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

[Shannon]

Shannon was trying to send me a tweezer to do a review video on, unfortunately there were issues with the item being returned due to the battery... so I still don't have it yet, I don't know if Shannon has given up trying to send it to me, I wouldn't blame them !

If Shannon is able to get a pair to me then I will do a review on them.

of course not, we have not given up

and the story is we sent it out to one transport company last week, but we got some unfortunate feedback, they need extra shipping cost, which is much higher than our expectation.

So the plan is this company will send back the package to me, and we will select the other transport company to help us send it to you
After we get the international tracking number, we will let you know.

[fkfaraz]

Hello dear Shannon
congratulation on releasing the tweezers.
on checking Ali, i am getting price of 183 USD instead of 109USD. is the price increased?

Because the $109 version is sold out (Plain ST42). Only the ones with leather case is available ($183). Either wait for him to restock or wait for kickstarter.

As thm_w said, we have sold almost all second batch of  st42, the price is not for early birds as you.

New batch of ST42 is on the way, we can update the numbers in store next week.
And the price will be changed to early birds mode

I will be waiting for the Updated Price then,   

[eliocor]

The ST42 arrived today:
PROS:
- fast measurements
- precise: checked against some reference capacitors/resistors
CONS:
- with some low value inductors it show the resistance, instead of inductance. Solution: maybe it needs just a little change in the FW
- the springiness of the two arms is too high and require a rather hefty force to bend them. This is rather annoying! Solution: no idea.
- several times I press on the joystick, I get a beep, but no changes in the menus. Solution: do not beep if the key is not accepted; eg: when in scroll mode, accept only Up/Down keys
- menu scroll: if the end is reached, it should restart to first menu item if Down is pressed.
SUGGESTIONS:
- Implement a menu item which try to test any kind of components: L/R/C/Diode/LED altogether. Even if the readings will be SLOWER, it will be better than roaming through the menus to change the test type.
- Offer in your shop (as a replacement part) the tips of the tweezers

[Shannon]

The ST42 arrived today:

Great to know you have revived ST42, and thanks a lot for your feedback
We have discussed some solutions for your feedback.
Let's talk about the Cons and suggestions only.

- with some low value inductors it show the resistance, instead of inductance. Solution: maybe it needs just a little change in the FW
  Do you remember the value of the inductor? we can try to repeat this issue.
  For a while, we were plagued by the problem that 0ohm resistors were easily judged as inductances, so we realized that near the limit.
  Small inductance and small resistance are very similar in physical properties.
  Therefore, for very small inductances, it is recommended to use the manual mode for measurement.
  We are still trying to figure out whether we can further improve the discrimination accuracy.
  if one day, we could use a 100kHz excitation signal, that could make big progress.

- the springiness of the two arms is too high and require a rather hefty force to bend them. This is rather annoying! Solution: no idea.
  The solution could be to use thinner or low springiness material as the arms of tweezers.
  if we receive more feedback for the springiness as you mentioned, we will upgrade it.

- several times I press on the joystick, I get a beep, but no changes in the menus. Solution: do not beep if the key is not accepted; eg: when in scroll mode, accept only Up/Down keys
   The logic of this case is that we want to remind the user that you did press the button, but the function is empty. . .
   We're trying to improve this, maybe it's a good idea to use left and right in menus for exiting and entering menus, as elsewhere.

- menu scroll: if the end is reached, it should restart to first menu item if Down is pressed.
   One tip is when you choose to long-press the joykey in the middle, you can exit the menu of this layer, without operating to the bottom, and choose to exit.
   For the cycle menu function, we will implement an instead function, long press "down" joykey, the menu will scroll to the end automatically.

SUGGESTIONS:
- Implement a menu item which try to test any kind of components: L/R/C/Diode/LED altogether. Even if the readings will be SLOWER, it will be better than roaming through the menus to change the test type.
  This function is in our TODO list for a long time, we need some time to implement and validate it.
   
