Greetings - Op amp problem...

[JohnB]

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?




The circuit is two negative feedback amps

 

[Nigel]

Where is the transmitter resitor tied (-12V or 0V)?

Any resitors in your transistor circuit?

What is the operating voltage/current of the relay coil?

When you do get this working it is also advisable to put a reversed diode in
parallel with the relay coil. This protects the transistor from damage from
reverse EMF generated when the coil is switched on and off.

My guess is that your have the emitter tied to ground. When you present 11V
at the base, the transistor switches on because of the B-E voltage exceeds
0.7V. If there is no base resistor, the circuit will try at set the emitter
to 10.3V but it can't because it is tied to 0V.

This means that the full 11V will be placed across the B-E junction which
will certainly cause the transistor to switch on but may damage the
transistor or have strange effects.

I suggest you put a resistor between the op-amp and the base. This will
carry the excess 10.3V and should be sized so that it provides sufficient
base current which when multiplied by HFE of the transistor creates a
collector current to meet the relay's requirements.

e.g Rb = 10.3 / (HFE x Relay Operating Current)

Make Rb a little smaller to allow for tolerances etc.

If my assumption is incorrect please provide some more info on the
transistor stage.

Cheers,

Nigel

"JohnB" <jbanion@jb.com> wrote in message
news:cb0b2c28.0310112329.6c4ed444@posting.google.com...

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?




The circuit is two negative feedback amps

 

[Gareth]

JohnB wrote:

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?

Do you have a diode from the transistor end of the relay coil to the
positive supply? A relay is inductive and when it is switched off
quickly current will continue to flow in the coil in the same direction
as before it was switched off, if you put a diode in, the current has
somewhere to go (in this case back to the supply rail). If you don't
provide a path for the current you could get a voltage spike which will
damage the transistor.

An alternative is to put a zener diode with zener voltage > the supply
voltage but less than the maximum Vce for the transistor from the 2n2222
collector to ground.

+12V----+-------|
) -
( ^ diode
relay ) |
coil ( |
--------+
|
|

Does your op amp output ever go negative? - this could also damage the
transistor.

--
-----------------------------------------------------------------------
To reply to me directly:

Replace the text after the@symbol with: totalise DOT co DOT uk

 

[Thomas C. Sefranek]

A resistor between the op amp output and the base is good.

"JohnB" <jbanion@jb.com> wrote in message
news:cb0b2c28.0310112329.6c4ed444@posting.google.com...

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?




The circuit is two negative feedback amps

 

[John Fortier]

"JohnB" <jbanion@jb.com> wrote in message
news:cb0b2c28.0310112329.6c4ed444@posting.google.com...

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?




The circuit is two negative feedback amps

Here's my suggested circuit:

First, between the final TL082 output and the base of the 2N2222, place a
diode to block negative going voltages. A 1N4248 is good here.

Second, place the relay in the emitter circuit, with a 100 ohm resistor in
series to ensure that the relay coil gets enough current to operate without
any chance of burning out.

Connect the collector to the positive supply and the resistor to earth.

I've checked this in PSpice, and it should work

May I suggest that a good move would be to go to the Cadence site,
www.cadenmce.com, and download the, free, student version of
PSpice/Schematics. You'll be able to try out all the various circuit
variations and not build until you are pretty sure that your circuit will
behave as you wish.

good luck,

John

 

[John Fields]

On 12 Oct 2003 00:29:34 -0700, jbanion@jb.com (JohnB) wrote:

Hello,

I am building a Video detector circuit with a TL082. The goal is to
detect a composit video signal and then turn on a relay. The circuit
consists of two negative feeback amps in series which works great. IE
the voltage goes to 11V when video is present and drops to zero when
there is no video.

My problem is as follows. I tie the output of the second amp into a
2n2222 NPN transister intending to use it to drive the relay. The
relay turns on but will not turn off when when the opamp tries to
return to zero volts. the emiter is grounded , the base attached to
the opamp output and the collector ties to the relay. The relay is
then tied to +12V. Did I mension I am using +12/-12 powerrsupply...
I want to keep things as simple as possiable.

Ideas or suggestions?

---
Yeah. First, download the TL082 data sheet at

http://focus.ti.com/lit/ds/symlink/tl082a.pdf

and take a look at the gain-bandwidth of the opamp.


Second, let us know what you mean by "composite video", and

third, post a schematic of what you're talking about to a website
somewhere or to alt.binaries.schematics.electronic.

--
John Fields

 

[JeffM]

Gotta spell those links correctly, John.

http://www.cadence.com/
http://www.cadencepcb.com/products/pspice/
http://www.cadencepcb.com/products/downloads/PSpicestudent/default.asp

 

[John Fortier]

"JeffM" <jeffm_@email.com> wrote in message
news:f8b945bc.0310121255.70624767@posting.google.com...

Gotta spell those links correctly, John.

http://www.cadence.com/
http://www.cadencepcb.com/products/pspice/
http://www.cadencepcb.com/products/downloads/PSpicestudent/default.asp

Too much reliance on psell checker!

