A general question about VNA calibration kit definitions/constants

[analogRF]

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?

[joeqsmith]

Why wouldn't you just ship them back for calibration?  They should be able to replace any damaged parts and provide you with new coefficients.  Unless you are setup with everything needed to check them, I wouldn't mess with them. 

[TheSteve]

A cal kit without the constants doesn't seem overly useful to me. I'd like to think another similar kit would get you pretty close though.
I like that the Keysight kits all use fixed cal constants that are the same with each kit(same with Maury Microwave for the most part).

You could also have someone else measure and calculate them if they have a good kit to measure against.

I once asked R&S for the constants for an older kit, they claimed they looked but no longer had any data from that far back. It never hurts to ask though, maybe Anritsu has better record keeping.

[analogRF]

Why wouldn't you just ship them back for calibration?  They should be able to replace any damaged parts and provide you with new coefficients.  Unless you are setup with everything needed to check them, I wouldn't mess with them.

I never said I want to mess with them. I would like to know how big of an error I have to expect
if I use the cal constant from another identical kit.

Are the manufacturing tolerances between different sets that big?

as for calibration by anritsu, I dont know if you have had to deal with them in the past or not, but in my 3 experiences with them
they are simply the biggest aholes in T&M compared to my similar experience with other manufactures. I'm afraid they are gonna charge me $2k-$3k for calibration if they even accept it from an individual/hobbyist. But I will contact them

[analogRF]

A cal kit without the constants doesn't seem overly useful to me. I'd like to think another similar kit would get you pretty close though.
I like that the Keysight kits all use fixed cal constants that are the same with each kit(same with Maury Microwave for the most part).

That's also my thinking. I dont think that manufacturing tolerances between different sets are that big to make a big impact
I am not looking for like metrology grade measurements!

as for keysight, I am not sure about that. As far as I know, the cal constants are tied to the set because they are measured individually.
It is usually written on the floppy that comes with the cal kit that it belongs to the kit with such S/N. Although there are
tons of these floppy data sets being sold on eBay without the cal kit ! 

if the Keysight ones are not tied to the set, then perhaps I am wrong about the Anritsu being tied to the set either.

[joeqsmith]

Why wouldn't you just ship them back for calibration?  They should be able to replace any damaged parts and provide you with new coefficients.  Unless you are setup with everything needed to check them, I wouldn't mess with them.

I never said I want to mess with them. I would like to know how big of an error I have to expect
if I use the cal constant from another identical kit.

Are the manufacturing tolerances between different sets that big?

as for calibration by anritsu, I dont know if you have had to deal with them in the past or not, but in my 3 experiences with them
they are simply the biggest aholes in T&M compared to my similar experience with other manufactures. I'm afraid they are gonna charge me $2k-$3k for calibration if they even accept it from an individual/hobbyist. But I will contact them

It's used.  How would you know how much error there would be without having some way to measure it.  Worse case, you were sold some damaged parts and now start inserting them into good connectors, damaging them.   Suddenly the cost goes up...  Who knows how many cycles have been on the used parts or the condition of the plating.  Not to mention all the other mechanical tolerances that need to be checked. 

I'm a little confused.  In your original post you mention an ENA and I just assumed it was a typo.    Now after reading your posts, it seems like you may have an Agilent/Keysight ENA and want to use these standards with it.   I've looked at buying used standards but the sellers want a third to half the cost of a new set.  It makes no sense to me as I could end up with more invested in the used set than buying brand new, plus getting the correct ones.   I looked on eBay at the low cost standards from drkirkby, $742.82 for the basic set.  For me personally, I would rather invest in a new Keysight set.   

With the HP/Agilent standards I have used, they are indeed held to very tight tolerances and the parts are able to be changed.  Of course, that comes at a price!  You plug one of those standards onto a low grade connector, that may be the end of it.   They are very easily damaged.   With the newer equipment using S-parameters, they can get away with lower material costs and may achieve better results (in the short term).   These lower end standards wouldn't be interchangeable.     There are some good papers on this subject available for free. 

