Hi all,

As part of my Elec. Eng course I have to design a circuit that will take an analogue tv signal (PAL) and display it on an oscilloscope. This isn't too hard considering we have some help, and we are allowed a LM1881 sync separator circuit that does all the nasty stuff for us. I already have a working design, although it's rather decadent in its use of components.

However, while the whole thing can be designed using op-amps provided, I would like to try and make mine a little more unique, if possible. We are given a components list to use which is mainly op-amps, BJTs, voltage regs, diodes and zerner diodes (and resistors and cap and all that). I know for a lot of the amplification parts it is possible though hard to swap out an op-amp for a BJT amplifier design of some sort, but I was also wondering if it was possible to replace the sync separator (or part of it) with something of my own (or someone else's) design.

I have seen a few projects doing this sort of thing on the Internet (with triodes, don't know what those are), but nothing so far that addresses what I am looking for. The reason I want to do this (apart from that it increases your marks) is that there is a class prize that I would like to win to make my parents proud.

Anyway, if anyone has any ideas or can point me in the direction of ideas that would be greatly appreciated. I know you can set up BJTs to act as logic gates and such, so I feel it is at least possible, even if it is implausible. As part of the lab you can submit an application for new components for use in your design so it's a possibility that I can use others too.

Jo

 

Hi all,

As part of my Elec. Eng course I have to design a circuit that will take an analogue tv signal (PAL) and display it on an oscilloscope. This isn't too hard considering we have some help, and we are allowed a LM1881 sync separator circuit that does all the nasty stuff for us. I already have a working design, although it's rather decadent in its use of components.

However, while the whole thing can be designed using op-amps provided, I would like to try and make mine a little more unique, if possible. We are given a components list to use which is mainly op-amps, BJTs, voltage regs, diodes and zerner diodes (and resistors and cap and all that). I know for a lot of the amplification parts it is possible though hard to swap out an op-amp for a BJT amplifier design of some sort, but I was also wondering if it was possible to replace the sync separator (or part of it) with something of my own (or someone else's) design.

I have seen a few projects doing this sort of thing on the Internet (with triodes, don't know what those are), but nothing so far that addresses what I am looking for. The reason I want to do this (apart from that it increases your marks) is that there is a class prize that I would like to win to make my parents proud.

Anyway, if anyone has any ideas or can point me in the direction of ideas that would be greatly appreciated. I know you can set up BJTs to act as logic gates and such, so I feel it is at least possible, even if it is implausible. As part of the lab you can submit an application for new components for use in your design so it's a possibility that I can use others too.

Jo

The most impressive final project in your group/class will be the one that works the best on the fewest/cheapest components. I bet you'd get bonus points, or at the very least, an " 'atta boy!" from a prof.

 

^^

This is true, there is an undefined tradeoff of picture quality vs. number of components vs. cost of components. He said he would only take marks off if you had way too many, and would give marks for minimalism or unique designs. I plan to quiz him more thoroughly on this one

 

^^

This is true, there is an undefined tradeoff of picture quality vs. number of components vs. cost of components. He said he would only take marks off if you had way too many, and would give marks for minimalism or unique designs. I plan to quiz him more thoroughly on this one

So, the name of the game is to minimize the product of these values:
crappy_picture_score x cost/number_score x crappy_quality_score

There is an optimum value! It will be interesting to see what happens.

Did they give you a microcontroller in that little kit?

 

Well I think that is only the general idea; if I am using 50 BJTs to make a really clever circuit I am sure I will come out on top of someone using 4 op-amps in the normal design they give you instead.

So that is what I am thinking of aiming for. And no there are no microcontrollers or anything like that.

 

You can use the circuit at the URL below as a very simple and robust sync stripper. It is for composite video with negative going sync. To make it work for inverted video, just change the transistor from PNP to NPN, invert the power supply and swap the capacitor to opposite polarity. That gets you composite sync on the output.

To you can use the leading edge of the sync pulses to generate horizontal triggers, and you can get vertical sync by either filtering and clipping, or better yet, trigger a one-shot from the leasing edge of the sync pulse, the one-shot being about 1/2 the width of the broad pulses (during serrations), and compare that with the width of the sync pulses with that of the one-shot. The sync pulse width will exceed the width of the one-shot only during vertical sync.

The stripper circuit:
http://www.freeimagehosting.net/hkzzk

 

...trigger a one-shot from the leasing edge of the sync pulse, the one-shot being about 1/2 the width of the broad pulses (during serrations), and compare that with the width of the sync pulses with that of the one-shot. The sync pulse width will exceed the width of the one-shot only during vertical sync.

The stripper circuit:
http://www.freeimagehosting.net/hkzzk

This seems very helpful (although I won't be sure until I do some IRL or software simulations). This means that it would be possible to completely bypass the LM1881 then? I assume sync stripper means what I think it does.

I am not too sure how I am going to do the stuff I quoted though... The horizontal sounds straightforward (now), but filtering out the vertical sync sounds harder, and I am not really sure about anything after 'one-shot' at all. I will need to re-read when it's not approaching midnight I suppose.
Also getting the odd/even interlace right sounds like a mountain of a task :/
Luckily I have some time to experiment I suppose.