EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: KE4CDR on July 11, 2022, 02:06:35 pm
-
My siglent spectrum analyzer is off calibration, apparently. From a fresh startup inputing a verified 50.75MHz at 1VPP (checked with my siglent oscilloscope (SDS1102cml) and function generator (SDG1032x) freq is right on) I'm getting a 50.83MHz on my SSA. Anyone have any ideas why this may be happening and what I can do about it? Is my setup wrong? Any help or suggestions as to what to look for would be much appreciated.
Update: At span 10KHz the peak is right on 50.75KHz. If I move the span up to say 20KHz or higher it goes off frequency. Seems like it shouldn't do this.
Update#2: Ok this getw weirder and weirder. So, it seems the 500Hz frequency addition is arbitrarily added and subtracted throughout the span range. For example at span 50KHz it will flip from 500Khz added to the actual frequency to the correct frequency reading several times going from 50KHz to 51Khz. There's no specific span difference that seems to initiate the 500khz addition. It seems random, but repeatable. The SSA reading of the peak at 50.75KHz goes from the correct freq reading to a reading with 500Hz added throughout the span settings. It seems just accidental that you set the span on the freq that will allow you to have the correct frequency reading. Anyone else seeing something like this? Do I need to call Siglent for a repair? I haven't had this thing more than 2 months and I always use a protection circuit on the front end. Any suggestions? Thanks in advance.
-
The frequency error depending on the span setting looks a bit like a software problem or maybe a resolution limitation. Especially with a rather wide span the bins may not be fine enough to have one every 500 Hz. One may have to adjust some settings for a slower scan.
I would not expect this to be broken hardware - more like old firmware or a rather embarissing bug. So a first point would be to check the frimware version and if there is possibly an update.
Broken hardware would be more like off by a certain factor (like 100 ppm) or as the main symptom too low an amplitude by a lot.
-
I am not familiar with that SA. Does it sweep continuously through the span or only measure at discrete points?
-
on SSA3021xPlus do not get this effect at all
-
I've just tried this with a Siglent SSA3021X about 2 years old. The 50.75 MHz signal is from a Marconi 2019 sig gen, checked against a Racal 9921 counter, checked against a GPSDO. I'd say it's accurate and stable to within a few Hz. I see nothing like this. I've changed the span from from 10KHz to 20KHz and then all sorts of values. Maybe I've misunderstood this and I'm not doing what you were.
Have you updated to the latest FW?
Is the signal source as accurate and stable as you assume?
I'd be more inclined to trust the frequency reference in the SA than the other two pieces of gear.
-
Ok, my initial estimate was incorrect. It is not off by 500 Hz but 43 Hz repeatedly. Sorry for the error, but, it is there. Maybe it's the firmware. I'm looking into that. I have a siglent 1032 function generator that I'm monitoring with a Siglent 1102 CML+ Oscilloscope. The scope shows the frequency of the sig gen at only a few hundredths of Hz off and steady. When the SA is showing the peak 43 Hz off, the scope Hz reading has not changed and is constant at 50.75KHz. So, it's apparent that something is going on in the SA. I use the same wiring harness for the SA as I use to monitor the FG with the Oscope. I've looked for problems in the wiring and the issue stays with the SA. I talked to Siglent this morning and they are "looking into it." Haven't heard back yet. When I do, I'll post their response. Thanks for all the comments. 73
-
I have a siglent 1032 function generator that I'm monitoring with a Siglent 1102 CML+ Oscilloscope. The scope shows the frequency of the sig gen at only a few hundredths of Hz off and steady.
Yes, if it was a constant error of less than two Hz, it could be the AWG. Not a huge error like this, that depends on the SA span randomly.
I talked to Siglent this morning and they are "looking into it."
Since none of the SSA3000X owners in this thread has been able to reproduce your problem, I very much doubt that Siglent will. You should have told them that it appears to be a specific problem with your unit only.
-
According to the data sheet, the frequency reference in the SSA3021X 10MHz <1ppm. My frequency counter says the GPSDO is 1Hz over 10MHz and the 10MHz output from the SA is about 1Hz under 10MHz. The SA says the GPSDO is at 1Hz over 10MHz. I trust the GPSDO, which has been left on for months. I don't see any change on the SA, apart from maybe 1Hz, when I change the span.
I have a Siglent SDS1102 CML+. It doesn't have an output for its frequency reference. The datasheet doesn't talk about the quality of the frequency reference, but says the timebase is accurate to +/- 50ppm. The frequency counter is stated as good for 6 digits. Feeding it an accurate signal at around 10MHz, depending on how they do things, it could be anywhere +/- 100Hz of 10MHz and it would say 10MHz on the nose. That's about 10ppm and about what you'd expect for a pretty good plain crystal. The frequency readout is very convenient for those times when a fairly good measurement of frequency is good enough and you don't want to drag out a proper frequency counter and give it half an hour to warm up.
I don't have a SDG1032x. The datasheet doesn't say much about the frequency reference but talks about an initial accuracy of +/-25ppm.
43Hz in 50.75MHz is <1ppm. The random changes seem odd.
You may have a hardware problem. You may have come across a firmware problem. If you haven't already done so already it makes sense to update to the latest firmware, as it's quick and easy. However, I suspect that you are judging the SA by assuming two pieces of kit with lesser frequency standards are totally accurate.
-
Hello,
KE4CDR: "50.75MHz at 1VPP (checked with my siglent oscilloscope (SDS1102cml) and function generator (SDG1032x) freq is right on) I'm getting a 50.83MHz on my SSA"
This is 80 kHz off not 500 Hz
KE4CDR: "Update: At span 10KHz the peak is right on 50.75KHz"
KE4CDR: "For example at span 50KHz it will flip from 500Khz added to the actual frequency to the correct frequency reading several times going from 50KHz to 51Khz"
KE4CDR: "The SSA reading of the peak at 50.75KHz goes from the correct freq reading to a reading with 500Hz added throughout the span settings."
What ist the true problem?
Best regards
egonotto
-
Your scope and function gen agrees on frequency and your SA does not show the same value for some cases.
How do you conclude that the SA is off frequency and not the scope and func gen?
I do not think there is anything wrong with your SA, just your perception on how the SA is supposed to work. Apologies for being blunt.
ps. I have an SSA 3021X and I get different results depending on span settings and if the SA uses external GPSDO reference or not.