Second hand VNA info required.

[trevwhite]

I have the chance to buy a R&S ZVRE or HP8753A for I think pretty reasonable money. I don't really know much about VNAs but basically thought either of these should enable me to tune antenna designs up to 2.4Ghz?

I understand the cal kit is very important.

Anyone used or own a ZVRE unit? Anyone got any thoughts on how they compare with the HP?

Thanks

Trev






Sent from my SM-G950F using Tapatalk

[TheSteve]

Does the HP unit include the S parameter test set?

[trevwhite]

Does the HP unit include the S parameter test set?

Yeah

Sent from my SM-G950F using Tapatalk

[TheSteve]

Would be no problem then to see return loss etc on an antenna.

[trevwhite]

That's good to know,  thanks. I was wondering if there was any advantage in choosing the R&S unit

Sent from my SM-G950F using Tapatalk

[hendorog]

The 8753A is pretty old now - no idea about the ZVRE but it looks quite a bit newer from what I can see on the internet with a colour LCD screen and a 3.5 inch floppy.

Weak points about the 8753A/B that might cost $$ are the display, and the RF source YIG and microcircuits.

Strong point is that you will probably find any part you need on eBay for the thing - at reasonable prices too if you are patient. Reasonably good amount of docs around, the schematics are available for that model, and the designer of it still replies to posts on the Keysight support website.
The extra options are available from another HP guy as well.

The RF components in the test set are actually good enough that you can get by as a beginner without a cal kit. Measurements are more accurate with a cal kit of course.
There are various options for getting cal kits := you can DIY one, buy a cheap one from SDR Kits, buy a good one from Kirkby Microwave, or assemble one from new or used parts. Don't bother searching eBay for a complete used cal kit, as they are always overpriced. I have never found a reasonably priced official kit in several years of looking. You do see individual parts from time to time, I forgot to bid on one auction this morning....

[Mechatrommer]

You do see individual parts from time to time, I forgot to bid on one auction this morning....

are you talking about that agilent 26.5ghz sma 50 ohm standard? ended about 7 hours ago? It was $50 at 1st,but i never took it seriously since its a one lone ranger kit, no box no brother and sister,the metrology could be off, the parm would be nowhere to be found, it can be just as good as diy kit imo.if only claimed >30db lost, i think i can struggle to get that through diy approach. It ended close to $80 iirc while i'm browsing the net and discussing with mr kirkby, hope it can find a good home.

[hendorog]

Yep thats the one.

There are no params for loads in the VNA, they are assumed to be 50 ohms.
So if you get a decent one like that from a good seller its probably going to be pretty good.

The params are for the open and the short only.

[G0HZU]

If used as a full 2 port analyser after a full 2 port calibration I'd expect the HP8753A to be quite slow in terms of screen update rate because it will be slow at calculating all the correction factors. I think each version of the 8753 got faster at this because the processor performance improved over the years.

The ZVRE does look to be a nice VNA with loads of dynamic range although it also looks very big and is probably very heavy. The user interface is likely to be a bit odd and it might be a bit fussier to operate remotely if you want to dump out data from it. But then again it might be fine. Some of our older R&S VNAs at work are not as easy to use as Agilent VNAs when it comes to remote operation. But the technical performance of the R&S VNAs are usually just as good as Agilent. The ZVRE looks to be a very nice VNA although it is probably older than it first appears.

[Mechatrommer]

then all is not lost, there are still many cheaper 50 ohm terminations claimed good to 18ghz...

[KE5FX]

If used as a full 2 port analyser after a full 2 port calibration I'd expect the HP8753A to be quite slow in terms of screen update rate because it will be slow at calculating all the correction factors. I think each version of the 8753 got faster at this because the processor performance improved over the years.

The ZVRE does look to be a nice VNA with loads of dynamic range although it also looks very big and is probably very heavy. The user interface is likely to be a bit odd and it might be a bit fussier to operate remotely if you want to dump out data from it. But then again it might be fine. Some of our older R&S VNAs at work are not as easy to use as Agilent VNAs when it comes to remote operation. But the technical performance of the R&S VNAs are usually just as good as Agilent. The ZVRE looks to be a very nice VNA although it is probably older than it first appears.

Speed isn't a problem during actual use, but the 8753A-C models are relatively slow at interpolating a calibration.  If you set up a four-parameter calibration at 1601 points and then try to narrow the frequency range later, it can take the better part of a minute to do all the number-crunching. 

Software support is a big deal with VNAs.  If the R&S model doesn't give you a way to save setups, calibrations, and .SnP files on your PC, that in itself is a good reason to go with the HP.

[hendorog]

then all is not lost, there are still many cheaper 50 ohm terminations claimed good to 18ghz...

Yep that is what I'm using. I got a cal kit from SDR Kits, but the load was 48 ohms or something, not even close 
So I got some Inmet loads and selected the ones which are closest to 50 ohms at DC.

[Mechatrommer]

thanks for good review. But at $76 kit, no coeff data will be provided.... thats suspicious.. we get what we pay for...

[hendorog]

There is a pdf with the coeffs somewhere where someone has modelled them on a proper VNA.

[Mechatrommer]

There is a pdf with the coeffs somewhere where someone has modelled them on a proper VNA.

yeah i read about the diy cal kit and modelling them in gnu octave or whatever http://www.qsl.net/in3otd/electronics/VNA_calkit/calkit.html. but that requires owning such working VNA and more importantly a certified brand name cal kit to calibrate the VNA in the first place. i mean each SDR kit should be provided with coeff disk/usb/link like DrKirkby did, i cant see that in SDR website other than some mechanical spec pdf.

