Spice simulation

Jump to Latest
#541
teodorom said:
As I have already said, in trying to get reasonable models starting from, e.g., OnSemi datasheets, I found quite impossible to get consistent SGP models parametes from the various informations therein.
E.g. I found that the IS, NF values one can get from VBE(on) versus IC (for IC<1), are not appropriate when trying to get the other parameters (BF, IKF, ISE, NE) from hFE versus IC.
Now ! the doubt ... are the OnSemi datasheets reliable ?
I looked at the MOSPEC datasheets (http://www.ortodoxism.ro/datasheets/mospec/2SA1302.pdf ) for the 2SA1302 that should be the same as the OnSemi MJL1302A: they provide different information.
Am I wrong ?
Thanks
Click to expand...

Hi Teodoro

I agree absolutely with you in all that you refer in your two posts -and not because your name it is of Byzantine origin - a thing that is obvious from my posts. Unfortunelly no one has the taste to give a response in you. They seems like to live in their "cosmara" (a tricky word in Greek which is the superlative degree of the word cosmos, in the same way we say a beautifull goal "goalara", a beatifull women "gomenara" and so on, generally we use the ending "ara" in everything we want to express it in the superlative degree). So many verbosity around the Spice! Hey guys attention in chandelier to not break it! And you Teodoro and me the Photius and some few others we are models -not SPICE- of a simulation with Jean Baptist Clemans (known also in Christians as "one who speaks loudly in the desert"-from people of course-) and i hope all of others not "ask our head on plate" such as made finally the queen Salome in him so he stop to speak.
I am not sorry if you consider this post out off topic

Cheers to you and in all beloved Italian neighbours
Fotios
 
#542
Linear Purpose said:
Hello guys,

I'm looking for goods models for:
- MJL1302A/MJL3281A
- MJL4281A/MJL4302A

Any body can help ?

Regards.
Click to expand...

There are not good models or bad models. There are only models! Logically thinking i wonder: Who is in place to confirm in me and to you that the model of OnSemi or of Fairchild (for example) for the same type of active device is most reliable?
Think about.

Fotios
 
#543
Re: Re: Re: Re: Re: Filters

Edmond Stuart said:


As long as you are playing with that buggy piece of bloatware from Orcad-Candence (list price $9,995.00), which apparently even lacks the tools for designing filters, I can barely see a good reason why you are whining about the price (starting from $0.00) of Micro-Cap.
Click to expand...

Homo Sapiens Non Urinat In Ventum, therefore I fully agree with you.
 
#544
Re: Re: Re: Re: Re: Filters


Hi Edmond

I looked your PGP amplifier in your webpage. Indeed it is very impressive with all the novelties included. Sincerelly i acknowledge that you are front from me by far. Nevertheless i have a remark on your project, but i will refer it only if i have your leave. Also your confirmation with each sincerity from you, that my reply will not cause a friction between us. I am tired from the costless criticism and the perpetual exchanges of posts without significant reasons.

With each respect to you and in your admirable effort.

Fotios
 
#546
Re: Re: Re: Re: Re: Re: Filters

fotios said:
Hi Edmond

I looked your PGP amplifier in your webpage. Indeed it is very impressive with all the novelties included. Sincerelly i acknowledge that you are front from me by far. Nevertheless i have a remark on your project, but i will refer it only if i have your leave. Also your confirmation with each sincerity from you, that my reply will not cause a friction between us. I am tired from the costless criticism and the perpetual exchanges of posts without significant reasons.

With each respect to you and in your admirable effort.

Fotios
Click to expand...

Hi Fotios,

First, thank you for your kind words.

If you have a remark, please feel free to let me know. As I'm not sure whether your comment fits the purpose of this thread (i.e. exchanging models and tips), perhaps it's better to send me an email.

Cheers, Edmond.
 
#547
Re: Re: Re: Re: Re: Filters

Edmond Stuart said:


As long as you are playing with that buggy piece of bloatware from Orcad-Candence (list price $9,995.00), which apparently even lacks the tools for designing filters, I can barely see a good reason why you are whining about the price (starting from $0.00) of Micro-Cap.
Click to expand...

Hi Edmond,

This is a straw-man argument. The one to compare with on price/performance/value is LTSPICE. I'm sure that there are a couple of things that MC does better, and it may have a slightly better user interface, but for the price ($0), LTSPICE seems superior for most of us.

Cheers,
Bob
 
#548
Re: Re: Re: Re: Re: Re: Re: Filters

Edmond Stuart said:


Hi Fotios,

First, thank you for your kind words.

If you have a remark, please feel free to let me know. As I'm not sure whether your comment fits the purpose of this thread (i.e. exchanging models and tips), perhaps it's better to send me an email.

Cheers, Edmond.
Click to expand...

