TH2830 vs TH2832

[hpw]

BTW:

Did someone measured a 0.1 ... 1uH SMD RF coil?

as may Hantek 1832C at 100kHz shows up only using 47uH, using an 4-pol Tweezers.. below as at lost 

Hp

[KungFuJosh]

Did someone measured a 0.1 ... 1uH SMD RF coil?

I'm waiting on a Mouser order with the following:
10nH
100nH
10uH
100uH
1mH
10mH

[Martin72]

Nano henry...wohoo...

[KungFuJosh]

Nano henry...wohoo...

Sounds like a superhero when you spell it out. 😉

BTW- I ran the test again, this time with an official Tonghui kelvin lead fixture using a CMF 1% 100kΩ resistor. Prior to each test set, and after each firmware flashing, I let the meter warm up a minimum of 30 minutes before performing new open/short/dcr corrections.

Results:



Did you test your ST2830 with 100 kΩ with frequencies besides 100 kHz?

I have a theory that the reason I have issues with the ST firmware is that because they're older, they're likely only compatible with C6 or A6 hardware, and won't work right on C7 or A7 hardware (beyond any bugs). That would be one possible explanation why I have issues with the same version of ST2830 fw as you...assuming you are able to read 100kΩ across all the frequencies with better results.

[Martin72]

Hi,

https://www.eevblog.com/forum/testgear/th2830-vs-th2832/msg4953937/#msg4953937

DC,100Hz, 1kHZ, 10kHz, 100kHz...

If you want other frequencies inbetween, sorry man, for the next 2weeks I couldn´t get the ST2830 for measure..

[KungFuJosh]

If you want other frequencies inbetween, sorry man, for the next 2weeks I couldn´t get the ST2830 for measure..

Screw that, go to work! 😉

Nah, I don't need anything else. Clearly the firmware works better on your A6 hw than on my A7 hw, assuming that's the responsible difference. I wonder what's actually different. Both of the mainboards were made in 2019 or earlier.

Although it's clear that my 2830 results were more consistent with the newer TH fw.

[Martin72]

Screw that, go to work!

Looking at your "results"...
This will probably be some correction data that does not work with your device.
The cross-check would be to load YOUR firmware on another device.

[KungFuJosh]

Screw that, go to work!

Looking at your "results"...
This will probably be some correction data that does not work with your device.
The cross-check would be to load YOUR firmware on another device.

My firmware was given to me from another user. I asked what his hardware ID is. If we're both A7, then that doesn't tell me anything, but if his is A6, and both work with the same newer fw, then that's at least a clue.

You could load both the TH and ST 2830 fws on your work meter if you want. You already know the ST version works fine. 😉

[Martin72]

If then something is going wrong, I´m dead..
Jokes aside, with every firmware "update" the previous calibration is void.
Another thing:

I'm waiting on a Mouser order

Part numbers please..

[KungFuJosh]

Part numbers please..

Attached is a list of what I'm using for testing the LCR.

[hpw]

Did someone measured a 0.1 ... 1uH SMD RF coil?

I'm waiting on a Mouser order with the following:
10nH
100nH
10uH
100uH
1mH
10mH

Nice ... may the question rises about the used tweezers as I use a Kelvin 4 pol one. As may this draws soon an issue.

But first lets see what you get.

Hp

[KungFuJosh]

Nice ... may the question rises about the used tweezers as I use a Kelvin 4 pol one. As may this draws soon an issue.

I don't like wired tweezers at all. I use an SMT fixture, or Shannon ST-42 tweezers for SMT stuff.

I actually disassembled the tweezers that came with my DE-5000 to make kelvin clips.

[hpw]

Nice ... may the question rises about the used tweezers as I use a Kelvin 4 pol one. As may this draws soon an issue.

I don't like wired tweezers at all. I use an SMT fixture, or Shannon ST-42 tweezers for SMT stuff.

I actually disassembled the tweezers that came with my DE-5000 to make kelvin clips.

Hmm, your looks like the TH26008A Magnetic ring fixture ...

as performance:

"When the test frequency is ≥100kHz, the capacity is ≤3pF, or the inductance is ≤1µH, this type of SMD fixture is suitable."

so below 1uH not supported nor possible for any measurements. 

My question is also on this "TH26008A Magnetic ring fixture", how to go in the supported MHz regions as the TH2830 max as 100kHz.

Hp

[KungFuJosh]

Hmm, your looks like the TH26008A Magnetic ring fixture ...

as performance:

"When the test frequency is ≥100kHz, the capacity is ≤3pF, or the inductance is ≤1µH, this type of SMD fixture is suitable."

so below 1uH not supported nor possible for any measurements. 

My question is also on this "TH26008A Magnetic ring fixture", how to go in the supported MHz regions as the TH2830 max as 100kHz.

You read that backwards. It says less than or equal to 1µH. We'll see what it does in practice. It's not the same as that anyway, it's a knockoff. Real world testing results may vary. 😉

[Martin72]

Nano henry...wohoo...