For your home use, maybe it doesn't matter.  I recently made a video for the NanoVNA V2 Plus 4.   I made a few changes to the low grade cal kit that was supplied with it but used the ideal model throughout the review.   For my own home use,  my standards are home made.  Below a GHz, even 1.5GHz, this hasn't been much of a problem.  I would like to play around at higher frequencies and sorted some low grade loads and got some help from one of the members here to characterize them.   This or course required a few friends loaning me some parts.   It may be possible for you to do something similar with your used set. 

I'm excited to see the next generation of low cost VNAs as they push for 6GHz.   Crazy....

[analogRF]

yes I would love to buy a brand new keysight kit too. 85033E -> C$5900 (incl tax)
Anritsu wants C$2000 (tax incl) only for calibration plus extra (God knows how much) for repair or replacement of each damaged part

I bought my (non working) Agilent ENA E5062A with option 250 (full two port S-param) for $1270 delivered and repaired it and
it is working pretty good now.
so i went and bought the Anritsu cal kit for $600 (probably a mistake??) which as I said is in pretty good shape at least visually (I dont have
precision gauges to measure the dimensions etc...) and it is complete and the pieces work fine and do the calibration but since i dont have the cal constants I know I am off.

So I was thinking if I get my hands on a floppy from another unit how bad it could be? although at the moment I have no source for that floppy either 

[analogRF]

any comments on the quality and durability of the Dr Kirkby's cal kit? does anyone here have experience with them?

[joeqsmith]

yes I would love to buy a brand new keysight kit too. 85033E -> C$5900 (incl tax)
Anritsu wants C$2000 (tax incl) only for calibration plus extra (God knows how much) for repair or replacement of each damaged part

I bought my (non working) Agilent ENA E5062A with option 250 (full two port S-param) for $1270 delivered and repaired it and
it is working pretty good now.
so i went and bought the Anritsu cal kit for $600 (probably a mistake??) which as I said is in pretty good shape at least visually (I dont have
precision gauges to measure the dimensions etc...) and it is complete and the pieces work fine and do the calibration but since i dont have the cal constants I know I am off.

So I was thinking if I get my hands on a floppy from another unit how bad it could be? although at the moment I have no source for that floppy either 

Looks like a nice setup.  They may still offer an e-cal for it.  Might be less expensive. 

A few things to consider.  Even if you had the coefficients they may not be readable by the ENA.  Even if the two played well together, you still have no way of knowing if the standards are good without having them checked.   I'm not surprised by the cost to have them checked.   

Kirkby had made a video showing calibration using bad standards and how easily you could fool yourself into thinking things are fine when in fact they are not.  It used to have more down votes than up.  I suspect this simple video demonstration went right over the heads of most viewers.   IMO, it was clear and to the point.   

[analogRF]

yes I had seen that video quite a while back. I am familiar with VNA measurements and calibration. I do these measurements at work with well maintained keysight instruments (PNA-X)
yes the ENA may not be able to read that floppy from Anritsu but I was hoping if I can find it, I can extract the data and enter them manually

ECal for ENA --> cheapest is 85093C -- C$10500 (incl tax) 

maybe a used ECal can be a better option??

still I would like to know if someone has any experience with Dr Kirkby's cal kits?

[joeqsmith]

If you had the disk and could read it, do they use the same coefficients as the Agilent?  Or would you need to somehow convert that?   Does your PNA support Touchstone format for the standards? 

Personally, I would be smuggling those standards into the QC lab at work.   After you make sure they are mechanically alright, with your PNAs you could easily characterize them.   Of course, you screw up one of their metrology grade connectors, they may terminate your employment.  I like those odds myself. 

