Here you go.
Both instruments cold and without external frequency reference. SSG running on previous FW.
10 dB attenuator fitted to SVA3032X input. Siglent 6 GHz rated N-N cable.
Thank You ![Smiley :)](https://www.eevblog.com/forum/Smileys/default/smiley.gif)
Not sure what happened with the Frequency accuracy on the low end to high end with respect to the output level ??
98.1333 MHz is most likely outside its spec?? when dialling-up 100MHz even if the device is 'running cold' - would have expected better Frequency accuracy ![Undecided :-\](https://www.eevblog.com/forum/Smileys/default/undecided.gif)
Having said that - we are assuming the SVA is doing its job and reporting the correct Marker frequencies ![Tongue :P](https://www.eevblog.com/forum/Smileys/default/tongue.gif)
Reason for the wide Frequencies test was to see precisely that - if the device is linear with respect to output level within its specified frequency range.
Now you're addressing 2 parameters, frequency and amplitude accuracy.
Let's examine each result in this case.
For me to report frequency accuracy I had 3 options:
1. Exactly as measured with cold instruments and unreferenced. Both mentioned.
2. Cheat. Link both instruments via their 10 MHz BNC's and then they would agree even if the frequency results weren't accurate to a real world measurement.
3. Connect both to an external 10 MHz ref.......I don't possess one ! ![Tongue :P](https://www.eevblog.com/forum/Smileys/default/tongue.gif)
So you got an honest screenshot and with qualified conditions........something too often missing on this forum.
I did use a Peak marker although I agree the frequency measurement is hardly ideal however I thought your primary interest was amplitude level flatness.
Amplitude accuracy.
Back a few posts you asked about: Level setting vs Level of performance
The screenshot results and associated comments explain this, along with the Datasheet spec and User manual.
To obtain flatter response a power level control sensor is required and SSG models support a few types. RTFM. ![Wink ;)](https://www.eevblog.com/forum/Smileys/default/wink.gif)
So if we need a guaranteed flatness of output we have 2 options, use the SSG at frequencies where flatness is specified to 'Level accuracy ≤0.7 dB (typ.)' or connect a power level sensor, for which level accuracy will then depend on the sensors spec.
![](https://siglentna.com/wp-content/uploads/2018/10/5.jpg)
Tautech
First of all thank you once again for your prompt response
![Smiley :)](https://www.eevblog.com/forum/Smileys/default/smiley.gif)
Yes you
are an honest broker – so your comments about the measurements you did are welcomed with respect
![ThumbsUp :-+](https://www.eevblog.com/forum/Smileys/default/icon_smile_thumbsup.gif)
Although my main concerns are Amplitude accuracy – I can accept a calibration
offset (something which can be corrected like the power coupler Siglent has as an option) but having a big drift in amplitude vs. frequency is far more difficult to control as making a perfectly linear RF signal generator is not an easy task.
Most of the RF signal generators seem to have amplitude responses < 20dBm
– so to get accurate figures at this level would not be easy as is already indicated in the specifications data.
The fact that it goes to 20dBm is a bonus
– but reproducible accuracy throughout the frequency band is not something one would expect – as you have confirmed.
I did have a close look at the review Defpom posted on the SSG3021X you loaned to him.
It was good to have a quick look inside the unit – to see the build quality.
I am not sure if the unit you made the tests on just now – was the same unit?
I ask, because I am also not sure if the unit in the review video was an early production device – if so perhaps there might be some reasons (HW or FW) for the ‘cold’ frequency offset - if indeed this was the SSG and NOT the SVA which was measuring
What made me think the above – was the multitude of scratch marks on the internal chassis that the mainboard was mounted inside the SSG.
When Defpom ‘opened the sealed unit’ – I could not believe the multitude of scratches inside.
Yeah – I know it makes no difference to the operation of the device – but I guess it’s some reflection on Siglent’s quality control.
That’s why I suspect that the review unit must have been from a pre full production run??
Anyway, with all factors aside, the SSG is certainly a nice bit of kit
– it sure is a handy device and VERY functional.
I like that it has an EXTERNAL modulation input (the competition, Rigol device has same)…
This is VERY HANDY for those involved with FM Radio RF amplifier design.
You can easily introduce a Stereo Encoder to generate external stereo signal with the 19KHz pilot as well as an RDS encoder into the external modulation input
– this will give you a FULL FM broadcast signal for testing FM receivers
with modulation ![Smiley :)](https://www.eevblog.com/forum/Smileys/default/smiley.gif)