Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
Semi-bad clone?
I received my NanoVNA last week, ordered from sqcase, a 'seller' here in
the UK. I think they are a Chinese supplier with stock in the UK but it meant I got it quickly and it came with a battery. All for GBP 46.99, tax included. My NanoVNA looks like the 'bad clone' in Hugen's photo. Key features are missing 'CH0' labelling on the front panel, no screening on the PCB and a plastic back panel. The short circuit sma was missing so I made a temporary one from a shorted PL259 and an SMA adapter. It seems to produce plausible results when antenna testing despite the temporary shorted calibration SMA and lack of screening but I plan to compare it with a more expensive VNA soon. 73 |
Hi John
I'll report back when I can but it may not be for a week or two. I know two other people with more expensive VNAs. One cost >?1k but is suspect, despite having just been 'repaired'. The owner of the other one is heading off on hols for a week. I hope you got the full calibration set btw. Mine was missing the short and the seller hasn't responded to my ebay message. :-( 73 Mike On Fri, 19 Jul 2019, 09:01 John Woods via Groups.Io, <g1euh= [email protected]> wrote: Hi Mike |
Hi John, Mike,
Several folks have done various measurements including myself. Measurements and records between big box VNA and the nanoVNA would be of interest. However, that said, my best guess is that the agreement below a few hundred MHz will be excellent. For the most part that is what I have found although I did not bother with making recordings of the sweeps between instruments. I expect as you approach 900 MHz that the deviation between measured on big box vna and the nanoVNA will be larger. The simple fact is due to the accuracy and records stored for the calibration pieces. The SMA short, open and 50 ohm load, I SUSPECT are treated as ideal elements. They are certainly not ideal. I am in the process of finding out from the developer how these calibration pieces are handled. Regards, Alan |
Hi Alan
toggle quoted message
Show quoted text
Agreed that comparison with a big VNA would be v interesting. I'm not aware that any of my friends have one of those though. Mind you at this price I wouldn't expect too much. The frequency steps seem quite large to me, so who knows what compromises have been made in calibration measurements? I'm beginning to think we need a Wiki to pull all the information we're gleaning about these VNAs into the same place. 73 Mike On Fri, 19 Jul 2019, 20:25 alan victor, <avictor73@...> wrote:
Hi John, Mike, |
Hi Mike.
I did extensive measurements on narrow as well wide band filters. Also a number of antenna out in the field. I have compared results with VNA big box, Vector Z Meters and the results at HF are essentially spot on. At the price and with 101 data points... This little box is a GREAT value. I bought 2 and I may buy a third just to mess with in the lab. Alan Victor, Ph.D. W4AMV ________________________________ From: [email protected] <[email protected]> on behalf of Mike Brown <mbmail@...> Sent: Friday, July 19, 2019 9:30 PM To: [email protected] <[email protected]> Subject: Re: [nanovna-users] Semi-bad clone? Hi Alan Agreed that comparison with a big VNA would be v interesting. I'm not aware that any of my friends have one of those though. Mind you at this price I wouldn't expect too much. The frequency steps seem quite large to me, so who knows what compromises have been made in calibration measurements? I'm beginning to think we need a Wiki to pull all the information we're gleaning about these VNAs into the same place. 73 Mike On Fri, 19 Jul 2019, 20:25 alan victor, <avictor73@...> wrote: Hi John, Mike, |
Hi Mike,
Here is a 3 pole helical filter centered at 465 MHz. The filter is "C" coupled so the skirt on the high side rolls more gentle than low side. The insertion loss and return loss as well center freq are pretty spot on with expected. The DR on the second pix shows we can see down at least 65 dB and if I set the reference point to the screen top, perhaps 70 dB. Next, I'll try the big box. I expect very similar results although if I were to read the R+jX values at each and every frequency point there will be delta values. Alan |
Hi Alan
toggle quoted message
Show quoted text
You've obviously got a good one there. I have been finding odd calibration issues and finally worked out it mine is happier up to 450 MHz and not going too low either, although the issue may be the span rather than the frequencies. I currently have it set to 1.5 to 450 MHz and am seeing repeatable results. I was able to quickly compare mine with a MiniVNA Pro (which works up to 200 MHz) earlier - another budget VNA that costs GBP 380. I wasn't able to capture screen shots but on the couple of things we tried the results looked fairly close. I've attached a photo showing what mine thinks of a dual band handheld antenna, which looks plausible. 73 Mike On Sat, 20 Jul 2019, 16:28 alan victor, <avictor73@...> wrote:
Hi Mike, |
Is the 4 trace a firmware thing, or hardware thing
toggle quoted message
Show quoted text
Frank On 7/20/2019 1:49 PM, Mike Brown wrote:
Hi Alan |
never mind. Just got back from vacation, and didn't read the other posts
toggle quoted message
Show quoted text
On 7/20/2019 2:00 PM, Frank S wrote:
Is the 4 trace a firmware thing, or hardware thing |
Hi Frank, The 4 trace or 4 track vs. 2 trace or 2 track thing is a FIRMWARE THING.