- Offer in your shop (as a replacement part) the tips of the tweezers
  No problem, we will add it to the shop.
  You are not the only one to give this suggestion.

[nez]

- several times I press on the joystick, I get a beep, but no changes in the menus. Solution: do not beep if the key is not accepted; eg: when in scroll mode, accept only Up/Down keys
   The logic of this case is that we want to remind the user that you did press the button, but the function is empty. . .
   We're trying to improve this, maybe it's a good idea to use left and right in menus for exiting and entering menus, as elsewhere.

Perhaps a quick double-chirp or other 'different' sound in the case of input that doesn't work?

I'm not totally sure it wouldn't bring more confusion instead of helping without trying in practice, but maybe.  'Rejected input' feedback behavior could be made into an option setting (if it's worth the effort and menu space).

[Shannon]

- several times I press on the joystick, I get a beep, but no changes in the menus. Solution: do not beep if the key is not accepted; eg: when in scroll mode, accept only Up/Down keys
   The logic of this case is that we want to remind the user that you did press the button, but the function is empty. . .
   We're trying to improve this, maybe it's a good idea to use left and right in menus for exiting and entering menus, as elsewhere.

Perhaps a quick double-chirp or other 'different' sound in the case of input that doesn't work?

I'm not totally sure it wouldn't bring more confusion instead of helping without trying in practice, but maybe.  'Rejected input' feedback behavior could be made into an option setting (if it's worth the effort and menu space).

Yes, different beep sounds could be used to inform the users. We will try to compare these schemes and choose the best solution.

[Shannon]

The new batch of ST42 is on the way, we can get part of them in this week.

So the numbers of ST42 on the Aliexpress platform have been refreshed. you can order now.

And the gold-plated tips are an add-on option as eliocor suggested.

[Brumby]

I've been following this development (with interest) since the start and have had several suggestions - but others have beaten me to the punch every time.  There have been a lot of other great ideas as well.

Over the last few months, I have been time challenged so have missed the early bird opportunity and now the subject of nuptials for my daughter has been hinted at.  Discretionary spending is on indefinite hold.


I shall have to resort to the vicarious alternative for a while - so, people, please post videos!

[Hydron]

Mine just arrived, shipping was a bit slow but that seems to be mostly down to luck these days unless you pay $$$ for a courier.

Haven't had much of a chance to play, but I can already tell that it's WAY better than the Mastech MS8911 I bought before this (some others on the forum think the Mastech is a good product, but personally I found it horrible to use). Biggest issue I've seen so far is that of getting a good solid contact with components on a PCB, but I think this is just a general limitation of using tweezer-style testers on already-soldered parts. The joystick is also a little fiddly to use, but again I think that's a form factor constraint.

I'll play around a bit more, but unless I hit any big issues I'll probably try and order a couple more for work. I look forward to seeing the FW develop more - if I think of or find any areas for improvement I'll make sure to make suggestions

[Grandchuck]

Have been using tweezers for years and the contact issue is the weak point.  Don't think there is a practical solution.

[bicycleguy]

Received LCR tweezers today about 36 days after ordering, not bad in todays environment.  Slightly odd that the sw version date was 5 days after I ordered.

The quality seems very good (almost) and the measurements are fast and very accurate or at least match my 121GW at 10X it's speed.  By the time I think I have a proper connection and can take my eyes off what I'm measuring the readings are already there.  Along with that the hold feature is implemented very well with repetitive beeps I think, letting you know it will hold the value of the last beep. 

The display is super crisp even with my old eyes, even without the magnifier I'm usually using for working with small parts.  I have no problem with the arm stiffness, I think its just right.  It's really nice to pick small chips with some quality tweezers!

The auto function works great so the menus aren't usually needed, however, now an issue:

Like @eliocor posted #277 sometimes I get beeps with no action, almost always with the down key.  Fiddling with this it acts like a bad switch, except for the beeps.  Sometimes if I squint just right it works but usually not.  This mostly effects the down button.  Inevitably when the down works it won't scroll back up.  As a programmer this seems like a software glitch, or maybe a crummy switch with less than optimal software dealing with its issues.  Perhaps the down and left or right key are engaging at the same time or the left or right key is engaging when it should be the down.

