TH2830 vs TH2832

[GnomeZA]

"Today" I got the new firmware for our ST2830 from Sourcetronic (the guy asked me if I knew about the new firmware through a certain "Josh"...  ).
I then asked again what exactly had been changed, the question was passed on and I may get an answer soon (which I will then post here).
This is interesting because I have to pay attention to the calibration.

Any chance you could also share this ST2830 firmware Martin?
I think having both is not per-se a bad thing

[Martin72]

Hi,
No problem - tomorrow..

[Martin72]

A little late, but not forgotten...

https://drive.google.com/file/d/1VnfhRBVc8n2ndyWcnriZIla9GFPnQXkD/view?usp=sharing

[KungFuJosh]

One of my contacts at Sourcetronic was kind enough to share how to do the Dissipation Factor test. He said:

For the D test, Cs-D fundamentally has the same information as Cs-Rs, as the loss factor is directly connected to the ESR Rs.
That also means that you can fake the D by putting actual resistors in series to the capacitor, and you get different sizes of D depending on the resistance you put up.

So I tried it out, and it works. I threw different value resistors in series, and it gave DF values similar to what was expected using math (eww, gross)...just kidding, I used a calculator. 🤣

Measured Rs of the cap was about 29Ω at 1kHz. So using some standard R values, I can do something like this:
29Ω = 0.00018 (cap only)
79Ω = 0.00049 (29Ω + 50Ω)
162Ω = 0.00101 (29Ω + 133Ω)
797Ω = 0.00506 (29Ω + 768Ω)
1609Ω = 0.01010 (29Ω + 1.58k)
16029Ω = 0.10071 (29Ω + 16k)
160029Ω = 1.00530 (29Ω + 160k)

I'm sure those numbers aren't perfectly accurate, but they should be close enough to test it on a new test board. 😉

Thanks,
Josh

PS. Abbreviating Dissipation or Dissipation Factor as just D, leaves some hilarious statements out of context.

[KungFuJosh]

Here's an update on the Dissipation Factor test values.

Tested at 1kHz:
1nF cap (no series resistor) -> 0.00010
50Ω (actually 49.9) -> 0.00050
133Ω -> 0.00100
768Ω -> 0.00500
1.58 kΩ -> 0.01000
16 kΩ -> 0.10000
160 kΩ -> 1.00000

Those values are pretty spot on for the cap I used. I used Panasonic 0.1% 25ppm SMT resistors.

Thanks,
Josh

[KungFuJosh]

In case anybody is curious, you can correct for cable length on these meters. Not yet?

[Martin72]

Try switching to 1m or more....does ist work? On our ST2830 not(It is even described in the manual that it won´t work), but we didn´t have the latest firmware.

[KungFuJosh]

Oh, that's both annoying and interesting. It says "no cable data installed." Maybe time to email ST and ask them if they know anything about it.

[Martin72]

Yes, I noticed this when I connected the fixture from Aliexpress to our ST2830 last week and saw 0m on the display.
But we always use the Kelvinclips, so I tried to switch it after the tests and it didn't work.

[mawyatt]

Our TH2830 doesn't allow cable compensation either. This likely entails some calibration parameters and setup with test fixtures, there is also the issue with cable propagation constants which may not be the same for all cables.

If we did the math properly, at 100KHz a 1M cable with Propagation constant of 0.66 has a phase delay of 0.1818 degrees.

Best,

[KungFuJosh]

Our TH2830 doesn't allow cable compensation either. This likely entails some calibration parameters and setup with test fixtures, there is also the issue with cable propagation constants which may not be the same for all cables.

If we did the math properly, at 100KHz a 1M cable with Propagation constant of 0.66 has a phase delay of 0.1818 degrees.

EastTester claims to have cable correction in their ET35 series. Maybe they're just pretending? 😉

[Martin72]

They have it and it works, sorry.

[KungFuJosh]

Oh, weird! Mine just starting working too! 😉😉

[mow]

Maybe time to email ST and ask them if they know anything about it.

Just that it doesn't work, sorry.

[KungFuJosh]

Maybe time to email ST and ask them if they know anything about it.

Just that it doesn't work, sorry.

Fair enough. I'm sure you don't want to email them for 2 years to get them to fix it either. 😉

[2N3055]

Our TH2830 doesn't allow cable compensation either. This likely entails some calibration parameters and setup with test fixtures, there is also the issue with cable propagation constants which may not be the same for all cables.

If we did the math properly, at 100KHz a 1M cable with Propagation constant of 0.66 has a phase delay of 0.1818 degrees.

Best,

Calculation is perfect.  c is not. 

360/(299792458*0,66/100e3)
= 181,944051926264754314e-3

 
Just joking...