Jump to content
secretspy711

Oscilloscope recommendation for video system troubleshooting

Recommended Posts

Can you capture the TV Line waveform directly at the camera output while it is still connected to the vTx?

BTW, did the Pyle video amp help? Also, have you tried your friend's UNO to see if it experiences less video bouncing than yours?

Some Comments:
No surprise that two identical receivers have different video levels. I have seen variations in video levels on many brands of FPV gear. A lot of users have no idea how bad their video signal really is. They could achieve better performance if this was not going on.  It seems to have gotten worse now that everything is made in Hong Kong / Mainland China (versus a few years ago when the popular choices were produced in Taiwan).

Fortunately most of the new 5.8GHz FPV gear has better video level accuracy (as long as brand mixing is avoided).  But there are still some bad apples even in the 5.8GHz market.

Edited by Mr.RC-Cam
Bimbo typo's.

Share this post


Link to post
Share on other sites

Yes, I did that the other day, here is that image (with the Vector bypassed).

Cam only with bypassed Vector using IRC 2.4vtx as load.png

I haven't tried the Pyle amp yet.  I was about to go fly 2 back-to-back flights to 4-5 miles, one with my Uno, one with my friend's, and see if there is any difference in stability in a real-world test.

BTW, I'm not sure if you picked up on this from the other thread, but when the Uno's first boot up, the waveform is significantly taller, around 1.2-1.3 vpp (and thus so is the sync.)  After it beeps 3 times to indicate the # of cells it's connected to, it drops.  Depending on which of my 4 vtx's I'm using it's slightly different, but in general mine goes to ~840-860 mV while my friend's goes to 920-930 mV.  So maybe there is a chance I can send my Uno back to immersion and have them tune it to put out slightly higher levels.  If my back-to-back flights show improvement with my friend's Uno, I will probably pursue that.

My Uno at startup:

brian uno startup.jpg

 

My Uno after the 3 beeps:

brian uno.jpg

Friend's Uno at startup:

randy uno startup.jpg

 

Friend's Uno after 3 beeps:

randy uno.jpg

Edited by secretspy711

Share this post


Link to post
Share on other sites

Thanks for posting the waveforms.

Quote

Yes, I did that the other day, here is that image (with the Vector bypassed).

The H-sync into the vTx is indeed Ok. Unfortunately that confirms the sync attenuation is happening inside the RF modules (video emphasis mismatch or video bandwidth related problem).

Quote

I haven't tried the Pyle amp yet.  I was about to go fly 2 back-to-back flights to 4-5 miles, one with my Uno, one with my friend's, and see if there is any difference in stability in a real-world test.

Might save you some pain if you did it on the bench using your attenuators and the special procedure you mentioned before. But a test flight would tell the most accurate story.

Quote

BTW, I'm not sure if you picked up on this from the other thread, but when the Uno's first boot up, the waveform is significantly taller, around 1.2-1.3 vpp (and thus so is the sync.)  After it beeps 3 times to indicate the # of cells it's connected to, it drops.

My interpretation on the power-up behavior is that the controller's initialization commands to the RF module are not completed until after the battery cell count reporting. The symptom could even vary depending on the rest time between power ups (due to the charge states in the RF chip's CMOS registers).

Quote

So maybe there is a chance I can send my Uno back to immersion and have them tune it to put out slightly higher levels.

If you don't see any pots inside the UNO2400 then calibrating the video is not as simple as making an adjustment. I suppose a patient tech could cherry pick an RF module that exhibits higher output levels. Sounds messy and time consuming.

Instead, maybe we can mod the vRx. I don't have a UNO2400 to peek inside; If you post some clear close up PCB images maybe I can help with hacking it so it can grunt out a higher overall video signal. No promises, but worth a shot.

 

Share this post


Link to post
Share on other sites

Ok I'm just back from flying, typing on my phone at the field while it's still fresh in my mind.

 

1st flight, using my own Uno, started getting a lot of bouncing at 3.5 miles.  Came back and landed.

2nd flight, using friend's Uno.  Saw one or two bounces on the way out at about 3.5 miles, but nothing to complain about.  Pushed on to 5 miles without any more bounces on the outbound leg.  Turned around, again saw one or two bounces on the way back, but in general was clear.

Just to be sure, I made a 3rd flight with my own Uno again.  Bounces started at about 4 miles, pushed on to 5 miles anyway with bouncing the whole way.  Turned around at 5, bouncing remained until I was inside of about 3.5 miles.  In general it seemed like the overall video quality was not quite as good compared to my friend's Uno.

I will post pictures of the PCB when I get home.  I would love to be able to coax a higher signal out of it.

Share this post


Link to post
Share on other sites

Full-size shots of the Uno PCB:

https://drive.google.com/folderview?id=0BxRugM4Cs1qIaUd0U1M4ZTBPRGs&usp=sharing

I'll let you study that while I work on wiring up the Pyle amp and see if that helps the levels.  If it works then maybe we don't have to mess with the Uno's guts, but I'm curious if you are able to spot anything obvious that might help.

