make things easier... How do I know if I have a new or old version?

[madires]

Simply try the first firmware. If you don't see any output on the display flash the second one. They use different pinouts for the display.

[eti]

Does a curve tracer option exist in this sea of firmwares? That would be nice.

[bffargo]

If the graphics are clear enough on and around the symbol, some screen space and visual noise could be saved by removing the redundant line of text/glyphs then?

[madires]

Does a curve tracer option exist in this sea of firmwares? That would be nice.

No, the circuit doesn't support that.

[madires]

If the graphics are clear enough on and around the symbol, some screen space and visual noise could be saved by removing the redundant line of text/glyphs then?

We could make the classic pinout an option, but the freewheeling diode is not part of the symbol.

[Feliciano]

Does a curve tracer option exist in this sea of firmwares? That would be nice.

No, the circuit doesn't support that.

Trully the circuit "as is" doesn't support that. But what if we suppose we have enough memory -or a dedicated firmware fork is created for this purpose for the higher resolution displays-, AND we add an external RC integrator to PWM output(s), tuned to the sampling speed? That way some rough IV curves could be slowly plotted for the <6mA range (at least some guy already did that with an Arduino). What do you think?

[madires]

I know, curve tracers are a hot topic, but the Transistortester isn't designed for that. How useful would be a curve tracer with a driving capabilty of just up to 5V and 7mA? This isn't sufficient, not even for a small signal BJT like the BC548.

[indman]

Coffee is made by a coffee maker, transistor curves are drawn by a special Curve Tracer, and the device in question already has a surplus of functions and options.

[rsjsouza]

I know, curve tracers are a hot topic, but the Transistortester isn't designed for that. How useful would be a curve tracer with a driving capabilty of just up to 5V and 7mA? This isn't sufficient, not even for a small signal BJT like the BC548.

I agree with you, although a very popular curve tracer (DCA75) does not go much further than that.

[bffargo]

I am having a problem with a new GM328A (red pcb) kit. I build an identical one successfully previously, and also have a blue pcb pre-assembled with SMD working fine; including upgrading crystals and firmware on the blue board.

This new red one will not complete self test as it doesn't see a 220nf cap as > 100nf using the stock 1.12K firmware (reports 0nf during calibrate).  Putting in something very large gets it through the self test but then all caps are metered very low. (100nf is < 50nf, 220nf is < 140nf). All tests perfect in the other self-assembled red board.  Due to this it also messes up other tests like a NPN shows up as a FET or something.

I pre-tested every resistor, and retested every one between the good red pcb and the bad red pcb now in circuit. The only thing I haven't done is pull all the caps and test them.  I made sure the 3x each 680R and 470K resistors were all identical resistance before installing.

When I swap the ATMEL328P between boards the problem stays with the main board so it's not a damaged microcontroller.   Before I pull all the caps any other stuff to test out of circuit, does anyone have any ideas what I should concentrate on?

I tried multiple firmware alternatives and 3 different clock speeds to match the firmware (thus the angled crystal leads--but this all failed when I started with the original 8 MHz soldered directly as well.) and all exhibited similar calibrate and testing problems.

Thanks in advance.

[Yuriy_K]

I am having a problem with a new GM328A (red pcb) kit.

I had a similar issue when upgrading from 1.12k to 1.13k. The problem is in the 7550 and the capacitance of 10 microfarads, to the right of the crystal. Change the 7550 or match the indicated capacitance with a large ESR within> 3 ohms