electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Articles  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links  |  Store
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
 Forum index » DIY Hardware and Software » Video synthesis and manipulation
CMOS Video Synth - easier using external sync generator?
Post new topic   Reply to topic
Page 1 of 1 [11 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Author Message
Top Top



Joined: Feb 02, 2010
Posts: 198
Location: California

PostPosted: Wed Feb 19, 2014 5:27 pm    Post subject: CMOS Video Synth - easier using external sync generator? Reply with quote  Mark this post and the followings unread

I've been grinding some ideas about making basic video synthesizer using CMOS, like maybe a basic black & white pattern generator to start.

I have a video mixer that outputs a sync signal and "reference" signal. I am wondering if I could simplify things a lot for myself by just allowing the DIY Video Synthesizer to be synced externally to my video mixer.

I am aware of how each line of composite video is structured (like this:)
Posted Image, might have been reduced in size. Click Image to view fullscreen.

But does anyone have any idea of what the video sync signal (as used by video mixers, studio cameras, analog editing equipment, etc), in isolation, actually consists of?

If the sync signal is simply a periodic 0VDC pulse (this is what it is within the composite signal), I am thinking that one or more oscillators could be split to an analog switch (such as 4016 or 4066) that is set to automatically send the output to ground when triggered by the sync signal.

Like this:

Posted Image, might have been reduced in size. Click Image to view fullscreen.
(note: there maybe would have to be an inverter between the sync signal and the 4016 control pin input, so that the sync signal was converted to turn the 4016 "on" when the sync reset portion, 0VDC, is present)

That way, different divisions of the input oscillators vs sync signal would generate different patterns on screen -- boxes, lines, diagonal lines, etc -- depending on how the frequency of the oscillator "beats" against the sync signal that resets each line.


Below are some NTSC video signal facts:

Composite Video Signal Luma Voltages:
• 0V DC = sync level
• .339V DC = black level
• 1V DC = 100% white level

Parameters of interest (source: http://www.ni.com/white-paper/4750/en/)
• # of lines/frame: 525 (this includes 485 lines for display; the rest are VSYNC lines for each of the two fields)
• line frequency: 15.734 kHz
• line duration: 63.556 µs
• active horizontal duration: 52.66 µs
• # active pixels/line: 640

Posted Image, might have been reduced in size. Click Image to view fullscreen.

_________________
A. MAGIC PULSEWAVE ELECTRONICS
Back to top
View user's profile Send private message
commathe



Joined: Jul 26, 2013
Posts: 108
Location: Beijing
Audio files: 5

PostPosted: Wed Feb 19, 2014 6:58 pm    Post subject: Reply with quote  Mark this post and the followings unread

I am not sure about the sync signal you are using, but with VGA you can apparently "steal" the sync signal from some source and then just feed 0-0.7V signals freely into the R G and B channels. It seems to be a sample and hold type deal with VGA
Back to top
View user's profile Send private message
Top Top



Joined: Feb 02, 2010
Posts: 198
Location: California

PostPosted: Thu Feb 20, 2014 9:53 pm    Post subject: Reply with quote  Mark this post and the followings unread

I already do a lot of experimental live analog video work. All of my equipment works with composite, so working with VGA isn't useful to me.

Also, I have a sync generator, I just don't actually know what the signal it spits out consists of. That's the main thing to find out for me, and then I think I could design a simple monochrome synth from there.

_________________
A. MAGIC PULSEWAVE ELECTRONICS
Back to top
View user's profile Send private message
commathe



Joined: Jul 26, 2013
Posts: 108
Location: Beijing
Audio files: 5

PostPosted: Thu Feb 20, 2014 11:50 pm    Post subject: Reply with quote  Mark this post and the followings unread

What does the sync generator accept as input? I don't really understand how it's working. Composite video sync is created by the signal dropping to 0V for 5 microseconds every 64.5 microseconds. I always kind of envision it like a "gating" or "tremolo" effect if that makes any sense. You can't really "mix" the sync signal in as it's sort of a "lack" of signal.

Have you tried the muffwiggler forums? They have a much more active video synthesis community. Some of the LZX industries team frequent it, they might be able to help
Back to top
View user's profile Send private message
Top Top



Joined: Feb 02, 2010
Posts: 198
Location: California

PostPosted: Fri Feb 21, 2014 12:41 pm    Post subject: Reply with quote  Mark this post and the followings unread

I have a video mixer and it outputs sync. I believe it generates the sync based on composite video coming in one of the channels (it must strip out the video signal and re-generate a new purely sync signal?)

From what I understand, there is a reference level, just below the "black" level of .3v. I don't really know what a purely "sync" signal would be, but my guess is it would sort of be like a composite signal with just black and sync pulses (the drops to 0v), and then the frame vertical reset part, which is a long 0V period with a .3V pulse.

Posted Image, might have been reduced in size. Click Image to view fullscreen.

That's actually what my question is, I don't know what a sync jack actually outputs, so I don't know if it is useful to attempting to make a video synth.

If it IS just the signal devoid of any picture info (just black levels and 0V sync pulses), then I could use an inverter logic gate to convert the 0V pulses to positive trigger pulses, which could be used to trigger the oscillators to "dip out" of the signal at the appropriate times (by dropping them to ground).

Thanks for the suggestion about muff wiggler. I will check it out.

_________________
A. MAGIC PULSEWAVE ELECTRONICS
Back to top
View user's profile Send private message
synchroton



Joined: Mar 02, 2010
Posts: 20
Location: Milwaukee, WI USA

PostPosted: Thu Mar 06, 2014 12:39 am    Post subject: Reply with quote  Mark this post and the followings unread

Yeah, you're on the right track. Composite sync or blackburst will have these reference sync pulses for both horizontal and vertical sync. Additionally the color subcarrier is included on color video equipment.

