02-23-2011 01:36 PM
Hello!
I am trying to simulate a isolated flyback with Multisim. Is this possible using a standard model? It's not in the sample files and the VM Flyback does not work isolated.
Thank you for the help -- in advance 🙂
Andrei
02-28-2011 09:16 AM
Hi andreidanescu
I'm unfamilliar with the Isolated Feeback that you are trying to simulate. Can you explain the concept/Item you are trying to simulate and I will see what I can do to help.
Regards
Richard
02-28-2011 12:54 PM
Hello,
The schematic is well described in this link: http://en.wikipedia.org/wiki/File:Flyback_conventions.svg
http://en.wikipedia.org/wiki/Flyback_converter
Main ideea would be that for an isolated flyback you need to float a part of the circuit. This seems to be my problem, multisim will not accept isolating a part of the circuit and not provide any common ground.
I tried to connect the schematic via a 1M ohm resistor but it's not a good ideea.
Thank you very much.
Andrei
03-01-2011 11:57 AM
Andrei,
Isolated circuits tend to cause some problems for simulators. A common trick, like you suggested, is to connect the floating side to ground (node 0) using some large resistor such as 1 or 10 Meg. It will not introduce any significant errors into the results.
Attach your circuit and I can help you as I've successfully simulated various flyback topologies.
Thanks
03-01-2011 12:51 PM
Hello.
The circuit I tried is the sample one from NI Multisim's database. SMPS average analisys - Flyback_VM.
Now, I really need to work with this, because I'm suppose to use the standard block for this simulation. The problem is that the model doesn't accept the solution with a 1M resistor.
It must use rectifed AC - 120 -350 V DC and output 5V.
I will try a classic approach soon, without this block but I must get this one working first.
How did you go about simulating the isolated flyback? Do you have any working example with isolation? If you can, please attach it, no matter what voltages. I am mainly interested to get the isolation than use the specified values.
Thank you very much,
Andrei
03-01-2011 01:39 PM
Andrei,
Note that SMPS circuits by their nature are difficult for simulators to tackle. Isolation exacerbates the problem. You are right - in the flyback isolation does lead to non-convergence.
However I am not clear on what simulation differences you expect to see between a flyback with isolated grounds and one without. I think many more issues before grounding generally need to be considered in the design of a flyback.
I've attached a very basic open loop flyback to assist you. The GND ground on the output side is connected to ground zero for simulation purposes.
03-01-2011 01:49 PM
i can't belive it:)
i did exactly the same thing this morning. used the "digital" ground to simulate - but ideally speaking this is not a true solution, as you mentioned.
I am verry well aware of the design of the unit, I actually finished all but simulations. It is university research work and they had this specific ideea with simulating the "isolated flyback", altough I do not expect any differences isolated/ not isolated and I understand the limitation of the simulator.
This will be the final solution and it clearly shows the way the converter is intended and how the isolation is to be provided. Thank you for all your help, it really cleared things up!
03-23-2011 01:48 PM
Does anybody have an air gap transformer model?
Thank you,
Andrei
02-18-2013 11:25 AM
Hi all,
I'm trying to design a flyback smps which takes in 12 VDC and output a 240 VAC. Attached is my schematic for the design.. I cant seem to get my desired output.
Comments and feedbacks are appreciated.
02-19-2013 09:33 AM
Well, you are not turning On your FET. You need a Vgs higher than Vt so your gate voltage amplitude (the way you're driving it), needs to be higher than the 12V power supply.
Also, a flyback is typically used for creating a DC voltage. You'll need a diode and a capacitor on the output side. You also need to ensure that the polarity of the transformer (dots) are on opposite ends, so that when the transistor is Off, the current from the magnetising current forward biases the diode on the secondary side.
Here's a nice article:
http://en.wikipedia.org/wiki/Flyback_converter