开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育

Jumping base line


 

Hi Erik

I was wondering what you think might cause the base line to jump up and down like this. I'm using your latest firmware as of today, but it was the same with your previous firmware version.

I've done the self test and the level calibration with a through lead. All OK.

Their are no nearby strong signals (that I can find), it's on it's own little antenna. It also sometimes does it with no antenna connected.

I've not changed any setting but set it to high input and the start/stop frequency. All other settings as default.


 

The sudden jumps are always 10dB up/down.


 

A scan of the PCB with the covers removed.


 

On Sat, Aug 15, 2020 at 02:21 PM, OneOfEleven wrote:
The sudden jumps are always 10dB up/down.
OneOfEleven,
? Try setting LEVEL->REF LEVEL_>MANUAL.? Your video shows the reference level changing when the level shifting occur.? In rapidly changing signal environment,? manually setting the reference level and attenuation level will eliminate the bothersome level shifting.

? Nice idea to include the video. "A picture is worth a thousand words".

- Herb


 

Thank you Herb.

Am trying it with manual ref-level (-10dB) set, all appears OK now.

The auto ref-level code might need some tweaking by the looks of it :)

My heads full of NanoVNA code at the moment so still not got around to looking through the TinySA code. Played with it a lot though :)


 

On Sat, Aug 15, 2020 at 02:35 PM, OneOfEleven wrote:
A scan of the PCB with the covers removed.
OneOfEleven,
? You probably noticed the following:
? 1. The VDD and BOOT pins are available as a last resort for a user that can't access DFU mode in the usual way.
? 2. No SanDisk pads are available, so DiSlords's SanDisk modification for the NanoVNA would require a steady hand and nerves of steel to implement.
? 3. Pads for applying additional shielding are present.? Erik did a ton of experimentation so I'm pretty certain leaving out the shields was not cost savings move.? Still, the pads are there if a user is in an RF rich environment and thinks adding additional shielding will help.
? 4. Lots of space to add a larger capacity battery if longer operating life is needed.
? 5. The sma connectors are soldered securely to the pcb so less chance of shearing off from over torqueing cable connections.
? 6. The display cable does not appear to be connectorized so changing out the display will require above average soldering skills.
? 7. A UART interface is provided for possible RasPi and wireless experimentation.
? 8. A linear regulator is used for less power supply noise.

? Unseen is that the tinySA uses serial console commands so developers familiar with the NanoVNA should face few problems porting tinySA specific versions of their programs (hint, hint).

- Herb


 

lol, I already started writing PC software for it. Which would also compile for all 3 main OS's - windows, nixies and macs.

I've done a lot of SDR 2D/3D real time panadapter/waterfall spectrum graph coding with intensity grading/fading etc over the years so can use that to do some real posh GL graph displaying you've yet to see ;)? anyone remember the QS1R ? ;)


 

On Sat, Aug 15, 2020 at 03:30 PM, OneOfEleven wrote:
, I already started writing PC software for it.
Make sure you use the "scanraw" command i.s.o scan or sweep as scanraw can do unlimited points and sends data in binary, much faster


 

On Sat, Aug 15, 2020 at 11:09 PM, Erik Kaashoek wrote:
On Sat, Aug 15, 2020 at 03:30 PM, OneOfEleven wrote:
, I already started writing PC software for it.
Make sure you use the "scanraw" command i.s.o scan or sweep as scanraw can do unlimited points and sends data in binary, much faster
? ?Besides the faster speed and unlimited points, the tinySA 'scanraw ' command transfers corrected data - unlike the similar command on the NanoVNA which transfers raw data and requires you to externally apply calibration factors.? That's the major reason I've never used the 'scanraw' command on the NanoVNA.

? The binary format of the data returned by the tinySA 'scanraw' command is easy for C programmers to convert to actual dBm values.? Programmers using higher level languages like Python, will struggle doing the conversion until some examples start appearing on the web.

? From notes provided to me from Erik:

? ? ?Scanraw command outputs:
? ? ? ? ?byte 1: '{'
? ? ? ? ?byte 2: 'x'
? ? ? ? ?byte 3: LSB of 16?
? ? ? ? ?byte 4: MSB of 16?
? ? ? ? ?.. repeats bytes 2,3 and 4 for all points
? ? ? ? ?byte end: '}

? ? ? ? ?level in dBm? = ((byte3 + (byte4) * 256) )? / 32 - 128

? - Herb