$20 LCR ESR Transistor checker project

[all_repair]

It's really a quite beautiful case and PCB!
It's currently running firmware 1.11k and it was not locked so was possible to save it unlike most others I've seen.

I had an idea to also fit a ATtiny44 SMD PCB inside for easy programming.
It fits great but I realize a problem: it interferes with the sense resistors and thus can't be in circuit while measuring.
So now I'm thinking to find a 5v relay to drive via the USB port too, so it switches the 4 signal wires over. (MOSI, MISO, SCK & Reset)
Currently considering Panasonic S4-5V. Will see how that goes, or anyone have some better solution?

Thank you for your firmware.  I have recovered my set.  I tried to program in-circuit by adding header pins for the MOSI, MISO and SCK signal which sit nicely on the R18, R19 and R20 respectively.  I just tap the Vcc from J6.V, Gnd from J6.G, and reset from J6.R.  But this did not work, with or without R18, R19 and R20 in, and with the LCD ribbon out.  I am not sure what is the problem.
 

[Seabird]

Thank you Indman.  That worked a treat.  Much appreciated as I thought I would have to buy another unit (and be more careful)!

[Per Hansson]

Thank you for your firmware.  I have recovered my set.  I tried to program in-circuit by adding header pins for the MOSI, MISO and SCK signal which sit nicely on the R18, R19 and R20 respectively.  I just tap the Vcc from J6.V, Gnd from J6.G, and reset from J6.R.  But this did not work, with or without R18, R19 and R20 in, and with the LCD ribbon out.  I am not sure what is the problem.

Strange, I did something very similar to read it out in circuit:
Cables tapped into R18, R19 & R20, I took reset at R2 closest to the Atmega, but I see you are correct that it's a direct connection to that J6-R point.
But then I used power on device: so I wired up ground to the battery ground and powered on the meter with it's own test button...

Later I also tried it with the small ATtiny44 PCB, the problem I have is that it can't be in circuit, so need to switch it out with a relay or mechanical switch.
That worked too with the caveat as described...

[Lenny]

I have uploaded the newest "Markus Firmware". Everything works fine...

But where can I see if the reference voltage is used - I have retrofittet an LM4040 and would like to know if it works.

Is it possible to stop refreshing after a proper value is read?

The tester is a very useful tool... thanks to the developers!

Lenny

[madires]

Select "show values" in the main menu to display the adjustment values and check if Vcc has changed. There's no dedicated indication that the external reference is used. Have you enabled HW_REF25 in config.h?

[Lenny]

Thanks for the quick respone, I will try the days - the settinngs I have made.

Do you know by chance where this is descriped with the repetitions?

[madires]

Do you mean the auto-hold mode? Start the tester with a long key press (>300ms).

[rwgast_lowlevellogicdesin]

Ok I replaced my blown up TC1 with an AY YT kit with color LCD. I seem to be having some subtle problems and Im not sure if there normal.

#1 The display always says my battery is weak @7.3v or lower even with a brand new 9 volt battery. I measured the vRef at 2.493 which seems normal enough, but the voltage regulator is putting out like 5.8v or something near that.. could that have something to do with why the tester thinks the 9v is always low?

#2 The LCD background is flickering, like there are white refresh lines moving around, no amount of re-flowing the cheap LCD module helped. It is only the background that flickers though not the foreground text, could this just be an issue with the 8mhz crystal?

#3 The tester reports the ESR of every cap at 0.00, unless it is an extremely bad cap which has an ESR of multiple ohms.

Lastly where do I find the newest trendy firm ware hex and fuse settings? I looked through a link to an SVN mentioned in this thread but the newest version didnt seem to match up with the version # posted in this thread and I still have no idea what fuse settings I need to use.

[Willem52]

The output of the LDO (voltage regulator) is too high!
All problems could be related just to that fact.

The output is expected to be like 4.95 - 5.05 VDC.

