开云体育

The tinySA does a double conversion with a first high IF at 433MHz for its low input IF? range (till 350MHz). Performance is more then sufficient for measuring spectral purity of HF transmitters. The high range, till 960MHz, is not a harmonic mode like nanoVNA but uses a separate direct conversion IQ mixer to a IF of 700kHz. Hence the very different performance.
An experimental tinySA using harmonic modes and SW mirror elimination was able to reach above 4GHz with reduced sensitivity but this is not ready for wide use.
More technical details in the wiki?

My guess would be limitations of the spectral purity of the local oscillator. (I haven't seen the hardware design.)

The two ranges suggest that it works similarly to the NanoVNA: a signal source is being used to generate frequencies up to 320 MHz, and the third harmonic of that source is used for the higher frequency range. Enabling an even higher range to 1.5 GHz would require using the fifth harmonic of the LO or switching to a different LO design.

As those of you have used the original NanoVNA (probably just about everybody here) know, the dynamic range is severely limited on the highest range, the one that extends coverage to 1.5 GHz. That's because you're going way beyond what the Si5351A is intended to do. It is still useful for many applications of a VNA, but a dynamic range of 40dB is too low for most of the things you use a spectrum analyzer for. It's not enough, for example, to use it to detect non-compliance with FCC spectral purity requirements for transmitters. TinySA will not be a lab grade instrument and as such is probably not adequate for full compliance testing; you'll still need verification by a better equipped lab for that. But it can serve as a way of finding non-compliant designs.

Hoping to shed some light on the tinySA possibilities I created a wiki measurement example on IQ balance tuning in a single side band transmitter.

Implemented, see wiki

Herb,

The commands you recognize are the same as the nanoVNA. The returned data per trace is only one number as the power measurement is scalar and not a vector
Next to the nanoVNA level commands there are a number of low level commands. These can be used to scan any amount of points in one scan.

If you enable the tinySA.exe output logging you will see the commands used.

On Mon, May 18, 2020 at 05:28 AM, John P wrote:

.. But wait, there's more! We are slowly but slowly implementing a complete serial command handler that when complete should allow you to do anything that you can do with the touch screen. We are also re-arranging the touch screen menus into a more logical hierarchy and will eventually provide the capability to set everything vis the web interface!

=======================================
? ? I'm uncertain how the help file John posted correlates to the console commands list that Erik provided by sending the "help" console command to the tinySA.

- Herb

On Mon, May 18, 2020 at 08:38 AM, <erik@...> wrote:

Can I propose some changes

Focus is not exactly the?same as "ZEROSPAN". I?do plan on putting that option back.

The "GRID" commands will be separate. The current structure makes the?two-word idea difficult!

There are two "REG" commands; "REGISTER" and "REGDUMP"; You only have to type "REGI" or "REGD" to distinguish them.

It's still a work in progress (as is even completed software)!


?
--
John P.
WA2FZW

John,

I guess you did see the tinySA wiki


W.r.t the console commands I agree single letter commands are not easy to understand.
Can I propose some changes?

Sweep Settings:

    FOCUS->ZEROSPAN  (zerospan is the regular technical term for this type of measurement)

Display Options:

    GRID REF->REF (one word)
    GRID SCALE->SCALE (one word)

Debugging and Troubleshooting Commands:

    REGISTER->REG (less letters to retype if you are doing register level debugging)
    REF_FREQ->CAL (is is actually more a power level calibration output)

On Sun, May 17, 2020 at 05:40 AM, Gyula Molnar wrote:

If I can see clearly, the one-letter commands can be identified from the help made by John WA2ZFW

But wait, there's more! We are slowly but slowly implementing a complete serial command handler that when complete should allow you to do anything that you can do with the touch screen. We are also re-arranging the touch screen menus into a more logical hierarchy and will eventually provide the capability to set everything vis the web interface!

Is clear, I was just too lazy to implement them yet,
Triggering in the frequency domain depends a lot on the scan speed and may not be useful for narrow RBW scan (scanning is slow)
The menu tree keeps getting bigger, wonder when I will run out of memory...

Greetings Erik,

? ? ? ? ? ?Looking good!!! This feature that would permit the screen capture of a sporadic burst IMHO could prove very useful. Terrific that you can do this for both time and frequency domain displays!

