- This topic has 5 replies, 2 voices, and was last updated 3 years, 8 months ago by
.
-
Posts
-
Hi everyone,
I seem to be having some overshoot issues on my test rig.
I have calibrated two different sound cards. Both are 192kHz, DC coupled and have no tilt. There is some HF roll-off on each but nothing dramatic.
I set the deviation to 75kHz with the Bessel Null calibration at 31187.6Hz
But MPX tool shows overshoot when modulated with audio, to the extent of 85-93kHz.
At first, I thought it could be an ultrasonic LF boost that wasn’t showing, but on a scope, the quick sweep is clean (with the exception of the intentional initial peak trace)
So I decrease the modulation so it sits at 75kHz, however, there are still overshoots up to 83kHz and when I switch to the Bessel Null it shows 67kHz deviation.
There is no analog STL path, this is direct analog audio from the sound card. It points towards the issue being with the exciter which is an NRG PLL PRO 3. Pre-emphasis is disabled on the exciter, so perhaps something in the audio stage, maybe a failing capacitor? Can anyone advise in this area?
Thanks
“I set the deviation to 75kHz with the Bessel Null calibration at 31187.6Hz”
bessel null is used to calibrate an MPX tool, transmitter deviation indicator or other monitor mode, not to set BAO.For this you need a spectrum analyzer or network analyzer (you have to visualize the RF carrier).
To set the BAO (if an MPX tool or other tool is calibrated) you first test the square response. Then you select 400Hz and adjust the MPX level until you read 75K on the MPX tool. If the transmitter’s amplitude and phase response is good (the same goes for the card) then it should not exceed 75K.
Thanks for clarifying. Sometimes a manual for BAO would be handy 🙂
With 400Hz level calibrated to read 75kHz deviation, avg deviation is 77kHz with peaks to 83kHz. Seems to be happening mostly on the deep bass, so it’s looking like the TX audio/PLL stage causing some tilt. I’ll need to contact the manufacturer or someone with some knowledge of that circuit. Unless there happens to be a TX guru around on here.
Thanks
most likely pll loop filter problem. more or less can be fixed using the Corrective PEQ option in BAO. Example for start: gain -2 at 20Hz width: .5 Oct. Be sure to use Phase linear options!
Now turn on the sine test and the frequency at 20Hz. Adjust the Eq gain until you get 75K. After that 30Hz. If greater than 75K, increase the Width option until you see a change to 30Hz. Now 40, 50 …. 100Hz everything must be at 75K.
Thanks pos1, very helpful. I have ended up with around -1.35dB 20Hz 2.87oct phase linear. Maximum peak is now around 78.7kHz. I set the sine test tone at 20, 30, 40, 50 and 100Hz and adjusted the PEQ width to bring them all to 75kHz (in that order, ending with 100Hz). A little more tweaking to bring max peak to 75kHz.
- You must be logged in to reply to this topic.