The fuses:
lfuse: 0xf7   hfuse: 0xd9   efuse: 0x04

[mebel]

In the readme with software 1.31m it is witten

The external 2.5V voltage reference should be only enabled if it's at least
10 times more precise than the voltage regulator. Otherwise it would make
the results worse. If you're using a MCP1702 with a typical tolerance of
0.4% as voltage regulator you really don't need a 2.5V voltage reference.

Does it mean that if I change voltage regulator to MCP1702-5002E/TO  (0,4%) then I shuldn't change reference voltage to any with 0,1% precision ?

[madires]

Yep, with a MCP1702-5002 you don't need any external voltage reference, unless you happen to have one with a tolerance of <= 0.05%. It would make more sense to measure the output voltage of the LDO with a 6-1/2 digit (or better) DMM and change UREF_VCC in config.h.

[mebel]

 So, what to do better - change voltage regulator to 0,4% or reference voltage to 0,1% ?

[madires]

I'd go for the 0.4% voltage regulator.

[robca]

What do people use to program a 2017 2578ay-at? Remove the Atmega328p and program it separately (which would be easy enough, but removing the 28DIP Atmega frequently risks causing problems)? Or would connecting to R3-1 (RST), R10-17 (MOSI), R11-18 (MISO), R12-19 (SCK) and VCC/GND work in circuit? If the latter, do you power the Atmega thru the USBASP connector, or connect GND and use a separate power for the AY AT?

Since the kit uses thru-the-hole components, putting a clip in the right resistor leg should work fine, and I wouldn't need to solder a more permanent ISP

I have programmed all types of Atmega328p both in circuit and standalone, just wondering what's the best option for this particular board (since I'm not sure how the other components could influence the in-circuit flashing)

Thanks in advance (and thanks to all the info provided in the thread, which I just finished reading start to end...)

[mebel]

I'd go for the 0.4% voltage regulator.

So if I mount 0,4% voltage regulator and leave chinese TL431 then I must disable the external reference by removing it or undefining HW_REF25 to get better results?

Thank you for help.

[madires]

Exactly! Either comment out "HW_REF25" or remove the reference. And for the k-firmware one would have to remove or comment out the check of the external reference in the source code, or physically remove the reference.

[madires]

What do people use to program a 2017 2578ay-at? Remove the Atmega328p and program it separately (which would be easy enough, but removing the 28DIP Atmega frequently risks causing problems)? Or would connecting to R3-1 (RST), R10-17 (MOSI), R11-18 (MISO), R12-19 (SCK) and VCC/GND work in circuit? If the latter, do you power the Atmega thru the USBASP connector, or connect GND and use a separate power for the AY AT?

Some simply pull the ATmega, some add a 6-pin ISP header. Do whatever you prefer. The 5V from the programmer are sufficient for programming.

[mebel]

For all beginners simpliest way is serial port programming with only 3 resistors and one jumper for reset pin, copy downloaded files in root c:\ (TransistorTester.hex and TransistorTester.eep ) I`m soldered pin header  from downside direct on atmega pins and program in three seconds .

In avrdude.conf change inverted reset to non-inverted

reset = 3

and with next 3 comands programm atmega

avrdude -p ATMEGA328P -c ponyser -P COM1  -U flash:w:"C:\TransistorTester.hex":i

avrdude -p ATMEGA328P -c ponyser -P COM1  -U eeprom:w:"C:\TransistorTester.eep":i

avrdude -p ATMEGA328P -c ponyser -P COM1 -U lfuse:w:0xF7:m -U hfuse:w:0xD9:m -U efuse:w:0x04:m

Fusebits are for default 8Mhz version.

How the reset should be connected to serial port??? On the photo I can se the closed jumper, where it shuld be connected? There are already 5 pins MOSI, MISO, SCK, AVCC and GND which are connected to serial port.

