I tried to do a fullwave rectifier in hspice to test my AC input buck regulator. However, It seems that my filter capacitor is not working because the output of the rectifier is still a pulsating DC. Is it because I have to use a polarized filter cap? Please help.
I guess, you rather have a wrong circuit or a silly part dimensioning. An unpolarized cap can always replace a polarized type (if it has a correct capacity value).
The rectifier with cap filter is so simple for one to make mistakes. I tried changing the cap values(i.e making it very large) and it still doesnt work. I tried it also on PSPICE and tried both polarized and non-polarized, and the rectifier only works for polarized cap.
The capacitor in Hspice, I believe, is not polarized. Because it can be used in either directions. When I tried to design in a schematic in Pspice, it worked. But I need to do it in Hspice. Maybe I need to get a Polarized capacitor Model. But still I don't know if there's such thing in Hspice.
I already don't understand what's the "polarized cap" you say to have used in PSPICE. The basic "C" part in PSPICE (as well as in all other SPICE variants I've ever seen) is a generic element with capacitance as only parameter, not needing a model. A user model can be optionally supplied.
I would be really suprized, if this is basically different with HSPICE.
P.S.: I verified in the Synopsys HSPICE User Guide, that a basic C element statement refers to a capacitor without a particular model:
Code:
C1 1 2 10u
If you are using schematic entry, you may want to check, if your tool generates additional (possibly incomplete) model statements, but the basic SPICE syntax clearly can work with HSPICE, as expected.
By the way: A frequent beginners error with some SPICE schematic entry tools is using a GND symbol, that isn't connected to node 0 by default. Posting the netlist used in your analysis can help to clarify this and similar issues.
The capacitor should be O.K., I rather expect a problem with the diode model. Interestingly, the diode voltage drop is much too high. Possibly the diode is discharging the capacitor.
You should monitor the diode and capacitor currents as well. Unfortunately, the simulator doesn't care if the diode reverse current is mA or kA!
Hi Andrew,
perhaps the endio_3 model needs a specification of its reverse breakdown voltage (BV), the default value in HSPICE being BV=0 (!), s. att. Hspice models quickref p. 13 (PDF p. 17). So try to change it:
My capacitor is now working. Instead of using solely capacitor to make the pulsating dc (from Bridge Rec) as constant DC, I used the Valley Fill (VF) Power Factor Circuit. My circuit would look like this: **broken link removed** (1st page). VF is composed from D3 to C10. and my Vbuck waveform is in the attached file (in RED, the Black one is the current for LED). My Problem is I want to make the DC as constant as possible but why it still looks like a pulsating DC just with a small-time constant dc. As you can see, during this very small constant Vbuck the LED's current looks very ugly.
My point is, I want to make the Pulsating Dc as constant as possible so that my current in LED will remain constant as well.
Please Help Please dont tell me to remove the Valley FIll Circuit and replace it by a single large value capacitor.
Your design specification is basically unclear. Generally, it's a matter of capacitor dimensioning.
You see the time constants, you have designed in.
In addition, the circuit behaviour should be fully understandable in terms of stored charges and energy. Perhaps you should start with pencil and paper method rather than using an apparently mysterious simulator.