Need help with optical security circuit.

[cbt711]

I am going to buy these parts and prototype it with real components, but since I have multisim, I thought it would be nice to create the circuit and maybe work through any issues I can virtually.

I need a circuit that takes 120Vac in, generates 5Vdc and 1.5Vdc power to drive optical transmitter and receiver.  I'm using a data port actually because it has great range and is pretty cheap.  Instead of sending binary code though I'm just sending out a stead light that is broken or not broken by doors and windows in my home.  The receiver then sees this as an input and controls a relay.

I tried several voltage regulators that come with multisim, but I get errors running my circuit.  Really I just can't get from the 120Vac to the power levels needed to operate the optics.

Alternatively I might want to run this on a 120Vdc system with battery backup, so throw a switch mode power supply from 120Vdc down to 20Vdc - but I haven't found a SMPS in the library that takes 120 as an input and 20 as an output.

Basic plan:  120Vac source -> transform to 24Vac -> Full bridge rectifier to ~20Vdc -> Capacitor voltage filter on input of two voltage regulators (1 at 5Vdc, 1 at 1.5Vdc)  - then the circuit from the two voltages to the transmitter and receiver. 

I just need help getting to the 5V and 1.5V, from there I know the real world circuit will work from component testing already done.  Thanks for reading.

[johnsold]

Why fight the world? Power supplies tend to come in standard voltages. The 20 V limit is the maximum those regulators can accept as input. At that level the power dissipation will also be near the maximum, depending on your load current.

My suggestion would be to use as your source a power supply rated 12 or 15 V. That will give you many choices of standard power supplies and reduce the dissipation in the regulators. You could even use a 5 V source for the 1.5 V output.

Lynn

[cbt711]

Thank you for the reply Lynn, are there any componants native to multisim I could throw in my circuit to get it working in the program that do this?  When I put 12V into the voltage regulators pictured above, my probe reads -70.9kV output!  I agree whole heartedly, the more standardized the better, but I want to get this all on one printed card, so I don't want to have a 24 / 12 / 5V supply as a buyout part separate from the circuit.  Thanks again for the reply.

[johnsold]

I have not run a simulation, so I am not sure where the 70 kV comes from.   Do not stand too close to your computer while running that - it could be a shocking experience!

Part of the problem may be that C1 is too small.  According to the datasheet the quiescent currents of the regulators are 3.6 and 6.0 mA (typ). That gives approximately an 8 V ripple at C1. The simulation may misbehave if the input voltage goes out of range.

What happens if you remove V1 and T1 and feed 16 VDC to the input of D1? 

Lynn

[cbt711]

That made the output voltage -200MV give or take 15 MV.

Yes that's a capital M for mega.  Haha.  The peak to peak voltage is close to 1.5V and 5V respectively for the outputs.  But it is showing 214 MVrms and 214MVdc with a frequency of 184Hz.  I attached a screen shot.

I am also going to try without grounding the downstream circuit from the rectifier on.  It gives me an error, but maybe the convergance error auto fix will change the parameters to stop the error from happening.

[cbt711]

I even made a simple circuit from the voltage regulator's cut sheet here:

http://www.onsemi.com/pub_link/Collateral/NCP1117-D.PDF

Page 2.

With 12Vdc input, and the basic circuit shown from the manufacturer's recommendation, it still doesn't work.  It's showing -4.12MV.

The load resister was 1kohm (covered by probe readout).  See attached.  I don't think this voltage regulator series is working in multisim.

[johnsold]

I would agree that the model for the regulator is likely the suspect.  I have never used that regulator, nor have I ever simulated a 3-terminal regulator, but those circuits are simple enough that it should not be a problem.

I have downloaded the model from On Semiconductor. I will take a look at the model to see I anythign strange appears in there.

Lynn

[cbt711]

If you know of any regulators that I would have in my multisim 12.0 library that would work, I am perfectly happy to try them instead.

[johnsold · Accepted solution]

I do not have Multism so I cannot advise you on compatible models. I ran the On Semiconductor model with slight modifications of format on my SPICE simulator which is based on Berkeley Spice 3f5. I had to change the semiconductor resistor model call format but did not change any values.

The output for your power supply 3 circuit (with the 5 V regulator, not the 1.5 V regulator) was 4.99995 V.

There are some posts on modifying published models for compatibility woth Multisim. You might search those to see if there are any suggestions on what you might need to fix in the model.

Lynn