Negative Video

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Negative Video

Postby Andrew Davie » Fri May 11, 2007 9:30 pm

I'd decided to try using a laptop as a supply of the NBTV video signal, instead of burning to CD and then via a CD player. Only the laptop I have is sending "negative video". I have one CD player that does the same.

I'm hoping someone might be able to contribute a circuit design that will convert "negative video" to "positive video" -- I tried switching the leads, but this did not work (I wonder... why?).
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1511
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Neg. Vid.

Postby Steve Anderson » Fri May 11, 2007 9:45 pm

I did in one of the threads post a diagram of a simple 'optional inverter', like yourself I have found that it's a 50/50 chance that domestic gear will give you correct polarity video. Throw in the fact that the computer nerds simply don't care...

If the 'inverting' position is feeding a low impeadance input I'll update the diagram with a simple emitter-follower.

Steve A.
Attachments
inverter_147.gif
inverter_147.gif (5.54 KiB) Viewed 10809 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Dave Moll » Sat May 12, 2007 3:57 am

With my (battery operated and thus electrically isolated) MUTR televisor I have taken the low-tech approach of simply feeding the inner terminal from line out to the outer terminal of the televisor input and vice versa. It seems to work - but does, of course, prevent me from using the audio feed-through.

The use of an inverter circuit is, of course, much more elegant - so I may build one to the above design in due course.
User avatar
Dave Moll
Just nod and pretend you understand me
 
Posts: 377
Joined: Tue Feb 27, 2007 9:11 am

Inverter circuit version 2.0

Postby Steve Anderson » Sat May 12, 2007 1:47 pm

Here's a modification of the previous inverter circuit where the load is of a low(ish) impedance. The previous circuit is OK with loads greater than 10k, but there will be a bit of attenuation in the 'inverted' position with loads of around 10k. (Around 30%).

Adding the second transistor provides a lower output impedance restoring the circuit to (almost) unity gain but with inversion. This will drive loads as low as 1k but C2 will need to be increased in value to retain the low frequency -3db point at around 2Hz. With a 1k load a value of 100uF will be needed.

As before Vcc can be anything from 12 to 24V, it doesn't need to be regulated, but it should be clean and ripple free, otherwise some of the noise/ripple/spikes will be added to the signal. At the higher volages of above 18V R5 should be a 1W resistor as it is dissipating just over half a Watt at 24V. TR2 will run warm, but it's within its ratings.

The circuit consumes about 36mA at 24V, half that at 12V. The transistors can almost be any general-purpose NPN device, 2N2222, BC107/8/9. I hope this is useful. I removed the switch just to make the circuit simpler.

Steve A.
Attachments
Inverter 2.gif
Inverter 2.gif (5.29 KiB) Viewed 10786 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Inverter circuit version 2.0

Postby Andrew Davie » Thu May 17, 2007 10:48 pm

Steve Anderson wrote:Here's a modification of the previous inverter circuit where the load is of a low(ish) impedance.


Thank you for this. I am about to start building this circuit, but first, where is the 'normal' and the 'inverted' output in the new circuit? Am I missing something?
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1511
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Postby AncientBrit » Thu May 17, 2007 11:38 pm

Andrew,

Steve is economising on switches!

S1 is of course still needed between TR1 collector/emitter and TR2 base.

Regards,

GL
AncientBrit
Green padded cells are quite homely.
 
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK

Switches.

Postby Steve Anderson » Fri May 18, 2007 2:43 pm

Ask any electronic circuit designer worth their salt and they'll say that it's best to avoid electromechanical components, pots, switches, connectors...anything with moving parts. Simply for the unreliable nature of these items compared to rest.

This is in the context of a mass-produced item, when cost, reliability and warrenty claims are paramount. This is in contrast to the auto industry where things, both mechanically and electricaly, seem to be getting more complex.

GL is quite right, if needed I can re-draw the circuit with switch.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Switches.

Postby Andrew Davie » Fri May 18, 2007 2:47 pm

Steve Anderson wrote:Ask any electronic circuit designer worth their salt and they'll say that it's best to avoid electromechanical components, pots, switches, connectors...anything with moving parts. Simply for the unreliable nature of these items compared to rest.

This is in the context of a mass-produced item, when cost, reliability and warrenty claims are paramount. This is in contrast to the auto industry where things, both mechanically and electricaly, seem to be getting more complex.