While writing this I found a way to make it work.  If I want to go down I have to kind of roll the switch from left to right as I pull down.  The attached picture shows the button seems to be down and to the left of center.  Some times it seems to look rotated also.

I think it's time to take it apart!

[Shannon]

Received LCR tweezers today about 36 days after ordering, not bad in todays environment. 

Like @eliocor posted #277 sometimes I get beeps with no action, almost always with the down key.  Fiddling with this it acts like a bad switch, except for the beeps.  Sometimes if I squint just right it works but usually not.  This mostly effects the down button.  Inevitably when the down works it won't scroll back up.  As a programmer this seems like a software glitch, or maybe a crummy switch with less than optimal software dealing with its issues.  Perhaps the down and left or right key are engaging at the same time or the left or right key is engaging when it should be the down.

While writing this I found a way to make it work.  If I want to go down I have to kind of roll the switch from left to right as I pull down.  The attached picture shows the button seems to be down and to the left of center.  Some times it seems to look rotated also.

I think it's time to take it apart!

Hey bicycleguy,

Thanks for your support and feedback

And let's follow the traditional way, we talk about the cons only.

“Slightly odd that the sw version date was 5 days after I ordered.”
The story is we send it out on the weekend, so before we send it out, we will update the firmware to the last version.
You know some guys may miss the firmware upgrade notification on our website.
We try our best to make the user experience better.

through you and @eliocor feedback, you all think we can make the menu and joystick better.
We are serious about this feedback, this is one key factor of the user interface.

BTW we are encouraged by your comments on the comparison with other instruments,
and the performance of accuracy/test speed/auto mode/hold function/mechanical stability and so on.

We will continue to make it better.
Thank you all

[bicycleguy]

....
I think it's time to take it apart!

... BTW we are encouraged by your comments on the comparison with other instruments,
and the performance of accuracy/test speed/auto mode/hold function/mechanical stability and so on.

We will continue to make it better.
Thank you all

So I took the front cover off to look at the switch and the display went random and then froze.  The switch still beeped.  Then the screen went black, but the switch still beeped.  Of course I thought I had trashed it.  Measured 3.2V at some pads near the top right of the display.  Didn't know at the time I could have measured the battery at the bottom right.  The battery had measured 80% on the display a few minutes before, so I thought it was ok.

Put the device on the USB charger and went to bed.  Sure enough, in the morning all is well, even the scroll down key seems fixed.

So the conclusions are:
1.  Charge the device, even though the battery says it's ok.  I should have known to do this but you may want to point it out for idiots like me.
2.  Perhaps the battery level indicator is off or maybe it dropped real fast since it was a month in shipping.
3.  Even when the switch was acting up the meter seemed to measure correctly.  Seems a little weird that a lowly switch would need more voltage than a measurement.  Perhaps the switch pulldown current in combination with the beep is enough to make the microprocessor glitch on low voltage.

[bicycleguy]

Don't know if this is by design or what but on my device one of the four cover screws was slightly different:
3 screws  10mm long, 2mm dia, .83pitch
1 screw   11mm long,  2mm dia,  .62pitch

Not noticeable until you put them side by side.  This small difference will surly mess up the self tapping holes if exchanged enough.  The longer one was on the top right looking at the back in sticker reading orientation on my particular unit.

[Shannon]

....
I think it's time to take it apart!

... BTW we are encouraged by your comments on the comparison with other instruments,
and the performance of accuracy/test speed/auto mode/hold function/mechanical stability and so on.

