I just started making a "Green Pre" model in Proteus/PSPICE...A few q's

[ENS Audio]

Hi there, I've been getting pretty good lately with PSPICE/Proteus VSM and was planning on finishing up a Green Pre Simulation in Proteus VSM/ISIS and possibly will be making more circuit simulations very soon.

If anyone can post links/tutorials related to PSPICE that would be great mainly because that im not an expert atm (all the "parameters","floating" terminology hasnt sunk into my brain as of yet  ???)

I have came to the conclusion that this is possibly the best stragety for troubleshooting my "real" circuits for that having an "ideal" reference to my measurements when applying OHMs Law to a circuit.

Please stop me from going any further if this seems to sound "counterproductive" to the troubleshooting process

Thanks,

 

[nickt]

Can't help but very interested in your results and the learning process.

 

[michal_k]

I'm not sure what distribution of pspice you're using, but i think that Orcad is one of the most popular. So if you happen to have that one have a look at manual at first:
http://www.eng.usf.edu/~amjoseph/docs/PSpice%20Manual.pdf

there is a lot of stuff but you should get some basics.

I remember I also used this documentation:
http://www.macspice.com/ug/
but berkley spice functions aren't fully compatible with orcad spice. again, it's good for basic stuff though.

There are also some tutorials on macspice site:
http://www.macspice.com/Tutorials.html
but they will be probably too basic for you.

I think i had pspice lectures in english so I'll post some slides if I still have them.

M.

 

[ENS Audio]

I'm not sure what distribution of pspice you're using, but i think that Orcad is one of the most popular. So if you happen to have that one have a look at manual at first:
http://www.eng.usf.edu/~amjoseph/docs/PSpice%20Manual.pdf

there is a lot of stuff but you should get some basics.

I remember I also used this documentation:
http://www.macspice.com/ug/
but berkley spice functions aren't fully compatible with orcad spice. again, it's good for basic stuff though.

There are also some tutorials on macspice site:
http://www.macspice.com/Tutorials.html
but they will be probably too basic for you.

I think i had pspice lectures in english so I'll post some slides if I still have them.

M.

Thanks for your reply, at the moment I already have those manuals but please post more links for they're may likely be some manuals/pdfs that I dont have atm.  Also I was lucky to find some "User Libraries" for just about every component that I'll need to do the simulation.

Also, I have already started with the simulation design for Proteus 7.2 with some custom libraries in addition to that program as well.  I do find Proteus easier to use for some reason, dunno why but I'd like to force myself to start getting cozy with Orcad instead for Orcad seems to be more popular and has more support forums and tutorials that are out there.

Say if you or anyone else happens to have any Proteus Tutorials or "Reference Manuals" that go over basic file extensions and the "1's and 0's" of the application would be super!! ;D

 

[ENS Audio]

BTW to all fellow newbies I STRONGLY recommend getting into learning any PSPICE/Simulation Program for it will not only make troubleshooting IMO a lot easier but I will probably help you have a better understanding of all this theory crap, that is if you have one of those of many ELEC project books/tutorials out there.

 

[jdbakker]

BTW to all fellow newbies I STRONGLY recommend getting into learning any PSPICE/Simulation Program for it will not only make troubleshooting IMO a lot easier but I will probably help you have a better understanding of all this theory crap, that is if you have one of those of many ELEC project books/tutorials out there.

Some would disagree.

SPICE has a few underlying assumptions which are necessary for a simulation environment but don't necessarily map well to actual circuits. Using it without having a feeling on how the circuit is supposed to behave will lead you on wild goose chases, especially if you're trying to 'fix' an actual circuit in the process.

As mentioned in the linked thread, one surefire way of getting a better understanding of all this theory crap is to analyze existing, known-good circuits. Print 'em out, and work out the DC operating points with pencil and paper. Once you feel confident doing that, take a stab at small-signal AC behavior. Treat it as your Sunday crossword puzzle.

JDB.

