Author Topic: Mini-teardown: Omicron B-WIC impedance test adapter  (Read 9199 times)

0 Members and 1 Guest are viewing this topic.

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Mini-teardown: Omicron B-WIC impedance test adapter
« on: December 30, 2017, 07:14:21 pm »
Ever since I was exposed to the Omicron Bode 100 system by forum member Pitrsek, I've dreamed of one materializing on my bench. It appears to be the cat's meow for low-level impedance measurement up to ~30MHz. I want one. But the price new is too dear. And they're snapped up immediately when they appear used.

Needless to say, a Bode 100 hasn't materialized. But a couple of their impedance test adapters did; the B-SMC and B-WIC. I was planning on making something like these, but these are better executed than I could make here. From what I've seen, they appear to be the best general-purpose impedance test fixtures you can buy. So the mission is, can I make them work with my Anritsu (or any other) VNA or FRA ?

First up is the B-WIC, designed for thru-hole components. Nicely thought out, with gold-plated contacts, a strong actuating spring for good contact, and sitting in a machined teflon carrier.

So what's inside, Johnnie? Five 1/4W metal film resistors. The circuit is attached below. I don't yet know the configuration of the Bode 100's inputs and output when using this adapter; I'm hoping Pitrsek can help us with that. It looks like they might all be 50 Ohms. What I do know is that the Bode 100 has an entirely separate mode when using the adapters, and requires full Open-Short-Load compensation sweeps be made before using them. So this is not a plug-and-play device as it sits.

The CH 1 output (to the reference channel) is attenuated by ~27dB. Maximum input to the fixture is stated as 1VRMS, so I'm guessing the analyzer has to have good dynamic range. The reference impedance is formed by the two 4R7's in parallel at the CH 2 output, followed by a 47 Ohm buffer R. So after applying the OSL corrections to T/R, the magnitude vector * 2.35 would give the impedance in Ohms, correct?
« Last Edit: January 20, 2018, 03:15:48 am by precaud »
 
The following users thanked this post: mcinque, r0d3z1

Offline petemate

  • Regular Contributor
  • *
  • Posts: 126
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #1 on: December 30, 2017, 08:21:09 pm »
I don't yet know the configuration of the Bode 100's inputs and output when using this adapter;

You can just download the software at https://www.omicron-lab.com/. It has a graphical representation of how everything is connected in various modes, which makes it easy to understand. Its also nicely represented in the manual: https://www.omicron-lab.com/fileadmin/assets/manuals/Bode-100-User-Manual-ENU10060503.pdf Check out page 37.
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #2 on: December 30, 2017, 11:15:26 pm »
Yes, I looked at the manual, the section on using the impedance adapters starts on pdf page 64. The internal configuration is not shown; they only say that it is "configured correctly".
 

Offline petemate

  • Regular Contributor
  • *
  • Posts: 126
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #3 on: December 30, 2017, 11:19:00 pm »
Yeah, thats why you download the software, set up the measurement and then click on the hardware configuration to see how its actually configured.
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #4 on: January 02, 2018, 07:02:08 am »
When used with B-SMC/B-WIC, Bode inputs are configured to 1Meg,
Output from the generator is 50Ohm
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #5 on: January 02, 2018, 12:44:57 pm »
Great, thanks. So it is basically two dividers in parallel, the DUT being a series element in one, the other one is just a 22.27X divider to better match signal levels for the analyzer. They can make up for it in the math, and probably use some preamp gain on both channels.

I need to search for bridge formulas for this configuration. I've only worked with DUTs as shunt elements when measuring Z.

When used with B-SMC/B-WIC, Bode inputs are configured to 1Meg,
Output from the generator is 50Ohm
« Last Edit: January 03, 2018, 03:11:47 pm by precaud »
 

Offline rx8pilot

  • Super Contributor
  • ***
  • Posts: 3634
  • Country: us
  • If you want more money, be more valuable.
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #6 on: January 02, 2018, 10:04:19 pm »
The Bode100 system is a really cool instrument - I just cannot justify the cost. I am sure they have plenty of customers that are not really price sensitive so that can charge what they want. Guessing the build cost is rather small and they are not interested in dealing with price shoppers like me.

