46 replies to this topic

[JMS]

Thanks Thomas, for the test. I believe the gain in performance comes directly from the reduced bandwidth. I am however, exploring some filters with far lower IL (typically 4-5dB) which would represent a 16dB improvement. Unfortunately these filters need to be custom made but it would be really interesting to see the results of such a filter.

Daniel

well it would go nicely with all these LRS floating around now.

[Old Man Mike]

First thing to do is to measure a stock 900MHz receiver. So, I grabbed a brand new one (fresh from the box) and prepared to measure it. I needed an accurate 900MHz fixed RF source that had about -80dBm signal levels that could be adjusted.

My solution was to use a 900MHz / 500mW (+27dBm) A/V transmitter and install a fixed 30dBm attenuator on it. That gave me a nice -3dBm signal source, which I verified on my Spectrum Analyzer. Next I added a 110dB step attenuator. The net effect is that had a verified signal source that could go from -3dBm to -113dBm. That's plenty of range.

Sorry to bring up an old subject but I am doing similar measurements and have a few questions about your approach:

1) How did you measure the video power level? Did you just measure the unmodulated carrier?

2) How did you measure the video bandwidth?

My measurement seem to show a lower sensitivity level. For example, with AGC just above minimum (1.28 Ghz Lawmate has AGC which increases with signal level) and quieted audio just beginning to have some noise (Audio subcarrier verfied at -20 db from center carrier), I measure -85 dBm. At -90 dBm there is still good enough video to fly by but it is quite noisy. This appears to be several dB better than I expected.

I've done a number of checks to make sure source leakage is not a problem. My HP-8594E Spectrum Analyzer is accurately calibrated as well as the step attenuator, etc.

OMM

Edited by Old Man Mike, 17 January 2010 - 05:57 PM.

[Mr.RC-Cam]

(1) RF power level verification from the video Tx was done on the raw carrier (without video). But if you want to do it with video modulation then be sure that your tests use a fixed test pattern so that you can reliably repeat the before/after results.

(2) Video bandwidth is determined with a lab freq generator set to 1VpkpK into the Tx's 75 ohm video input. Video bandwidth was measured to the -3dB point on the receiver's video signal (via a 75 ohm termination and o-scope).

My measurement seem to show a lower sensitivity level.

On my 900MHz test Rx I measured -81dBm after the SAW upgrade. But you measured -85dBm on your 2.4GHz Rx. So your 2.4GHz Rx has a *higher* sensitivity than my 900MHz, not lower.

[Old Man Mike]

(1) RF power level verification from the video Tx was done on the raw carrier (without video). But if you want to do it with video modulation then be sure that your tests use a fixed test pattern so that you can reliably repeat the before/after results.

Done it both ways with the same result. Actually don't need a fixed test pattern since the HP will list a repeating table of maximum values.

(2) Video bandwidth is determined with a lab freq generator set to 1VpkpK into the Tx's 75 ohm video input. Video bandwidth was measured to the -3dB point on the receiver's video signal (via a 75 ohm termination and o-scope).

What frequency did you use for the generator?

On my 900MHz test Rx I measured -81dBm after the SAW upgrade. But you measured -85dBm on your 2.4GHz Rx. So your 2.4GHz Rx has a *higher* sensitivity than my 900MHz, not lower.

First, I meant to say receive a *lower* signal level since -85dBm is a *lower* signal level.

Second, I'm sure you meant to say *1.28 Ghz* - not 2.4 Ghz.

By the way, have you tried using a 2.33 Mhz, 1VpkpK sine wave for the video input to set the video gain? It sure beats any other method I've seen.

OMM

Edited by Old Man Mike, 17 January 2010 - 10:22 PM.

[Mr.RC-Cam]

What frequency did you use for the generator?

I just use the freq gen to find the -3dB points. The frequency that these occur at will depend on the video link's performance. On the test system the high end was measured at ~4.9MHz before and after the SAW upgrade.

By the way, have you tried using a 2.33 Mhz, 1VpkpK sine wave for the video input to set the video gain?

I always use a composite video generator to set the video levels. All my A/V systems get this tweak (the mfg's factory adjustments are not reliable).

[Old Man Mike]

I just use the freq gen to find the -3dB points. The frequency that these occur at will depend on the video link's performance. On the test system the high end was measured at ~4.9MHz before and after the SAW upgrade.


I always use a composite video generator to set the video levels. All my A/V systems get this tweak (the mfg's factory adjustments are not reliable).

I've also seen a lot of variation in the mfg's factory settings. Using a 2.33 Mhz input to set the Bessell null is quite nice since it does not require a composite video generator and "standard" receiver so there are less variables in the process.

One more question and I'll stop pestering you. Have you tried pre-emphasis/de-emphasis for an improved S/N?

OMM

[Mr.RC-Cam]

Have you tried pre-emphasis/de-emphasis for an improved S/N?

I indeed have been working on that. Custom PC boards are built and partially debugged. My gut feeling is that using it with the SAW filter will double the range on a 900MHz - 1.3GHz system. Fingers are crossed!

[Old Man Mike]

I indeed have been working on that. Custom PC boards are built and partially debugged. My gut feeling is that using it with the SAW filter will double the range on a 900MHz - 1.3GHz system. Fingers are crossed!

6 dB is what I would expect too. I hope your results are better than mine. I just did not see the expected improvement but I did it rather quickly with passive filters on the TX and RX. It is possible that I made a mistake in the build and/or test. I'll be interested in hearing your results and may revisit mine later this week.

OMM

[ssassen]

6 dB is what I would expect too. I hope your results are better than mine. I just did not see the expected improvement but I did it rather quickly with passive filters on the TX and RX. It is possible that I made a mistake in the build and/or test. I'll be interested in hearing your results and may revisit mine later this week.

OMM

Great minds think alike? I was just about to solder in these pre/deemphasis networks and do some further measurements. Unfortunately I'm still waiting for my SAW filters to arrive, but once they do I'll be doing the same measurements both of you have done. I've flown out to >4km with a 500mW 1.2GHz (1180MHz) Tx (http://www.vimeo.com/8715397), which shows plenty of link budget left at that distance (RSSI values noted are for the EzUHF RC control link, but video is crystal clear). Since then I've plotted the RSSI curve and verified the signal levels actually reflect a useable image and am confident that with an omni (0dBi) on the Tx and a 8dBi flat patch on the Rx I should be seeing >10km of range, if these SAW filter and pre/deemphasis mods work out that should about double.

Cheers,

Sander.

Edited by ssassen, 19 January 2010 - 04:15 AM.

[Old Man Mike]

Great minds think alike? I was just about to solder in these pre/deemphasis networks and do some further measurements. Unfortunately I'm still waiting for my SAW filters to arrive, but once they do I'll be doing the same measurements both of you have done. I've flown out to >4km with a 500mW 1.2GHz (1180MHz) Tx (http://www.vimeo.com/8715397), which shows plenty of link budget left at that distance (RSSI values noted are for the EzUHF RC control link, but video is crystal clear). Since then I've plotted the RSSI curve and verified the signal levels actually reflect a useable image and am confident that with an omni (0dBi) on the Tx and a 8dBi flat patch on the Rx I should be seeing >10km of range, if these SAW filter and pre/deemphasis mods work out that should about double. If that all holds true, all that is left for me to do is to make a custom 1.2GHz 500mW Tx design which uses the ATV channels so can be legally used by HAMs.

Cheers,

Sander.

Hi Sander,

Nice 1.28 Ghz video signal at that range. I've been approaching the omni coverage issue a bit differently as you can see here:

http://www.rcgroups.com/forums/showthread.php?t=1147430

You can also see my new antenna design on page 7 of that thread which provides a circular polarized omni with a 20 dB LHCP to RHCP ratio over the entire radiating sphere. This allows much reduction of multipath distortion/loss which is a problem for long distance, low height quad copter flying.

OMM

[Mr.RC-Cam]

I'll be interested in hearing your results and may revisit mine later this week.

Update: A bit disappointed, but not discouraged. I finished up the measurements today and sadly must report that the pre-emphasis trickery does not improve my 900Mhz test system.

I suspect there is a non-standard pre-emphasis network already in the system I tried it on (900MHz/500 mW Tx and Rx, www.dpcav.com). However, this should not stop anyone from continuing on the quest since there may be some A/V systems that would benefit from something like this.

Below is a photo of the project (pre-emphasis board on Tx, de-emphasis board on Rx). Both boards had video level compensation to ensure proper amplitudes through the emphasis circuity. Also, active buffer-amps in the video chain ensured that the LC networks saw exactly 75 ohms, regardless of what was connected to them.



