A general question about VNA calibration kit definitions/constants

[virtualparticles]

"The calibration coefficients for a mechanical calibration kit are the same for every piece with the same model number."
----------------------------

As a starting point - yes.  That's why most instruments have standard cal kit definitions pre-loaded so all the operator has to do is choose the correct kit number.

--------------------------

"The manufacturer does not measure each one and create new coefficients"
-------------------------------

No.  The manufacturer or a calibration lab does in fact measure the individual calibration components.  If they didn't - what would be the point of sending a new disk?  The cal lab would simply send a certificate that says the kit is within spec.

All the components of a particular part number share the same design specifications so that a lost or damaged part can simply be replaced by another with the same part number.  But, the reality is that nothing is perfect.  Real parts will have minute deviations from the design target.  These deviations can be characterized by a cal lab.  That's the point of the disk (or - today - likely a USB stick)

Using that data is the difference between a generic calibration and a traceable calibration.

No doubt this is inconsequential for most applications - especially for those of us working with instruments that have an upper frequency limit of 3 GHz or 6 GHz.  So, for the original question.  If the mechanical definitions for the Anritsu kit can be located, a cal kit file can be created in the format recognized by the target instrument.

Here is an Agilent diagram that illustrates the difference:

That's why I mentioned the "data based" approach to a calibration kit. So I'll state again, NO the manufacturer does not measure each standard kit and create new coefficients for each one. Yes, it is possible to obtain "data-based" data for a kit. I've created these data-bases here in the lab so I do actually know what I'm talking about.

[Joel_Dunsmore]

Just ran across this thread. I can't speak for all companies, but for Keysight (and this should not be construed as an official response), all kits of a particular model number have the same calibration coefficients, so all 85033E kits have the same C0, C1, C2....  There is another kind of kit called a verification kit, and those have verification artifacts that do have individualized measurements of each part.  On sending a new disk: not sure, I'll have to check but from time to time we do change the values of the constants if we have a refinement in the calculation of the model, so we might send a new disc if the definition changed. They don't change often. I will note, for example that a kit specified to 6 GHz might have different constants than a kit specified for 18 GHZ even if the physical part is identical, because the open circuit capacitance cannot be adequately described by 5 values (the C values and delay). The polynomial does not exactly fit the Bessel function curve that mostly describes the open circuit capacitance (which is the key thing in the kit, along with the shorts' offset delay.  The set of coefficients for the 6 GHz kit fit the 6 GHz performance best but are not good at 18 GHz. The 18 GHz fit well over the whole band but are slightly worse at 6 GHz.  Above 50 GHz, the polynomial fit falls apart rather badly so we use databased fits.  But all the data-based numbers are the same.  We do have very specialized kits called CD (characterized device kits) we use internally and occasionally sell to NIST or NPL or PTB.  They are individually characterized.  For the most part, the specifications are derived from the mechanical dimensions of the kit which are controlled so well that the constants for each kit are essentially the same (precision Swiss screw machines then precision plating).  The biggest thing that changes the performance is damage to the slotless connectors, and physical distortion due to over-torqueing which can change the pin-depth and cause the model to not be correct.

[Bud]

Good to have you joining the forum, Mr. Dunsmore!   

[analogRF]

Just ran across this thread. I can't speak for all companies, but for Keysight (and this should not be construed as an official response), all kits of a particular model number have the same calibration coefficients, so all 85033E kits have the same C0, C1, C2....  There is another kind of kit called a verification kit, and those have verification artifacts that do have individualized measurements of each part.  On sending a new disk: not sure, I'll have to check but from time to time we do change the values of the constants if we have a refinement in the calculation of the model, so we might send a new disc if the definition changed. They don't change often. I will note, for example that a kit specified to 6 GHz might have different constants than a kit specified for 18 GHZ even if the physical part is identical, because the open circuit capacitance cannot be adequately described by 5 values (the C values and delay). The polynomial does not exactly fit the Bessel function curve that mostly describes the open circuit capacitance (which is the key thing in the kit, along with the shorts' offset delay.  The set of coefficients for the 6 GHz kit fit the 6 GHz performance best but are not good at 18 GHz. The 18 GHz fit well over the whole band but are slightly worse at 6 GHz.  Above 50 GHz, the polynomial fit falls apart rather badly so we use databased fits.  But all the data-based numbers are the same.  We do have very specialized kits called CD (characterized device kits) we use internally and occasionally sell to NIST or NPL or PTB.  They are individually characterized.  For the most part, the specifications are derived from the mechanical dimensions of the kit which are controlled so well that the constants for each kit are essentially the same (precision Swiss screw machines then precision plating).  The biggest thing that changes the performance is damage to the slotless connectors, and physical distortion due to over-torqueing which can change the pin-depth and cause the model to not be correct.