I saw one come up on eBay, but only for a moment. It was sold really fast.
Factory400 - the worlds smallest factory. https://www.youtube.com/c/Factory400
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #7 on: January 03, 2018, 03:08:00 pm »
The Bode100 system is a really cool instrument - I just cannot justify the cost.

Same here. But I do like their attention to detail, and willingness to think outside the box to optimize measurements. These fixtures are an example of that. In their defense, their support is excellent. That has to be built into the price.

A question for Pitrsek: When you're using the Bode 100 in "Fixture" mode, do you run a "Through" before doing the Open-Short-Load sequence? If so, is the Through setup done with the fixture connected and with no DUT (same as Open) ?

My previous assessment of the measurement setup was basically correct. It is a "Series-Thru" topology, with tweaks to optimize the dynamic range. The basic topology is discussed in good detail and compared to other impedance measurement topologies in this Agilent app note:

https://www.keysight.com/upload/cmc_upload/All/ChallengesandsolutionsforImpedance.pdf

The Series-Thru circuit is shown on page 66 (the Gain-Phase one). They describe this technique as best used at mid-to-high impedances, from a few Ohms and up. So why the uncertainty at lower impedances? The problem mostly centers around the fact that it uses the 50 Ohm termination inside the analyzer as its reference impedance. That choice makes it generic and portable from one instrument to the next, but it also inserts a high (and inconsistent) amount of series R and L, and shunt C, from cables, connectors, relays, pcb traces, etc, between the DUT and reference R. Simply by moving the reference R next to the DUT where it belongs, optimizing its value, and changing the math to match it, you can extend the usable range of this technique down quite a bit, into the milliOhm range. That's what Omicron did. Simple, clever, and it appears to work.
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #8 on: January 04, 2018, 05:07:35 am »
OK, I've modified my Z measurement software for the MS420 to accept the B-WIC. The results are mostly good, with one remaining issue.

Here is a plot of three current sense R's. Each one is plotted twice; one with Open-Short-Load (OSL) compensation applied, and one without, so we can see what difference the OSL makes.

The formula (polar coordinates, HTBasic syntax) to apply to the measured S21 magnitude at each freq to get correct impedance value is:
RealS21 =  10^( (S21 - 26.956) / 20.)                    ! subtract the fixture's CH 1 divider from the MS420's dBr, and convert to a voltage ratio
Zmag_Ohms = ( ( 1. - RealS21) / RealS21) * 2.35    ! Series-thru impedance with 2.35 Ohm shunt R
It is worth noting that the phase term plays no part in these equations, and is carried through unchanged. All they're doing is scaling the magnitude vector.
OSL corrections are applied after this. When they are being measured, they are also processed by the above before being stored.

It should be noted; I do not have the standard short-load card for the B-WIC yet. One is on the way. I used a 3/4"-wide gold-plated shorting bar, and an 1/8W 0.1% 100 Ohm axial resistor for the OSL measurements. The difference between them and the factory-supplied standards could be impacting the results. I'll know in a few days when the standard one arrives.

Some observations on the results:
: The OSL correction is what makes it useful below 1 Ohm or so. Without it, the impedance accuracy is really poor down there.
: My OSL-corrected values run about 2% high. This could be because I didn't include the 50 Ohm source R of the generator in the divider value. And/or the 1% tolerance parts used in the fixture. Easy to correct for.
: Above a couple Ohms the accuracy becomes increasingly good enough to not use the OSL. This agrees with Agilent's findings with generic Series-Thru. At 10 Ohms the curves were identical up to the 10MHz plot limit.
: Overall, I would agree that, with OSL, 20mOhm is the lower limit for this topology. Below that, the trace gets too noisy to be useful. It is already too noisy at 25mOhm for my needs.
: These sweeps were done with a 10Hz IF bandwidth. To see if I could clean the trace up, I dropped the IF BW to 3Hz. The difference was minimal. Not enough to offset a tripling of sweep time. I then increased drive level by 10dB and dropped receiver sensitivity to match. No difference. The S21 short sweep is only 60dB down so this isn't a dynamic range problem.
: Below a couple Ohms, the Z curves with OSL are not accurate at high freqs. The 25m and 100m resistors are axial current sense types and should show more inductance than they do.
: The phase curves are not accurate above approx. 100kHz , with or without OSL, going capacitive while the Z was going inductive. The native sweeps on the analyzer showed it too, so it's not a math issue; it's coming out of the fixture that way. I did not run a "thru" sweep on the analyzer before doing these, the curves without OSL show a ~ -7ยบ phase mismatch between the channels. This is not in the analyzer; it is coming out of the fixture. OSL corrects the mismatch but the high-freq error remains. Perhaps Omicron applies an inductive constant correction factor?

