20 replies to this topic

[Mr.RC-Cam]

I've mentioned in the past that the video levels on some wireless systems are often poorly calibrated. Even systems from our favorite makers. Although a typical monitor or recorder can mask the problem, we just cannot achieve the best video performance if the levels are incorrect. So if you want the best dynamic range and colors from your video, don't rely on what was shipped to you. Get out the o-scope and check it!

I published some tips about video level adjustments a few years ago, as seen here: Lawmate Video Tricks and Tips.

The problem with this simple task is that not many hobbyists have access to an o-scope. The PC sound card based solutions just don't have the performance to do the job either. But, the $50 digital storage scope produced by Jyetech does a good job when it comes to measuring video levels. This new test tool is now distributed by dpcav.com. That's where I work and is how I got my hands on one.

Truth be told, I was not expecting this scope to work well with video signals. Although it does not have TV-Triggering features, I discovered that it's trigger functions do just fine. Honestly, I was very excited about that. Below is a photo of the scope I am talking about.

To get you on the right path, continue reading to learn some tips.

[Mr.RC-Cam]

Let's start off with looking at the age old graybar pattern from a common video test generator. This is what it looks like to our eyeballs.


Now let's see what the Jaytech scope show us. In the setup I have used alligator leads to parallel a 75 ohm resistor (the required terminator for video) at the input of the scope. The scope is setup for DC coupling, 0.2V scaling, and 10uS timebase. The scope's three slide switches are as follows:
SW2 = X2
SW3 = 0.1V
SW1 = DC
The trigger level has been set to midpoint of the horizontal sync. Trigger mode is Norm and polarity is Rising Edge. I used the H.Pos feature to horizontally move the waveform over so that one full video line was seen. The V.Pos feature was used to move it vertically so that the bottom of the sync was just touching the bottom of the screen.

Here is what the annotations on the photo are telling us:
S = Sync region. Trigger is set to mid point of this.
BLK = This is the Black region of the graybar.
WHT = This is the White region of the graybar.
63uS = Sixty-three microseconds is one horizontal video line, sync to sync.


Now let's measure the video levels. Keep in mind that each vertical graticule is 200mV. Here is what the annotations are telling us:
V = Video level, 1.1V at full white area (spec = 1.0Vpkpk).
S = Sync level, 320mV (NTSC Spec = 285mV, PAL Spec = 300mV).


In this case the video level is slightly high. A bit too high is usually much better than a bit too low. But, tweaking it for 1.0V is a fine thing to do to. By the way, you won't be able to individually set the sync level. But, if you see a sync level that is grossly wrong then that may cause problems in some applications. So, I always check it for sanity and if it is bad I do my best to find out why and solve it.

[Mr.RC-Cam]

Ok, so let's take what we learned so far, but adapt it so that you can use your existing video camera as the video test signal. This is a bit tricky, but with patience and practice it works out fine.

The main problem with using your video camera for the test signal is that cameras have an AGC (auto-gain-control) feature. The AGC prevents bright video from saturating the image. However, a saturated white image is exactly what we need. That is, we want to use a bright light source to create a full white image. However, the camera is going to fight us on this.

There are a couple solutions. If you have one of those fancy OSD cameras, then you can turn off the AGC. But, most of us aren't so lucky. As a workaround, I found that I can trick most cameras. What I do is slightly turn down the overhead room lights and use a flashlight (torch to you English blokes). I start with the flashlight pointed away from the camera. When I'm ready, I quickly aim it directly at the camera lens. For a short moment I can see the top part of the video go to its maximum level and sort of flat line there. That is the saturated video level, which represents full white. Within a moment the AGC will catch up and reduce the level, so you have to look fast. Using this cat and mouse game I can easily identify the peak white level.

The video level we need to measure is from the bottom of the sync to the top of the saturated white. Because this little 0-scope is a digital storage scope, you can press the scope's Hold button to freeze a waveform of interest. It takes practice (and luck) to catch the one you want, but is not impossible to do. Below is an example of a saturated waveform that I was able to catch with the digital storage Hold feature. This captured video level shows that mine is a tiny bit too high. But that's the way I like it.

Success with using this method will depend on your camera. Besides the issues with the AGC, it requires that your camera adhere to the video standards. If you have an R/C telemetry OSD then you must disable it while you measure/adjust the video. Do not allow the OSD telemetry text to appear during your measurements.

[W3FJW-Ron]

Nice. Great little tutorial. Please keep 'em up as you have time either here or on your website. This old TV repairmen & fledgling FPVer appreciates you reminding me about the things I've forgotten since Tesla was born.

My scope is in the mail. Should have it in two or three days.

Might I make a suggestion?

Another forum dedicated to tips like this with a topic for each tip/tutorial. No general chatter allowed. It would, I think, make it easier to find specific information without having to read through many forums or threads. RCG has a lot of useful info but it's a pain digging through each loooong thread trying to find the meat. Sort of like trying to find the chicken in a bowl of Campbells chicken noodle soup.

[Mr.RC-Cam]

Please keep 'em up as you have time either here or on your website.

Normally I would put this sort of How-to info on the main web site. But someone sent me a PM today asking if the scope could be used to adjust video. So, to quickly share the info I posted it here. When I have time (wishful thinking), I plan to create a fancy page on the rc-cam projects page.

... has a lot of useful info but it's a pain digging through each loooong thread trying to find the meat. Sort of like trying to find the chicken in a bowl of Campbells chicken noodle soup.

Creating more sub categories for a How-to is not really needed. Keep in mind that the rc-cam forum has a lot more chunks of quality meat, so the good stuff is easy to find.

