$20 LCR ESR Transistor checker project

[indman]

Vitaly_Ne,at the beginning you correctly wrote the message in English. This is an English-language forum.

[Vitaly_Ne]

Hello.
Sorry, I fixed it.

[Yuriy_K]

Hello @Yuriy_K.
Your firmware came to my copy of the tester (on page 277 - ST7735_BGR_en_1).
Everything was installed without any problems. only there are no points:
- Voltage
- Front color
- Back color
- C + ESR @ TP1: 3

But this is exactly what I would like to have in the firmware - are such fixes possible?

Added to the menu - Voltage, memory at the limit. Items - Front color, - Back color removed as unnecessary. Item - C + ESR @ TP1: 3 is not needed, since capacitance measurement at similar inputs duplicated this item.

[stani9898]

Hi, i bought the LCR-T4 tester but unfortunally has arrived with a broken lcd . So i have think to utilize the tester by reading the values on a pc, for example with the serial monitor. I  want know also if i can reprogram the tester in order to use an 16x2 lcd display (1602a). I don't know how to do that, so if you have any tips.. . Thank you very much 

[indman]

stani9898,there were a lot of people with that problem. Here is one of the solutions. In the include manual you will find the pins for the correct connection of the 1602 display to the ATMega328.

[stani9898]

Is it only necessary to connect the pins correctly,as in the pdf? Or i have to modify also the hex file? Thanks

[indman]

Is it only necessary to connect the pins correctly,as in the pdf?

Just that!

[Vitaly_Ne]

Thank you very much!
Everything works fine on this firmware.

Probes (15cm with clamps at the ends), capacitors 25nF and 145nF for calibration are the same (film) as previously used (on V1.12), no changes in the type of soldering or replacement of parts were made (only a 6-pin cable is soldered for firmware).

1. What has changed (compared to the original firmware 1.12):
- the supply voltage shows more accurately (8.263V multimeter RM219 - 8.28V device) +/- 0.03V which is fine, it used to be +/- 0.2V
- subjective - it looks prettier in color)
- offhand measurements of capacitors, resistors and transistors are about the same (more precisely, then I will compare with old records)
- measurement speed is approximately the same

2. What raises questions (for me personally, although maybe I still don't understand something or haven't figured it out to the end):
- The ESR of many capacitors has become .00, although it used to show 0.03 - 0.07 .... for those that showed up to 0.2-0.8 or more, it also became almost half less ...
- when calibrating in the first steps, it shows R .00 .00 .00 although this may be the influence of the wires of the probes ... but should it be shorter with jumpers? I will understand and try))), although it measures the resistances relatively accurately (compared to RM219)

3. What I would like (if of course possible):
- completely remove auto shutdown, or make it as long as possible, or what delays to set so that it does not turn off itself ... to turn it on and until the battery is completely discharged))))
- a little bit even brighter to make the inscriptions ... otherwise it seems pale to me.

p.s. I recently had a tester, so don't judge strictly, I'm still learning to work with him)))

[Yuriy_K]

p.s. I recently had a tester, so don't judge strictly, I'm still learning to work with him)))

Take out all your amateur performance if you want to have an accurate meter, not a show meter. Carry out all measurements on the standard connector, unsolder the wires for firmware, leave the pins for connecting an external cable at most. Calibrate with short shorting contacts. R0 must be different from 0.00. All the settings were made just for such measurement conditions. Read about the brightness setting on the page where the firmware -277 is published ...
Added a photo as an example of measuring SMD inductance 100 nH.

[stani9898]

Thank you very much!! It works 

This is the guide for flasing the firmware.
https://www.instructables.com/TransistorTester-Firmware-Flash-With-Arduino/

[madires]

@madires, @indman: you've been PM'd.

I've placed both FAQs in the repo: https://github.com/madires/Transistortester-Warehouse/tree/master/Documentation/English

[Vitaly_Ne]

I will take into account the comments, thanks.
- I read about increasing the brightness by replacing the resistances. I shorted the resistor on the board (and also on the display) with tweezers - yes, it works. I just thought, maybe you can just fix something in the settings WITHOUT soldering to increase the contrast and brightness ... but since it is impossible, then the question is closed.

- about resistance: I soldered 3 pins from the comb (I put heat shrinkage for easy pulling out).
And .... the result is the same ... R .00 .00 .00, although measured on the same tester just as resistance shows 0.1 ohm for both.
I measured 0.08 and 0.09 ohms with a multimeter.

I made a calibration and measurements with probes, and then a calibration with a jumper and measurements directly in the ZIF-panel and ...
- there is practically no difference...

[indman]

- about resistance: I soldered 3 pins from the comb (I put heat shrinkage for easy pulling out).
And .... the result is the same ... R .00 .00 .00, although measured on the same tester just as resistance shows 0.1 ohm for both.
I measured 0.08 and 0.09 ohms with a multimeter.

Replace the Chinese +5V stabilizer with a quality one and compare the calibration results!

[Yuriy_K]

I will take into account the comments, thanks.

I made a calibration and measurements with probes, and then a calibration with a jumper and measurements directly in the ZIF-panel and ...
- there is practically no difference...

I will repeat again, remove all amateur activities from the board. I pointed to the firmware cable in red. The maximum that can be left on the board without compromising work was shown by an example on my Hiland, shown by a green arrow ...