We will continue to make it better.
Thank you all

So I took the front cover off to look at the switch and the display went random and then froze.  The switch still beeped.  Then the screen went black, but the switch still beeped.  Of course I thought I had trashed it.  Measured 3.2V at some pads near the top right of the display.  Didn't know at the time I could have measured the battery at the bottom right.  The battery had measured 80% on the display a few minutes before, so I thought it was ok.

Put the device on the USB charger and went to bed.  Sure enough, in the morning all is well, even the scroll down key seems fixed.

So the conclusions are:
1.  Charge the device, even though the battery says it's ok.  I should have known to do this but you may want to point it out for idiots like me.
2.  Perhaps the battery level indicator is off or maybe it dropped real fast since it was a month in shipping.
3.  Even when the switch was acting up the meter seemed to measure correctly.  Seems a little weird that a lowly switch would need more voltage than a measurement.  Perhaps the switch pulldown current in combination with the beep is enough to make the microprocessor glitch on low voltage.

Hi bicycleguy,

Thanks a lot for your feedback.

Have you upgraded your firmware version?
Highly recommend that upgrade the firmware to the last version, here is the upgrade instruction https://shannontweezers.top/docs/upgrading-firmware/
We have released the low power protection in version 1.3.9

As you mentioned, we found low battery voltage protection is necessary after Dave received the ST42 with no power.
It is not that ST42 silent current makes the battery die, we have put it alone for two months, and the battery is still healthy, and we can use it immediately.
When our package is on the way, sometimes there are some other packages pressed on ST42, and the key is always pressed, so the battery is exhausted.

the other strange phenomenon is that you found the battery monitor function has some issue, we have not met before.
so before you teardown them, the battery voltage shows 80%,
after that, the battery is 3.2V tested by DMM.

Is the 80% displayed on the screen when you charge ST42?
We will try to repeat this issue.

if you have not upgraded the firmware, please update the firmware firstly,
and if you found this phenomenon again, Pls help us use the joystick switch to System->Service->Battery, and observe the battery voltage.
That will help us find out the root cause.

Really appreciate it.

[Shannon]

Don't know if this is by design or what but on my device one of the four cover screws was slightly different:
3 screws  10mm long, 2mm dia, .83pitch
1 screw   11mm long,  2mm dia,  .62pitch

Not noticeable until you put them side by side.  This small difference will surly mess up the self tapping holes if exchanged enough.  The longer one was on the top right looking at the back in sticker reading orientation on my particular unit.

To be honest, we have tested different long screws.
And we found that 11mm and 10mm are all suits for this application, we may mix one 11mm screw in the 10mm screw package.

As you said, if the screw length is different, it is worth noting that it will easily affect the effect of the assembly.
And we need to optimize the assembly process as well.

[thm_w]

Received today, took about 20 days to arrive.
Need to play with it more, but, in Auto mode, it seems to be calibrated correctly and read a 10mR resistor close (14mR). However in resistance mode, it was not applying the offset calibration? It reads 25mR, which is high.

Tried running short CAL once, ran it again immediately after and now the screen is stuck displaying "--SHORT CAL--" there is no way for me to exit the screen. Might have to take it apart to reset it?
Maybe it needs an emergency reset or shutdown option, like holding button down will trigger a reset. Currently, as others noted, it just repeatedly beeps when you hold a button down.

Added some red and black heatshrink, can see in photo below.
I do like the sound buzzer level adjustment option.


Steps to upgrade firmware:
- Unplug USB
- Enter menu -> Shutdown, wait for shutdown
- Hold nav button UP, plug into USB, "bootloader" shown on LCD
- Use "fw_upgrade.exe comxx" in command line

[Shannon]

However in resistance mode, it was not applying the offset calibration? It reads 25mR, which is high.

Tried running short CAL once, ran it again immediately after and now the screen is stuck displaying "--SHORT CAL--" there is no way for me to exit the screen. Might have to take it apart to reset it?
Maybe it needs an emergency reset or shutdown option, like holding button down will trigger a reset. Currently, as others noted, it just repeatedly beeps when you hold a button down.

Hi thm_w,

thanks a lot for your feedback, and you help us find out a big bug.