 

[ENS Audio]

BTW to all fellow newbies I STRONGLY recommend getting into learning any PSPICE/Simulation Program for it will not only make troubleshooting IMO a lot easier but I will probably help you have a better understanding of all this theory crap, that is if you have one of those of many ELEC project books/tutorials out there.

Some would disagree.

SPICE has a few underlying assumptions which are necessary for a simulation environment but don't necessarily map well to actual circuits. Using it without having a feeling on how the circuit is supposed to behave will lead you on wild goose chases, especially if you're trying to 'fix' an actual circuit in the process.

As mentioned in the linked thread, one surefire way of getting a better understanding of all this theory crap is to analyze existing, known-good circuits. Print 'em out, and work out the DC operating points with pencil and paper. Once you feel confident doing that, take a stab at small-signal AC behavior. Treat it as your Sunday crossword puzzle.

JDB.

Most likely I'll sound "dense" for asking this...when you mean printing out known good circuit you are just referring to any Schematic design thats confirmed to be correct and just start with reviewing formulas, nodes, branches, e.t.c??

There is a pdf book that I have which( dunno the name off hand) has all these easy to do projects/tutorials for understanding how to apply Ohm's Law, e.t.c in a real-time scenario.


Btw, my Green Pre Schematic/Simulation is almost finished and if someone to whom knows how to translate a Proteus DSN file ext to >>> [xyz Schematic/Simulation program] I'll be more than happy to post my results, when then would reduce the amount of crap and confusion thats out there about troubleshooting and circuit analysis.

If I can do my part to help prevent days worth of wasted time for those newbies out there!! ;D

 

[jdbakker]

BTW to all fellow newbies I STRONGLY recommend getting into learning any PSPICE/Simulation Program for it will not only make troubleshooting IMO a lot easier but I will probably help you have a better understanding of all this theory crap, that is if you have one of those of many ELEC project books/tutorials out there.

Some would disagree.

[...]

As mentioned in the linked thread, one surefire way of getting a better understanding of all this theory crap is to analyze existing, known-good circuits.

Most likely I'll sound "dense" for asking this...when you mean printing out known good circuit you are just referring to any Schematic design thats confirmed to be correct and just start with reviewing formulas, nodes, branches, e.t.c??

You make it sound so complicated. Print out the schematics and start with getting ballpark estimates. Let's take the Green pre schematic, for example. A resistor divider sets the voltage at the + inputs of U1a and U1b. Schematic says resistances are 2.2k and 10k, supply voltage is 15V, so said inputs sit at 10k*(10k+2.2k)*15V = 12.3ish Volt compared to ground.

Meanwhile, any op-amp that's hooked up properly will do its darndest to keep the voltage between its input terminals equal. We know the Green works, so it's a safe assumption that it is successful in doing so. (For now, never mind how its feedback loop achieves this, just remember that the op-amp can only achieve this by changing the voltage at its output, which in a real circuit can never swing beyond the supply rails). So the voltage across the 2.2k resistors connected between +15V and the transistors' collectors (and the op-amps' - input) is equal to the voltage across the 2.2k resistor of the voltage divider connected to the + inputs. This divider sees a total of 15V, of which ~12.3V was found to appear across the 10k resistor, leaving 15-12.3=2.7V across said 2.2k resistors. Ohm's law yields that the current through those resistors must be (2.7V/2.2k) or about 1.2mA. That 1.2mA has to go somewhere, and as only a negligible amount of current flows into or out of an op-amps input terminals, all of it has to flow into the transistors' collectors.

One simplified view of a transistor is to treat it as a current amplifier, which takes a small base current to steer a larger collector current. The current amplification or hFE differs between transistor makes, transistor current and temperature, but 'somewhere over 100' is a good bet for a small-signal transistor made in the last 20 years. If you want to be particular, the datasheet for the BC550B in the Green schematic lists hFE@2mA as 290 typical, 200 min, 450 max. So in the typical scenario, both transistors have a base current of 1.2mA/290 = 4uA (or 6uA for a specimen with hFE on the low side of acceptable, vs 2.7uA for the overachiever in the batch). So what's the voltage on the transistors' bases? As we're just doing a DC analysis you can forget all about those pesky capacitors for now, that leaves a 10k and a 47k resistor in series between the bases of each of the transistors and ground, for 57k total. Ohm again with the base currents: the bases sit at 57k*4uA = 230mV typ, 350mV for a min hFE part, 150mV for max hFE. All this is below ground, so to be proper I should list it as -230mV/-350mV/-150mV.

