Topspice simulation - experts advice needed - RF Cafe Forums

 Post subject: Topspice simulation - experts advice needed

Posted: Wed Aug 23, 2006 1:50 pm 

Simulating a Voltage Regulator spice model - Performing Transient analysis with Voltage regulator loaded with current source set with 20KHx pulse input. Voltage regulator is a fixed 5V regulator.
When I measure the regulator output after the simulation it seems to be stable.

When i load the regulator with a LPF and then run a similar simulation I see the regulator output to be ringing?

can anyone throw some light on whats happening in this situation?

Any help in this regard would be much appreciated

Thanks

 Post subject: TopSPICE & VR output ringing

Posted: Mon Aug 28, 2006 11:03 pm 

Hi!

There are two common structures for voltage regulators. In one, the output comes from the emitter of a transistor (NPN for a positive voltage, PNP for a negative voltage). In the other, the output comes from the collector of a transistor (PNP for a positive voltage, NPN for a negative voltage). The first one is more widely used, but the second is the structure of choice for LDO's (Low DropOut voltage regulators). These two behave rather differently, because feedback is more local for the emitter output, and more global (read "slower") for the collector output.

So it's important to know which style of regulator you have.

BJT's are often the active element of choice, due to die size considerations.

Also, just as in RF, the input impedance of your LPF (the one whose connection leads to ringing) is important - is it capacitive or inductive at the ringing frequency?

The oscillating emitter follower is a famous and rather widespread problem, so the simulator is probably not lying.

So let us hear more from you!

Good Luck,
Fred

 Post subject:

Posted: Tue Aug 29, 2006 9:43 am 

Thanks Fred for all the inputs.

The regulator model that I have been using is a UA7805 fixed positive voltage regulator from TI. I believe it is a emitter follower. Not sure about the input impedance of the filter. Shall get back to you on this one.

Can you throw some light on why oscillations occur in a emitter follower?

Thanks for your time

Best regards
Balaji

 Post subject:

Posted: Tue Aug 29, 2006 11:18 am 

Fred

The oscillation frequency of the ripples were at 25 - 30 MHz.
I looked at the S11 for the Filter ( 1MHz to 100 MHz), At the lower end it was inductive and moved over to capacitive. @ the oscillation frequency its looks like an open. Its more like a crossover point from being inductive to capacitive.

Looking forward for your feedback

regards
Balaji

 Post subject: Emitter Follower Oscillation

Posted: Fri Oct 20, 2006 2:57 pm 

Hi again!
Sorry for the delay - I got swamped at work.

Here's the best discussion I've found so far:

http://www.analogzone.com/col_1017.pdf

Let me know what you think!

Good luck,
Fred