However in resistance mode, it was not applying the offset calibration? It reads 25mR, which is high.
During the calibration, we did the AC mode calibration only, DCR short calibration have not been implemented.
We will add this feature to the coming firmware version these days.

Tried running short CAL once, ran it again immediately after and now the screen is stuck displaying "--SHORT CAL--" there is no way for me to exit the screen. Might have to take it apart to reset it?
the first response in my mind is "WHAT"? we never met it before.
But the second step we try to repeat this phenomenon on our own tweezers, fortunately, doing "--SHORT CAL--" the second time immediately may make the display hold.
There is a firmware logic issue there.
We will fix this issue as soon as possible, and try to add a force restart function as well.

The recommendation is to put the tweezers on your desk, and the tweezers will consume the power of the battery and restart.
This process may spend a long time.
Tearing down the tweezers, the screen installation is a little difficult to put in the slot, which may lead to the screen breaking.
We have broken some screens during the assembly

Hope this method could help you recover ST42, and we will come back soon with new firmware version.

[bicycleguy]

We have released the low power protection in version 1.3.9
...
the other strange phenomenon is that you found the battery monitor function has some issue, we have not met before.
so before you teardown them, the battery voltage shows 80%,
after that, the battery is 3.2V tested by DMM.

Is the 80% displayed on the screen when you charge ST42?
We will try to repeat this issue.
...
and if you found this phenomenon again, Pls help us use the joystick switch to System->Service->Battery, and observe the battery voltage.
That will help us find out the root cause.

Really appreciate it.

In order above:
Version currently at 1.3.8, will update.
Sorry, I was not clear and learned more details after the post.  The 3.2V was measured between two solder pads at the top right above the display.  I realize now this is probably the regulated voltage, not the battery voltage.  It was 3.2V when the key problems existed and after charging.  I didn't know about the battery mV readout as the menu issue was difficult but did write down 3 of 4 bars.  The 80% was with USB charging.  This morning, after charging overnight and with it  working correctly, I noticed the battery display readout you mentioned displayed 4039mV.  I removed the cover again to measure the battery.  I measured 4025mV a few seconds later at the wire connections to the pcb.  While the cover was off the display froze.  I soon figured out a slight press to the flex connection to the OLED was required.  Without the display freezes, breakups and eventually goes black.  Pressing on the flex fixed it.  Putting the cover back on I noted the tight fit you mentioned. 

So in conclusion:
I don't know what the battery display mV or the actual battery V was when I had scroll key issues.
The screen issue I had was from the flex connection from taking the cover off.  It didn't immediately work when I put the cover on that night.
The fact it worked the next day and still does may have been better seating of the flex over time?  Is there supposed to be a pad or something?

On the screw issue, my concern is the thread pitch difference, not the length.  All four screws should be the same pitch so you don't have to keep track of where it came from.  I'd go with the course pitch.

[tautech]

On the screw issue, my concern is the thread pitch difference, not the length.  All four screws should be the same pitch so you don't have to keep track of where it came from.  I'd go with the course pitch.

Agreed, it is poor form to use different thread pitches within a group of screws holding anything together.

[Shannon]

On the screw issue, my concern is the thread pitch difference, not the length.  All four screws should be the same pitch so you don't have to keep track of where it came from.  I'd go with the course pitch.

Agreed, it is poor form to use different thread pitches within a group of screws holding anything together.

Sorry for this issue, we will unify the model of screws and throw away those screws that are purchased for test.

Thanks bicycleguy as the first one who helps us point out these issues, and give us the advice to optimize them:
screws are not unified

Let's follow the traditional way, we will send you some gifts to thank you for your contribution to st42.

we will catch you through eevblog PM 

[Shannon]

...
So in conclusion:
I don't know what the battery display mV or the actual battery V was when I had scroll key issues.
The screen issue I had was from the flex connection from taking the cover off.  It didn't immediately work when I put the cover on that night.
The fact it worked the next day and still does may have been better seating of the flex over time?  Is there supposed to be a pad or something?

