$20 LCR ESR Transistor checker project

[madires]

Hi.. how do I adjust the value of the measurement result? I'm using firmware 1.13k and the capacitor measurement results have to increase 0.5pf while the resistor measurement results have to reduce 0.05 ohm

If you really want to add/subtract those values you can modify the firmware, e.g. the output function. However, for those low values it doesn't make much sense to go to the extreme because the Transistortester isn't a precision instrument.

[Yuriy_K]

Hi.. how do I adjust the value of the measurement result? I'm using firmware 1.13k and the capacitor measurement results have to increase 0.5pf while the resistor measurement results have to reduce 0.05 ohm

Sample case,
I measure Capacitor 1pf, then the measurement result is 0.51pf

Sample case,
I measured a 0.33 ohm resistor, then the measurement result is 0.38 ohm.

How do I adjust the measure value in code? Calibration has been performed but does not solve the problem.

The measurement accuracy depends on many factors, voltage stability, input wiring, measurement frequencies, and so on. What true meaning are you targeting?
Here is an example of measuring capacitors using different methods on different devices. As you can see, the readings of the devices do not always correspond to the inscriptions on the capacitors. A similar example with resistance measurement, but measurements are made with direct current and the readings differ slightly ...

Show examples of your measurements ...

[anwfeb]

Hi.. how do I adjust the value of the measurement result? I'm using firmware 1.13k and the capacitor measurement results have to increase 0.5pf while the resistor measurement results have to reduce 0.05 ohm

If you really want to add/subtract those values you can modify the firmware, e.g. the output function. However, for those low values it doesn't make much sense to go to the extreme because the Transistortester isn't a precision instrument.

can you help me to adjust the output so that the measurement results can match?

[anwfeb]

Hi.. how do I adjust the value of the measurement result? I'm using firmware 1.13k and the capacitor measurement results have to increase 0.5pf while the resistor measurement results have to reduce 0.05 ohm

Sample case,
I measure Capacitor 1pf, then the measurement result is 0.51pf

Sample case,
I measured a 0.33 ohm resistor, then the measurement result is 0.38 ohm.

How do I adjust the measure value in code? Calibration has been performed but does not solve the problem.

The measurement accuracy depends on many factors, voltage stability, input wiring, measurement frequencies, and so on. What true meaning are you targeting?
Here is an example of measuring capacitors using different methods on different devices. As you can see, the readings of the devices do not always correspond to the inscriptions on the capacitors. A similar example with resistance measurement, but measurements are made with direct current and the readings differ slightly ...

Show examples of your measurements ...

for the measurement I want is how it approximates exactly, it doesn't have to be right in the final digits of the 3rd digit but if 5.6pf looks on the screen is 4.8pf it means I will assume this capacitor is 4.7pf if there is no writing on the physical body of the capacitor

[anwfeb]

or, if the resistor is 0.33 ohm, if on the physical body there is no color then I will assume the resistor is 0.39 ohm

[Yuriy_K]

In your photo, I do not see a way to connect the part to be measured with the measurement contacts. It is impossible to understand what the Ttester readings are in the absence of the measured part.
In my photos, I specifically show the way to connect the part to be measured ...

[anwfeb]

In your photo, I do not see a way to connect the part to be measured with the measurement contacts. It is impossible to understand what the Ttester readings are in the absence of the measured part.
In my photos, I specifically show the way to connect the part to be measured ...

I connected directly to the ZIF socket because the photo with the app collage was cut off the bottom of the ZIF, the component was connected directly without a probe cable, and when without a component the value was 0.01 after calibration (not -0.5) I understood the calibration function so that the value without components was 0.00 or close to 0.

[indman]

anwfeb,show the measurement results of the same capacitors, but insert the part into pins 1-2 and 2-3.
I would also like to see the T1-T7 hardware test results on your clone.

[anwfeb]

anwfeb,show the measurement results of the same capacitors, but insert the part into pins 1-2 and 2-3.
I would also like to see the T1-T7 hardware test results on your clone.

Hi.. indman, I have done some experiments, the last time I saw a difference in the ZIF socket, when I calibrated with the ZIF Locked position, the 1pf Capacitor measuring value is now 1.2x pf. but if I calibrate with ZIF socket unlocked, the result of measuring 1pf capacitor is 0.5x pf. I already know that different measurements when calibrating with ZIF socket locked/unlocked make different measurement results, I don't know if this is the same for other devices? or my ZIF socket is broken? but I have realized that it affects the difference in the measurement results. however, for resistors measuring in 0.22 ohm still shows 0.28-0.31 ohm, I haven't figured out what the problem is.

i have researched further for TL431 at the foot of the PA3 pin I measure the voltage is not up to 2.5v but only get 2.48v is this a problem?

T1-T7 test results will follow soon I will inform

[indman]

I already know that different measurements when calibrating with ZIF socket locked/unlocked make different measurement results, I don't know if this is the same for other devices?

Yes, this is a prerequisite for the correct calibration of all devices if you want to obtain reliable results with small capacitance values. As open contacts ZIF introduce an additional error during calibration.

i have researched further for TL431 at the foot of the PA3 pin I measure the voltage is not up to 2.5v but only get 2.48v is this a problem?

No.Install a quality MCP1702 stabilizer in your clone and the external TL431 can be removed from the circuit altogether. Firmware does not need to be changed, just change the 2.2kOhm resistor to 47Kohm.
Also select the same pairs of 680Ohm,470kOhm measuring resistors. Refer to the T1-T7 test results. This is 99% likely to help you solve your resistor measurement problem, although even now your error is no more than a reasonable value for TESTER.

