12AC control circuit

[Walter Harley]

Hi,

I need to be able to silently switch the output of a transformer 12V ac -
on/off. Load is 50W. My current attempt has been to use a triac although it
gets fairly hot, sorry, that should read very hot and needs a large
heatsink. Maybe someone knows of a noiseless relay or a solid state device
that can operate at those voltages.

Any ideas would be appreciated.

Search for "solid state relay". They're very common and easy to use these
days.

 

[Tweetldee]

Hi,

I need to be able to silently switch the output of a transformer 12V ac -
on/off. Load is 50W. My current attempt has been to use a triac although it
gets fairly hot, sorry, that should read very hot and needs a large
heatsink. Maybe someone knows of a noiseless relay or a solid state device
that can operate at those voltages.

Any ideas would be appreciated.

Thanks, Paul.

Controlling only 50W (or did you mean V-A?) is not difficult at all, and
with a proper triac, should need little heatsinking.
Describe your circuit. What kind of load? What kind of triac? Do you switch
the triac on at zero-crossing or somewhere later in the cycle?
--
Tweetldee
Tweetldee at att dot net (Just subsitute the appropriate characters in the
address)

Time is what keeps everything from happening all at once.

 

[Spehro Pefhany]

Controlling only 50W (or did you mean V-A?) is not difficult at all, and
with a proper triac, should need little heatsinking.

The heatsink will need to be for about 4W, which is substantial
enough, especially if you need high Ta operation.

Describe your circuit. What kind of load? What kind of triac? Do you switch
the triac on at zero-crossing or somewhere later in the cycle?

Lower voltage drop (and smaller heatsink) would require something like
MOSFET switching. You'd need two MOSFETs back-to-back to switch AC. A
photovoltaic isolator could provide the drive. For example, a couple
of 0.028 Ohm logic level MOSFETs and a TLP190B driver might cost
around $6 US in one-off and would not require any heatsinks for even
fairly high Ta.

Best regards,
Spehro Pefhany

 

[Fred Bloggs]

Hi,

I need to be able to silently switch the output of a transformer 12V ac -
on/off. Load is 50W. My current attempt has been to use a triac although it
gets fairly hot, sorry, that should read very hot and needs a large
heatsink. Maybe someone knows of a noiseless relay or a solid state device
that can operate at those voltages.

Any ideas would be appreciated.

Thanks, Paul.

Move the TRIAC into the transformer primary circuit- and use a small
1-2VA xfmr for the 5V supply if that's how you derive it.

 

[PT]

Hi,

I need to be able to silently switch the output of a transformer 12V ac -
on/off. Load is 50W. My current attempt has been to use a triac although it
gets fairly hot, sorry, that should read very hot and needs a large
heatsink. Maybe someone knows of a noiseless relay or a solid state device
that can operate at those voltages.

Any ideas would be appreciated.

Thanks, Paul.

 

[Adrian Jansen]

A triac is not the way to go here, nor is a solid state relay ( which is
just a triac and a driver cct ). The ON state voltage drop of a triac is
about 1.5 volts, so at 50 W / 12 volts = 4 amps, will dissipate around 6
watts. You dont really want to lose the 1.5 volts either - this will
severely drop the brightness and efficiency of your lamps.

Either move the triac to the mains side of the circuit, or use a mechanical
relay - is the click of a relay that noisy ?

--
Regards,

Adrian Jansen
J & K MicroSystems
Microcomputer solutions for industrial control

 

[Fritz Schlunder]

Thanks for that - it seems like the best solution. The problem of putting a
triac into the primary side is that the exposed switch and wiring (it's in a
cupboard) would also have to be on the primary side and that presents a
risk. Or just put your faith into earth (ground) leakage protection.

I believe that MOSFET'S have a much lower voltage drop across them vs
triac's, so would be ideal. Using them to control AC hadn't entered my mind
but I would love to see a circuit diagram or a link if that's possible.

Thanks for any help.
Paul.

I suggest reading a thread titled "Switching Problem" authored by Robbie
Banks in the google groups archive (groups.google.com). We covered this
basic problem in that thread as well as perhaps a couple of other threads
earlier.

http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&safe=off&threadm=3f28159e
_5%40corp.newsgroups.com&rnum=1&prev=/groups%3Fq%3Dfritz%2Bschlunder%2Bphoto
voltaic%26hl%3Den%26lr%3D%26ie%3DUTF-8%26safe%3Doff%26selm%3D3f28159e_5%2540
corp.newsgroups.com%26rnum%3D1

That is kind of a long link, it may be easier to search for it manually and
look at the complete thread.

I've previously recommended reading this:

http://www.irf.com/technical-info/appnotes/an-1017.pdf

And doing what you see in figure 6.

I recommend lower on resistance MOSFETs than the 28 milliohm type though.
Without a heatsink such a device may run very near to the maximum
temperature rating. Since MOSFETs are quite cheap I recommend something a
little beefier.

Something like the IRF7811W might be a good choice since it offers low on
resistance and is easily available from Digikey at a decent price. Also
this device is low logic level capable which is nice since many of those
photovoltaic isolators produce only something like 6V or so of drive each.
Datasheet at:

http://www.irf.com/product-info/datasheets/data/irf7811w.pdf

Be careful with switching excessively capacitive or inductive loads without
adding protection. With your low voltages capacitive loads of pretty much
any sane sizes (surely less than one farad) won't likely be much of a
problem, but inductive loads could be. An adequately sized transient
supressor placed accross both MOSFETs should protect them fine.