First of all, welcome to the forum  Your experience and insight in RF measurements and specially VNA measurements are highly valuable and greatly appreciated.
It's hilarious that under your eevblog user ID it says 'Newbie'   

Also thanks for your comprehensive and illuminating answer.

So does this mean that if I e.g. choose 85033E kit on my ENA and then use a 85033E kit for calibration, I do not need to actually have the floppy disk that should have come with the kit (because I suppose the cal constants are saved on ENA already and they are universal for this kit)?

[joeqsmith]

Just ran across this thread. I can't speak for all companies, but for Keysight (and this should not be construed as an official response), ...

Welcome to the forum.  I hope you continue to contribute.   I can't promise I won't ask you about your thoughts on Carly.

[virtualparticles]

I like how Joel shows up as "newbie" on here... *snort* 

[eb4fbz]

Attached photo from my 3750R kit

[analogRF]

Attached photo from my 3750R kit

but do you have (or do you need) the full cal data on floppy? or is this printed data all you need to set it up on a VNA?

[virtualparticles]

I have received an Anritsu Cal Kit for my ENA which is in pretty good shape but the floppy containing the cal constants (coefficients) is missing.
Now I have three questions

1-I know the cal constants are tied to the particular set with particular S/N for obvious reasons, but in practice, how much do I suffer
if I just use the data from another identical cal kit? I am not measuring anything above 4GHz (possibly to 6GHz at most)

2-Do you think it is possible to get the cal data for my cal kit from Anritsu?

3-does anybody have the cal definitions for Anritsu 3750R cal kit to share?

You can fill in the capacitance and delay for the open based on this data and put in the delay and zeros for the inductance of the short. The load is assumed to be perfect. Keep in mind that the Anritsu and Keysight coefficients are scaled differently so you'll have to make a few changes to the numbers.

[Joel_Dunsmore]

"Does this mean.... I do not need to actually have the floppy disk that should have come with the kit (because I suppose the cal constants are saved on ENA already and they are universal for this kit)?"

Yup. Means exactly that. You only need the 5 values for the open.  I don't remember if the ENA even used inductance for the short (8753 did not, but 8720 did, although the difference was de minimus).  PNA does, but that's just because the the guy that did the code was a bit pedantic. 

[Joel_Dunsmore]

Keysight uses delay as a time, not a distance.  so convert cm to meters, then meters to seconds (delay in meters/speed of light).

[joeqsmith]

I have been looking into buying a used Anritsu 3750R kit because I wanted a second 3.5mm cal kit of at least half decent quality. For that reason I saved the attached picture from an eBay auction. No idea if the data applies to your kit and what Option 1 is. It seems that the shorts are only corrected by an offset length, with no offset loss or frequency dependent inductance. There is no offset loss for the open as well.

The member who helped me with my home made standards had procured a kit from Kirkby.  ...

I think that was me. 

Yes sir, you were very helpful indeed.   

You may want to have a look at:
 
https://www.amazon.com/Economical-Professional-VNA-Calibration-CALKIT-08/dp/B08LD2JYNL/ref=sr_1_9?dchild=1&keywords=vna+calibration+kit&qid=1621253612&sr=8-9

They have a website that has a fair amount of data for the kits they offer.  It may be a step up from the Kirkby kit and a little less cost.  They characterize each set.  Amazon no longer shows the 6GHz kits but they did offer them as well.  Maybe they sold out.     

[analogRF]

I have been looking into buying a used Anritsu 3750R kit because I wanted a second 3.5mm cal kit of at least half decent quality. For that reason I saved the attached picture from an eBay auction. No idea if the data applies to your kit and what Option 1 is. It seems that the shorts are only corrected by an offset length, with no offset loss or frequency dependent inductance. There is no offset loss for the open as well.

The member who helped me with my home made standards had procured a kit from Kirkby.  ...

I think that was me. 

Yes sir, you were very helpful indeed.   

You may want to have a look at:
 
https://www.amazon.com/Economical-Professional-VNA-Calibration-CALKIT-08/dp/B08LD2JYNL/ref=sr_1_9?dchild=1&keywords=vna+calibration+kit&qid=1621253612&sr=8-9

They have a website that has a fair amount of data for the kits they offer.  It may be a step up from the Kirkby kit and a little less cost.  They characterize each set.  Amazon no longer shows the 6GHz kits but they did offer them as well.  Maybe they sold out.   

these seem to be pretty good for the price and they do provide quite a bit of data
does anybody have any experience with this company?