"Jasen Betts" bravely wrote to "MIKE ROSS" (15 Mar 04  07:40:20)
 --- on the heady topic of "{at}%^{at}#$%^ veroboard"

 JB> Hi MIKE.

 JB> 14-Mar-04 09:56:40, MIKE ROSS wrote to Jasen Betts

 JB>> the final circuit was something like this.
 MR>  JB>
 JB>>         ___________________________________________________________ +12
 JB>>         |          |      |       |                |           |
 JB>>        ???        10k    _L_     _L_100n   ,-[1M]--)----.      |
 JB>>         |          |     /^\     ~T~       |  R7   |    |  .---|---.
 JB>>         |     _____|  || ~T~ . .  |        |     |/     |  \  _L_  |
 JB>>         |  |/'     |--||--+--|<|--+-[100K]-+-----|~ Q3  `-|/  /^\  RL
 JB>>  ,-[1M]-|--|      /   ||     ` `     R6          |\        \  ~T~  |
 JB>>  |      |  |\| |/'    1n                           |        `--|---'
 JB>> _L_     |   ~`-|  Q1,2                             |          /'
 JB>> ~T~     |      |\|                                 |       |/'
 JB>>  |10p   |       ~`\                                |-[10K]-|   Q4
 JB>>  |      |          |~~~~|                         10K      |\|
 JB>> _|     2M2        3K1  _L_1u                       |        ~`\,
 JB>> video   |          |   ~T~                         |           |
 JB>> input   |          |    |                          |           |
 JB>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0V
 MR>  JB>

 JB> Thanks.

Welcome!


 JB>  relays have
 JB> inherent in themselves a measure of mechanical positive feedback.

Yes a "holding current" rating. It's due to a mechanical inertia that
the relay motor works against to both turn on and then to turn off.
That's not exactly a "feedback" but the electronic equivalent would
indeed require a positive feedback to behave in that way.


 MR> might the addition of R7 have possibly made the circuit a little
 MR> susceptible to oscillation ?

 JB> after installing the circuit and the relays together that's exactly
 JB> what I found happened,

 JB> another move to stop that could have been to put a shield between the
 JB> relay and the electronics. or to re-position the components closer
 JB> together.

I don't think that is necessary as long as you set the pot to a
position that gives reliable triggering and not at a critical point.


 MR> The original task really boils down to 2 requirements: 1-triggering,
 MR> 2-timing.

 JB> timing?

Well, this is just my opinion but I would have solved the problem of
detecting the video signal using the presence of the sync signal
rather then the whole waveform by using a DC level restoration as an
on/off trigger for a timing circuit. I think alternatively your
circuit depends on detecting 2 voltage levels for the on/off signal.
The timing would simply be added to ensure a little turn-off delay to
prevent relay chatter (which is what you had to do using R7). In this
way the sync itself would be the logic signal that turns on the relay.
There are many ways to get to Rome... ;-)

 Mike
 ****

... Inductor designers do it in the gap.
--- Blue Wave/DOS v2.30