need help with a question

[ranjith]

I need help with a homework question rather, any help will be appreciated

I've attached the little circuit diagram, the question was "Describe how the input impedance behaves at extremely low and high frequencies"

edit: pic is more visible when enlarged

[efzauner]

Well, I dont think that is fair if we give you the answers, that would be cheating. However, what do you know about inductors, resistors, and capacitors at very very low frequencies and very very high frequencies? (it's a trick question for the resistors) If you know this answer, then you can solve the circuit. If you do not know the answer, then you need to study your class notes some more.

[ranjith]

OH no no, I didn't mean to sound like I was trying to get an answer for an assignment or anything. My roomate and I were studying were for an exam we couldn't get the question I posted right so I was just looking for an explanation. But yes, I've got it figured out since I posted it yesterday. Thanks

[WMcD]

At high frequencies the cap is a short. Therefore the input impedance is just R2. The cap shorts out the right side R and H and they have no effect.

At low frequencies the cap is an open. The inductor is a short. Therefore the input impedance is R1 plus R2.

I'm a bit uncetain of the midrange. You have a loossy tank, I think the impedance will have a peak..

It is fun to make such guessses and then put the circuit into P-Spice and see what comes out.

Gil

[ranjith]

Hello Gil,

So this Pspice program, is it available anywhere for trial or as a limited-functioning edition? At school they give us a program called Multisim 7, its a very nice piece of software for when working with circuits. Very user friendly, has a load of features... we were only introduced to them recently, it'll be awhile before I find my around to doing everything right on it.

Also has anyone used both, if yes which do you find is better?

[djk]

Workbench/Multisim is OK for simple stuff. P-spice has too high of a learning curve.

For analog there is no substitute for really learning and building a prototype.

I've had ID ten T types that tell me my circuit doesn't work(they sim'd it). I just point to the working prototype and say: go ahead tell it.

I've had people try an get me to sign off on a sim'd design that I know will blow up over temperature.

GIGO, and no model is complete.

[Al Klappenberger]

ranjith,

The explanation Gil gave earlier is the right answer. Here is an actual computer plot of the inpedance. As Gil expalined, at DC you see r2+r1 = 2300+1100= 3400 Ohms as the inductor is a short circuit. At infinite frequency, where C1 is a short circuit, you see only R2=2300 Ohms.

On the plot, notice that the top lines is 3600 Ohms. The plot is down 1 division at DC (200 Ohms / division). At roughly 100 KHz (at the marker) the impedance is nearly 2300 Ohms. At true infinity, it would be exactly 2300 Ohms.

Al K.

[ranjith]

Gil and Al, Thanks a bunch. That graph was pretty much exactly what I wanted to see, it gives me a better idea. I should learn to use Multisim or Pspice better.

djk, thanks for the comparison. I've never used P-spice before so I was wondering, I found a student version for download online I'll go ahead an try it.

[WMcD]

BTW is that really a nano Farad? It looks to me like you'd hae an interesting circuit if it was just a a micro farad. I have not tried either.

In any event, the lesson is to just dope out circuits visually. This gives you a start on design.

I took some EE classes at IIT. We used a student, DOS version of P-Spice. Spice itself is a mainframe program set up by UC Berkley. P-Spice ran on a PC. One old version would run on a PC with no numberical processor. I got that running on an HP-200 LX palm top. It was very slow. DOS, student P-Spice also had a post processor which made, or makes, an analysis of the data far more easy.

I'll throw out an example. Spice has no way of generating impedance (maybe you can make up a sensor but I've not gone that far). So one way of finding input impedance is to feed the circuit wiith a 1 amp constant current source. That way the voltage across the source is the impedance of the circuit.. This is because Z = V / I and I is always 1. The post processor just required some commands..

In SPICE and P-Spice, in those days, you had to make up a net list describing the circuit with nodes and devices. You also described the analysis. E.g. frequency sweep. Or transiient analysis.

I learned a lot. A short coming of the student versions is that they limit the number of transmission lines to 2. I wanted to build horns. A bunch of mated TX lines would do that. I eventually got some good results using caps and inductors to simulate TX lines. I was able to get the same results which the Olsen study of finite horns show.

Later I ran across Winspice on the Internet. (You can Google.) It is freeware, or begware. This is pretty much Berkley Spice for Windows. It is text based except you can get graphs. I've used that to simulate horns. It is not easy but can be done.

The more recent programs of this type start with a program to build a schematic, graphically. Then there is a Spice based engine to do the analysis. There are visual lab equipment, like an oscillosope, to display what you want to see.

I bought the student version of Electronic Workbench but didn't have the time to figure it all out. Also, it seems to have a limit on TX lines. The schematic program seemed cranky to me. It is probably just a learning thing and need for a better computer. I went back to Winspice.

Please excuse the history and war stories above.

Certainly you will HAVE to learn some version of the modern Spice based simulators. They start with schematic entry, Spice simulation, and then circuit board design. They have analog and digital modes, and mixed.

The problem, for me and perhaps starters, is that many programs are very complicated and there is a fair amount to learn before getting a simple circuit working. I have no doubt that if you are emersed in these for a while, you will get over the hump of initial learning. Then you can do wonderful things.

In any event, use what the profs suggest and what your classmates are using. When in Rome . . . learn Latin.

Best,

Gil

[ranjith]

Yeah that was a nano F.

Well, I'm in a laptop based univ, and the default program for my EE class was Multisim. Being in a mechanical engineering course with specialization in mechatronics, this EE is just my beginners course since its only 2nd year. I believe we have more advanced classes in 3rd year such as Control Systems, Robotics and a few that get into the more fun part- applying the theory. What I found frustrating about my EE was that, I think we spent an hour so far this entire semester on how to use Multisim. The TA made a couple of very basic circuits, with resisters capacitors and inductors just as an example. But we were never told how to evaluate a circuit design by bringing up graphs of any function and such, I've been trying to learn using the Help function but its certainly not as easy as someone teaching or showing us how its done. With school exactly a month away from completion of this academic year I have no leisure time on my hands, but when the summer comes around I plan on getting myself acquainted with P-spice. I know we probably won't go into a lot of advanced EE in my major but its just part of my interest in electronics, especially audio. Eventually I'll understand what Al, Dean and Craig argue about all the time