Edited by secretspy711

Share this post


Link to post
Share on other sites

I rigged up the Pyle (quick and dirty with long cables I had lying around).  Here's a shot with it adjusted to 1Vpp.  I am still able to produce some bouncing using my ground test procedure:

Uno to Pyle adjusted single line.jpg

 

And here is a shot with it adjusted to its maximum.  I am NOT able to produce bouncing at this setting :)  Overall level is high compared to the standard, but image looks ok in the goggles, at least indoors.  I need to clean up the wiring to the Pyle and perform a flight in this configuration, but that will have to wait, maybe tomorrow.

Uno to Pyle at MAX single line.jpg

Edited by secretspy711

Share this post


Link to post
Share on other sites

The Pyle Video Amp sounds promising!

Even better, it looks like we can hack your UNO2400 and skip the external amp. At the moment I'm taking a break from tiling my master bath and won't be available for a couple more hours. I'll try to get the mod's details posted tonight, but worst case by tomorrow morning.

 

Share this post


Link to post
Share on other sites

Change the UNO's 75 ohm resistor at location R15 to 100 ohms. This will increase the overall video level.

UNO_video_mod.jpg

The nice thing about this hack is that it does not disturb the 75 ohm impedance on the three A/V outputs. That is because the dirty work occurs before the three +6dB buffer amp IC's. BTW, increasing the R15 resistor beyond 100 ohms could diminish the sync clamp performance in the buffer amps. So hopefully 100 ohms provides the larger video levels that your setup needs. Fingers crossed!

One more thing. Try all the 2.4GHz channels to see if any of them offer higher levels, especially in the sync region.

Share this post


Link to post
Share on other sites

Fantastic!  Does it need to be an SMD resistor or can it be a metal/carbon film type?  Any idea what to expect the peak to peak voltage to be with this mod?  25% higher?

Edited by secretspy711

Share this post


Link to post
Share on other sites

For this hack I would grab a 0805 size 5% carbon SMD. But it can be any common resistor you have on hand. Video level will be roughly 15% larger (but may vary depending on the RF module's output configuration).

Edited by Mr.RC-Cam
Corrected expected video level after mod

Share this post


Link to post
Share on other sites

I have an assortment of metal film resistors but not any 100 ohms.  I could get close by using a couple of them in parallel but I decided to just order a few of the correct ones for a "cleaner hack".  These are what I got: http://www.mouser.com/Search/ProductDetail.aspx?R=ERJ-T06J101Vvirtualkey66720000virtualkey667-ERJ-T06J101V

I wanted to clarify something you said earlier.  You said the entire waveform from the Uno was poor/disappointing with reduced bandwidth.  What exactly are you looking at when you say this?  Is the RMRC receiver better or worse in this regard?

Share this post


Link to post
Share on other sites
Quote

I could get close by using a couple of them in parallel but I decided to just order a few of the correct ones for a "cleaner hack".

While you wait for the "clean hack" resistor I recommend you validate the proposed mod using whatever you have on hand. Test using both ~100 ohms and ~120 ohms. With the 75 ohm terminator on the UNO2400, capture TV-line and TV-Field waveforms at the video output and also at the bottom side of R15 (raw vRx module's video signal). This information will help confirm the mod has a chance of working for you.

Quote

You said the entire waveform from the Uno was poor/disappointing with reduced bandwidth.

The disappointment was because of the IRC vTx/UNO combo's attenuated sync level, low overall video level, and some SAG tilt (see my red boxed TV Field photo). These are problems that I've seen on other brands. I'll let you review your waveforms and decide for yourself if RMRC is better/worse.

 

Share this post


Link to post
Share on other sites

I test-flew the Pyle @ max setup this evening.  It was a much more punishing 5 miles than my earlier tests due to flying low over many rock formations that would've been in the fresnel zone.  The sync instability is much reduced, and was about on par with my friend's Uno as before, with a few bounces here and there, but overall it's much better and I'm pretty happy with it now.  The picture also seems clearer at range compared to the RMRC receiver.  The few bounces I saw were almost like a "pre-warning" that starts before the fuzziness you get when you start to fly a behind something, out of LOS.

Haven't had a chance to swap out the resistor yet, there's only so many hours in the day and frankly soldering small things like that intimidates me a little bit.  I don't want to screw it up before the annual FPV meet next weekend.

Share this post


Link to post
Share on other sites

Sounds like the Pyle video amp has saved the day!

If SMD soldering worries you then practice removing and re-installing some 0805 size parts on a junk-box electronic device. YouTube is full of SMD How-To videos too.

Share this post


Link to post
Share on other sites

Hack successful!  I measured 832mV immediately before the hack, and 948 after.

I think I had better return the scope for now -- someone else may want to use it.  This has been an enlightening adventure :)

Share this post


Link to post
Share on other sites

Glad to hear the resistor swap increased the video level. Hopefully this reduces the bounce issue; If not then come back and we can discuss what to do next.

 

  • Like 1

Share this post


Link to post
Share on other sites

Did the resistor mod solve the video bounce problem that occurs on your long distance flights?

