$20 LCR ESR Transistor checker project

[indman]

The Chinese have several variants of the model T7- they use different displays. There are options with black and white screen, with color screen on ST7735 (14pin interface) and also screen on ST7735 (6pin interface). If you bought an option on 6pin, it is suitable for a firmware from TC-1, which was adapted by madires or the original firmware, a link to which I had previously offered.

[Andbro]

Or it's better to buy the TC1 than TC7 ?

Best regards

[theleakydiode]

Holy markup batman!

A very reputable electronics retailer is flogging these testers for £50, doesn't even come with an enclosure!

[timelessbeing]

yup design looks the same. No open source attribution   

[madires]

Or it's better to buy the TC1 than TC7 ?

Beware of U4 and some poorly chosen components in the power supply section.

[brunosso]

Hi to all!

I have this version of the Transistor Tester



Could you please tell me wich version i need to flash?!

Thanks

[madires]

k-firmware: try mega328_2X16_menu
m-firmware: change display to HD44780 4-bit parallel and try default settings

[madires]

Still looking for xmas presents? I have one for you! And it's free.

v1.38m
- Optional rounding of temperature value for DS18B20 (UI_ROUND_DS18B20, suggested by Obelix2007@EEVblog).
- Support for DHT11, DHT22 and compatible sensors (SW_DHTXX, thanks to indman@EEVblog and Obelix2007@EEVblog for testing).
- Added two thin cyrillic fonts (Thanks to Andrey@EEVblog).
- Changed output of BJTs to show V_BE and hFE also in case of a B-E resistor. Adapted remote commands accordingly as well.
- Added alternative Czech texts and several fonts with Czech characters (thanks to Bohu).
- Added tools for monitoring R/C/L on probes #1 and #3 (SW_MONITOR_RL and SW_MONITOR_C, suggested by indman@EEVblog).
- Added option for trigger output for event counter (suggested by Bohu).
- Updated Czech texts (thanks to Bohu).
- Added option to disable hFE measurement with common collector circuit and Rl as base resistor (NO_HFE_C_RL) to cope with some testers (issue reported by Obelix2007@EEVblog).
- Added option to output Zener voltage in high resolution (ZENER_HIGH_RES, suggested by Andbro@EEVblog).
- Improved OneWire_Probes() to minimize misdetection.
- Updated Russian texts (thanks to indman@EEVblog).
- Updated Spanish texts (thanks to pepe10000@EEVblog).

[bitseeker]

Free? Priceless!

[Messtechniker]

Just got myself two "HILAND" transistor testers with the 644 chip. One to use and one for mucking programming around. Currently updated to 1-13k 809 because I found in the SVN sutable .hex and .eep files. This editing and "make files" thing is a bit beyond me.

Thus my question. In addition to adapting to firmware 1.38m the existing Windows programs (see Transistor Tester SVN - Folder "Messtechniker"), what else might be of interest?

The rotten bit here is that the RxD and TxD ports on the 644 PA are not easily accessible.  Some direct soldering to the 644PA chip (not much SMD soldering expertise here) will be required when wanting to use the external Windows software.

[hpmaxim]

Dumb question, I've seen a lot of people talk about replacing the 680/470k resistors, or the TL431 or the voltage regulator to improve accuracy.  A resistor is a resistor.  It strikes me that the value of the resistance isn't that important, the main issue is KNOWING what the resistance is.  How hard would it be to measure the values and put them directly into the firmware rather than just assume they are the nominal value.  I looked through some code on github and it appears the resistance value is pre-defined in the header file, but only once not for each resistor.    I'd think you could do the same with the TL431 and regulator.  The other thing that could be done would be to replace the resistor with two resistors in parallel (1M + 820k) and (1200 + 1500), which would give nearly the same values (you'd have to adjust obviously) but would likely reduce error by a factor of 4, particularly if you could bin one high resistor with a low resistor.