And for good reason...
The nice thing about known components is that you have/get data sheets for them.
So you can verify whether the values are correct.
To be on the safe side in this respect, you measure as the manufacturer had done when specifying the values, so far so good and clear.
And now comes the "Wo-hoo"...
I rummaged through all the data sheets that could be found for your parts list and listed the respective test frequencies that have to be used to arrive at the same values.
And that's where it usually gets sporty with an LCR measuring device that can "only" do 100khz/200kHz....
This is just information for the case that you measure "strange" things with your TH2830/32.

Martin

[mawyatt]

We have the handheld Tonghui TH2609B tweezers, they work well but are not as good as the dedicated SMD fixtures we've found for repeatability of measurement, even the two no-name SMD fixtures we have (copies of the 120MHz TH26008A) perform better (especially after some internal mods & adjustments).

Awhile back we did a direct evaluation of these no-name fixtures, mainly with our Tonghui TH2830 and Hioki IM3536 LCR meters. The test was simple and designed to verify IF these no-name SMD fixtures could be reliable in measurement. Various quality SMD components were compared between the two instruments with the same fixture transferred, and proper warm up and short/open calibration invoked. During this we discovered the fixtures lacked some internal shielding to the case an corrected such, then found the attachment to the plungers were loose and could introduce measurement variations, these were tightened and all the test repeated. This proved the modified and corrected fixtures were reliable and produced almost identical results between the two instruments with a variety of SMD devices over the frequency range of 50Hz to 100KHz (limits of the TH2830), then went beyond with the IM3536 (4Hz to 8MHz) just to "see" if any anomilities showed up in the measurements, none did.

So we concluded these SMD fixtures were good enough for our use, of course YMMV

Note also verified the handheld DE-5000 did well when using these SDM fixtures (created an adapter for such).

Edit: Also, while on the discussion of fixtures and such, we've found the common Kelvin type clips with the individual cables to BNCs work fine at DC, and very low frequencies (like for use measuring electrolytics), but don't behave well with DUTs that have high impedances (small caps for example) as frequency increases. If you note the quality Kelvin clips like the TH26011CS they only recommend to 100KHz, and the 4 individual Kelvin leads are ~250mm long and come from a box which mechanically terminates each lead into a combined single cable which is the major part of the ~1M length back to a fixture which hosts the 4 BNC connectors. This is more stable mechanically and electrically than the common 4 individual Kelvin leads types with BNCs, especially considering these are usually only a single shield cables which are not 100% effective shielded (here double shielding would be desired, but cost more and less flexible).

For leaded DUTs we've found the TH26048A works well and is speced to 13MHz.

Best,

[KungFuJosh]

So we concluded these SMD fixtures were good enough for our use, of course YMMV

Well, yeah, it's your fault I bought the fixture. 😉

For leaded DUTs we've found the TH26048A works well and is speced to 13MHz.

I like that too. I haven't had any issues with the Th26011CS either.

[mawyatt]

Well, yeah, it's your fault I bought the fixture. 😉

Yeah, we generally don't recommend things, usually just state what we have and use and leave it at that!!

However, these inexpensive SMD fixtures work so well (after mods/fixing), even with handhelds like the DE-5000 (with adapter), we wanted others that are on a budget, either hobbiest or small business, to benefit from such with better, more repeatable measurements.

If we had discovered these before we purchased the somewhat expensive (~$180) TH26009B Kelvin tweezers, well we could have saved some $, nothing wrong with them but they aren't used much anymore.

Best,

[KungFuJosh]

Calling @tv84 - we have a new toy to play with. 😉

This one has a date of 2021-06-25, and is supposed to be compatible with the newer hardware.

Once again, Sourcetronic customer service wins. They contacted "the firmware developer" who sent them this.

Thanks,
Josh

[Martin72]

They contacted "the firmware developer" who sent them this.

A-haaa...

[tv84]

A-haaa...

No, it was not me! 

[KungFuJosh]

A-haaa...

No, it was not me! 

It is now. 😉😉

[tv84]

It is now. 😉😉

Here.

Header: 41424344F07D0700A03E1400
Signature: "ABCD"  (MAGIC OK)
Block1 Size:  00077DF0   [00000018-00077E07]
Block1 Checksums: 3BF0   [00077E08-0007B9F7]
Block2 Size:  00143EA0   [0007B9F8-001BF897]
Block2 Checksums: A1F5   [001BF898-001C9A8C]

[KungFuJosh]

Here.

Thank you!

All the previous bugs seem to be gone so far. It remembered the previous state upon startup too. I'll do a full test soon when I receive my test PCB.

Version:



Quick test:

I did not give the meter the required 30 minutes before running correction. Probably not even 30 seconds lol. I'll do it properly when I do the full test.

[Martin72]

I'll do it properly when I do the full test.

And then with open/short calibration...

I'll do a full test soon when I receive my test PCB.

Could be interesting what you´ll measure..I´m curious.

It´s itching in my fingers to take the file and.....No.