I'll withhold further comment until the short-load pcb arrives.
« Last Edit: January 04, 2018, 03:21:45 pm by precaud »
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #9 on: January 04, 2018, 10:37:11 am »
The Bode100 system is a really cool instrument - I just cannot justify the cost. I am sure they have plenty of customers that are not really price sensitive so that can charge what they want. Guessing the build cost is rather small and they are not interested in dealing with price shoppers like me.
Actually when we have been searching for the FRA/Impedance analyzer, the Bode100 was the most affordable one. AP300, FRA from Venable, Agilent... all are times more expensive. I agree tat for one man shops it is pricey, but I'd much rather have FRA/Impedance analyzer and average scope, than high end scope and no analyzer (my personal opinion with regard to stuff that I do, YMMV).  The service and support is top notch  :-+.

@precaud - with fixtures there is just OSL

 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #10 on: January 04, 2018, 02:07:32 pm »
OK, thanks @Pitrsek. I agree with your assessment re: cost. The Bode 100 is less than half its competitors. Has much better software. And good accessories/fixtures. It looks expensive for us bottom-feeders, but for equivalent capability, it's a very good value proposition. (I didn't include Cleverscope FRA in this because its not a complete system, but it is very attractive too.)

The mind was crunching away at this inaccuracy problem even while I slept. The thought is, that the phase error is caused by:
a) my short reference is not good enough (I saw this with the shunt-thru), and
b) the different ways that the two outputs to the VNA create their nominal 50 Ohms output.

The Ref channel does it by the shunt R of the divider (which divides everything, not just the R component), while the test channel does it through a buildout R, which leaves reactive elements intact. Swamps them, yes, but they are still intact. So my R divider attenuation correction is valid at low freqs, but needs an inductive constant to be added back in to be accurate at high freqs.

I would bet my boots that Omicron applies such a constant. (If I am wrong, I would like to keep my boots until winter is over, please...)

When measuring Z/phase at high freqs, these little details make a huge difference...
 

Offline BartSchroder

  • Newbie
  • Posts: 3
  • Country: nz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #11 on: January 04, 2018, 06:58:35 pm »
Hi,
As a comparison, I checked our CS328A-FRA with a 10 mOhm resistor. The frequency at which the impedance doubled (ie Zl = Zr) is 2.9 MHz. So that suggests the fixture series inductance is L= 10m/2 pi 2.9M = 0.54nH.

Looking at the Bode 100 graph up above, witha  25 mOhm resistor, I guessed the doubling at about 1 MHz, so L= 25m/2 pi 1M = 4nH.  However, as Precaud pointed out, the Bode plot is capacitive, which I would not have expected, so take this with a grain of salt.

So some notes:
- The CS328A-FRA test jig has about 7x lower series inductance than the Bode 100 one.
- The CS328A-FRA response is much less noisy
- The CS328A-FRA should be usable to quite a bit lower than 10 mOhm, all limited by the series inductance. With 0.54nH, the doubling for 1 mOhm would be F = X/2 pi L = 1m/2 pi 0.54n = 294 kHz. So you could use it with 1 mOhm up to about 300 kHz or so.
- I did this with a bandwidth of 100 Hz. So 'd be guessing that it should complete way faster than the Bode 100.