? ? ? ? ? ?IMHO if it is practical to implement multiple modes of triggering that would be desirable. Oscilloscopes typically offer Auto, Normal and Single trigger modes. Situations occur where one needs a Single?trigger that captures and holds the view. For recurring cyclic situations Normal? triggering where the capture is repeated each time the trigger point is reached is very useful. Auto?triggering simply provides a recurring sweep mode whether or not a trigger point is reached. Auto is useful for obtaining a trace whether or not there is a signal present which is the typical spectrum analysis mode which you have no doubt already implemented.?

Tom, VA7TA

Here they are:
version reset freq saveconfig clearconfig data frequencies scan sweep test touchcal touchtest pause resume trace marker capture vbat vbat_offset help info color if attenuate rbw mode spur selftest x y i v a b t e s m p w o d threads

The single letter commands will need a bit of explaining :-) t.b.d.

Erik,
? ?I combined this request with another topic so you may have missed it.??When you get the time could you post what the "help" console command returns?? I'd?like to know what commands are available for use with the tinySA.? Just a list is fine, no need to explain.? I'd like to eventually contribute to the group by producing a console command document like the one by Larry on the NanoVNA groups.io.

- Herb

Here is a picture of the most simple setup



The tinySA at the bottom right is doing the sweeping. Its tracking output from the high connector is connected to the LO port of a mixer.
A generator at 433.8MHz at 1dBm (I used a tinySA at the top right as I do not have a generator) is connected to the RF port and tracking output is coming from the IF port where is goes into a DUT. output of the DUT goes into the low input port of the tinySA.
Instead of a generator at 433.8MHz you may also use a generator with square wave output at 144.6MHz and use the 3rd overtone.

With this setup you can measure the transfer function of a frequency translating device by offsetting the generator.

Yes, that was the intention.

You may use this setup:
The external tracking mixer has three ports, LO, RF and IF. I'm assuming the IF port is connected to the diodes in the mixer so it can go down to zero Hz
TinySA in low input mode with tracking output enabled.
The external tracking mixer LO port is connected to the high output of the tinySA.
The mixer drive of the tinySA normally is set to +9dBm but if needed for the tracking mixer you can increase the drive to +15dBm.
You have an external generator that is set to the IF of the tinySA (433.8MHz normally) and is connected to the RF port of the external mixer.
The IF port of the external mixer will now output (amongst other) the signal tracking the scan frequency of the tinySA.
You may clean up the IF output by using a LPF connected to the IF port. With a 300MHz LPF you will also reduce the leakage from the RF port thus increase the dynamic range.
Make sure your external generator output is below mixer drive minus 10dB to avoid overloading the external mixer.
The output signal of you tracking generator will be at external generator output level minus 7-9dB. You may add a amplifier after your LPF but it is not needed as the tinySA is sensitive enough.
?

And then it occurred to me I can do the same trigger trick to pause a sweep once a signal above trigger level has been seen so here is the frequency scan of the 433MHz RF remote control.



Not so clean but the at least we know the frequency of the transmission (433.8MHz)

A rather basic trigger mechanism is implemented,
If a signal is above a settable level a capture is done.
Here a 433MHz RF remote control transmission



I need to add some info on the horizontal time scale and decide if it is going to be trigger once or trigger always as now you have to pause sweep to prevent further captures

Erik,
? ? Can a tracking generator be created using an external mixer and LPF driven by the tinySA LO?? I have a WJ 0.2-500MHz mixer and 300MHz LPF?on hand. I'm not clear what I.F.? is normally used to drive the mixer, and whether a preamplifier is necessary after the LPF.?

? ? Just wondering if a set-up like that would be an alternative to purchasing two tinySA's to enable the tracking option.??

- Herb

On Fri, May 15, 2020 at 12:09 AM, <erik@...> wrote:

The VID/PID are identical to the nanoVNA (for now)
The virtual serial port (when using it with windows) becomes? a COM port so no need for VID/PID there

The info command output starts with "tinySA" with version info added (currently v0.2), rest is similar to nanoVNA

Erik,
? ?Thanks for the replies.? When I have more than one serial device connected, the VID/PID helps my scripts automatically identify the device I want to communicate with.? The "tinySA" in the "info" command output will let my scripts differentiate the tinySA from a NanoVNA.

? One more request.? When you get the time could you post what the "help" console command returns?? I'm interested in seeing the commands available for use with the tinySA.

- Herb