[robca]

If you get an AY-AT Clone, the following is a list of hardware corrections to order if you would like to optimize it's operation on M-Firmware. If you normally order parts from Mouser or Digikey all information and links are provided. (No Supplier Affiliation/No Financial Affiliate Program Links) This is simply a resource to make the information easier for the next person to find.
AY-AT Clone - The Proper Mods
1.) 0.1% Probing Resistors
2.) 20mhz Crystal
3.) Correct LDO Vreg
4.) Correct Vref

Manufacturers Part Number=MPN
Mouser Part Number=MoPN
Digikey Part Number=DkPN

(3×)470k Resistors
TE Connectivity
MPN: H8470KBZA
Metal Film Resistors - Through Hole H8 470K 0.1% 100PPM
MoPN: 279-H8470KBZA
$0.97each (8/2017)
$2.91 (×3)
DkPN: PTF475KCCT-ND (not stocked)
Dk Alt:
475k 0.1% 1/8W
Vishay Dale
MPN: PTF56475K00BYEB
DkPN: PTF475KCCT-ND
$1.64each (8/2017)

(3×)680ohm Resistors
TE Connectivity
MPN: H8680RBYA
Metal Film Resistors - Through Hole H8 680R 0.1% 15PPM
MoPN: 279-H8680RBYA
$1.14each (8/2017)
$3.42 (×3)
DkPN: H8680RBYA-ND(not stocked)
Dk Alt:
681ohm 0.1% 1/4W
Note: listed as discontinued by digikey (w/available stock 8/2017)
TT Electronics/Welwyn
MPN: RC55Y-681RBI
DkPN: 985-1036-1-ND
$1.70each (8/2017)

[...]

2.5 Volt 0.1% Precision Reference
Texas Instruments
MPN: LM4040AIZ-2.5/NOPB
Voltage References Prec MicroPwr Shunt Vtg Ref
MoPN: 926-LM4040AIZ25NOPB
$1.72 (8/2017)
DkPN: LM4040AIZ-2.5/NOPB-ND
$1.73 (8/2017)

5v LDO Voltage Regulator
Microchip
MPN: MCP1702-5002E/TO
LDO Voltage Regulators LDO w/ Low Quiescent
MoPN: 579-MCP1702-5002E/TO
$0.52each (8/2017)
DkPN: MCP1702-5002E/TO-ND
$0.49each (8/2017)

Total
Mouser: $10.88 (8/2017)
(+USA Econ Shipping@$4.99)
Digikey: $14.56 (8/2017)
(+ Shipping I'm too lazy to look this up)

First of all, thanks for putting this list together, really awesome

I'm a cheapskate (I think that spending ~$14+shipping to improve a ~$10 tester kinda defeats the purpose  )

So I tried looking for cheaper alternatives to the components you suggest ordering via Digikey, which has much cheaper shipping for small orders (usually $3.39 USPS first class in the USA)

I plan to get the 16MHz crystal and Atmega via eBay ($1.8 for the Atmega and $0.79 for the 10 crystals... I decided to go with 16MHz, btw, since I won't need the higher frequency and I know aht 16MHz is rock solid even for marginal parts). But getting those thru Digikey would make no real difference

I then looked for resistors similar in value to the default ones, but cheaper than the ~$1.7 each for the ideal match. I have read many times that matching the resistors is more important than the absolute value, so I was thinking about

750 Ohm instead of 680 Ohm (quite a lot, 10% difference, but all 3 resistors would be closely matched)
https://www.digikey.com/product-detail/en/yageo/MFP-25BRD52-750R/750ADCT-ND/2059145

499k Ohm instead of 470k (this one is close enough, methinks),
https://www.digikey.com/products/en?keywords=A105891CT-ND

Plus the LDO and reference voltage:
https://www.digikey.com/products/en?keywords=LM4040AIZ-2.5%2FNOPB-ND
https://www.digikey.com/products/en?keywords=MCP1702-5002E%2FTO-ND