[W3FJW-Ron]

Yes. I agree about this forum. I was referring to the RCG forums.

[jaywalker]

Nice. Great little tutorial. Please keep 'em up as you have time either here or on your website. This old TV repairmen & fledgling FPVer appreciates you reminding me about the things I've forgotten since Tesla was born.

My scope is in the mail. Should have it in two or three days.

Might I make a suggestion?

Another forum dedicated to tips like this with a topic for each tip/tutorial. No general chatter allowed. It would, I think, make it easier to find specific information without having to read through many forums or threads. RCG has a lot of useful info but it's a pain digging through each loooong thread trying to find the meat. Sort of like trying to find the chicken in a bowl of Campbells chicken noodle soup.

[jaywalker]

Ron , I am also new to FPV using Electrostik to carry camera , goggles and recorder. Also an old TV Repairman, N5KBL is my call sign. I was interested in the small scope you mentioned. I am using OSD with a GPS attachement to give location, altitude, speed . I mounted a Hi-gain omni ant on a tripod on top of my RC plane trailer. Using the 2.4 Futaba . If you don't mind let me have some info on scope. I worked TV repair in Oregon and Texas. I could see the throw away stuff coming in and got out in time. I still miss working on the 6BK4 shunt regulators, heh,heh.

[Terry]

That little scope is super (awsome as you American guys would say)

Terry

[W3FJW-Ron]

Hi Jaywalker. The link Mr RC posted at the top has all the details. http://www.dpcav.com...productid=16330

I bought one from them and still have it setting here. Bought a plastic 8.5 x 6 x 2.5 box to build it into (scope, pwr supply& test lead storage etc.). Need to build a ckt board for my two inputs, termination, & switches (75 & 50 ohms). Still trying to decide on input jax, BNC, F59, and a few other things. Don't have room yet for a test/work bench in the RV so haven't finished packaging it.
Great little unit judging from my preliminary look-see.

[tascheri]

This is what I found in my e-bay sony 1/3 ccd camera. Looks like it´s more like 2.2 Vpp. What should I do?
1 div = .5V / 10 uS

Edited by tascheri, 29 April 2011 - 04:34 AM.

[Terry]

is it causing you a problem?

if so then adjust it if it has a pot you can adjust or make an attenuator with a pair of resistors.


Terry

[Mr.RC-Cam]

A 2.2V measurement is a sure sign your test setup is missing the required 75 ohm terminator on the video output of the Rx. So install the terminator on the Rx and use your scope to measure across the terminator.

[tascheri]

A 2.2V measurement is a sure sign your test setup is missing the required 75 ohm terminator on the video output of the Rx. So install the terminator on the Rx and use your scope to measure across the terminator.

This was measured with the camera hooked up to the Airwave 633 TX module and the measurement is the video signal of the camera. I believe the TX will supply adequate load to the camera´s output, or not? Funny thing is the camera works fine with the matching ebay TX/RX module but when I hooked it up to the Airwave module, full white screen when pointing the camera outside or to a bright light.

[Mr.RC-Cam]

Airwave Tx modules usually provide an approximate load of 50-80 ohms. You can use your ohmmeter to see what your module is providing.

Regardless, if you are getting 2.2V at the input of the Airwave module then either a mistake was made with the measurement or something is wrong with camera.

Most importantly, the video level out the Rx must be checked too. So you need to measure it and confirm you have 1Vpp with a 75 ohm load on the output of the Rx.

when I hooked it up to the Airwave module, full white screen when pointing the camera outside or to a bright light.

If your signal is really 2.2Vpp then this amplitude is much too high and over saturated video will occur.

[tascheri]

Can I hook up a resistor voltage divider? or would it be better to use a 1/2 gain Op amp?

[Mr.RC-Cam]

A lower value termination on the camera's video signal will reduce the amplitude going into the Airwave module. Just use a resistor value there that provides 1.0Vpp at the output of the Rx (when the Rx is terminated with exactly 75 ohms).

But before you do this I advise that you try another camera to see if the problem goes away. If it does not go away then there is something suspicious about your measurement or your circuit.

[tascheri]

Coming from me I would bet on the latter... but then, this camera hooked up to the airwave system gave me a VERY saturated video, like I could only see white when pointing the camera towards the window... I did check on the scopes calibration signal to validate my measurements ( I always do )...

Thank you very much sir and have an excellent Sunday. Tomorrow only flying, no lab stuff.

[Nandox7]

Appologies for digging out a old thread, but the information shared here is great.
I spent some time reading it and another related to the tuners in the video RX's.

Question, would it be ok to use a generated video signal? Example use a simple arduino
to generate the required full white video signal?
Couldn't that produce better results as we know the incoming signal is the one expected?

Thanks.

Edited by Nandox7, 28 April 2012 - 12:44 AM.

[Mr.RC-Cam]

Using a calibrated video source is ideal and it will provide the most accurate results. I use a professional video generator, that has a variety of test patterns, and it is awesome for calibrating wireless video systems.

If you build your own video generator then pay very close attention to the NTSC (or PAL) video standards. The signal levels are very well defined and whatever you create needs to conform to these standards. Just keep in mind that many DiY microcontroller based video designs do not provide industry accurate video signals. BTW, to keep it simple you can follow RS-170 standards (set to 1Vpp), which is the equivalent to NTSC without the color (monochrome video).

If you don't have time to build a video generator then eBay sometimes has some good deals on used color bar generators. Or, just use the camera method that has been discussed here.