Share this post


Link to post
Share on other sites

I got a similar problem with my IRC Uno. It occurs as soon as I use some kind of rebroadcasting. I tried different cameras, VTx, rebroadcasting VTx. It only occurs with the IRC Uno. I also tried to analyze the signal with a scope but lack knowledge to judge it except from it being a bit weak compared to the output of other receivers.

Glad I found that thread. I had already ordered (but not assembled) a video amp but I will try the resistor mod first. Will report back to you. It's easily reproducible with that rebroadcasting setup actually, no need to make range, just power the tx and watch the video for a few minutes ...

Share this post


Link to post
Share on other sites

OK so I did the mod. Unfortunately it didn't help. I've now got an original and a modded Uno2400, both perform the same way.

First to demonstrate the problem I did a video from inside my goggles where you can see the flickering happen:

 

Again notice that the problem does not occur if attaching the goggles via cable, it also does not occur with other cominations of Rebroadcasting VTx and VRx and it also does not happen when attaching the camera to the rebroadcasting Vtx directly.

 

I measured the output at the scope and what I found out that whatever Vtx I use for rebroadcasting, it seems to pull down the output video signal of the Uno2400 significantly. Have a look at the two screens. Here you find the output with the rebroadcasting Vtx attached:

 

 

IMG_1316.jpg

Edited by dmpriso

Share this post


Link to post
Share on other sites

And here without the Vtx attached.

 

image1 (1).jpg

 

BTW my latest test setup was without any OSD etc. involved. Just a IRC 2400 VTx with 700mw, the IRC Uno2400 "NexWave" version, and a TBS Unify Pro for rebroadcasting. Same-same with a Foxeer 25mw Tx for rebroadcasting. Had a TBS Groundstation running for reference and picture was clear there all the time without flickering.

 

Another information: Using the TBS Groundstation rebroadcasting works perfectly, and the rebroadcasting Vtx also doesn't cause that amplitude drop on the groundstation.

Edited by dmpriso

Share this post


Link to post
Share on other sites

1. This link will give basic information on how to measure the video:
https://www.rc-cam.com/forum/index.php?/topic/2825-using-the-50-digital-scope-to-measure-video-levels/

2. If you disconnect the rebroadcasting vTx from the Uno the Uno's output will increase 2X. With the rebroadcasting vTx removed you must install a 75 ohm resistor on the Uno vRx's video signal to properly terminate it. Then you can measure the video level of the Uno.

3. The video example looks like multipathing interference to me.  That will happen a LOT inside a building. Take everything outside, away from all structures, and then test it. If the video is more stable then the problem indoors involved multipathing.

Share this post


Link to post
Share on other sites

Thanks for the link and the video. I will repeat the measurements outside when temps drop a bit; currently we do have 35 deg. C

However I know that the problem also arises outside during flight.

Edited by dmpriso

Share this post


Link to post
Share on other sites

I did some measurements (in the basement still, cause too hot outside).
It seems to be a combination of several factors.

The camera outputs about 1.3vpp with a sync pulse of about 300mv.

The Uno2400 Rx outputs a signal of about 800-900mv with a sync pulse just above 170mv. When fed to the goggles via cable, this is still enough to get a stable picture inside my DomV2 goggles.

Interestingly, the TBS groundstation outputs a video signal of about 4vpp! No wonder that if I feed that into the rebroadcasting Vtx I don't have any sync issues .. however one can see the signal saturation on the receiving side.

With the rebroadcasting set up and the NexWave receiver in my goggles I measured the video signal in my goggles. With the NexWave 2.4 Module without rebroadcasting, the signal look similar to what the Uno2400 puts out. However, with rebroadcasting and the 5.8 Nexwave module, the vpp is down to about 600mv, with a sync pulse of less than 160mv. This is where I start getting occasional rolling or black/white image.

 

Edited by dmpriso

Share this post


Link to post
Share on other sites

That's a lot of things that are contributing to your problem. Too bad our FPV gear makers aren't more careful with the video signal levels.

The symptoms include indications that you have a severe video emphasis compatibility issue or other modulation problem. That often occurs when mixing vTx and vRx brands. Since the goggles have a Nexwave vRx module (an ImmersionRC development), there's a chance that a ImmersionRC Nexwave branded vTx would improve your rebroadcasting link's sync and colorburst levels.

 

Quote

However, with rebroadcasting and the 5.8 Nexwave module, the vpp is down to about 600mv, with a sync pulse of less than 160mv. This is where I start getting occasional rolling or black/white image.

First on the list is to restore the overall (full-white) signal level to 1Vpp. Maybe a different resistor in the UNO will provide a cheap workaround. Otherwise,  a video amp with a gain of about 1.6X is needed on the UNO's video out. Still a bandaide, but might get you back on your feet.

- Thomas

Share this post


Link to post
Share on other sites

Swapped out the ImmersionRC Uno V3 for a ImmersionRC Duo V3 and everything is perfect. The video signal looks much better, the sync pulse is about 300mv and the problem is gone :)

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...