Tracking Generator for the tinySA Ultra Spectrum Analyzer

[@PRRGG1]

Nicely done. I recently purchased a TinySA Ultra and I have been watching your videos on it's use while waiting for the UPS guy.

[@PapasDino]

Another option is to use a inexpensive noise source but they're probably limited in frequency range. Thanks for the great ideas Kerry!

[@InteraliaTony]

Dino makes a good suggestion. I have home brewed simple zener based noise generators and they are useful in the HF region. The output level is not at all constant though. But a couple of years ago I bought (off Amazon) a very much better noise generator for around 15 quid. KZbin doesn't always allow links in comments, but if you search for "Noise Source Simple Spectrum External Generator Tracking SMA Source" you will probably find it. Obviously, the noise source is not a tracking generator at all, but that description will find the device. Kerry, Your video was very interesting and thanks for posting it for us to share.

[@jean-louisbezombes7135]

When using noise source, the limitation is the dynamic range, and possible saturation for wide band active systems.

[@skywave12]

I like the Guitar pick. Very good demo. It is good to use weak signals for inputs to the SA. Even a hand held transceiver behind you can overload the front end of a SA or station monitor if a antenna is connected to it's input. Keep up the good teachings.

[@bigal159]

Excellent Kerry, just the analysis I was looking for. Appreciate you putting it together.

[@ovalwingnut]

Really GR8T information. I was hanging on by a thread but I'm hoping that by just listening and not fully understanding will help me pass my exams on Monday. If not, I'll be sorting 1/4 watt resistors for the rest of my life ;O)

[@Mark-ce9xh]

What are you studying?

[@josemesquita603]

Thank you! Very impressive results.

[@anandarochisha]

Superb work. Thank You !

[@izsakitt3711]

Sure. NanoVNA of course provides phase and impedance information. Amazing times for the engineers who thinker in their mancaves.

[@ruhnet]

Great video! One thought: did you try to tweak the 977MHz IF input frequency from the 8642B? It could be that the SA IF isn’t exactly 977MHz but maybe something like 977.54321MHz, and tweaking the LO input to the mixer could give you additional boost on the tracking generator output. The HP 8444 tracking generator made for use with the 8555 and 8568 analyzers had a fine adjust knob that would tweak the output frequency going into the mixer with the LO, and a small adjustment makes a large difference in the trace.

[@KerryWongBlog]

It was pretty much 977 MHz spot-on. I did vary the frequency a bit during my testing and it appeared that 977 MHz gave the maximum response.

[@gixerags750]

Do you think it wont be long until the Tiny SA Ultra does a new version with a tracking generator? Cheers.

[@vincei4252]

Nice. Wouldn't it be awesome if there was a way for the Tiny-SA to stream out its measurements to a serial port. You could do the normalization of the thru measurement and subtract from the actual measurements among many other things.

[@KerryWongBlog]

I recall there is a "serial output" option buried in the menu somewhere for the tinySA Ultra, I am wondering if that is for sending data out.... now I am curious...

[@dougbas3980]

@@KerryWongBlog Yes the serial port can be hooked to a PC serial port and it can be controlled and exchange data files and screen shots.

[@MaDrone]

Hi Kerry. Is it worth finding a way to generate a (stable) LO, add a mixer, and thus lose the portability of the tiny instrument? (it was nice to see an old HP8566B still in use, an analyser that was in my daily workbench, with the 8756 scalar analyser and HP436a power meter... in the mid 80s)

[@KerryWongBlog]

Well, I demonstrated it was doable, but whether it's worthwhile that is an entirely different story :-) It's incredible to see how well these old HP instruments work even today.

[@MaDrone]

@@KerryWongBlog They were top of the range instruments and I doubt they will ever be surpassed, although the extra features were extremely limited (just the max hold and the HP-IB port to plot the graphs... on paper). To give you an idea, I used such bench to calibrate and test telemetry transponders in on-board telecommunications and extremely expensive scientific satellite equipment.

[@jimwolsiffer4397]

Great video! When I test I check flatness with both cables connected via a barrel where the DUT would be placed. Thinking your blue cables have alrsady been tested for flatness though.. Also to increase dynamic range I use a pre tested amplifier when needed. Would like to learn about the serial data out if you work that out. 73, Jim

[@KerryWongBlog]

Well, looking at the documentation: CONNECTION activates a sub menu for selecting the connection to the tinySA USB selects the USB interface for connection SERIAL selects the internal serial interface for connection SERIAL SPEED enables the selection of the speed of the serial interface tinysa.org/wiki/pmwiki.php?n=Main.USBInterface, it appears that the USB/Serial modes are for interactive controls only.

[@VasyVasii]

I'm using a Nanovna to analyze RF filters. Is this a good approach, or should I be using a tracking generator and spectrum analyzer instead?