Attached is the plot.

cheers, Bart
« Last Edit: January 04, 2018, 07:13:52 pm by BartSchroder »
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #12 on: January 04, 2018, 07:46:26 pm »
Thanks for weighing in, Bart. I guess I should make clear:
: I'm not evaluating or comparing the performance of impedance measurements systems.
: The plots I posted were not made with a Bode 100 system.
: The B-WIC is a fixture for thru-hole components, so the resistors being tested are axial- and radial-lead types, which have higher self-inductance than SMD types.

The purpose of this thread is to evaluate the B-WIC fixture and figure out how to use it with other VNA's and FRA's (including yours). It is a work-in-progress, and the results to this point are not final.

I'll be inviting the Omicron guys to chime in (if they wish) once I get their standard short-load pcb and update the measurements to reflect its use.
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #13 on: January 05, 2018, 01:05:28 am »
I think I found the problem. See attached graph. It shows the freq response of the CH1 and CH2 outputs separately, with DUT terminals open and shorted.

The CH2 output (yellow) behaves as one would expect; a flat response becoming increasingly inductive when open (capacitive leakage through the fixture, I would guess).

The freq response of the CH1 output (cyan), the Reference channel, the channel with the ~27dB L-pad attenuator, varies as a function of the DUT's impedance. It is flat only when the DUT's impedance is near the output resistance of the L-pad, around 45 Ohms. For DUT's below 45 Ohms, this "reference" sweep becomes increasingly inductive. When the reference is inductive, the phase of the DUT is going to appear capacitive whether it is or not.

You can correct for this if you have access to the raw data for each channel before doing the T/R division. I couldn't get it to work by doing it after the fact. I tried adding inductance and removing capacitance from the measurement to get the phase to match the impedance curve. For a 100mOhm resistor it took adding 40nH of inductance to get the phase to match the Z curve at 1MHz. But there was still a "capacitive" rolloff starting at 5MHz.

The logical thing to do it put the CH1 into a 50 Ohm load, change the L-pad attenuation so the impedance values are correct, and try again.

Then again, it could just be the inductance of the long leads of the 1k0 series R in the divider causing all of this.

This has turned into more than I bargained for...

EDIT: Changing CH1 to work into 50 Ohms made not difference.
« Last Edit: January 05, 2018, 01:25:04 am by precaud »
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #14 on: January 05, 2018, 04:42:10 am »
I tried the fixture with another analyzer (an old AP 102B) and I'm not seeing nearly as much of the odd phase behavior with it. So either the R channel on my Anritsu has an issue (I just cal'ed it last month and it was fine...), or else its floating inputs are a problem with this topology. The 102B has common ground for inputs and output and the phase looks better for CH 1's magnitude curve, though the transfer function with shorted DUT still does show negative phase shift. I'm temped to delete the posts with the errant measurements until I get this sorted out. My apologies...
« Last Edit: January 05, 2018, 05:04:29 am by precaud »
 

Offline r0d3z1

  • Regular Contributor
  • *
  • Posts: 116
  • Country: it
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #15 on: January 12, 2018, 01:21:57 pm »

The formula (polar coordinates, HTBasic syntax) to apply to the measured S21 magnitude at each freq to get correct impedance value is:
RealS21 =  10^( (S21 - 26.956) / 20.)                    ! subtract the fixture's CH 1 divider from the MS420's dBr, and convert to a voltage ratio
Zmag_Ohms = ( ( 1. - RealS21) / RealS21) * 2.35    ! Series-thru impedance with 2.35 Ohm shunt R
It is worth noting that the phase term plays no part in these equations, and is carried through unchanged. All they're doing is scaling the magnitude vector.

I am not sure to have understand well this 2 equation, but are you sure that Omicron is using T/R measurement, couldn't that be a I-V measurement ?
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #16 on: January 12, 2018, 02:00:49 pm »
I am not sure to have understand well this 2 equation, but are you sure that Omicron is using T/R measurement, couldn't that be a I-V measurement ?