[hendorog]

There is a pdf with the coeffs somewhere where someone has modelled them on a proper VNA.

yeah i read about the diy cal kit and modelling them in gnu octave or whatever http://www.qsl.net/in3otd/electronics/VNA_calkit/calkit.html. but that requires owning such working VNA and more importantly a certified brand name cal kit to calibrate the VNA in the first place. i mean each SDR kit should be provided with coeff disk/usb/link like DrKirkby did, i cant see that in SDR website other than some mechanical spec pdf.

I'll try and find it.

[hendorog]

Here is the sheet with the offsets:
https://www.sdr-kits.net/documents/Rosenberger_Fairview_Male_Cal_Standards.pdf

This zip file contains more docs with the coefficents. The comparision was between a proper HP cal kit and the SDR cal kit. My memory failed me as it looks like it was done on a VNWA though, not on an HP VNA as I thought.

https://www.sdr-kits.net/downloads/2014-Rosenberger-Fairview-male-female-Cal-standards.zip

My kit is the Rosenberger male one, but I don't use the load, instead I use one of the Inmet loads I selected which is very close to 50 .

[trevwhite]

So how good are the DG8SAQ-VNWA-3EC-Low-Cost-Vector-1-3-GHz-Network-Analyzer for antenna tuning?

http://www.ebay.co.uk/itm/DG8SAQ-VNWA-3EC-Low-Cost-Vector-1-3-GHz-Network-Analyzer-/232269549705?_trksid=p2141725.m3641.l6368

Does it all come down to the cal kit? I read the following line about them and wondered how much of an impact it would have on tuning antenna at 915Mhz?

"covering 1 kHz to 1.3 GHz and powered from a PC USB-bus, the VNWA 3EC offers a dynamic range of 90dB up to 500 MHz and better than 50dB above 500 MHz."

The ZVRE seems to have a better spec but for antenna tuning do I need such a big lump on my desk or would I get good results from the DG8SAQ?

[Mechatrommer]

So how good are the DG8SAQ-VNWA-3EC-Low-Cost-Vector-1-3-GHz-Network-Analyzer for antenna tuning?

you said you can afford the ZVRE if its this http://www.ebay.com.my/itm/Rohde-Schwarz-ZVR-ZVRE-Vector-Network-Analyzer-10-kHz-to-4-GHz-/382103107700?hash=item58f71eb874:g:oH0AAOSwRXRZTeyX it can do full 4 parameters measurement from the look, but its $6K+, not sure about the HP8753A, whats sure is it will need external bridge to further measurement VWSR etc. now you are going backward to ham designed USB PC stuff from real engineer R&S stuff. if you can afford $6K+ and would like to go down the price, there is another better and portable standalone alternative in the middle range, but i hesitate to make recommendation since i'm not the right guy and you've limited the discussion to the 2 VNA.. if the case of ZVRE vs DG8SAQ, ZVRE is a no match clear winner. if the case of ZVRE vs HP8753A, i will still say ZVRE due to mentioned color vs phosphor CRT, clearly a 4 parm measurement (not just antenna tuning S11) with internal bridge (from the look of it), mine can only do 2 s-parm measurement. if you can have ZVRE affordable like $500-$1K affordable, you may pass it to me or other if you decide to pick the DG8SAQ-VNWA , the only stopping factor for me will be the shipping cost.

The ZVRE seems to have a better spec but for antenna tuning do I need such a big lump on my desk or would I get good results from the DG8SAQ?

in the future you may use your VNA for other purpose, you dont want to stick antenna pole as your career for the rest of your entire life do you? buying ZVRE is like buy now and once for all, good for long term investment imho, ymmv..

[trevwhite]

Hi, and thanks for the responses. I am not looking at the ones on Ebay, they are silly prices. I have nowhere near that budget.

I am going to have a look at the unit later in the week/weekend I think. Hopefully I can scrape the cash together to buy it. It does sound like the ZVRE is my best option. I think I can even plug an external monitor into it. The only concern I have is that I would need to clone the hard drive in case it packs up. Hopefully, everything else might be quite robust.

Trev

[Bud]

I agree with other posters that the 8753 is well understood, repairable and the designer is still around and accessible via the Keysight forum and can give clues if something needs to be troubleshooted. Also cal coefficients, s-parameters measured data and screenshots can be exchanged via GPIB and the programming doecs are available. Do not know how much of that can be gotten with R&S units.

[trevwhite]

Hi Bud, that is quite compelling regarding the 8753. All I know about the R&S unit is that it has no options installed and there is no way to get any of the options.

With regard to repair jobs on a unit. I would prefer not to have to but it is a good point that I reckon most of the faults on the HP have probably been detected and people know how to fix them.

Does anyone have any views on just how much better the specs are on the R&S unit? I think via the floppy drive I will be able to transfer data from the R&S unit. It has a colour screen as well. Plus connection for an external screen which is nice. I think its running Windows NT. I believe its a stronger spec'd unit but I do not really know what I am looking at. Was hoping there would be someone out there that has used one or both and can offer some opinion.

My current leaning is that the R&S machine is superior in every way except if it breaks down.

Trev

[Mechatrommer]

assuming... if either one broken, you need to open and repair them either way... what is certain is what features you will get from day to day use, and i believe you need to do the homework yourself, ie asking the owner, googling the net for manuals etc... i just like to comment on old system... maybe floppy disk you can hack to usb drive with floppy to usb adapter board, but i heard unsuccessful attempt due to incompatibility, not sure why is that possible. but will be a nice hack attempt nonetheless.. the old HDD usually will be parallel ATA, you have to find a PC that can still read it, my ~10yrs old PC here still can. a conversion to SSD with SATA to ATA adapter might be a good hack as well, but not sure. it all depends on your economy or if you can stand damage risk ymmv.