AC-AC Optocoupler

Thread Starter

momotee

Joined Jan 18, 2012
14
Hi,

I am wondering if a AC-AC opto-isolator actually exist?
I know that a transformer is a good example, but what about dual in-line package IC chip?

Thank you!
 

Brownout

Joined Jan 10, 2012
2,390
Optoisolators do exist to isolate AC. Transformers are something completely different. I think a transformer can be in a dual in-line package.
 

t06afre

Joined May 11, 2009
5,934
Hi,

I am wondering if a AC-AC opto-isolator actually exist?
I know that a transformer is a good example, but what about dual in-line package IC chip? Thank you!
Can you tell more about what you want to do. And what kind of voltages/current magnitude we are talking about. Another important parameter is the bandwidth(speed) of your signal
 
Last edited:

debjit625

Joined Apr 17, 2010
790
Transformers works on inductive coupling ,opto-isolator use light coupling.And about opto-isolator for AC I am not sure... may be they exists,as its just an electronic circuit which use light for coupling between two circuits.

Good Luck
 

Thread Starter

momotee

Joined Jan 18, 2012
14
sorry i'm new to forums. i guess i have to resize the picture.
what i'm trying to build is a dc-ac inverter with a closed loop controller.
i have successfully implemented the inverter in open loop.

the problem i'm facing now is this.
as i cannot connect the ground of Vout with the ground of the (Sine and op amp subtractor) together, i have to use some kind of electrical isolation.
however, when i connect the transformer directly in parallel with the Load resistance, the output voltage Vout drops.
Thus the output voltage is not a true reflection of the inverter capability.

I require an output voltage that is not affected, as well as a replica of the output voltage to be used to subtract the Sine wave with some isolation.
 

Thread Starter

momotee

Joined Jan 18, 2012
14
[/IMG]

when i use this method, the current thru the transformer is too small to detect any voltage on the primary side, let alone the secondary side.

i am trying to step up the voltage so that i can voltage divide the secondary voltage to replicate the Vout across the inverter Load resistance.


thanks for the replies!
 

Brownout

Joined Jan 10, 2012
2,390
when i use this method, the current thru the transformer is too small to detect any voltage on the primary side, let alone the secondary side.
Remove the resistors in the primary.

i am trying to step up the voltage so that i can voltage divide the secondary voltage to replicate the Vout across the inverter Load resistance.
Is the Load a voltage divider?
 

Thread Starter

momotee

Joined Jan 18, 2012
14
Remove the resistors in the primary.

I cannot remove the resistors as the Vout will experience a loading directly from the transformer and the voltage level of Vout will drop. Thus giving a misrepresentation of the actual output voltage of the inverter.

Is the Load a voltage divider?
Nope. The load is drawing most of the current. The voltage divider is just to draw as little current as possible but still be able to produce a voltage from the transformer to sense.
I realise that the current through the transformer is too low to "drive" the transformer. Thus, making the transformer unable to sense the voltage.

I was advised to look for an AC-AC optocoupler but can't seemed to find any.:)
 

crutschow

Joined Mar 14, 2008
34,281
..................
I realise that the current through the transformer is too low to "drive" the transformer. Thus, making the transformer unable to sense the voltage.
A transformer of the proper characteristics should work fine for your requirement. The magnetizing current is very small as long as it is operating at or above its design frequency and voltage. You should not have to place any resistor divider in series with a transformer of the correct design.

What kind of transformer is it?

Edit: If you want to use an optocoupler just rectify and RC filter the output to get the average value of the AC and send that dc value across the optocoupler. You don't need to send AC.
 

Brownout

Joined Jan 10, 2012
2,390
I cannot remove the resistors as the Vout will experience a loading directly from the transformer and the voltage level of Vout will drop. Thus giving a misrepresentation of the actual output voltage of the inverter.
Well, they shouldn't be in there. You'll need to solve that problem another way, such as ensuring the transformer is properly rated for the input voltage and frequency.
 

crutschow

Joined Mar 14, 2008
34,281
If the transformer is properly rated, its load will be vary small, essentially just the reflected load from the secondary.

It would appear that you are unclear as to how transformers work. For an ideal transformer, primary current is equal to the secondary current, due to any secondary load, times the turns ratio. A real transformer will also have magnetizing current but that is very small in a properly designed transformer.
 

crutschow

Joined Mar 14, 2008
34,281
i was searching and found an isolation amplifier.

not sure if this could work? comments please :) thanks!
That should work also, but be aware that both sides require a separate, isolated power supply voltage. So you will need to generate a DC voltage from your output to power the isolator.
 

praondevou

Joined Jul 9, 2011
2,942
i was searching and found an isolation amplifier.

not sure if this could work? comments please :) thanks!

I used low-power (a few VA) transformers for the feedback loop of DCAC converters. Works very well. I guess your transformer is much to powerful. At the output you will only need a few mA at most.

The isolation amplifiers also work very well, even though I never used the one you posted but this one:HCNR201. They have internal feedback and therefore high linearity. They all have a disadvantage though, you need a power supply for them on the primary side. And in the case of the one I mentioned some opamps on primary and secondary side.

You really should get this working with a transformer, it's much easier.
 

Thread Starter

momotee

Joined Jan 18, 2012
14
hi everyone, thank you for the comments and advice. my professor sat down with me and went through some testing and thinking about new solutions.

turns out that a simple LM741 op amp configured to be used as a differential amplifier solved the problem! now i can have a signal with no grounding problem!

the one in blue is a separate grounding from the input signal. thank you for the help!

[/IMG]
 

crutschow

Joined Mar 14, 2008
34,281
A differential amp, such as you show, requires that the two grounds be within a few volts of each other or the circuit won't work. If that's ok for your purpose then it's fine. But that does not provide true isolation between grounds.
 
Top