Hi,

I am trying to model a simple 230V/50Hz to 5VDC power supply to drive a downstream circuit. I am using a transformer to get 12VAC (modelled as V1); this feeds a bridge rectifier (D2-D5), then filter (C1) before being passed into a voltage regulator (U2, C2, C3).



I get a perfect output as can be seen below (blue trace showing a stable 5V).



However, when I place a load on the generated DC (in this case, it is a high resistance potentiometer (R1), the results become very messy. I don't understand why this is happening and would appreciate any advice out there.

 

Try putting a load of at least 10mA on the output.

You're simulator might not be working correctly with no load.

 

Try putting a load of at least 10mA on the output.

You're simulator might not be working correctly with no load.

thank you very much; that seems to be helping!

 

One thing I can see that may be an issue is that the pin numbers for the 7805 don't match usual convention, though there certainly is no hard and fast rule about pin numbering for 3-lead packages. Usually 1 is input, 2 is ground and 3 is output. But that should make no difference as long as the model knows which pin number corresponds to which function. It shouldn't care if you call the pins 1, Fred, and 666.

Also you should verify if your simulator interprets M as 10^6 or 10^-3. Some will interpret M and m as milli and require use of Meg. This is easily confirmed by putting a 1M resistor across a 1 volt DC supply. Or try 1000000 in your circuit.

Some simulators will let you use diodes at grossly above their rating without showing either smoke and flames or what the realistic forward voltage would be (it is OK to use a diode far above its average or RMS rating as long as the time is short and the duty cycle low).

Looking at currents in simulations is often helpful.

The roughness in the waverforms may be due to the calculation interval, in terms of time, being too long.