EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: KD0RC on February 27, 2014, 04:48:54 am
-
I tried a new method for calibrating my Rigol DSA 815 this evening. I attached an antenna to the SA input and tuned the 815 to 10 MHz to view WWV directly. I set it to a 10 Hz RBW and a 1 Hz VBW and turned on video averaging. I turned on the frequency counter and set it to a 1 Hz resolution.
Next, I put the 815 into Service mode and adjusted the Ref Dac a few points at a time until I got the counter to show 10.000000 MHz most of the time, and the signal to show perfectly centered on the display. Every now and again the display shows a distorted sig, so I presume that is either atmospherics or interference. By turning on video averaging and letting it perk for a while, I get a really good looking trace, and the counter sits for long periods of time at exactly 10 MHz with occasional jumps up or down by 1 Hz.
So here is my usual question... Is there any reason that this is not a good way to do this? Not being an engineer, I may be falling into some of Dave's 'traps for young players' and not know it.
Thanks again to Alan, W2AEW, for his very enlightening video on the difference between RBW and VBW. Without that understanding, I would not have been able to get this to work!
Thanks,
Len, KD0RC
-
Interesting technique. I hope some of the spectrum analyzer experts see this and give an assessment.
-
I've used this technique, or variations of it, to calibrate the reference oscillator in various radios and transceivers, as well as adjusting some old frequency counters. I've never used it to adjust a spectrum analyzer though. There are other factors that may play into the SA accuracy though, such as RBW filter accuracy, sweep accuracy, etc.
Here's a video I did that talks a bit about how to check/adjust a frequency counter using WWV and a receiver.
How to zero-beat WWV to check or adjust a Frequency Counter's accuracy (https://www.youtube.com/watch?v=QCJ4cQGOQLI#)
And, here's a link to a little write-up I put together to show how to use WWV to adjust the reference oscillator in a HF receiver:
http://www.qsl.net/w2aew/adjustrig.htm (http://www.qsl.net/w2aew/adjustrig.htm)
-
I tried a new method for calibrating my Rigol DSA 815 this evening. I attached an antenna to the SA input and tuned the 815 to 10 MHz to view WWV directly. I set it to a 10 Hz RBW and a 1 Hz VBW and turned on video averaging. I turned on the frequency counter and set it to a 1 Hz resolution.
I'm not sure you need more than the hardware counter. The display slows down when the counter is on, presumably because it takes time out to tune to the marked frequency and count whatever signal it finds there. The question is do the display settings have any effect on how the counter tunes to the marked frequency or is it just based on the marked frequency and set counter resolution.
-
Thanks for the reply Alan. I have watched your videos (and given thumbs up!) on calibrating HF transcievers. I use a very similar technique. Here is a description of what I do and my reasoning for it working.
To calibrate my K2, I receive WWV on my TS 2000 in AM mode and feed the audio to Spectrogram (soundcard based audio spectrum analyzer). This shows me the precise 500 and 600 Hz tones.
I then tune the K2 to the same WWV sig, but in LSB or USB and feed the audio to the other channel in spectrogram. Now I can line up the 500 or 600 Hz tone with the corresponding tone from the TS 2000. I can get it to within about 1Hz for short periods of time (drift is the problem...).
This works because the 500 and 600 Hz tones from the AM receiver are referenced to the WWV carrier, and are not affected by the dial accuracy of the receiver. The SSB radio under test has the tones referenced to its own BFO. Lining up the tones by adjusting the K2 internal oscillators aligns it with WWV (i.e compared to the AM receiver tones). Accuracy of the PC sound card is not a factor because any drift will affect both audio signals equally. It just provides a visual reference (much like the beating S meter) so that it is easy to see when the signals are aligned.
Am I fooling myself that I have a level of accuracy in the 1 - 2 Hz range?
-
Thanks for the reply Alan. I have watched your videos (and given thumbs up!) on calibrating HF transcievers. I use a very similar technique. Here is a description of what I do and my reasoning for it working.
To calibrate my K2, I receive WWV on my TS 2000 in AM mode and feed the audio to Spectrogram (soundcard based audio spectrum analyzer). This shows me the precise 500 and 600 Hz tones.
I then tune the K2 to the same WWV sig, but in LSB or USB and feed the audio to the other channel in spectrogram. Now I can line up the 500 or 600 Hz tone with the corresponding tone from the TS 2000. I can get it to within about 1Hz for short periods of time (drift is the problem...).
This works because the 500 and 600 Hz tones from the AM receiver are referenced to the WWV carrier, and are not affected by the dial accuracy of the receiver. The SSB radio under test has the tones referenced to its own BFO. Lining up the tones by adjusting the K2 internal oscillators aligns it with WWV (i.e compared to the AM receiver tones). Accuracy of the PC sound card is not a factor because any drift will affect both audio signals equally. It just provides a visual reference (much like the beating S meter) so that it is easy to see when the signals are aligned.
Am I fooling myself that I have a level of accuracy in the 1 - 2 Hz range?
Without thinking too hard about it, it sounds reasonable to me. Of course it requires a second receiver and two channel spectrogram. But of course if you've got the, use them! Nice!
-
Here is a video that I made of the procedure:
http://youtu.be/E82NL086VrE (http://youtu.be/E82NL086VrE)
-
Excellent presentation!
-
Very neat method. Checked my few month old 815 and it was off by 3 or 4 Hz. Tweaked it to 'spot on'.
I have a nice Agilent 33250A (80MHz arb function generator) that I scored from a place I consulted at (they shutdown and gave me first dibs on the equipment they had). Anyhow, the last calibration was many years ago. So, with accurate 815 I checked the Agilent and it was off my a long way (30Hz or sp) at nominal 10MHz. Entered into its calibration system and ran just the TCXO calibration steps. Now it's spot on too.
Thanks for the video that prompted me to check my Agilent and bring at least the TCXO back into spec (now that it has aged for quite a few years).
cheers,
george.
-
Excellent presentation!
Thanks!
Very neat method. Checked my few month old 815 and it was off by 3 or 4 Hz. Tweaked it to 'spot on'.
I have a nice Agilent 33250A (80MHz arb function generator) that I scored from a place I consulted at (they shutdown and gave me first dibs on the equipment they had). Anyhow, the last calibration was many years ago. So, with accurate 815 I checked the Agilent and it was off my a long way (30Hz or sp) at nominal 10MHz. Entered into its calibration system and ran just the TCXO calibration steps. Now it's spot on too.
Thanks for the video that prompted me to check my Agilent and bring at least the TCXO back into spec (now that it has aged for quite a few years).
cheers,
george.
Excellent! Do you have the 10 HZ RBW option enabled or were you able to make it work at 100 HZ RBW?
-
Excellent! Do you have the 10 HZ RBW option enabled or were you able to make it work at 100 HZ RBW?
Yes, I've enabled the 10Hz 'option'.
cheers,
george.