[timelessbeing]

Indeed you can specify your resistor and vref values explicitly in the config.

[madires]

You can select three matching resistors for Rl (680\$\Omega\$) and Rh (470k\$\Omega\$) and update the default values in the header file. A single value for the test resistors Rl and Rh simplifies a lot of calculations and reduces flash usage. And you can also update the default voltage values for Vcc and the external 2.5V reference. The external voltage reference helps if you have an inaccurate voltage regulator. But with an accurate regulator like the MCP1702 the external reference is unnecessary.

[masterx81]

 :wtf:Hi! Someone please know what firmware i need to use on this tester? I have 1.11k by now...
328p mcu, ams1117 5.0 voltage regulator, graphic display...
Thanks!

[madires]

The tester seems to be the standard design. Any idea about the LCD controller? Monochrome or color display?

[masterx81]

Uhm, monochrome, i think that is a jlx 12864g, uc1701x driver

[madires]

The ST7565 driver might work, possibly mega328_st7565.

[masterx81]

Thanks! I will try it! 

[hpmaxim]

I assume ALL modern SMD film resistors are laser trimmed from the factory, and it's probably no harder to trim to 1% than 5%.  Not sure if its harder to trim to .1% or .01% -- although they certainly do charge more.  If we assume that if you buy X% resistors, they have a gaussian distribution with X% std deviation with a cut off at 1 standard deviation, I think you'd need to buy about 16-20 resistors to find 3 resistors within .1 * X% of each other.  Is my expected distribution reasonable?   Is there a certain level of accuracy beyond which you are limited by other factors? 

It's hard for me to understand why having multiple calibration values would be that painful for the code, but I'm not going to change the software.

[pierreraymondrondelle]

Beware the temperature stability: precision resistors are in a range below 25 ppm/K. Standard ones are far above.

[madires]

I assume ALL modern SMD film resistors are laser trimmed from the factory

AFAIK, thin film resistors are laser trimmed or etched. Thick film resistors are cheaper to produce (1 and 5% tolerance).

It's hard for me to understand why having multiple calibration values would be that painful for the code, but I'm not going to change the software.

But it would be the best way to understand the drawbacks.

[timelessbeing]

it's probably no harder to trim to 1% than 5%.

I would imagine that it works the same as in printing, photography, LCD display, chip fab etc.
To obtain a higher yield of low defect product, you have to produce more, and discard more, which drives up the price.

find 3 resistors within .1 * X% of each other.

By that logic you could use the cheapest carbon resistors.  But absolute value is not the only thing desired in resistors. For measurement, you also want consistency. Temperature stability, low drift, low noise etc.

[SMDfan]

Mr. Clone Ttester(indman),

I came across your archive related to LCR ESR tester. I am also trying to build a tester of my own for hobby. Do you happen to have the source codes for these testers? Thanks

SMDfan

Hi,

Clone Ttester is a Russian who has made differents firmwares for many ESR. He has many model not included in MicroKontroller.net.
The problem, it's in Russian.

See the link: https://yadi.sk/d/yW8xa5NJgUo5z

Thanks

Hi! It is my archive.
As not only Russian-speaking people use this archive, I decided to add new firmwares from k-firmware in English for different clones.
In the next several days I will add archive thus, as shown in a screenshot. You monitor changes in archive.
If dear madires to me allows to use its work, I can add to archive new compiled firmwares UI_ENGLISH for m-firmware?

[hpmaxim]

So... I just replaced the 7550 with a MCP1702 and replaced the 680 and 470k resistors with 0.1%.


Only remaining thing to do is remove the TL431?  Its firmware 1.13k.  I only ask because the TL431 is very close to the socket for the Mega328.  I think it would be easy to remove but VERY hard to solder back in, at least using a soldering iron.  Do I need to reprogram the AVR or does it autodetect the presence of the 431?

[timelessbeing]

Your answer is in the readme file