It is O.K., and has not relation with the theoretical part. My aim it is to give you a help (if you need it). It has only relation with the practical implementation of PCB. It has a nice and very carefully placed GND star point. My remark it is about the orthogonal drawed tracks. They don't have the need of mittering the corners? Or for better the rounding of corners? As we know from the basic physics of Faraday's laws, the charge or the negative ions have the inclination to concentrated at peaks such the 90 deg. corners of the tracks on PCB and thus generating unpredictable & unwanted parasitic capacities.
Only that.

Regards and many thanks for your gentlness

Fotios
 
#549
Re: Re: Re: Re: Re: Re: Filters

Bob Cordell said:
Hi Edmond,

This is a straw-man argument. The one to compare with on price/performance/value is LTSPICE. I'm sure that there are a couple of things that MC does better, and it may have a slightly better user interface, but for the price ($0), LTSPICE seems superior for most of us.

Cheers,
Bob ]
Click to expand...

Hi Bob,

Take it easy, Ovidiu is teasing me with "straw man arguments" and I'm teasing him back with the same kind of BS.

BTW, for DIYs it's not the price that matters, rather the "availability". 😀


Cheers, Edmond.
 
#550
Re: PGP

fotios said:
It is O.K., and has not relation with the theoretical part. My aim it is to give you a help (if you need it). It has only relation with the practical implementation of PCB. It has a nice and very carefully placed GND star point. My remark it is about the orthogonal drawed tracks. They don't have the need of mittering the corners? Or for better the rounding of corners? As we know from the basic physics of Faraday's laws, the charge or the negative ions have the inclination to concentrated at peaks such the 90 deg. corners of the tracks on PCB and thus generating unpredictable & unwanted parasitic capacities.
Only that.

Regards and many thanks for your gentlness

Fotios
Click to expand...

Hi Fotios,

Sure, the PCBs does leave room for such improvements. As a matter of fact, it was a haste job (because of a time limited special offer of the PCB manufacturer).

Cheers, Edmond.
 
#552
teodorom said:
Now ! the doubt ... are the OnSemi datasheets reliable ?
I looked at the MOSPEC datasheets (http://www.ortodoxism.ro/datasheets/mospec/2SA1302.pdf ) for the 2SA1302 that should be the same as the OnSemi MJL1302A: they provide different information.
Am I wrong ?
Click to expand...

Well, if we assume the datasheets are wrong, then the whole universe implodes! 🙂 (joke).

But seriously, one can never really know without trying to duplicate their measurements. That requires lots of equipment that I do not have and cannot afford to own. Faced with that situation, the only option I have is to trust the datasheets. This is certainly far from ideal, but I don't know of a better alternative.

In trying to get the best possible models, there are still some things I haven't explored. Pete B. mentioned the variation of Rb with base current. Also, some of the measurements are taken with Vce=5 Volts. This may be low enough that quasi-sat is playing a part. I really don't know, and can only speculate. In trying to fit all these parameters, the data required to do so is often just not there.

It ends up being a case of just finding the best possible compromise given the limited data available in the datasheets. I've found the model parameter extraction to be a trap that can take up huge amounts of time. It can be so much time that it gets in the way of design.

Regarding the MOSPEC vs. OnSemi datasheets, it appears these are quite different devices, despite the similarity of part numbers. Breakdown voltages differ, and the fT of the MOSPEC devices looks to be inferior to the OnSemi devices by quite a lot.
 
#556
So ... I was wrong ! they are not the same transistor, so the datasheets are different.
My problem with the parameter extraction is that the values for (IS, NF, ISE, NE, BF, IKF, RB, at least ...) should produce good fit both (simultaneously) for IB(VBE) and for IC(VBE) (or, equivalently, for hFE(IC)).
There are transistors for wich I get quite good results, other for wich situation is more desesperate. Strange enough when I'm able to get reasonable results with my methodology, I can get good results using PSPICE Model Editor.
Imagine Kepler discovering that its laws worked for Hearth but not for Jupiter ...
In any case, what makes me crazy about the MJL1302A datasheets is the flatness of hFE(IC) for low values of IC. The 2SA1302 are less flat, so that the SGP model fits better.
 
#557
... After few search on gummel Poon extraction topic it seams that it is a quite serious job - ie there are some research publications on the suject....
One approach that seams to me quite interisting (I found that on the web) is the use of genetic algorithms. It is probably quite efficient to avoid local extremas and an overall accuracy.
 
#558
The literature is full of "solutions" of this problem. I have access to IEEE, so I have collected a lot of papers where the authors stated they have solved the problem. Unfortunately it is a lot of achademic work, nothing really operative.
I guess that the "service" companies that offer SGP parameters extraction have their own instruments and tools. On the other side I guess that the information offered in the datasheets are not enough to do that task (when they are not contradictory).
GA's ? yes, a lot of people say that they have successufully used them. Yes, they say.
Me too I tried to use GA's (with Mathematica): after two hour run I stopped the search. And I'm not even sure that a global minimum (even properly constrained) is a "good" minimum.
I asked for some help. Unfortunately people (excluding andy_c) here seems more interested in (often fake) models. Sorry about that.
 
You must log in or register to reply here.
Top Bottom