Design a New Precision LCR Tweezers with Much Lower Cost 【Shannon Tweezers ST42】

[ceut]

I was a little away from Eevblog because of some personal problems.

Nice work!

Thank you very much 

Well I went into this thread once again, this time because I was having problems updating my ST42 so it's funny that you say "hope you are the only unlucky guy who has this problem" because I'd like to introduce myself: Hi, I'm another unlucky guy, it seems :/ I'm getting exactly the same error, and I'm not really happy to hear it's a hardware problem. Is replacing the chip the only solution?

EDIT2: I would like to add that I remember updating the firmware when I was having the screen glitches (which I also reported here). And I don't remember the process giving me errors that time. Maybe this is not necessarily a hardware problem (in my unit at least) but rather that the sofware gives the same error under different conditions (including faulty hardware).

EDIT3: Ok, "solved" the issue and I'd like to keep a record here despite being a very silly mistake from my part. Apparently the intimidating error shown by the updater is the same in case of any problem communicating with the device. In my case it was either a driver issue or a cable issue, because changing the cable and usb port I got it to work.
BUT I also got the same error during one attempt in which I didn't enter into bootloader mode correctly. The COM port appeared on my computer but it wasn't in *actual* bootloader mode.

So I'd like to leave this post here as a warning for others: Even if you get the same error, it does not mean it's a hardware error. As obvious as it is now, the firmware updater will give the same error in any case in which the device is not responding back.

I'm happy to read that, as it was finally only a software/setting problem 
For me, the RX pin of the chip was shorted to ground  Since I put the new one, no problem at all 

For @shannon: I have checked the new diode signal: it is now super clean 
I have not used it a lot since my fix, so I don't know what is new in the 1.54 version ?
@Shannon: Do you have a changelog or something like that ? With that, I can continue to suggest some ideas I could have 

[ceut]

@Shannon how difficult do you think it would be to implement a special electrolytic capacitor testing mode? Electrolytic caps are polarized, and thus the correct way to test them is (I think) to bias the test signal to the Vpp/2 level, so that it always stays above zero instead of the standard AC signal of +/- Vpp/2.

I'm not sure what difference it will make though, but it just seems to me that applying a reverse polarity voltage to electrolytic caps, even if low and only for a half period, isn't quite right -- might be from the cap damage or the measurement precision perspective, or both. There are some LCR meters that implement this biased signal mode, e.g. East Tester ET432/433. I'm not able to find any good info on this topic, only vague suggestions that it's "better" to test the electrolytics with a positively biased signal, and it gives better precision.

It would be great to have that.
Also, as I was looking for a LCR Tweezer before Tautech told me about Shannon's one, here is the LCR-MPA function table, which I think may be added for some of them (the blue one would be great  ) ?

[KungFuJosh]

No need to wait any longer, just update your tweezers right now.

Is there a new firmware release? Your website still says 2023/5.

Just upload the new firmware a momenta ago, the website information has not been update.
And there are some other modification is under processing.

Hey Shannon, any update regarding the newer firmware? The OTA tool still downloads the older firmware, and the website also says 5/23.

Also, it's probably a good idea to add Shannon ST42 Tweezers to this thread title by now. 😉

[kladit]

I managed the update on my linux now.

My tweezer is detected by the kernel (type dmesg to see) as:

usb 1-4: Product: USB Serial
ch341 1-4:1.0: ch341-uart converter detected

lsusb shows it as:
ID 1a86:7523 QinHeng Electronics CH340 serial converter

udev mapped it to (type dmesg to see) :
usb 1-4: ch341-uart converter now attached to ttyUSB0

Now take the shutted down tweezer, press its navigator button upwards and put in the usb-connector
of the usb-cable already conected to your computer.

Now cd to the dir where you have the unzipped  fw_update-linux programm and type :

mkdir ./tmp; export TMPDIR=./tmp; ./fw_update-linux  ttyUSB0

Now the firmware gets loaded over the internet and is flashed to the tweezer.

[m.abdelwanis]

Any info about ESR table or calibration ?

[Shannon]

Any info about ESR table or calibration ?

Could you give us more details about your idea?

ST42 has the independent ESR measurement mode, so what does the ESR table mean?

Does it mean if ST42 show a ESR measurement table after testing many DUT?

And ST42 has four calibration mode: "self-cal" "short-cal" "open-cal" "factory", for the calibration process, please refer to

[Shannon]

Hey Shannon, any update regarding the newer firmware? The OTA tool still downloads the older firmware, and the website also says 5/23.

Also, it's probably a good idea to add Shannon ST42 Tweezers to this thread title by now. 😉

We will put the release information on the website.

And thanks a lot for your advice, the title has been updated.
Once again, it proves that our business sense is far away from good enough

[Shannon]

@Shannon how difficult do you think it would be to implement a special electrolytic capacitor testing mode? Electrolytic caps are polarized, and thus the correct way to test them is (I think) to bias the test signal to the Vpp/2 level, so that it always stays above zero instead of the standard AC signal of +/- Vpp/2.

I'm not sure what difference it will make though, but it just seems to me that applying a reverse polarity voltage to electrolytic caps, even if low and only for a half period, isn't quite right -- might be from the cap damage or the measurement precision perspective, or both. There are some LCR meters that implement this biased signal mode, e.g. East Tester ET432/433. I'm not able to find any good info on this topic, only vague suggestions that it's "better" to test the electrolytics with a positively biased signal, and it gives better precision.

It would be great to have that.
Also, as I was looking for a LCR Tweezer before Tautech told me about Shannon's one, here is the LCR-MPA function table, which I think may be added for some of them (the blue one would be great  ) ?

@ceut @shapirus Thanks for your information

The effect of DC bias on capacitance is a standard spec for those ceramic capacitors, electrolytic capacitors, etc.
It may not be easy to implement because a table is usually used to describe it. Taking ceramic capacitors as an example,
The abscissa of a chart is the DC voltage value, and the ordinate is the loss of capacitance. Therefore, when we consider providing DC bias, we need to implement an adjustable and wide-range DC bias voltage, which is reasonable.

However, as a low-power high-precision device, ST42 is basically unlikely to provide a higher DC bias, because the operating voltage of the core device is 3.3V.

We also checked the parameters of the EST43 series and found that the internal bias is adjustable, but the adjustable range is 0~500mV. Such a bias range is also difficult to meet real test requirements.
The most common application in my mind, what is the capacitance of the capacitor under 12V/5V/3.3V/1.8V DC bias.

But you all know, when you know one cap value of two serial cap and the serial cap total value, you can calculate the other cap value.
This principle could help you measure the DUT cap under any DC bias voltage. This test method makes me feel exited

[shapirus]

However, as a low-power high-precision device, ST42 is basically unlikely to provide a higher DC bias, because the operating voltage of the core device is 3.3V.


Well, my point was about implementing DC bias that's not high or adjustable (even though that would be nice to have), but only just so high as to shift the test signal above zero, so that it does not change polarity, which, as I assumed, might make sense for electrolytic caps.

Maximum Vpp swing of the test signal in the capacitance mode is only 1.0 V, so it's swinging from -500 mV to +500 mV. With a DC bias of +500 mV it would swing from 0 V to +1.0 V. That should still be well within the capabilities of the device operating at 3.3 V.

As it is now, electrolytic caps are tested with a signal that changes between forward and reverse polarity every half cycle. The level is probably too low to damage the capacitor under test, but my concern was about whether it's ok to use signal with alternating polarity to measure polarized caps (which by definition should never be reverse biased). Maybe it is, I don't know. But there's probably a reason why some LCR meters have dedicated modes for electrolytic caps.