All of the nanoVNA devices as far as I know have the same hardware and share the same system schematic. Although there were mentions of the receiver mixer, SA602 vs. SA612. Alan Yes Mike, If I want to do accurate HF stuff, I would do a cal dedicated to 1 MHz-30 MHz. Period. If VHF is your thing, perhaps, 100-500 MHz. And so on. I'll look again and see if I have any networks that can take this VNA close to 900 MHz without going past 900 MHz. I had some dielectric multipole filters but their center freq was 902 MHz. ________________________________ From: [email protected] <[email protected]> on behalf of Frank S <ka2fwc@...> Sent: Saturday, July 20, 2019 6:00 PM To: [email protected] <[email protected]> Subject: Re: [nanovna-users] Semi-bad clone? Is the 4 trace a firmware thing, or hardware thing Frank On 7/20/2019 1:49 PM, Mike Brown wrote: Hi Alan |
On 7/20/2019 4:13 PM, John Woods via Groups.Io wrote:
Hi MikeGood luck with that!?? I also had the Short missing and didn't think the vendor would even know what I was talking about, but it was pretty easy to make one with a PC mount male SMA and a scrap of brass foil left over from making the absent shields on mine. While I was at it I made a good Load with two 100 Ohm 0.1% 805 Rs on another SMA.? This is a lot closer than either of the Loads that came with mine. For what is seemingly "a bad clone" I'm pretty happy now with it's performance. Dan - ac6ao |
Dr. David Kirkby from Kirkby Microwave Ltd
On Sun, 21 Jul 2019 at 01:52, Dan Rae via Groups.Io <danrae=
[email protected]> wrote: On 7/20/2019 4:13 PM, John Woods via Groups.Io wrote:Hi Mikevendor,. Well, if something is missing that should be there, then if bought from eBay, I would stick in an ¡°item not as described ¡° claim. My company produces VNA calibration kits, so I know a fair bit about how they should be designed. Since the nanoVNA doesn¡¯t allow one to specify a delay of the short, any attempt to make a short the way you are doing will be worst than having a proper short. The short supplied with my kit looks like it is an Amphenol one, which are not very good, but will be better than you can make the way you describe. I have not verified it yet, but I think that you will be better not using the open standard at all, as that will add unwanted fringe capacitance, so make the open less like the idealised open I believe that the code expects to see. At 900 MHz an SMA connector will not radiate, so there¡¯s no need to shield the conductor. While I was at it I made a good Load with two 100 Ohm 0.1% 805 Rs on another SMA. This is a lot closer You really need to test that at RF frequencies though. I have some loads with a measured return loss of at least 45 dB at 900 MHz, although the uncertainty in the measurement is fairly high as the loads I used for calibration, which are the loads in my 26.5 GHz 3.5 mm HP 85052B VNA cal kit, have a return loss of at least 48 dB. Unfortunately I could not sell our loads at price of a nanoVNA, so it¡¯s unlikely nanoVNA uses would want them. For what is seemingly "a bad clone" I'm pretty happy now with it's Good. Dan - ac6aoDave, G8WRB. -- Dr. David Kirkby, |
Good luck John. Not too difficult to make your own short as long as you
have a spare SMA plug of course. Shame Maplin are no more! 73 Mike On Sun, 21 Jul 2019, 00:13 John Woods via Groups.Io, <g1euh= [email protected]> wrote: Hi Mike |
to navigate to use esc to dismiss