Tom W8JI 2023/07/30 05:31
The video is factual useless nonsense. A GDT will not protect a VNA of any type, it will barely help a much more robust communications
receiver. Adding a GDT and ground strap is a total waste of time and money.
I doubt there is anyone who would suggest that any GDT would protect a VNA in of itself. Of course, this not what the video shows.
There are two reasons a VNA is damaged:
1.) Steady state voltages either from local signals or bias voltage on the device under test.
.....
2.) Static or residual charge from a large high antenna and feed line capacitance being charged, or perhaps a capacitor in the system
accidentally left charged from some previous bias voltage.
I had mentioed a third where the ham connects the VNA directly to their PA.
The three most effective ways to handle this and protect the VNA are:
1.) Don't connect to an active circuit unless you know what you are doing, things like a bias-T can be a disaster
2.) Short the external load or have a static drain like a resistor or choke across the external load before connecting
3.) Don't test antenna systems in inclement weather
Good practices was covered during the video. In addition I talk about the use of ESD matts and wrist straps.
This thread goes on and on with occasional good and a lot of bad suggestions. Early on there was a long video link with a mix of a couple minutes of good information in with a lot of meaningless "protection" like the assumption a small resistor will clamp at the resistor's rated voltage.
Sorry you lost me. I don't believe I have ever suggested a resistor would act as a clamp.
Ten volt and higher protection devices and devices with a lot of capacitance really do not belong in the protection system.
As I demonstrated, at the frequencies I was working with (<300MHz), the capacitance of the parts chosen had minimal effect on the measurements.
These are not expensive devices so we just want some reasonable protection that does not compromise the instrument". I bought a sample collection of these from Amazon and have been using them in place of my old trusty but limited range MFJ analyzers and have not damaged one yet. First I follow some simple rules.
I am not too surprised. As I stated in that video, following good practices should help avoid most of the common failures.
I'm not fan of attenuator pads because the transient suppression, at least below the surprisingly high catastrophic mortality point of even a small resistor, is only by pad attenuation. If we use a 3dB pad, which then takes at least 3dB from the dynamic range of measurements, we reduce unwanted voltage to about 71% of the original port voltage. A six-dB pad is only a ~50% voltage reduction. The resistors are not remotely as good at fusing as real fuses, and fuses are notoriously slow and unreliable enough in transient response. We sure don't want to depend on such a meaningless protection return for a hit on the already too-low signal levels of these devices. The silliest thing is where resistors appear as glitch fuses in amplifiers, and yes I have measured their protection contribution. If they are unreliable as fault "fuses" in a simple rugged tube amplifier, they are an even worse fuse in a fragile semiconductor system. They are first and foremost just resistors.
As I talked about in the video, there are two attenuators. One between the GDT and TVS. The other between the TVS and port 1. Obviously you are giving up dynamic range. I demonstrated the effects of adding various attention had on the SWR measurement. But again, I have heard from hams suggesting they need to measure their SWR out three places beyond the decimal. As I talk about in the video, any design is a give and take. The circuit proposed wasn't suggested as an end all to protection but rather just going through the motions on what to consider if a ham were to attempt to design their own. Yes, I know hams no longer enhance the art. Well maybe the 1%ers are still out there..
The attenuators are there only to give the TVS something to work against. They are not their to act as a fuse as you suggest. Coax makes a good capacitor. Once the GDT fires, the stress on the first stage attenuator is gone. You could try creeping up on the GDT with a DC voltage and damage the attenuator. In that case, it may very well act as a fuse but it is not the intent. Again, what was presented it's not a one size fits all and based on what some of the hams have posted, I doubt you could add enough protection to save them in all cases.
GDTs are off the charts in voltage breakdown but can handle current in the hundreds or kilo-ampere range, and just look like a few pF of capacitance when not ionized. They are great for protecting high voltage high power systems.
They vary like any device, which is why there are so many and data sheets. One place you may find them is multi-meters where they commonly sit behind a surge rated resistor and PTC which limits the current they see.
About the only choice with fast response and low enough voltage is a simple TVS diode. They are readily available in low voltage-threshold and very low capacitance devices, like the Littlefuse SP3022 series.
Pin diodes are also common. Again, it all depends. As I mentioned during the video, I have seen neon bulbs being used in some equipment.
I use them on my expensive bench VNA, so I am sure they will work with a nanoVNA. They are not nearly as robust as a GDT and unsuitable in high power circuit protection (like linear amps), but are the component of choice in low voltage systems.
The LiteVNA has clamps built-in. For my home equipment, I don't add any protection. Normally I follow good practices. I did manage to damage an amplifier a few months back when I decided to not wear a strap. I touched a horn (antenna) which was gounded via several feet of coax back to the amp.
I would also use a suitable series capacitor between the DUT port and the TVS diode, or make it a symmetrical T with two suitable value MLCC (multi-layer ceramic capacitors) and a shunt TVS diode in the center.
I had a DC blocking cap in that design I show and went over it. I've had hams suggest that a blocking cap was enough to protect their equipment. The idead that the capacitor will present a low impedance to high frequencies and that transients, like those from an ESD event have fairly high frequency content goes over their heads.
Beyond this we are likely doing more measurement and cost harm than we are protection and failure cost good.
Oh, I can assure you we are giving up performance. Again, there is a reason I don't add any proction when working with my equipment. Most of my equipment is 20 years old or more and still the expense of repairs could easily be in the $1k and up. So, wrist straps and other practices are a must.
I've connected two or three unprotected nanoVNAs to various things without issue, but I use some common sense precautions.
There have been several posts from hams who have damaged theirs which is what prompted the video in the first place. Don't assume hams have any engineering background. I suspect had I ran those grill starter ESD transients directly into that VNA, it would have been game over. I saw no reason to demonstate that. The pulse from that grill starter is fairly weak compared with the IEC standards. I could have ran that same test using this generator without any concern. Higher energy transients like you may see with a charged section of coax may have been more interesting to demonstrate but the video was getting long.
... My biggest problem with the nanoVNAs are overload and the difficult time getting S-parameters into the computers. But if I use it like a fancy MFJ-259 it is great, and common sense when connecting protects it.
I don't see the VNA as an antenna analyzer like many of the hams market them and can certainly believe a purpose built antenna analyzer would out perform it for measuring SWR. I can even believe a basic SWR meter may do a better job (certainly more robust and easier for the hams to setup and learn to use).
As far as getting the data into the PC, for me that's the least of my problems. I started writing software for VNAs when I bought my first one a few decades past (and not to measure antennas). Porting over my software to support these various low cost VNA has not presented any real problems. Once the LiteVNA was released, I pretty much stopped using any of the others. Dislord has done a wonderful job with the firmware and the hardware, for the $120, is a tough combo to beat.
I'm do not have a CB or ham license and no interest in them as a hobby. I noticed you built some sort of drag car. I would have far more interest in that. I've been in the 7's myself in the 1/4.
***
Bad formatting