 

[Spehro Pefhany]

Thanks for that - it seems like the best solution. The problem of putting a
triac into the primary side is that the exposed switch and wiring (it's in a
cupboard) would also have to be on the primary side and that presents a
risk. Or just put your faith into earth (ground) leakage protection.

I believe that MOSFET'S have a much lower voltage drop across them vs
triac's, so would be ideal. Using them to control AC hadn't entered my mind
but I would love to see a circuit diagram or a link if that's possible.

Something like this:

In+ +-------+
o | | o
.---|---|--. | |
| | V | | ||-+D
| | - | | ||<- Eg. IRLZ44
| | . | +------+-||-+S
| V . | | |
| - . | | |
| | | | | |
| | V | +----|----+
| | - | | | |
'---|---|--' | | |
| +---------+ +-||-+S
| ||<-
.-. ||-+D
| | |
| | o
'-'
o
In-


With the type of MOSFET I suggested, the static power dissipation
would be around 1W total split evenly between the two MOSFETs at 4A
vs. roughly 4W in a single device using a typical triac.

If the 12VAC is completely isolated from your control circuit, you
maybe able to do away with the photovoltaic isolator completely. Or by
adding some additional circuitry.

Best regards,
Spehro Pefhany

 

[PT]

Lower voltage drop (and smaller heatsink) would require something like
MOSFET switching. You'd need two MOSFETs back-to-back to switch AC. A
photovoltaic isolator could provide the drive. For example, a couple
of 0.028 Ohm logic level MOSFETs and a TLP190B driver might cost
around $6 US in one-off and would not require any heatsinks for even
fairly high Ta.

Thanks for that - it seems like the best solution. The problem of putting a
triac into the primary side is that the exposed switch and wiring (it's in a
cupboard) would also have to be on the primary side and that presents a
risk. Or just put your faith into earth (ground) leakage protection.

I believe that MOSFET'S have a much lower voltage drop across them vs
triac's, so would be ideal. Using them to control AC hadn't entered my mind
but I would love to see a circuit diagram or a link if that's possible.

Thanks for any help.
Paul.

"it's the network..." "The Journey is the reward"
[email protected] Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers:

http://www.speff.com

 

[PT]

putting in


I suggest reading a thread titled "Switching Problem" authored by Robbie
Banks in the google groups archive (groups.google.com). We covered this
basic problem in that thread as well as perhaps a couple of other threads
earlier.

http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&safe=off&threadm=3f28159e
_5%40corp.newsgroups.com&rnum=1&prev=/groups%3Fq%3Dfritz%2Bschlunder%2Bphoto
voltaic%26hl%3Den%26lr%3D%26ie%3DUTF-8%26safe%3Doff%26selm%3D3f28159e_5%2540
corp.newsgroups.com%26rnum%3D1

That is kind of a long link, it may be easier to search for it manually and
look at the complete thread.

I've previously recommended reading this:

http://www.irf.com/technical-info/appnotes/an-1017.pdf

And doing what you see in figure 6.

I recommend lower on resistance MOSFETs than the 28 milliohm type though.
Without a heatsink such a device may run very near to the maximum
temperature rating. Since MOSFETs are quite cheap I recommend something a
little beefier.

Something like the IRF7811W might be a good choice since it offers low on
resistance and is easily available from Digikey at a decent price. Also
this device is low logic level capable which is nice since many of those
photovoltaic isolators produce only something like 6V or so of drive each.
Datasheet at:

http://www.irf.com/product-info/datasheets/data/irf7811w.pdf

Be careful with switching excessively capacitive or inductive loads without
adding protection. With your low voltages capacitive loads of pretty much
any sane sizes (surely less than one farad) won't likely be much of a
problem, but inductive loads could be. An adequately sized transient
supressor placed accross both MOSFETs should protect them fine.

Hi,

Excellent work! This is the way to go. I have checked out the previous
threads as well. Power loss across each MOSFET would be 4 * 4 * 0.009 =
0.144W. Looks too good to be true. It also gives me a chance to do a soft
start by tying in a ramp drive circuit.

Thanks for your help, Paul.

 

[PT]

Something like this:

In+ +-------+
o | | o
.---|---|--. | |
| | V | | ||-+D
| | - | | ||<- Eg. IRLZ44
| | . | +------+-||-+S
| V . | | |
| - . | | |
| | | | | |
| | V | +----|----+
| | - | | | |
'---|---|--' | | |
| +---------+ +-||-+S
| ||<-
.-. ||-+D
| | |
| | o
'-'
o
In-


With the type of MOSFET I suggested, the static power dissipation
would be around 1W total split evenly between the two MOSFETs at 4A
vs. roughly 4W in a single device using a typical triac.

If the 12VAC is completely isolated from your control circuit, you
maybe able to do away with the photovoltaic isolator completely. Or by
adding some additional circuitry.

Best regards,
Spehro Pefhany

http://www.speff.com


Thanks for all time and input. I am not familiar with the above shorthand
drawing but I found a similar circuit that works well. It's at
http://www.discovercircuits.com/M/mosfet.htm under "Charge Coupled Mosfet
Relay". Do a little adjustment here and there and it will work on my
project.

Cheers, Paul.