I would like to have an e-cal.  I had looked at the one from NI and wondered if they documented how to control it.  I run the VNAs with a PC anyway so adding support for an e-cal wouldn't present a problem.   I had also looked at PICO.  They have a low cost 6GHz VNA.  Their mechanical calibration kit wasn't too bad price wise.  I think I saw a brand new one for under $500.   No idea how many cycles they rate it for and if they support the Agilent coefficients. 

The member who helped me with my home made standards had procured a kit from Kirkby.  Other's here have also purchased them.  Have you tried the search?   

Looks like there is one video review but nothing more than an unboxing.   

[Bud]

Kirkby had made a video showing calibration using bad standards and how easily you could fool yourself into thinking things are fine when in fact they are not. 

I found this ironic, because by his own admission he used to work for a company where they made antennas with overstated specifications close to BS.

[Bud]

I'm excited to see the next generation of low cost VNAs as they push for 6GHz.   Crazy....

One should not allow themselves to be fooled by the high frequency limit the seller claims. Always look at Calibration standards characterization, if the seller supplies properly characterised cal kits. Otherwise you will be buying a brick or a gizmo with a much lower useful frequency range.

[Bud]

any comments on the quality and durability of the Dr Kirkby's cal kit? does anyone here have experience with them?

He started from selling  cal kits he was buying on ebay and get them refurbished. From this perspective those kits where as good as refurbished original kits could be. Not sure what he is selling these days. I think he is putting fair efforts in doing good job though.

[analogRF]

Kirkby's cal kits seem to be properly characterized at least it should be good for the 7 or 8GHz he claims
I am reluctant because I am not sure how rugged they are and how many times they can be used with negligible degradation
If they turn bad after like 20 times then of course it's BS. I am better off with a used non characterized Anritsu then

where did he claim he worked for a company that they made antennas with overstated specifications?

[Bud]

where did he claim he worked for a company that they made antennas with overstated specifications?

On one or more Yahoo Groups forums he was a member of.

[TheSteve]

A cal kit without the constants doesn't seem overly useful to me. I'd like to think another similar kit would get you pretty close though.
I like that the Keysight kits all use fixed cal constants that are the same with each kit(same with Maury Microwave for the most part).

That's also my thinking. I dont think that manufacturing tolerances between different sets are that big to make a big impact
I am not looking for like metrology grade measurements!

as for keysight, I am not sure about that. As far as I know, the cal constants are tied to the set because they are measured individually.
It is usually written on the floppy that comes with the cal kit that it belongs to the kit with such S/N. Although there are
tons of these floppy data sets being sold on eBay without the cal kit ! 

if the Keysight ones are not tied to the set, then perhaps I am wrong about the Anritsu being tied to the set either.

Here is the page you can download the constants for pretty much every kit they make, all values are fixed per kit. They include the floppy but they are all the same.
http://na.support.keysight.com/pna/caldefs/stddefs.html

I have a Maury Microwave 3.5mm 26.5 GHz kit sold by R&S that I will be selling if you're interested. They use standard values you can download from the website for all popular VNA's.

I can also measure your kit if you want, I can't say how great we can do calculating values but I have come very close to the proper Keysight values measuring some 9 and 26 GHz kits after calibrating with a very nice ECAL unit.

[joeqsmith]

Kirkby had made a video showing calibration using bad standards and how easily you could fool yourself into thinking things are fine when in fact they are not. 

I found this ironic, because by his own admission he used to work for a company where they made antennas with overstated specifications close to BS.

Ironic in what way?  I don't understand where he worked, what job he held, or what the company he worked for has to do with this particular video.   

I'm excited to see the next generation of low cost VNAs as they push for 6GHz.   Crazy....

One should not allow themselves to be fooled by the high frequency limit the seller claims. Always look at Calibration standards characterization, if the seller supplies properly characterised cal kits. Otherwise you will be buying a brick or a gizmo with a much lower useful frequency range.

Even with using the ideal model with the low cost NanoVNA V2+4, I was able to get some decent data off it.    It's impressive. 