[Vitaly_Ne]

...I pointed to the firmware cable in red. ...

I never would have thought that these 6-wires for firmware somehow affect the readings ... they then just "hang" in the air ...
Their length is about 10cm, they are not connected to each other ... strange. Is the device really so "capricious" that pickups from wires (which are not connected anywhere) can affect?
But I will try - remove them.
I still see that you can't do without a soldering iron ... shorten the resistance for brightness, remove these wires, and they also advised to replace the 5V stabilizer ...
Thanks for the tips, I will try ...

[Vitaly_Ne]

Replace the Chinese +5V stabilizer with a quality one and compare the calibration results!

Thanks for the advice. Now the stabilizer gives out 5.04V
Do I need to look for a stabilizer that gives exactly 5.00V? Can you tell me the model? It's just that it will probably be difficult to pick up to hundredths of values ... after all, the microcircuits still have a spread.
By the way, Yuriy_K in the screenshot above showed 4.94V, it turns out you need to select a stabilizer, not higher than 5.00V?

[N9XYP]

Added a photo as an example of measuring SMD inductance 100 nH.

What are you using for the contacts on your jig?  It looks really nice. I have a piece of pcb, some craft sticks, pointy bamboo skewer, and a rubber band 

[Yuriy_K]

What are you using for the contacts on your jig?  It looks really nice. I have a piece of pcb, some craft sticks, pointy bamboo skewer, and a rubber band 

I use spring contacts from very ancient connectors. It is possible, if you have some locksmith skills, to use the contacts from the mains socket. These contacts have a large thickness, it is necessary to reduce the thickness on the anvil by forging. And cut strips 2 mm wide

[Feliciano]

I wonder whether it would be possible to add 8x10 font to devices with graphic display (not sure how it would look, and what space the bitmaps would take), anyways as starting pont I decided to draw the std 8x8 characters in order to understand how the rendering works. While doing so I found a little detail I think it can be improved, it's about the single quotes. AFAIK there is "'", "´", and "`" (i.e. tilted to the left, vertical, tilted to the right). On the font header file I read:

Code: [Select]

0x0C,0x0C,0x04,0x08,0x00,0x00,0x00,0x00,   /* 0x17 ´ */
.
.
.
0x0C,0x0C,0x04,0x08,0x00,0x00,0x00,0x00,   /* 0x50 ` */Being currently the same picture for both symbols, I think it could be changed to something like:

Code: [Select]

0x30,0x30,0x20,0x10,0x00,0x00,0x00,0x00,   /* 0x17 ´ */

Edit:
Additionally, shouldn't the lowercase "q" be shifted 1column to the left as well?

Code: [Select]

0x00,0xFC,0xC6,0xC6,0xC6,0xE6,0xDC,0xC0,   /* 0x61 q */ <- original
0x00,0x7E,0x63,0x63,0x63,0x73,0x6E,0x60,   /* 0x61 q */ <- shouldn't be aligned to the left as others?

I attach pictures of what I'm talking about.

I leave my suggestions for your kind consideration.

[madires]

An 8x10 font is possible for graphic displays which don't use banks, e.g. ILI9xxx or ST7735. If you create any new font files or improve current fonts please send me a copy and I'll add them.

[Feliciano]

Thanks for the info, it would be the same for a 6x10 font, right? My current candidates, are a couple of ST7565, and a ST7735, and for what I've tried today, the 6x10 looks easier to draw.

I attach the improved font I was looking as a reference.

Regards,

[Yuriy_K]

I leave my suggestions for your kind consideration.

Authors use fonts with different orientations and sizes for different displays. To adjust and create your own fonts, you need to use different graphic editors. Here are examples of such work ...

[Feliciano]

Yes. Please don't get me wrong. I'm not trying to draw every character from scratch, just the needed ones and fixing the artifacts consequence of the combinations of a given source font and the tool used. In fact for optimizing that part, I'm writing a tool to allow me see the font rendering before flashing the equipment (so I can zoom in and out and judge where I need to polish more). Right now I think I will try some 6x10 as it looks easier to generate. (I'm focusing on the bitmaps first, the coding will be the second stage).

The other part of the post (the one you quoted) is about some minor details I noticed on the font I was studying. I already made some improvements and shared it for future upstream integration.

Regards,

[madires]

Thanks for the info, it would be the same for a 6x10 font, right? My current candidates, are a couple of ST7565, and a ST7735, and for what I've tried today, the 6x10 looks easier to draw.

Yep! ST7735 is fine but ST7565 uses banks/pages. The problem with banks is that they create 8 bit high bands across the display because it's the way the dots are addressed. If the height of a font matches a multiple of 8 then everything looks fine. A 6x10 font needs two banks (1 bank = 8 dots, 2 banks = 16 dots) and you'll get 6 rows of unused dots. You'll also half the number of text lines (from 8 to 4). The workaround would be to use a graphics buffer in the MCU's RAM and to copy that to the display. That's slower and we don't have sufficient RAM. The ST7735 uses an address window which can be freely set.

[Feliciano]

Dears,

Today I found an oddball: a 2N4990 SUS. The 1.13k firmware detects it as kinda the holy grail of PNP transistors ; Older cloned firmware(s) detect it as completely different things  .

Pictures attached for your notes.

Edit:
Updated to reflect the 1.44m firmware identifies as a PUT, which I think it's better than showing a PNP or two diodes.