John

 

[JohnB]

Thanks for your reply. I think the problem was that I usually use
2n2222 as switches with TTL logic, IE much lower voltage. I do have
the emiter tied to ground and I think the 11V accross the BE junction
was too much. I replaced the 2n2222 with a TIP 122 and it works great
now.


"Nigel" <nigel@NOSPAM.the-heathers.nildram.co.uk> wrote in message news:<3f892085$0$65577$65c69314@mercury.nildram.net>...

Where is the transmitter resitor tied (-12V or 0V)?

Any resitors in your transistor circuit?
No. Not yet.

What is the operating voltage/current of the relay coil?

12V

When you do get this working it is also advisable to put a reversed diode in
parallel with the relay coil. This protects the transistor from damage from
reverse EMF generated when the coil is switched on and off.

I know. Actually Im using a 100uF cap because the relay was bouncing
when being turned off. Is that a bad idea? Works nice.

My guess is that your have the emitter tied to ground. When you present 11V
at the base, the transistor switches on because of the B-E voltage exceeds
0.7V. If there is no base resistor, the circuit will try at set the emitter
to 10.3V but it can't because it is tied to 0V.

Correct.

This means that the full 11V will be placed across the B-E junction which
will certainly cause the transistor to switch on but may damage the
transistor or have strange effects.

Definatly damaged.

I suggest you put a resistor between the op-amp and the base. This will
carry the excess 10.3V and should be sized so that it provides sufficient
base current which when multiplied by HFE of the transistor creates a
collector current to meet the relay's requirements.

I will try this.

e.g Rb = 10.3 / (HFE x Relay Operating Current)

Make Rb a little smaller to allow for tolerances etc.

If my assumption is incorrect please provide some more info on the
transistor stage.

Cheers,

Nigel

Thanks!

 

[JohnB]

Thanks for your reply.

I have replaced the wimpy 2n2222 with a TIP122 and it works ok now. I
think the 2n2222 might have worked if I had biased it correctly.

John Fields <jfields@austininstruments.com> wrote in message news:<354jovg1flqq140iuae8cujbcj9pbo9ihb@4ax.com>...

On 12 Oct 2003 00:29:34 -0700, jbanion@jb.com (JohnB) wrote:

---
Yeah. First, download the TL082 data sheet at

Already have it.

http://focus.ti.com/lit/ds/symlink/tl082a.pdf

and take a look at the gain-bandwidth of the opamp.

bandwith is not an issue. I rectify the ac with a small signal diode,

resister and cap, before It gets to the first stage of the opamp.

Second, let us know what you mean by "composite video", and

I believe the more technical term is NTSC. IE the type of video

signal you would find on the back of a typical US VCR or DVD player.

third, post a schematic of what you're talking about to a website
somewhere or to alt.binaries.schematics.electronic.

Thanks again.

 

[Nigel]

Good to hear it is working now.

The diode is the more traditional solution, smaller and cheaper than the
capacitor, but if the you're happy with the capacitor and it works then I
guess stick with it.

I would still recommend a resistor between the op amp and the transistor
base though.

Cheers,

Nigel

"JohnB" <jbanion@jb.com> wrote in message
news:cb0b2c28.0310131144.44503718@posting.google.com...

Thanks for your reply. I think the problem was that I usually use
2n2222 as switches with TTL logic, IE much lower voltage. I do have
the emiter tied to ground and I think the 11V accross the BE junction
was too much. I replaced the 2n2222 with a TIP 122 and it works great
now.


"Nigel" <nigel@NOSPAM.the-heathers.nildram.co.uk> wrote in message
news:<3f892085$0$65577$65c69314@mercury.nildram.net>...
Where is the transmitter resitor tied (-12V or 0V)?

Any resitors in your transistor circuit?
No. Not yet.

What is the operating voltage/current of the relay coil?

12V

When you do get this working it is also advisable to put a reversed
diode in
parallel with the relay coil. This protects the transistor from damage
from
reverse EMF generated when the coil is switched on and off.

I know. Actually Im using a 100uF cap because the relay was bouncing
when being turned off. Is that a bad idea? Works nice.


My guess is that your have the emitter tied to ground. When you present
11V
at the base, the transistor switches on because of the B-E voltage
exceeds
0.7V. If there is no base resistor, the circuit will try at set the
emitter
to 10.3V but it can't because it is tied to 0V.

Correct.

This means that the full 11V will be placed across the B-E junction
which
will certainly cause the transistor to switch on but may damage the
transistor or have strange effects.

Definatly damaged.

I suggest you put a resistor between the op-amp and the base. This will
carry the excess 10.3V and should be sized so that it provides
sufficient
base current which when multiplied by HFE of the transistor creates a
collector current to meet the relay's requirements.

I will try this.

e.g Rb = 10.3 / (HFE x Relay Operating Current)

Make Rb a little smaller to allow for tolerances etc.

If my assumption is incorrect please provide some more info on the
transistor stage.

Cheers,

Nigel

Thanks!