The math doesn't work for any I-V method I tried. Those measure across a standard resistor, with the DUT shunted to ground. If you look at the circuit diagram in the first post, and the Keysight paper I referenced, you'll see it is a "classic" Series-Thru setup. The only difference is, Omicron adds the ~27dB divider in the Ref channel. This equalizes the output levels for each channel which optimizes levels for the ADC that follows. It's the same as what Test channel sees when DUT is a short (50 / 2.35 = 27dB). That's how they get an impedance measurement range that is better-than-usual from this setup. It's a simple and clever solution. But it seems to fall apart with large capacitors. Choosing a smaller ref resistor would extend the range but throws away even more level for low impedances and you'd need more make-up gain in the preamp. It's something I may experiment with.

I'm still waiting for the short-load card to arrive to verify my results. It is supposedly in the mail...
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #17 on: January 12, 2018, 02:32:08 pm »
I believe that  r0d3z1 is onto something... With series trough, you have 50Ohms to ground, with B-WIC the 50 on the right side is just termination and leads to 1Meg input. There is 4R7//4R7 to ground. Basically just sampling current. I've attached measurements of 1000 and 2200uF capacitors. The wobble in phase is probably result of low signal magnitude/low measured impedance.
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #18 on: January 12, 2018, 03:11:02 pm »
I believe that  r0d3z1 is onto something... With series trough, you have 50Ohms to ground, with B-WIC the 50 on the right side is just termination and leads to 1Meg input. There is 4R7//4R7 to ground. Basically just sampling current. I've attached measurements of 1000 and 2200uF capacitors. The wobble in phase is probably result of low signal magnitude/low measured impedance.

Yes, with "classic" Series-Thru you have 50 Ohms to ground. But I think that is only because most VNA's have 50 Ohm inputs. So it's only a change from 50 Ohms current sense to 2.35 Ohms current sense. That's how they can measure below 1 Ohm.

If anyone sees it different and can flesh it out, I'm certainly open to it. But as of now, I see no reason to think it is other than a "tweaked" Series-Thru setup.

Pitrsek, thanks for posting your measurements for 1000 and 2200uF caps. They are in the range one would expect. So no issues with large capacitances....


 

Offline r0d3z1

  • Regular Contributor
  • *
  • Posts: 116
  • Country: it
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #19 on: January 12, 2018, 03:13:46 pm »

The math doesn't work for any I-V method I tried.


Zmag = ( (V1/V2) - 1) * 2.35 with V1 and V2 RMS value, the phase could be measured as delay between source voltage and current. This is like what you do with signal generator and scope, isn't it ?

However, precaud could you please explain your formula in a more clear way, I have few experience with Scattering parameters and VNA.
 

Offline precaudTopic starter

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
    • LinearZ
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #20 on: January 12, 2018, 04:32:55 pm »
Zmag = ( (V1/V2) - 1) * 2.35 with V1 and V2 RMS value, the phase could be measured as delay between source voltage and current. This is like what you do with signal generator and scope, isn't it ?

Yes, but that is a different topology, with series reference R and the DUT to ground. This topology is different; the DUT is in series and the reference R is to ground.

I refer to the Agilent paper because it shows all the topologies and gives the basic formulas for each one:
https://www.keysight.com/upload/cmc_upload/All/ChallengesandsolutionsforImpedance.pdf

The V/I topology you refer to is shown on p. 26. The formula for Zmag is the same as the one you cited.

Quote
However, precaud could you please explain your formula in a more clear way, I have few experience with Scattering parameters and VNA.

The "Series-Thru" method is shown on p. 66 . It is the same as the Omicron fixture's topology, except: 1) it shows the VNA's 50 Ohm input as the reference R, where Omicron uses 2.35 Ohms; and 2) Omicron adds a 22.276X L-pad in the V1 channel. The formula is given as:
Zmag = (50 x 2) x ((1 - S21) / S21)    where S21 = V2/V1 (or Vt/Vr)    (I wonder if they made a typo, and it should be (50 || 50) ?)