[Mechatrommer]

any comments on the quality and durability of the Dr Kirkby's cal kit? does anyone here have experience with them?

i've been using Kirkby SMA version for 4 years... quality? okaish but looks similar to chinese quality, maybe better i dont know its just eyeballing grade of evaluation. the load is pretty close to 50 ohm, not 51 ohm members figured out on cheap NanoVNA cal kit. data-based characteristics profile are plotted until 12GHz on my 6GHz spec cal kit. durability? no one knows unless they have another super undestructable, purely non oxidizing non changable characteristics higher end cal kit to make comparison with. Kirkby claimed to also made cal kits for universities and professors and they are happy, so he claimed. i got a feeling he's telling the truth. ymmv.

[joeqsmith]

i've been using Kirkby SMA version for 4 years... quality?...

Four years is a decent amount of time.  Without a reference,  you could still determine how much they have changed.  Do you keep a record of how often you have used them?  I would imagine if you are torquing them, they are wearing faster.  All good data to have.   I keep track of the number of insertions for the new standards I made and saved off the data when I first characterized them just to get a feel how they change.    If you purchased them with the verification parts, even seeing how much this has changed would be of interest.   Maybe you saved the original SOLT calibration over a wide range and could reload it and measure the parts today.   

Even if you have no records,  I would be very interested in having these returned to them to be characterized and post the the original and new results.   Do they offer a calibration service? 

[Mechatrommer]

4 years but only for a couple of times. i'm not full time rf engineer, just a hobbiest, maybe i've snapped in the cal kit into vna/sma cable for like 10-20 times during the time, i dont keep record but i treat them like when handling a bowl full of superheated oil.. even if cheapest cal kit (with data) out there, $500+ is still something for me. i dont use torque wrench for reason that i mentioned in another thread sometime ago (to reduce wear and tear), working below 1-3GHz doesnt demand 0.1dB accuracy but thats me. it comes with verification attenuator kit with respond plot from Kirkby's HP VNA but its not going out soon today as my bench and room are full of things that i have to carefully step my foot or swing my arm otherwise i'll hit something to the ground. i think the cal kit is doing fine, i've yet to see any strange behavior while operating them. ps: someone PMed me few weeks ago about his Kirkby CAL kit order is not arriving, i told him Kirkby is real, not a scam, you just need to choose trusted courier service with track number and then do the work at tracking to ensure security arrival of your own toy, cheers.

[joeqsmith]

10-20 may show up something.  If you get a chance to retest them with the included attenuator, I would appreciate it.

[virtualparticles]

The calibration coefficients for a mechanical calibration kit are the same for every piece with the same model number. The manufacturer does not measure each one and create new coefficients. If you have these from another unit, you are good to go. If a manufacturer were to go to the trouble of looking at each one, they would create a "database" for it, which would be much more accurate. They wouldn't go to the trouble of deriving the coefficients.

[xmo]

"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:

[rf-messkopf]

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

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.  Other's here have also purchased them.  Have you tried the search?   

I think that was me. 

still I would like to know if someone has any experience with Dr Kirkby's cal kits?

I have Kirkby's model 85054 type N calibration kit. It seems to be properly characterized and I have never found any inconsistencies while using it. I cannot comment about its accuracy though since I have no trustworthy type N cal kit against which I could check it. I do not use N connectors with the VNA very much and so Kirkby's kit does not see a lot of wear, but the connectors seem to be of an okay quality. I recognize the manufacturers of some parts of the kit; the thru, for example, is this one: https://media.telegaertner.com/orig/J01024A0004Kp.pdf. I use Telegärtner adapters in my home lab when no lab grade accuracy is needed, and have no complaint about them. Other bits seem to be from https://www.jyebao.com.tw/.

The kit is certainly limited by the loads. The 6 GHz loads are, in fact, terminations, even though they are of good quality. Since I do not use my kit very much I decided that they are good enough for me. Kirkby offers to ship the kit with new or used calibration grade loads. This would surely lead to a great improvement in accuracy.