1. Home
  2. Electronics Design
  3. Re: SED Question: Looking for two-input comparator with hysteresis - SED2InputComparatorQuestion.pdf

Re: SED Question: Looking for two-input comparator with hysteresis - SED2InputComparatorQuestion.pdf

>

> >>John Fields wrote: >> >>>>>> >>> >>> >>> >>>>I want to build a circuit which produces 12 V pulses once per minute, of >>>>alternating polarity, in order to drive a large WWII German slave clock >>>>which I want to hang on my wall. >>> >>> >>>--- >>>Is this what you want? (View in Courier) >>> >>>+12V _ _ >>> | | || | >>> | | | | | >>>0V___| |____________ ____________| |_______ _ _ _ >>> | | >>> | | | >>>-12V ||_| >> >>Yes. 1/2 second long pulses into a 2785 ohm coil. I know there are >>various ways of accomplishing this with a quartz oscillator, but I've >>become interested in driving the system from the 50 Hz mains frequency, >>so I want to do it that way now. > > > --- > OK. > > I'll post a schematic for you, here, sometime tomorrow.

Thanks John. Don't go to a lot of trouble, though. It's only a hobby project.

Having looked at optocouplers, I'm wondering if they are the way to go. I even found an optocoupler which includes a Schmitt trigger, and the data sheet suggests that it can be used for "A.C. to TTL conversion - square wave shaping":

formatting link

The only problem is that the maximum reverse voltage which the optocoupler's LED can withstand is 6 V. I need to connect it to 6 V AC RMS, or 8.5 V peak. Could I put a resistor in series with the LED to give the correct forward current, and a diode in inverse parallel to prevent the reverse voltage building up?

Many thanks,

Chris

Reply to
Christopher Tidy
Loading thread data ...

Yes. The series resistor is required, and the antiparallel diode is the "standard" way to protect the LED from reverse volts.

Good Luck! Rich

Reply to
Rich Grise

Thanks for the very useful input people have given me. It seems we still have several possible solutions to the problem, so I think I'm going to have to dare to pick one, even if Fred kills me for it.

At the moment my preference is for the optocoupler. It solves the problem of having different logic ground levels and provides a Schmitt trigger all in one package. In this case I would ground the centre tap of the transformer secondary and connect the optocoupler's LED between one end of the secondary and the centre tap, using a series resistor of a few hundred ohms (the precise value doesn't seem too critical) to set the correct current, and a diode in inverse parallel to limit the reverse voltage across the LED. It seems very neat, but if I'm missing something which is going to stop the circuit working, please do say!

I'm off on holiday tomorrow so I won't be able to read the group messages immediately, but I'll read them on my return.

Thanks very much for all the help.

Best wishes,

Chris

Reply to
Christopher Tidy 

--
Yes, you could, but I\'ll show you a much better, cheaper way to do
it tomorrow.  

Sorry I didn\'t get the schematic posted today, but I got the circuit
design done and I\'ll post the schematic tomorrow.
Reply to
John Fields 

--
Including a killer noise/spike filter. :-)
Reply to
John Fields

How about a 0.5 RPM synchronous motor, a cam, and two microswitches?

John

Reply to
John Larkin

Okay John. I'm off on holiday in the morning, but I'll be interested to look at your schematic when I get back.

Best wishes,

Chris

Reply to
Christopher Tidy

I thought about it seriously. It doesn't seem as neat as a solid state solution, and I doubt it would be as reliable, which is a big consideration as I want this clock to run permanently. Plus I couldn't find a 1/2 rpm synchronous motor at a decent price. I could buy 1 rpm, and I have 1/12 rpm sitting in a box, but not 1/2 rpm. And it isn't such an interesting project :-).

Chris

Reply to
Christopher Tidy 

--
A little klunky for my taste, but if it gets you off, sure, why not?

Is this what you had in mind?


+12V--->\\    -->/
Reply to
John Fields

Reply to
John Larkin

On Fri, 26 May 2006 18:14:28 -0700, John Larkin Gave us:

No maybe. ANY solenoid should have a diode or transzorb across the coil. Any relay of any kind, in fact.

Reply to
Roy L. Fuchs

Well, not precisely ANY.

John

Reply to
John Larkin

On Sun, 28 May 2006 12:39:24 -0700, John Larkin Gave us:

Any COIL based relay. I wouldn't see a need for a millivolt triggered solid state relay to have one, but those are not what we are talking about here. We were talking about magnetic avalanche abatement, and that happens on coil energization strokes, and de-energization. ANY coil based device that has a DC voltage placed across it will exhibit a back EMF when the voltage is removed, if such removal is all at once as the magnetic field standing on the coil avalanches back down. If coils were energized with a slowly increasing voltage, and de-energized with a slowly decreasing voltage, the problem would not occur.

The word for today is "slew rate".

Reply to
Roy L. Fuchs

What about big ole AC contactors? I've never seen one there. Car starter solenoids? Fast-operating gadgets like dot-matrix printer solenoids or fuel injectors?

Last relay thing I did, I needed very fast response, < 1 ms on/off, so I put a resistor across the coil.

John

Reply to
John Larkin

Make that a transzorb - the OP wants to apply pulses to the solenoid with alternating polarity.

Reply to
ian field

On Sun, 28 May 2006 22:52:21 GMT, "ian field" Gave us:

A dual diode then? (dual transzorb, that is).

Reply to
Roy L. Fuchs

The triac appnote I was reading the other day mentioned transzorbs, and seemed to imply they were like a pair of back to back high voltage avalanche diodes. The schematic symbol also suggests this, but I've also seen a similar symbol used for a varistor back in my tube TV days.

Reply to
ian field

You can buy unidirectional or bidirectional transzorbs. They're just large-area zeners that can absorb a lot of transient power, numbers like 500 watts peak for the small ones.

John

Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.