GL is quite right, if needed I can re-draw the circuit with switch.

Steve A.


Come now, let's have some focus here! This is a mechanical television we're talking about. The more moving parts the better. Better still if they're noisy!

As I'm not totally sure where the switch should be located, I may go with the first circuit, first... then move on to the second one when I'm more confdent.

Cheers
A
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1511
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Switched version.

Postby Steve Anderson » Fri May 18, 2007 2:59 pm

Andrew,

Here's the switched version....very hastily re-drawn!

'Inverted' is in the position of the switch as shown, 'normal' is in the other position.

Steve A.
Attachments
Inverter 3.gif
Inverter 3.gif (4.73 KiB) Viewed 10727 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby DrZarkov » Fri May 18, 2007 3:49 pm

BTW: Which software do you use for drawing the circuits? I was looking around, I found some programs between total crap or 3000 EUR-software. But no easy to use (or even usable) free- or shareware for drawing a simple circuits. It can be MacOS X or Windows.
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1037
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

A 'deluxe' version.

Postby Steve Anderson » Fri May 18, 2007 5:15 pm

Gents,

For those that not only need a choice of signal inversion or not, a modest amount of gain and a low-impedance (down to 1k) drive capability here's a circuit that will do the job.

OK, it uses four transistors, but you don't get something for nothing! Again it's fine with a supply of 12 to 24V and will consume the same as those above.

R4 & 7 set the gain, where VR1 allows you to 'turn it down'.

If both 1k, maximum gain is 2.
If both 470, maximum gain is 4.
If both 180, maximum gain is 10.

Supply needs to be clean as before (as they should be anyway). At 12V it will output a signal of 6V peak-to-peak, more at higher voltages. The bandwidth goes to well beyond 1MHz.

Transistors can be almost any general-purpose NPN devices, BC184, 2N2222, BC107/8/9. C2 will need to be up to 100uF if you're feeding a 1k load. The switch this time is drawn in the 'normal' position.

If you don't need the low outut impedance omit TR4 and R11, connect the positive of C2 to the common of the switch. But I don't recommend this for loads less than 20k or so.

..as for the drawing package I use, it's AutoCad from Autodesk. It costs US$3,500, but I have to use it for my work, it's the default in virtually all industries.

Steve A.
Attachments
Inverter 4.gif
Inverter 4.gif (5.39 KiB) Viewed 10717 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Inverter circuit version 2.0

Postby Andrew Davie » Sun Jun 03, 2007 11:09 pm

Steve Anderson wrote:Here's a modification of the previous inverter circuit where the load is of a low(ish) impedance. The previous circuit is OK with loads greater than 10k, but there will be a bit of attenuation in the 'inverted' position with loads of around 10k. (Around 30%).

Adding the second transistor provides a lower output impedance restoring the circuit to (almost) unity gain but with inversion. This will drive loads as low as 1k but C2 will need to be increased in value to retain the low frequency -3db point at around 2Hz. With a 1k load a value of 100uF will be needed.

As before Vcc can be anything from 12 to 24V, it doesn't need to be regulated, but it should be clean and ripple free, otherwise some of the noise/ripple/spikes will be added to the signal. At the higher volages of above 18V R5 should be a 1W resistor as it is dissipating just over half a Watt at 24V. TR2 will run warm, but it's within its ratings.

The circuit consumes about 36mA at 24V, half that at 12V. The transistors can almost be any general-purpose NPN device, 2N2222, BC107/8/9. I hope this is useful. I removed the switch just to make the circuit simpler.

Steve A.


Tonight I put this circuit together and gave it a trial run. No luck, but after a bit of checking pinouts on the transistor alternate, I realised I had the transistors in backwards -- fixed that, and voila!! Yes, this circuit works as advertised -- teriffic! I used BC549C as the transistors.

I'd been putting this task off for ages -- and it was necessary to allow me to use my 'you beaut' CD portable originally loaned to me. This one is the one I wanted to use because not only is it shiny (unlike the one I've been using, which is 'beat up'), it also handles MP3. I've very much wanted to do some testing with MP3 source, but haven't been able to because the shiny new CD player also has negative video.

So, the circuit built, now I could test it out.

Frankly, I'm appalled. There are several reasons, but firstly though the shiny new CD player has a "Line Out" I discovered it also has a volume dial associated with it, and there's no headphone out. In other words, it's not really a line out after all. So, I'll be using "headphone out" for my testing with this. Not a problem, the old beat up CD player had a headphone out which I had tested and which seemed to produce good pictures.