In the formula, I replace the 50x2 Ohms with 2.35 Ohms.
Then we must adjust the V1 magnitude to correct for the L-pad divider in the fixture (1kOhms into 47 Ohms = -22.276 or -26.956dBr). The Anritsu gives V2/V1 in dBr, so I subtract 26.956 from it
So the formula becomes two steps:
The first step, X = 10^ ((S21db - 26.956) / 20.) gives correct V2/V1 as a voltage ratio.
Then Zmag = 2.35 * ((1 - X) / X)  calculates Series-thru impedance scaled to the 2.35 Ohm reference R.
When combined with open-short-load compensation (which the Omicron also uses), this gives correct results within tolerances for components (except for large capacitors at low freqs...).
The only thing I have not accounted for is the 50 Ohms source impedance. Maybe the first number should be (50 || 2.35) = 2.244 instead of 2.35 ?
I welcome your comments.
« Last Edit: January 12, 2018, 05:25:25 pm by precaud »
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #21 on: January 13, 2018, 06:56:30 am »
If I go from the schematic in 1st post:
 :-/O
U1=Uref*Att (attenuation factor of the divider)
U2=Uref*2,35/(DUT+2,35)

Uref=U1/Att
Uref=(DUT+2,35)*U2/2,35

U1/Att=(DUT+2,35)*U2/2,35
U1*2,35/(U2*Att)=DUT+2,35

DUT=(U1/U2)*(2,35/Att)-2,35
Att=47/(1000+47)

DUT=(U1/U2)*2,35/(47/1047)-2,35
DUT=(U1/U2)*52,35-2,35

What about this one?
EDIT:
Since the voltage is sampled at same node for both DUT and U1, I disregard the internal impedance of generator. Since both inputs are 1Meg Ohm, I disregard input currents as well.

What are your results without calibration? If you had a wrong formula, and make a calibration, wouldn't the calibration fix your results to a certain degree(slopes, offset)? So it might work, till you run out of signal gain/precision?....
Just a speculation on my part. I'll take a look at the S parameter definition...   
« Last Edit: January 13, 2018, 07:10:29 am by Pitrsek »
 
The following users thanked this post: precaud, r0d3z1

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #22 on: January 13, 2018, 07:32:39 am »
work in progress

EDIT:
 I'd like to mention that speaking about S parameters in systems that are not 50Ohms might make some people grumpy and may be not technically correct... but for the sake of augmenter let's live with it.

From the agilent paper:
Z=(50 x 2) x ((1 - S21) / S21)
Where 50 is the system impedance, ie the impedance on which you are measuring the reflected power/sampling the voltage that is proportional to current. This is an assumption on my end.
And S21=U2/U1 - in linear scale, NOT dB!!

Z=(2*Zref) * ((1 - S21) / S21)
Z=(2*Zref - S21*2*Zref) / S21)
Z=((2*Zref)/S21 - 2*Zref)
Since there is attenuator in U1 path, we need to boost fix S21 by 1047/47
lets say our Zref is 2.35
Z=((2*Zref)*(1047/47)/S21 - 2*Zref)
Z=104.7/S21-4.7 - if we divide by two
Z/2=52.35/S21-2.35
So our impedance is precisely half the previous result  :wtf:, what a coincidence...
My guess is that it is result of the 1Meg inputs, ie. the original formula is compensating 50R/50R divider somewhere on the way. But it's a wild guess. I need to dig a little bit deeper...

Any input?


« Last Edit: January 13, 2018, 08:46:13 am by Pitrsek »
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #23 on: January 13, 2018, 08:36:28 am »
In the formula, I replace the 50x2 Ohms with 2.35 Ohms.
Why did you that? What was the reasoning?
 

Offline Pitrsek

  • Regular Contributor
  • *
  • Posts: 171
  • Country: cz
Re: Mini-teardown: Omicron B-WIC impedance test adapter
« Reply #24 on: January 13, 2018, 09:06:30 am »
You can correct for this if you have access to the raw data for each channel before doing the T/R division.
You can't get Anritsu to spit out separate data for each channel(you have only S21 available)?
Not even for the 50Ohm inputs?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf