Doofer

FR632 5.8GHz diversity receiver sync problem

32 posts in this topic

5.8GHz seems to 'need' diversity less than 2.4GHz in terms of dropouts, but even so I thought I'd try the new FR632 diversity receiver for a local plus high gain arrangement.

Bench tests seemed okay, but on fitting it into a real system I found picture roll on some screens, intermittant 'interference' across the middle of the screen on others. Swapping the part out for an Airwave module or a 5.8GHz single-input 'scanning' receiver gave a perfect picture, so I suspected a sync problem in the FR632.

On the 'scope' the outputs are different (see attached) - FR632 - Airwave - Scanner.

Everything else was the same - okay, there was a OSD message flashing on and off in all three cases, which probably explains the slight extra 'detail' in the video trace for the Scanner.

The Airwave and Scanner outputs, although at different references to earth, look fine, with a clear sync pulse at the beginning. In practice these receivers perform well. The FR632 looks odd, and is awful.

Any suggestions how I might fix this? Online reviews of the FR632 are full of praise, so presumably it's just this one...

post-1834-0-91676900-1431366219_thumb.jp

Share this post


Link to post
Share on other sites

The FR632 has been optimized for the 32-Channel vTx's produced by Skyzone, Boscam, and similar low cost China brands. The suppressed sync suggests that your vTx is not compatible. What exact 5.8GHz vTx are you using?

Either that, or the FR632 is set to the wrong RF band or maybe it is defective.

Share this post


Link to post
Share on other sites

Airwave AWM661TX.

Share this post


Link to post
Share on other sites

Trying a 10mW FPV Hobby Tx, the picture looks alright, although the sync pulse still looks a bit odd

post-1834-0-47574800-1431415461_thumb.jp

Share this post


Link to post
Share on other sites

Unfortunately the de-emphasis used in the FR632 is not compatible with Airwave. That is the reason for the poorly shaped syncs.

The FPV forums are full of advice about how you can freely mix and match vTx with vRx brands as long as the frequency is the same. It is poor advice. RF Frequency, video bandwidth, video emphasis, and audio subcarriers all have to be compatible for correct performance. There is no industry standard on these things, so that is why unpredictable performance can occur when mixing brands.

Edited by Mr.RC-Cam
Corrected vRx number

Share this post


Link to post
Share on other sites

I guess I was hoping that if it was a known problem, someone would suggest some component in these things that could be altered...

Mind you, looking in detail at single shots on the scope, the sync pulse outputted from my FR632 contains bursts of something at a much higher frequency, totally absent with other receivers. I am wondering if it's just a faulty one; I get this even with a FT951 transmitter, which is from the same manufacturers as the FR632! Also, given that no-one else is moaning about this receiver, I'm inclined to think it may just be a faulty one.

Share this post


Link to post
Share on other sites

Rats - thought I'd nailed it. With the FT951 (same manufacturer) Channel B has the hairy sync pulse, Channel A is clean - faulty Rx syndrome. But then back to Airwave transmitter, and both Channels have the noise on the picture,

I'll probably see if a local supplier will let me have one on 'success or return'...

Share this post


Link to post
Share on other sites

I guess I was hoping that if it was a known problem, someone would suggest some component in these things that could be altered

That is possible but not practical. It would be an inductor and/or cap in the vRx's de-emphasis circuit buried inside the sardine canned module.

Also, given that no-one else is moaning about this receiver, I'm inclined to think it may just be a faulty one.

I suspect that some of those users have a compatible 32-channel vTx because they are cheaply priced, which goes well with the bargain priced 32-channel FR632.

BTW, the pre-emphasis used in the 7-channel Fatshark vTx is "reasonably" compatible with the FR632 so the video problems are less obvious. The ImmersionRC 7-channel vTx would be border line compatible with the FR632 (because it is more closely matched to Airwave's video emphasis) so success can be random with it.

I'll probably see if a local supplier will let me have one on 'success or return'...

There is enough slop in the component tolerances that if you test enough vRx's you will find one that will work reasonable well with your Airwave's pre-emphasis. But full compatibility would be unrealistic to expect so 100% performance would be unlikely. But with luck your displayed image might not show the problems that exist in the native video signal.

There is more to compatibility than just RF frequency. The video emphasis, video bandwidth, and audio sub-carriers specifications must also be compatible. There is NO standard in the industry for these things so mixing brands can result in compatibility issues. There are naive FPV'ers on the inter-web that claim that only the frequency needs to match and my point is that such statements are only part of the story. Those that claim only freq matters may have had "success" with their mixed and matched components, but their good fortune may not be the same as yours.

Practically speaking, there used to be two different common RF standards within the 5.8GHz FPV equipment choices. Airwave / ImmersionRC / Fatshark *versus* the Mainland / Hongkong Chinese brands. But now ImmersionRC and FatShark have drifted away from the Airwave standard (either by accident or intentional). So the lines that separate the compatibility issues have become blurred and now it's more of try-it-and-see if it works sort of thing.

At this point my best advise is to not mix 7-channel designs with 32-channel 5.8GHz designs. From my observations those tend to have lower success rates, and even when they work there will be performance loss (most often mal-formed syncs and video bandwidth issues). Some users don't see the effect of the compatibility problems in their FPV image so they believe that these technical issues do not exist.

Edited by Mr.RC-Cam
corrected Rx number

Share this post


Link to post
Share on other sites

I can always use my Oracle and a couple of Airwave RX's... Did you ever make a smaller version? Even stripped down to the board + servo sockets for connections, it's a fair footprint...

Share this post


Link to post
Share on other sites

I can always use my Oracle and a couple of Airwave RX's.

The RF sensitivity (-85dBm) on the 5.8G Airwave vRx's has fallen behind the latest offerings available to FPV'ers. You can get significantly more range if you use newer vRx technology that has -90dBm sensitivity.

If want to continue using the Airwave vTx and upgrade your existing Airwave vRx modules for higher performance, then the latest Fatshark 5.8G vRx modules have -90dBm sensitivity. The Fatshark modules are similar in size to the Airwave vRx module and have nearly the same wiring requirements. Or you can retire your Airwave vTx and use a vTx that is compatible with the FR632.

Did you ever make a smaller version? Even stripped down to the board + servo sockets for connections

I don't have a new diversity design that replaces the Oracle. But several months ago I published a DiY open source FPV diversity ground station project that I thought the rc-cam community would enjoy developing. The hardware design was posted and the open letter invite asked interested FPV'ers with Arduino programming experience to create firmware for the features that *they* wanted. It was intended to be a community driven project but sadly there was zero interest. It was a painful misstep on my part and yet another reminder that my DiY projects can be a bit too unusual for general consumption.

Share this post


Link to post
Share on other sites

This has been driving me nuts for months. Current set up is:

rmrc moniter

Boscam tx5826 module diy tx

skylark tiny osd

sony super had camera

Boscam fr632

quanum v2 pro goggles.

image on rmrc moniter has been ok, with the picture sometimes yellowing. Flyable

image on quanum has intermittent rolling/tearing of picture, so I assumed there was something wrong with the goggles.

Checked signal on Samsung TV and I get the same tearing of picture. So not a monitor issue.

Finally found my way here and read about possible incompatibility of 32 band and 7 band. 

Built a receiver using rx5808's that I was going to use for arduino based rssi tracking, and got an instant improvement in picture quality. It is now perfect. Checked range next to fr632 and get same range.

Would have thought boscam modules would play together? 

Doesn't Fr632 use similar modules to rx5808's anyway?

Will use fr632 rssi for tracking (it can at least do that) and add diversity switch to diy reciever.

what IS a compatable Tx for the fr632? 

Share this post


Link to post
Share on other sites
Quote

Would have thought boscam modules would play together? 

Hard to know why your Boscam modules did not play well together without some investigation. But keep these things in mind:

1. The Quanum V2 goggles are finicky eaters and can experience exaggerated video display problems with imperfect video signals. It is known to have unreliable video with the FR632 vRx. See this: http://www.rcgroups.com/forums/showthread.php?t=2641715

2. It appears that there are cloners in China that are making copies of the RF modules. As with many China produced goods, it is hard to know what is real and what is an unauthorized copy that does not work well.

3. The China produced FPV parts have random quality control. So what's great one week may be a bad choice another time.

4. The Boscam DiY modules are available in different 5.8GHz frequency bands. The sellers are clueless about this and often provide the wrong frequency version. This can cause some compatibility issues and/or frustration to the end user.
 

Quote

Doesn't Fr632 use similar modules to rx5808's anyway?

The RF modules all share a similar hand-me-down reference design. But the component and mfg quality greatly varies.

 

Quote

what IS a compatable Tx for the fr632? 

I don't have any recommendations for a vTx that would be ideal for the FR632. Unfortunately vTx / vRx purchase decisions have become a buy/try/pray situation. Also, it might be a mistake to choose the FR632 given the reported issues with using it with the Quanum V2 goggles. But if you are only using its RSSI signal (and ignoring its video) then it should be fine to do that.

 

Edited by Mr.RC-Cam
Added comment about RSSI.

Share this post


Link to post
Share on other sites

Points 1-4 noted. 99% it's not 4 (wrong frequency). Initially I thought is was a problem with with quanum goggles as well, but after viewing the exact same issue when the fr632 is plugged into a 40inch HD tv, I have to give it the benefit of the doubt. My very homemade rx5808 reciever works perfectly with the quanums as well, I think I must have a dud fr632. If the tx5826 should play with the fr632 think it must be a QC issue somewhere.

I have a spare tx5826 which I might connect to see if that fixes the problem --- long shot, and then cut my losses. Will use fr632 for tracking and use diy 5808 for diversity. The opposite of what i was intending.

If everything worked, i wouldn't get to spend hours fault finding! That's half the fun(frustration) of the hobby.

Share this post


Link to post
Share on other sites

DISCLAIMER: I am providing the modifications as-is. While it worked for me, the device can be damaged by these actions. I am in no way responsible for any damage done to any concerned devices.

Hello guys. I've actually found the issue with this receiver and was able to fix it...

So after I’ve bought the FR632 40ch diversity receiver I encountered the issues with distorted image on quick light changes. Seller confirmed it’s faulty and sent me an replacement, but the replacement receiver had the same issue (as for a lot of us). I got my money back, and after some search I’ve seen that quite a lot of people have the same issue (not everyone, not with all vTx’es, fpv cameras and screens).

So I disassembled one of the receivers and after some poking I’ve found out that the output of RX5880 modules inside is almost not distorted. Recently I was able to borrow an oscilloscope and it confirmed the case- the signal was altered/distorted after the RX5880 module and even after the diversity circuit (it was my first suspect). At each of the 3 outputs of the FR632 receiver there is a tiny transistor marked “1AMw” that does nothing (as far as my knowledge of electronics goes) but modifying, or, actually, distorting the signal.

After quick desoldering and bridging of the contacts the image distortion almost disappeared, it is now low enough to not even observe it when flying.

Here's the explanation video: 

 

Really hope I helped someone :)

Edited by Sulya
Please add the disclaimer
1 person likes this

Share this post


Link to post
Share on other sites

Thanks for sharing the mod.

The component you removed is a NPN transistor (2N3904) configured as an emitter follower. Its job is to buffer the video signal so it can safely drive the 75 ohm terminated video. Bypassing the transistor will increase the video amplitude by about 35%; Your image distortion problem involved low video amplitude so the increased video level helped it out. It is NOT the same problem reported by Doofer that started this topic on the FR632.

Although part of the blame is on the FR632's marginal design, the vRx's processed video level is also affected by the vTx that is being used. A reliable video image is also affected by the display device's tolerance to out-of spec-video. So some FR632 users will not experience your video issue.

FWIW, bypassing the transistor means that the video amp IC that feeds it now has to drive 37 ohms on its own. The video amp is house marked JJR but its official part number is unknown. Without a datasheet it is not known if it should be allowed to drive the low impedance video output by itself. But even so, if the choice is to trash the vRx or hack it, then hacking wins.

By the way, after bypassing the transistor I think it would be wise to remove its emitter sink resistor that is no longer needed (this will reduce the video amp's drive current burden and further increase the video level). It's the 75 ohm shown in this image:

fr632_mod.jpg

 

Edited by Mr.RC-Cam
Added comment that sulya's video level mod is not the same problem reported by Doofer.

Share this post


Link to post
Share on other sites
11 hours ago, Mr.RC-Cam said:

Thanks for sharing the mod.

The component you removed is a NPN transistor (2N3904) configured as an emitter follower. Its job is to buffer the video signal so it can safely drive the 75 ohm terminated video. Bypassing the transistor will increase the video amplitude by about 35%; Your image distortion problem involved low video amplitude so the increased video level helped it out.

Although part of the blame is on the FR632's marginal design, the vRx's processed video level is also affected by the vTx that is being used. A reliable video image is also affected by the display device's tolerance to out-of spec-video. So some FR632 users will not experience your video issue.

FWIW, bypassing the transistor means that the video amp IC that feeds it now has to drive 37 ohms on its own. The video amp is house marked JJR but its official part number is unknown. Without a datasheet it is not known if it should be allowed to drive the low impedance video output by itself. But even so, if the choice is to trash the vRx or hack it, then hacking wins.

By the way, after bypassing the transistor I think it would be wise to remove its emitter sink resistor that is no longer needed (this will reduce the video amp's drive current burden and further increase the video level). It's the 75 ohm shown in this image:

fr632_mod.jpg

 

Thanks for all the explanation!

I also removed the 75 ohm resistors as you advised. I compared the video level output to one with the resistor still present, and with the oscilloscope I have there was no visible difference. But it is indeed good idea to unload the video amp a bit, thanks for the tip. Maybe I should add this to the video explanation.

Now that transistor is identified it's maybe also possible to find an alternative for this transistor that would produce higher level output signal.

Anyway, at this point I am leaving my FR632 receiver with both transistor and resistor removed, as for me it works quite good this way. I will report if it dies because of increased video amp's load.

2 people like this

Share this post


Link to post
Share on other sites
Quote

Anyway, at this point I am leaving my FR632 receiver with both transistor and resistor removed, as for me it works quite good this way. I will report if it dies because of increased video amp's load.

With further investigation I was able to identify the "JJR" marked component that drives the video buffer transistor. It is a Chinese copy of the TS5A3159A SPDT Analog Switch.  The good news is that it can safely handle the video signal's drive current after removing the buffer transistor. But this means that the RX5808 / RX5880 vRx module's shared video outputs are no longer isolated or buffered.

So I have two strong recommendations when the buffer transistor is removed / bypassed:
1. Remove the 75 ohm resistor shown in my previous post.
2. Do NOT remove the buffer transistors from the other video outputs. Only mod the Diversity output.
 


Additional Mod idea:

Be aware that the FR632's video outputs are very low impedance and DC coupled (instead of using the traditional 75 ohm AC coupled methods). This sort of penny saving trickery can reduce compatibility with some display devices. A partial remedy could be achieved by adding a AC coupling cap to the vRx's video signal.

The AC coupling cap would not be difficult to add. The required value is 330uF to 470uF (electrolytic or Tantalum, 5V or higher rating). Essentially it goes in series with the video signal (between vRx and display with positive side towards vRx). For example, instead of shorting the two pads as shown in your YouTube demo you could replace the shorting wire with the cap as shown in the image below.

fr632_mod2_1000.jpg

 

If the cap is too large to fit inside the vRx then it can be installed externally. Long story short, keep this AC coupling cap mod in mind if the display device is still having problems with the modified FR632.
 

 

1 person likes this

Share this post


Link to post
Share on other sites

Hi there, glad I came across your solution guys.  Same issue here and I'd really like to mod it but I'm not sure where I'd find a coupling capacitor, especially the one in spec you recommended.  Nothing seems to come up on ebay or any cheap electronics site.  Could you tell me where I could find one cheap?

 

Thanks

Edited by SpagettiConfetti

Share this post


Link to post
Share on other sites

You don't need to try adding the cap unless you still have problems after Sulya's mod. It is a common part that can be purchased from eBay or any electronic component supplier: eBay 470uF Cap

1 person likes this

Share this post


Link to post
Share on other sites

Oh!  Here I was thinking it was some sort of special capacitor.  I have exactly 470uF 16v caps. Not sure if that's too out of spec.  I could always go out and purchase a 5v one.  Would putting in the cap as well be ideal, further problems or not?

 

Thanks for clearing that up

Edited by SpagettiConfetti

Share this post


Link to post
Share on other sites

Any rated voltage 5V or higher is fine. If you want to add the video coupling cap right away then go ahead. But I doubt it will fit inside the vRx cabinet; That is why it is best to hold off and only add it if the basic mod does not fully solve the video signal level problem.

 

1 person likes this

Share this post


Link to post
Share on other sites

Well if the video signal stands to gain anything from it then I'll try to figure a way to fit it all together,  I don't mind it sticking out if there's a quality gain to be had.

 

Thanks, hope I'm understanding correctly

Share this post


Link to post
Share on other sites

Did mod. Removed resistor, transistor and added capacitor. Early signs promising. No tearing of image or rolling of picture. Fingers crossed that it will stay that way. Plane in pieces at the moment so will be a while before I can do a proper test. 

Share this post


Link to post
Share on other sites

On further testing (20 min constant transmission), no tearing or image distortion was noted after the mod. Hooked up a 30db attenuator to simulate long range and image stays stable right up to "snow distance". A 470uF Cap does fit in the casing, but extension leads are needed to solder it in.

So happy this has been resolved.

Thanks Sulya and Mr.RC-cam.

2 people like this

Share this post


Link to post
Share on other sites

Awesome to hear man. Thanks for sharing.  I'm waiting until I get confirmation that they're giving me my return first before I mod.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now