[anwfeb]

No.Install a quality MCP1702 stabilizer in your clone and the external TL431 can be removed from the circuit altogether. Firmware does not need to be changed, just change the 2.2kOhm resistor to 47Kohm.
Also select the same pairs of 680Ohm,470kOhm measuring resistors. Refer to the T1-T7 test results. This is 99% likely to help you solve your resistor measurement problem, although even now your error is no more than a reasonable value for TESTER.

yes, I'll try again, it's true this is still within the reasonable threshold of measurement, but I just want to improve for a more precise measurement then that is the best.

[madires]

Good news regarding the frequency counter issue with the 16:1 prescaler and ATmega644. Karl-Heinz has added a configuration switch (NO_FREQUENCY_SWITCH=1) to disable the external prescaler. Please see https://github.com/kubi48/TransistorTester-source for the latest version.

[anwfeb]

Good news regarding the frequency counter issue with the 16:1 prescaler and ATmega644. Karl-Heinz has added a configuration switch (NO_FREQUENCY_SWITCH=1) to disable the external prescaler. Please see https://github.com/kubi48/TransistorTester-source for the latest version.

thanks info madries, I'll try with the new firmware.

[anwfeb]

anwfeb,show the measurement results of the same capacitors, but insert the part into pins 1-2 and 2-3.
I would also like to see the T1-T7 hardware test results on your clone.

I noticed T6 RH+ changed 4x in 4 seconds,
first second 0 0 0
2nd second -3 0 0
3rd second -2 0 0
fourth second -2 0 -1

[indman]

Why is there such a difference in the resistance of the closed contacts of 0.18-0.27?! This resistance should be approximately the same, check the reliability of the contacts of the ZIF panel. Also pick up the 680Ohm resistors.

[anwfeb]

Why is there such a difference in the resistance of the closed contacts of 0.18-0.27?! This resistance should be approximately the same, check the reliability of the contacts of the ZIF panel. Also pick up the 680Ohm resistors.

I have check ZIF, and here are the test results now

[indman]

0.24-0.34!!!0.1 Оhm distinction!!! Check the measurement results of the low resistor, e.g. 0.33 Ohm, on the different pins 1-2,2-3,1-3.

[anwfeb]

0.24-0.34!!!0.1 Оhm distinction!!! Check the measurement results of the low resistor, e.g. 0.33 Ohm, on the different pins 1-2,2-3,1-3.

Please check

[indman]

Yesterday you showed the same resistors, but the results today are different and more reliable. What's changed since yesterday?

[anwfeb]

Yesterday you showed the same resistors, but the results today are different and more reliable. What's changed since yesterday?

I don't know what changes break the measurement value,
now i use "k" firmware with freq counter fix (not use yesterday firmware) then i fix 5V LDO now my LDO shows 5.04V as you said i have to replace better LDO.

however, the resistor measurement after 1 hour I repeat the measurement without calibration then, the result will change. every low resistor measurement I have to calibrate, or I have to do a 3pin jumper probe but when there is a request for selftest mode... ? I looked at it and saw the results of the short probe 0.00 0.00 so I did a resistor check and the results were appropriate. but if I do not calibrate and see the 3 pin probe jumper then the result is 0.07 0.00 this will damage the current value of the measurement. can you find out where this error lies?

the change in the value of short check pin 123 occurs after a few minutes or after a few hours, so I have to calibrate again to get the result 0.00 - 0.00, as if eeprom didn't save the last calibration result.

[indman]

The first thing you need to do, as I wrote before, is to provide a stable and noise-free power supply to the processor. The word stable should be understood not only as a constant value of output voltage +5.000 V, but also the ability of the regulator to respond very quickly to changes in current and short-term pulses of the load. All this is achieved if you install a QUALITY! regulator, as well as apply the same QUALITY!  filter capacitors to the input and output. Only when these conditions are met can you hope to get reliable readings when testing different parts. Is that clear?

[anwfeb]

The first thing you need to do, as I wrote before, is to provide a stable and noise-free power supply to the processor. The word stable should be understood not only as a constant value of output voltage +5.000 V, but also the ability of the regulator to respond very quickly to changes in current and short-term pulses of the load. All this is achieved if you install a QUALITY! regulator, as well as apply the same QUALITY!  filter capacitors to the input and output. Only when these conditions are met can you hope to get reliable readings when testing different parts. Is that clear?

Yes, I will try again later to find another LDO that can stabilize conditions where there is a change or the load short-term pulses of the remains at 5.00v
I understand how important the regulator is to get the right and accurate measurement value

[CatalinaWOW]

It is kind of funny watching the change in attitude over the course of this thread.  Back when it started there was a sense of wonder that an inexpensive device could provide a variety of usable measurements.  Over time that has morphed into outrage that this same device doesn't provide instrument grade results.  In a sense it is recapitulating the history of test equipment.  Initially crude measurements refined over time through attention to detail and clever engineering.

[indman]

I Initially crude measurements refined over time through attention to detail and clever engineering.

Not only that, software and its improvement by the authors plays a big role. After all, the circuitry of this device is almost unchanged (not much)

[madires]

And some don't understand that the Chinese clone manufacturers use whatever parts are on sale in Shenzhen. If you're lucky your tester is spot on. But it could be also a few percent off. With the right parts you can get about 1% accuracy for a few measurements which is the limit of what is possible with the current design. The Transistortester was never meant to be anything more than a versatile tool. Karl-Heinz and I are sometimes amazed what can be achieved with that simple circuit.