[@KerryWongBlog]

VNA is the standard approach as it already has a tracking generator builtin. This video meant to demonstrate what you can do with a spectrum analyzer.

[@caltech-wirewizard]

You may have stated what kind of filter you used, but I’m missing it. Based on the SA sweep this looks like a Notch Filter / BandStop filter. Could you confirm?

[@longjohn526]

Couldn't a guy use a Nano VNA as the 977 MHz signal source for the mixer for a low cost tracking generator? (It would also justify buying both ....)

[@WarrenPostma]

I think the point isn't to get a single frequency but to get something in phase, "tracking" the other signal.

[@вечная_мерзлота]

Please make a video how to see phones outgoing calls/radio wave reception on Spectrum Analyzer

[@ЛайнТелеком]

Спасибо!👍

[@RobB_VK6ES]

So the solution to the lack of TG is to get the NanoVNA :) Have SA Ultra in the mail. Hope to use it to calibrate my home brew power meter.

[@KerryWongBlog]

Yes, VNAs are better suited for measuring S21. But if you only get an SA you could scrape by and use the method mentioned here.

[@willthecat3861]

Getting an S.A. with a tracking generator... was only if you couldn't afford to get at least a scalar network analyzer, or a vector network analyzer... for the cost of good broadband mixer (or a few of them, if you are going to characterize different filters)... I think you can get more than enough NanoVNA (or variants). I don't know why anyone wants to characterize filters with a Tiny SA.

[@w8lvradio]

What about using a PAIR of TinySA Ultras, where one is configured as the frequency generator with attenuation.and the other one as the Spectrum analyzer? All the Best! 73 DE W8LV BILL

[@Conjugate-Match]

Soooo ... all we need is an rf signal generator, costing thousands, to create a tracking generator for our $130 TinySA. It's a fun tech exercise for those who have the gear I suppose, but totally impractical otherwise. I do learn from your vids though, thank you for your efforts.

[@caltech-wirewizard]

Just an FYI... you don't need to spend thousands of dollars to create a tracking generator for the TinySA Ultra. A 0Hz - 8GHz Signal Generator can be gotten for under $150. That has both; an Oven Controlled Crystal Oscillator or a 10MHz External Source. I made this using a ZFM-4-S+ Mixer & an 8GHz Signal Generator works perfect & portable ! Total cost: $208.33

[@johnazar3712]

can a tinySA Ultra generates at 0.5Hz? i try but it only generate at even numbers, Also i cant seem to able to find frequencies in my area from 0hz to 5hz or simply wants to see whats around me between 0hz-1hz only and block them. Thanks

[@todaysfigures598]

❤

[@dougbas3980]

Thank you Kerry. As an MSEE, I found this very interesting. Of course, having an accurate LO source is more expensive than the TinySA by far and leads to modest results. So just interesting and not very practical.

[@KerryWongBlog]

Yeah, it's not really that practical, just wanted to demonstrate it could be done. BTW, I am an MSEE as well.

[@amirb715]

I think your connection of the mixer is not correct for this application. In this particular use of the mixer the 977MHz is acting as your LO and must be (at least) 7dBm and you should connect the LO output of the tinySA to the RF port of the mixer and being -10dBm is not a problem.

[@KerryWongBlog]

I tried that also, it didn't really make much difference. If I get my hands on a mixer in the correct frequency range I will be sure to test again.

[@ErikKaashoek]

Best to use a 20dB amplifier between the LO output of the tinySA and the LO input of the external mixer. This will increase the dynamic range, also because it reduces reverse leakage of the 977MHz going into the RF port of the external mixer

[@willthecat3861]

@@ErikKaashoek And that's the issue... to build a 'useful' tracking gen... you need a good mixer, good cables/connectors, and probably... as you suggest... a good amp too. Good stuff costs good money... and cost of the good stuff adds up... why not just use a VNA.

[@ErikKaashoek]

@@willthecat3861 With a tracking generator you can create a tracking signal at an offset and measure the conversion loss of frequency translation devices such as mixers. Can't do that with a cheap VNA

[@willthecat3861]

@@ErikKaashoek There's a way to do it, with a cheap 2 port VNA without an offset... if you want to do a relative measurement... and at a single frequency at a time. You need two other mixers, besides the DUT... one of which you trust the specs on. You can do it with expensive VNAs (same as expensive SA, with expensive tracking option) ... either a 4 port VNA, or a 2 port VNA, and a calibrated sig gen. I don't know about making a 'good' tracking gen... versus having a cheap VNA, a reasonable mixer, and a good commercial one (probably used... or a pull from a parts mule.) Using a tracking gen is just another way to do it.

[@SteveWrightNZ]

There's too much rambling on and random data, and its too difficult to follow you.

[@bussi7859]

Really crappy way to use for filters.use a VNWA instead