...and I'll bet you that's not exactly what SPICE said.

If you use SPICE as your test bench without understanding both the circuit and the SPICE models you use, you'll spend a lot of time chasing your tail. Analyze the circuit with Mr. Ohm and a few rules of thumb (most of which I mentioned earlier) and you'll know when a measured voltage somewhere in your circuit is within acceptable tolerances or not.

If you're even semi-serious about this: buy the Art of Electronics.

JDB.
[typing this with a fever and a headache, so some of the math may be off]

 

[ENS Audio]

BTW to all fellow newbies I STRONGLY recommend getting into learning any PSPICE/Simulation Program for it will not only make troubleshooting IMO a lot easier but I will probably help you have a better understanding of all this theory crap, that is if you have one of those of many ELEC project books/tutorials out there.

Some would disagree.

[...]

As mentioned in the linked thread, one surefire way of getting a better understanding of all this theory crap is to analyze existing, known-good circuits.

Most likely I'll sound "dense" for asking this...when you mean printing out known good circuit you are just referring to any Schematic design thats confirmed to be correct and just start with reviewing formulas, nodes, branches, e.t.c??

You make it sound so complicated. Print out the schematics and start with getting ballpark estimates. Let's take the Green pre schematic, for example. A resistor divider sets the voltage at the + inputs of U1a and U1b. Schematic says resistances are 2.2k and 10k, supply voltage is 15V, so said inputs sit at 10k*(10k+2.2k)*15V = 12.3ish Volt compared to ground.

Meanwhile, any op-amp that's hooked up properly will do its darndest to keep the voltage between its input terminals equal. We know the Green works, so it's a safe assumption that it is successful in doing so. (For now, never mind how its feedback loop achieves this, just remember that the op-amp can only achieve this by changing the voltage at its output, which in a real circuit can never swing beyond the supply rails). So the voltage across the 2.2k resistors connected between +15V and the transistors' collectors (and the op-amps' - input) is equal to the voltage across the 2.2k resistor of the voltage divider connected to the + inputs. This divider sees a total of 15V, of which ~12.3V was found to appear across the 10k resistor, leaving 15-12.3=2.7V across said 2.2k resistors. Ohm's law yields that the current through those resistors must be (2.7V/2.2k) or about 1.2mA. That 1.2mA has to go somewhere, and as only a negligible amount of current flows into or out of an op-amps input terminals, all of it has to flow into the transistors' collectors.

One simplified view of a transistor is to treat it as a current amplifier, which takes a small base current to steer a larger collector current. The current amplification or hFE differs between transistor makes, transistor current and temperature, but 'somewhere over 100' is a good bet for a small-signal transistor made in the last 20 years. If you want to be particular, the datasheet for the BC550B in the Green schematic lists hFE@2mA as 290 typical, 200 min, 450 max. So in the typical scenario, both transistors have a base current of 1.2mA/290 = 4uA (or 6uA for a specimen with hFE on the low side of acceptable, vs 2.7uA for the overachiever in the batch). So what's the voltage on the transistors' bases? As we're just doing a DC analysis you can forget all about those pesky capacitors for now, that leaves a 10k and a 47k resistor in series between the bases of each of the transistors and ground, for 57k total. Ohm again with the base currents: the bases sit at 57k*4uA = 230mV typ, 350mV for a min hFE part, 150mV for max hFE. All this is below ground, so to be proper I should list it as -230mV/-350mV/-150mV.

...and I'll bet you that's not exactly what SPICE said.

If you use SPICE as your test bench without understanding both the circuit and the SPICE models you use, you'll spend a lot of time chasing your tail. Analyze the circuit with Mr. Ohm and a few rules of thumb (most of which I mentioned earlier) and you'll know when a measured voltage somewhere in your circuit is within acceptable tolerances or not.