Something like the LM1881 sync stripper is a very handy tool to be able to translate these small sync pulses into usable logic levels.

Often a CMOS switch e.g. 4053 is used to omit signals during the blanking interval.

Alternately you could gate your oscillators and shut them off during that time.

The schematic you have drawn suffers from two issues - the sync signal doesn't have an appropriate voltage for the analog switch, and both your oscillator and reference will be pulled to ground when the switch is active. You could use 2 sections of the 4016 which are toggled with inverse logic signals, or you could use an analog mux (like the 4053)

Come over to the muffwiggler video synth forums! There are not many of us doing video synth DIY but that is probably the best place to find conversation like this!
Back to top
View user's profile Send private message
Top Top



Joined: Feb 02, 2010
Posts: 198
Location: California

PostPosted: Thu Mar 06, 2014 8:17 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hey man,
Thanks, I came to some of those conclusions myself, although it does seem that the LM1881 is able to trigger cmos inputs.

I have something working now. Some of what is working, I am a little uncertain why it is working, but it's working. I find that the gating of the signal to ground by the 4016 stabilizes the image, even though I understand theoretically what you are saying.

What I ended up is a pattern generator with two oscillators, each with pulsewidth and frequency control, and a divider with variable resets that can be clocked by either oscillator.

My main goal was to have something I could use with the downstream key on my video mixer to generate shapes alongside my regular setup. Here's a short clip of the pattern generator in combination with the video mixer and a feedback loop.




_________________
A. MAGIC PULSEWAVE ELECTRONICS
Back to top
View user's profile Send private message
synchroton



Joined: Mar 02, 2010
Posts: 20
Location: Milwaukee, WI USA

PostPosted: Fri Mar 07, 2014 4:59 am    Post subject: Reply with quote  Mark this post and the followings unread

excellent! Glad to see you made some progress!

When you are tying the oscillator to GND you're effectively shutting it down (I'm assuming your oscillator isn't buffered). So it starts oscillating at the same point each line or frame, because of the sync pulse. That's how you're accomplishing the stabilization. Looking at your circuit again I'm seeing that you actually should be able to accomplish what you were looking to do, pulling BOTH the black reference and oscillator to GND during the sync interval. Though I'm not sure how you're actually mixing those signals together to achieve the appropriate bias & signal amplitude.

You're probably noticing that certain frequencies will appear hazy/flickery and some appear solid - this is due to the interlaced nature of the video signal, and only signals which are identical during the even and odd fields will appear "solid" and non-interlaced, signals which are nonidentical during even/odd can still be in sync, but across the entire frame rather than between fields - so each field is 'out of sync' with the other so to speak. Maybe that makes sense.

Also, the DSK input on your video mixer may behave differently than other inputs since it's generating an overlay (what video mixer are you using, by the way?) In older video mixers I've found that they can be very fussy about sync - you can get weird chroma shifting and erratic behavior. I'm not sure if the color shifting is because of the inclusion of your circuit or because of the feedback...

anyway,
Keep up the good work!
Back to top
View user's profile Send private message
synchroton



Joined: Mar 02, 2010
Posts: 20
Location: Milwaukee, WI USA

PostPosted: Fri Mar 07, 2014 5:04 am    Post subject: Reply with quote  Mark this post and the followings unread

By the way, you could also try driving the counters off of the sync pulses to do line divisions. That might be useful. However those will be on during the sync interval sometimes, so rather than mixing directly you might use AND gates and combine them with your oscillator outputs... just a thought. That all depends on your ability to pull those CMOS outputs to ground - and I guess would be related to the resistor you're using as well as the on resistance of the 4016. Looks to be ~300 ohms from a random datasheet.
Back to top
View user's profile Send private message
brock



Joined: May 26, 2011
Posts: 96
Location: Canada

PostPosted: Tue Mar 11, 2014 10:41 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hi Tip Top,

Just checking out what you have been doing, looks good.

I don't know why synchroton thought the LM1881 won't drive the 4016, the levels match up fine as you noted. I think the two of you were talking about different things regarding the stabilization of the signal. He was talking about your pattern generators free running without the sync signal and I suspect you were talking about the video signal itself not being stable when viewed, likely rolling or tearing, something more like a glitch system than a pattern generator. You are essentially clamping the sync tips to ground with your setup which is probably necessary to restore them given the low bias voltage you have.

I have a question, are you generating blanking signals or does your pattern generator stuff affect the entire non sync signal?
Back to top
View user's profile Send private message
synchroton



Joined: Mar 02, 2010
Posts: 20
Location: Milwaukee, WI USA

PostPosted: Wed Mar 12, 2014 8:25 am    Post subject: Reply with quote  Mark this post and the followings unread

Sorry, I was a bit confused as to what was being suggested. I thought you were suggesting running the 'sync' output of your video mixer directly into a CMOS inverter. That was the logic level I found problematic, and why I suggested an LM1881. A CMOS logic 'high' input for a device running off a 12v supply (another couple of assumptions) should be 8v.

I'm still a bit unsure about what you're trying to accomplish - are the HF oscillators being reset or free-running?
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic
Page 1 of 1 [11 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
 Forum index » DIY Hardware and Software » Video synthesis and manipulation
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
e-m mkii

Please support our site. If you click through and buy from
our affiliate partners, we earn a small commission.


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use