In case there is any possible confusion from the casual readers: this pre-emphasis stuff has nothing to do with the original topic's SAW filter modification. So, I apologize for pushing the discussion off-topic.

[Old Man Mike]

Update: A bit disappointed, but not discouraged. I finished up the measurements today and sadly must report that the pre-emphasis trickery does not improve my 900Mhz test system.

I suspect there is a non-standard pre-emphasis network already in the system I tried it on (900MHz/500 mW Tx and Rx, www.dpcav.com). However, this should not stop anyone from continuing on the quest since there may be some A/V systems that would benefit from something like this.

Below is a photo of the project (pre-emphasis board on Tx, de-emphasis board on Rx). Both boards had video level compensation to ensure proper amplitudes through the emphasis circuity. Also, active buffer-amps in the video chain ensured that the LC networks saw exactly 75 ohms, regardless of what was connected to them.



In case there is any possible confusion from the casual readers: this pre-emphasis stuff has nothing to do with the original topic's SAW filter modification. So, I apologize for pushing the discussion off-topic.

Well, that makes me feel a little better about my test with the passive filters. I had wondered if an active pre-emphasis & de-emphasis would do any better but did not want to put the effort into it. Those are VERY nice looking PWBs and it is a shame that the results were not better.

It does make me wonder why we did not see the improvement.

OMM

[ssassen]

Well, that makes me feel a little better about my test with the passive filters. I had wondered if an active pre-emphasis & de-emphasis would do any better but did not want to put the effort into it. Those are VERY nice looking PWBs and it is a shame that the results were not better.

It does make me wonder why we did not see the improvement.

OMM

I can confirm that the pre/deemphasis doesn't really pay off. I couldn't see any increase in sensitivity or noise levels that would stand out from the margin of error in the test configuration. The SAW filter upgrade however did pay off nicely. I've seen a nice 4 tot 5dB sensitivity increase with a sampling of two receivers modified (BSS479 type). I did however do away with the whole baseband video decoding and reworked that entire part of the circuitry after some extensive modelling in LTspice. I now have video quality that's indistinguishable from the best 2.4GHz receivers I have here. The audio portion is next, but that requires a little more effort.

Ps. getting the SAW filter out was a piece of cake in my experience. Use solder wick to remove solder on the input/output pins and make sure they can move freely inside the hole. Use a soldering iron >50W for the ground pin with a broad/flat tip. Heat the ground pin and with a small screwdriver (slightly larger than the pin) just push the pin down after a few seconds of heating the pin. That'll pop the SAW filter right out without any effort!

Cheers,

Sander.

[lvsupertech]

I have replaced the saw filters in 900 MHZ RX. Before the swap out, I have noticed at distance 2 miles + some noise possible from transmitter close to frequency I was on. Now eliminated with the ECS-D480A New saw filter. Works old was F 480 2

Edited by lvsupertech, 26 December 2010 - 11:44 PM.

[JetPilot]

Do you think an LNA in addition to the saw filter modification would increase the performance of this 900 MHZ video receiver ?

Mike

[galaxiex]

Made a little video showing my method of removing the SAW filter from these A/V receiver modules.

I know some folks may think this a crude method...
but I find this method very easy to do, and this has worked for me every time on the 6 Rx's that I have modded so far.

The Rx in the video will be the 7th one I have done, and it is now all back together and working just fine.

Edited by galaxiex, 24 April 2011 - 04:43 PM.

[Mr.RC-Cam]

Thanks for the video. I use a SMD rework hot air wand and heat up the SAW case, then lift it out with tweezers.

Don't forget to wear eye protection!

[galaxiex]

Aaaaahh yes.... I can only wish to have a hot air rework station...

Eye protection.... ya I was wearing glasses... but not safety glasses... my bad!
I do normally wear safety glasses when soldering.

[galaxiex]

SAW filter install...

http://www.vimeo.com/23095429

Edited by galaxiex, 30 April 2011 - 07:22 PM.

[tascheri]

Just got one that has a filter part # D479.5B Should I change it??

edit: 3 db pass band = 26.7 Mhz typ... will soon be replaced!

Edited by tascheri, 03 May 2011 - 01:45 PM.