If you're even semi-serious about this: buy the Art of Electronics.

JDB.
[typing this with a fever and a headache, so some of the math may be off]

I appreciate your advice and even as I type this ealier I've been getting more of a better idea as far as what i'm doing in SPICE.
By the time I start getting my formal education, i'll be "ahead of the game" that is if my brain doesnt shutdown due to boredom.

 

[PRR]

> There is a pdf book that I have which( dunno the name off hand) has all these easy to do projects/tutorials for understanding how to apply Ohm's Law

"please post ...links for they're may likely be some manuals/pdfs that I dont have"

> when you mean printing out known good circuit you are just referring to any Schematic design thats confirmed to be correct and just start with reviewing formulas, nodes, branches

For a guy with a fever, JDB sure went overboard. He should be doing soup and bed-rest.

Start with something SIMPLE. If I had you in my class, you'd be doing this before looking at circuits with more than 3 pieces.

What is the collector voltage? +/-20% is plenty good. Assume no-leakage Si parts and no Vbe breakdown.

1) R1=10K, R2=1Meg, Beta=100.

2) R1=10K, R2=1Meg, Beta=50.

3) R1=1 ohm, R2=100 ohms, Beta=100.

4) Same as 3; What will SPICE say?

Yes, #3 and #4 are trick questions which may expose a difference between Human and SPICE. (Altho some newer SPICE-result tools may have a clue.)

 

[ENS Audio]

> There is a pdf book that I have which( dunno the name off hand) has all these easy to do projects/tutorials for understanding how to apply Ohm's Law

"please post ...links for they're may likely be some manuals/pdfs that I dont have"

> when you mean printing out known good circuit you are just referring to any Schematic design thats confirmed to be correct and just start with reviewing formulas, nodes, branches

For a guy with a fever, JDB sure went overboard. He should be doing soup and bed-rest.

Start with something SIMPLE. If I had you in my class, you'd be doing this before looking at circuits with more than 3 pieces.

What is the collector voltage? +/-20% is plenty good. Assume no-leakage Si parts and no Vbe breakdown.

1) R1=10K, R2=1Meg, Beta=100.

2) R1=10K, R2=1Meg, Beta=50.

3) R1=1 ohm, R2=100 ohms, Beta=100.

4) Same as 3; What will SPICE say?

Yes, #3 and #4 are trick questions which may expose a difference between Human and SPICE. (Altho some newer SPICE-result tools may have a clue.)

Very good example...(but im sure the compliments of your "know-how" must get old after awhile  ;D)

Let me ask you..for those like me to whom are ridden with ADD or multitask 24/7 and do not have adequate time to know all the formulas/tables via photographic memory;

Do you have recommendations to a website, book or section of a book that focuses on tables and formulas for solving equations that pertains to caculating values for resistors, transistors, diodes, e.t.c.?? (Hope that I make sense atm)

The Mini Engineers Ref book seems to be very useful...

There are a few others (forget the name) that explain everything very well such as "Practical Electronics for Inventors" which is the main book i've used to really understand Elec. Theory 101.....

BUT.....

I'd like something a bit more "bare bones" in practicality (less "in detail" of theory) which may have very simple examples (word problems) I dont mind using a few different books on the same subject as long as it can serve mainly as a "Reference Guide"

I'll probably remember the rest of the names of the books that Im looking for...(Had 2 1/2 hrs of sleep last night, moving equipment around and "reorganizing" my "Bench" and Sorting out files on my PC...Eeek!!!)


BTW...I aready promised myself not to create new threads or posts when in a "brain-dead" state of mind from too many hrs spent working on the "Bench" ???

 

[ENS Audio]

Now that my brain has finally processed what Jdbakker was talking about, I understand almost 100% IMO it seems SPICE in theory is just like any sort of calculator.  If you INPUT the wrong info, you'll get the wrong OUTPUT.  Also SPICE seems more practical for Prototyping purposes if im not mistaken.