Page 1 of 1

2.4GHz Diversity Video Receiver

Posted: Thu 26 Feb 26 2009 5:36 pm
by Daniel Wee
Most video diversity units out there rely on sync-detection or RSSI in order to determine the best antenna/receiver choice. Both these systems are weak though, with the RSSI method being marginally better than the sync-detection method. This project uses a novel method of assessing the best signal. This build blog continues from where I left off in the IF forums.

Re: 2.4GHz Diversity Video Receiver

Posted: Thu 26 Feb 26 2009 5:40 pm
by Daniel Wee
The board currently works but preliminary tests suggests that it might be good to shield the detection circuit as it is very sensitive to changes. Future board layouts should be more symmetrical in terms of trace lengths for the signal paths. The linear regulator is getting pretty hot so we should probably look at using a switch-mode power supply instead. The sync-detection circuit may or may not be necessary, given that we can probably switch the video at any time seamlessly.

Daniel

Re: 2.4GHz Diversity Video Receiver

Posted: Sat 07 Mar 07 2009 10:05 pm
by Daniel Wee
Some photos of the project:-
IMG_0972a.JPG
IMG_0972a.JPG (77.62 KiB) Viewed 17585 times
IMG_0973a.JPG
IMG_0973a.JPG (97.82 KiB) Viewed 17585 times
IMG_0974a.JPG
IMG_0974a.JPG (79.35 KiB) Viewed 17585 times
IMG_0976a.JPG
IMG_0976a.JPG (89.02 KiB) Viewed 17585 times
Some work needs to be done on the output amplifier to get the levels up a bit.

Daniel

Re: 2.4GHz Diversity Video Receiver

Posted: Thu 16 Jul 16 2009 2:18 am
by Daniel Wee
Okay, today I got down to figuring out why the received picture was not stable. As it turns out - we should not have put the output AC-coupling capacitor before the video amplifier because this caused the signal to be zero biased and only the positive side of the waveform was getting amplified. This needs to be DC-coupled to the video amplifier. The fix was a simple matter of removing the capacitors and jumpering them. This got us a nice image with the correct voltage. Problem solved.

We are also now seeing some interference in the video image - probably coupled through the power supply. If we are to re-design this, we will have to be a lot more careful about the layout and power supply isolation. Not a major problem.

The biggest problem we're seeing, though, is the fact that one of the channels doesn't seem to have proper variance in the signal evaluator. Regardless of how good or bad the signal is, it's level seems roughly the same. It is unclear at this point why this is the case but it is possible that there is a problem in the signal evaluator. Unfortunately we will need an SA to actually evaluate this. Sensor asymmetricity has always been a concern - be it arising from differences in the receiver, or from component tolerances.

The power supply is also woefully inadequate and is getting uncomfortably hot. This is one of those applications which require a proper switching PSU.

Another problem with this prototype construction is that the metal cased modules had been soldered direct onto the board without regard for the possibility of the casing shorting out against the vias. There should have been some spacing, even the width of a piece of paper would have done.

Daniel

Re: 2.4GHz Diversity Video Receiver

Posted: Tue 11 Aug 11 2009 8:11 pm
by Daniel Wee
Today I did some more testing and probing of the board and found the reason for the signal level discrepancy between the two channels. The problem seems to be arising from the modules themselves, and not from the detector circuitry. I think that the modules are picking up interference from the MCU due to the proximity of the antenna port to the MCU. This would also explain the interference in the received video. The key now is to isolate this interference and/or keep it away from the receiver modules so that we can have a clean signal to start with.

Connecting the inputs of both receiver modules together did not resolve this issue so it is also possible that the signal corruption is coming via the power supply. The culprit seems to be the +5V power line running under the the MCU and parallel to one of the LCD serial driver lines.

Daniel