That would bring the total to $5.84 + $3.39 shipping and another ~$2 for the Atmega and crystal(s)

Would it impact the measurements if I used 750 Ohm 0.1% resistors instead of 680 Ohm, and 499K Ohm 0.1% instead of 470k Ohm? Those cost significantly less than the closer matching ones

[rwgast_lowlevellogicdesin]

Ok so im adding a ti .05% 2.5vref to my AY YT along with better resistors and a 16mhz crystal. Is a 20mhz possible with the make file? Anyways Looking through the manual at all the config options when building the firmware my mind is blown, especially at all the LCD stuff... im not even sure if I have an ST or IL LCD or how to tell the difference all I know is i have some spi screens with and without the 4050 buffer... Is there a good way to easily test exactly which driver chip each lcd I have is using (Some are from the TC-1 testers)? Also is there anyway possible to get a default config file for each tester variant as a reference for the default options in there firmware?

One reason I want to move to a higher crystal speed is from the manual it sounds like the inductance measurement resolution will go down but I will be able to measure smaller inductors in turn, is this correct? Right now my AY-AT is not able to measure under 22mH while the TC-1 measured well in to the uH range.

[Willem52]

Just tested my 8MHz (k 1.12 firmware mod.) AY-AT tester and it measures a 47 uH as 0.05mH (last digit +/-1)
and my 16MHz (k 1.12 firmware mod.) AY-AT tester and it measures a 47 uH as 0.04mH (last digit +/-1)

So it seems to me that your AY-AT should do that too, there must be something wrong there.

The M-firmware can do till 20MHz.

[madires]

If you want to measure low value inductances the k-firmware's SamplingADC would be the better option. The m-firmware includes a file named "Clones" which lists the settings for a few clones.

[rwgast_lowlevellogicdesin]

Maybe I am doing something wrong, I am using sampling mode. I have to connect to pins 1 and 3 or it doesnt test inductance at all, using pins 1 and 3 it first tests resistance and then it tests inductance, with anything under 22mH it just blinks the inductance test and then reports the resistance only. What could cause this issue?

I still want to upgrade to a 16mhz xtal no matter if thats the problem with the inductance or not, the thing is just to slow with a color screen.. so what is the best way to figure out any of the config options im unsure about when it comes to specific hardware?

[madires]

A missing cap Please read Karl-Heinz' excellent documentation https://github.com/svn2github/transistortester/blob/master/Doku/trunk/pdftex/english/ttester.pdf.

[robca]

So I tried looking for cheaper alternatives to the components you suggest ordering via Digikey, which has much cheaper shipping for small orders (usually $3.39 USPS first class in the USA)

[...]

I then looked for resistors similar in value to the default ones, but cheaper than the ~$1.7 each for the ideal match. I have read many times that matching the resistors is more important than the absolute value, so I was thinking about

750 Ohm instead of 680 Ohm (quite a lot, 10% difference, but all 3 resistors would be closely matched)
https://www.digikey.com/product-detail/en/yageo/MFP-25BRD52-750R/750ADCT-ND/2059145

499k Ohm instead of 470k (this one is close enough, methinks),
https://www.digikey.com/products/en?keywords=A105891CT-ND

Plus the LDO and reference voltage:
https://www.digikey.com/products/en?keywords=LM4040AIZ-2.5%2FNOPB-ND
https://www.digikey.com/products/en?keywords=MCP1702-5002E%2FTO-ND

That would bring the total to $5.84 + $3.39 shipping and another ~$2 for the Atmega and crystal(s)

Would it impact the measurements if I used 750 Ohm 0.1% resistors instead of 680 Ohm, and 499K Ohm 0.1% instead of 470k Ohm? Those cost significantly less than the closer matching ones

Re-asking my question, hoping someone (Markus? ) knows the answer... how bad would it be to use 750 Ohm instead of 680 and 499K Ohm instead of 470k?