On the screw issue, my concern is the thread pitch difference, not the length.  All four screws should be the same pitch so you don't have to keep track of where it came from.  I'd go with the course pitch.

we will update the screen installation process, it's easy to make the screen unstable and broken.

the detail is putting some foam sponge glue between the screen and the flex PCB firstly, and then installing the top cover.
This should make the flex board more stable, and decrease the broken risk.

We have tested this process on a few parts, and it works well, so it will be the standard screen installation process on the next batch of ST42.

[thm_w]

The recommendation is to put the tweezers on your desk, and the tweezers will consume the power of the battery and restart.
This process may spend a long time.

I already have them on overnight, and they are still running! Can't imagine how little power it is using.
I won't dare to reset the micro as I don't know the pinout. So I'll keep waiting for battery to die.

If the battery was ever not enough it looks like there is ~2-3mm vertical height for a larger battery. But it should be fine.

edit: tweezers just shut down (~16h), plugged them into USB again and its charging now. Will test them once battery has recovered.
edit2: seem to be working well after another CAL. Seriously impressed at how good these are at reading low value resistances and capacitance. Higher value resistors are usually good but occasionally off depending on what else is in circuit, eg one 10k resistor was reading 10k on fluke but ~6k on these in Rdc mode.
There is a menu option Measure -> Level -> 0.1/0.5/1.2V but this doesn't seem to affect Rdc mode. Rdc is fixed at 0.5V?

[Shannon]

edit: tweezers just shut down (~16h), plugged them into USB again and its charging now. Will test them once battery has recovered.
edit2: seem to be working well after another CAL. Seriously impressed at how good these are at reading low value resistances and capacitance. Higher value resistors are usually good but occasionally off depending on what else is in circuit, eg one 10k resistor was reading 10k on fluke but ~6k on these in Rdc mode.
There is a menu option Measure -> Level -> 0.1/0.5/1.2V but this doesn't seem to affect Rdc mode. Rdc is fixed at 0.5V?

Hi thm_w,

We did the test at the same time, the working hour is around a whole day, which is similar to your test.

RDC is a voltage test mode, so the 1.2V/0.5V/0.1V voltage is the AC mode peak 2 peak amplitude, which is for the Auto Mode and L/C/R mode.

1. have you checked if the DUT res is 10kohm? you can use ST42 0.5V auto mode to do the On Board test.
2. Could you help us check the voltage on the 10kohm res when you use the RDC mode and Fluke doing the test?

The result could help us check the difference.

One reason may be the ST42 generate a higher voltage on the resistor, and trigger other components, the parallel res is smaller than the DUT.

It's a good recommendation to implement RDC onboard test function as well, we will discuss this feature, and add to our TODO list.

I will catch you by eevblog forum PM, and thanks for your contribute to ST42

[thm_w]

RDC is a constant current test mode, so the 1.2V/0.5V/0.1V voltage is the AC mode peak 2 peak amplitude, which is for the Auto Mode and L/C/R mode.

1. have you checked if the DUT res is 10kohm? you can use ST42 0.5V auto mode to do the On Board test.
2. Could you help us check the voltage on the 10kohm res when you use the RDC mode and Fluke doing the test?

Yeah I think you are right, its across FET gate, then 1k series resistor into an IO pin. So maybe one of those is slightly turning on. The Auto mode works very well most of the time, and here it detects as 10k exactly, so I should be in that mode instead.

For RDC:
The ST42 voltage across goes from 0.570V to 0.508V during its test.
The Fluke only delivers 80mV when testing the 10k. It may not be possible to achieve this low voltage.

One other Auto example, inductor of 4.9uH 100mR resistance:
- in 100/120Hz it will read as a resistor
- in 1kHz, 10kHz it reads as an inductor
Maybe intended, OK for me, as I should be in 10kHz mode when measuring inductance.