So, I switched the circuit over to "negative video", played a few sample clips (WAV) -- most particularly the test pattern #36 on CD1. This looked OK on first glance, but I noticed that whole chunks of the inner circle black surround are simply... missing. The signal coming from the CD is obviously so horrible that it's cutting out large and important bits of information from the picture. I don't think it is the "negative video-er" doing this.

Putting that to one side, now I could at least test MP3 tracks. I really am appalled. It's like going from perfect TV reception to something barely watchable with snow and interference all over it. It's terrible terrible, not just slightly degraded. MP3 video is most definitely out, at least when viewed using this "shiny new" (useless for NBTV) CD player. Ah well -- I wouldn't have been satisfied had I not confirmed this to my own satisfaction.

The conclusions (assuming the "negative video" corrector isnt' the fault)...

* MP3 video is bad, but it's terrible when played via a headphone out.
* there are CD players, and there are CD players. The same thing on similar players can produce a totally different result. On my original player, headphone out is fine. On the "new" one it's very far from fine.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1511
Joined: Wed Jan 24, 2007 4:42 pm
Location: Hobart, Tasmania, Australia

Postby Klaas Robers » Mon Jun 04, 2007 3:23 am

Andrew, use your oscilloscope, use your first CD-player, synchronise on the syncs of track 36, the Grant Dixon test chart, try to familiarize on the wave form that is displayed. You will recognise the different parts of the lines, although they are written on top of each other.

Then switch to your "new" CD-player and repeat the things. Look before and after the neagtive circuit and you can see whether the CD-player or the circuit does the bad job. May be that you turned up the volume too far and the line-out signal becomes too strong for the circuit. It is also possible that when you decrease the volume of the CD-player and increase the contrast setting of your monitor that thing are getting better.
User avatar
Klaas Robers
Frankenstein was my uncle.
 
Posts: 1453
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Negative video.

Postby Steve Anderson » Mon Jun 04, 2007 12:34 pm

Tonight I put this circuit together and gave it a trial run. No luck, but after a bit of checking pinouts on the transistor alternate, I realised I had the transistors in backwards -- fixed that, and voila!! Yes, this circuit works as advertised -- teriffic! I used BC549C as the transistors.


Good, pleased to hear you had success. The trannies you chose are fine. The transistors are OK too.

As for the 'appaling' performance of your 'you beaut' CD player....there is a limit as to how large a signal the inverter circuit can handle. Assuming that its the one I posted on the 18th May, at 12V it should handle up to 5Vp/p.

Check with your scope the the signal coming out of the player is less than this otherwise it will overload the inverter. This where chips win hands-down, a half-way decent op-amp will handle twice this amount. But I know your aversion to using them, that's why I have kept things discrete.

There have been items in the newsletter regarding using compressed audio for NBTV (Vol. 28 No. 2) done by Klaas is an example.

Good luck Cobber/Sport/Blue,

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Yet another....

Postby Steve Anderson » Mon Jun 04, 2007 5:44 pm

If you really need to handle signals in the region of 10Vp/p at 12V, here's a circuit that will do it. It's unity gain, but inverting and will drive 1k loads. It's not as 'elegant' as some of the others, but it does the job.

It's just an inverter, there is no 'normal' output, you either need it or you don't.

What I don't like is the introduction of yet another capacitor, I prefer to keep all my circuits DC-coupled, but with discrete components and a single supply that's not easy.

It's also mildly temperature sensitive, but from 0C to 50C it should be fine. We're not looking for military applications here.

Steve A.

What I should also mention is that it needs to be configured for the supply voltage you are using, as shown it's OK for a 12V supply. If your supply is 18V then R1 & R2 should be 100K, if 24V then they should be 150k. Pro-rata in-between, but it's not that critical. Higher voltages will also allow larger signals to be handled, almost pro-rata.

If you really need some gain then adjust R1, but I don't suggest going below 22K. With the 12V version you'll get a gain plus inversion of about 3 with R1 = 22k. the gain is set by (mainly) the ratio of R1 to R2, keep R1 above 22k, adjust to suit the supply voltages mentioned above.
Attachments
Inverter 5.gif
Inverter 5.gif (4.68 KiB) Viewed 10631 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 3843
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Next

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 21 guests