|
Re: Warning: Multiple definitions of model ...
开云体育No doubt you can carry out an optimization, but it requires a deep knowledge of what each of the model's parameters determines, which is far from straightforward. It's unlikely that a model with
produce curves that closely match those in the data sheet, and
determining whether a match is 'close enough' also requires
deep understanding unless the 'match' is a gross mismatch. ======================================================================================
Best wishes John Woodgate OOO-Own Opinions Only Rayleigh, Essex UK I hear, and I forget. I see, and I remember. I do, and I understand. Xunzi (340 - 245 BC) On 2023-07-10 16:51, Tony Casey wrote:
Good stuff. For accurate model assessment and optimisation, I also have a method of importing the datasheet data into LTspice for direct comparison. It's this latter bit that's onerous, because it requires digitising the datasheet graphs, which can then be overlaid on the simulated results. You can then derive an error function between the two curves which can be minimised by model parameter optimisation. |
|
Re: Warning: Multiple definitions of model ...
开云体育Good stuff. For accurate model assessment and optimisation, I also have a method of importing the datasheet data into LTspice for direct comparison. It's this latter bit that's onerous, because it requires digitising the datasheet graphs, which can then be overlaid on the simulated results. You can then derive an error function between the two curves which can be minimised by model parameter optimisation.--
Regards, Tony On 10/07/2023 17:37,
aburtonline@... wrote:
Fabulous guys. I had thought it might require me to write my own test setup but it's good to hear that these have been made available. I'm not talking about anything exotic here. I only mean BJTs, JFETs, MOSFETs, valves(?) and the like.? I'm off for a dig around in the files! |
|
Re: Warning: Multiple definitions of model ...
开云体育What you say about this is true, hFE does have significant statistical spreads, but the datasheet limits are probably set at 3 sigma (any outside this being rejected, being only a 0.3% loss). 3 sigma covers 99.7% of the total distribution, whereas 1 sigma covers 68%. 95% of the distribution is contained within 2 sigma, so you are very likely to get something not too far away from the typical (or nominal), and the chance of a random device being near the limit is vanishingly small.BTW, you should also search for "testjig", as it's more often used than "curve tracer". --
Regards, Tony On 10/07/2023 15:34, John Woodgate
wrote:
You put your device and its model into a curve-tracer .ASC and set it up to produce (if it will) the same curves as in the data sheet. Compare the results. You can find curve-tracer .ASCs by going to Files on the web site and Searching for 'curve tracer'. For a bipolar, I would start with collector current as a function of collector voltage with base current as parameter. But remember the large variation of current gain between samples, even for graded devices like BC847x. |
|
Re: Warning: Multiple definitions of model ...
开云体育You put your device and its
model into a curve-tracer .ASC and set it up to produce (if it
will) the same curves as in the data sheet. Compare the
results. You can find curve-tracer .ASCs by going to Files on
the web site and Searching for 'curve tracer'. For a bipolar,
I would start with collector current as a function of
collector voltage with base current as parameter. But remember
the large variation of current gain between samples, even for
graded devices like BC847x. ======================================================================================
Best wishes John Woodgate OOO-Own Opinions Only Rayleigh, Essex UK I hear, and I forget. I see, and I remember. I do, and I understand. Xunzi (340 - 245 BC) On 2023-07-10 13:59,
aburtonline@... wrote:
I have searched for this and can't find anything that is generic and at my level but, is there somewhere information on a sensible validation procedure for the simple classes of model?? I mean something like at best a tutorial video going through the process, or a description of the process and what needs to be validated, or maybe at least a simple step by step guide.? I do hear what you experts say on the models being flawed and it would be good to verify the ones we each use regularly.? It's an area I have al;ways meant to dip into but I've never been able to address the first step. |
|
Re: Periodic signal from PWL file
开云体育That’s what I was initially thinking of (and also posted it). What you are mentioning is BNF (Backus-Naur form of descripion of higler level programming language syntax).Only I was missing the left part of the or-sign in the Wiki. Maybe there were meant to be time/value tuples in angle brackets and they were fallen victim to the HTML conversion? — Christoph
|
|
Re: Warning: Multiple definitions of model ...
Alexander Bordodynov's greatly expanded standard.xxx libraries have been available for years. Alex is a highly valued member of the group. I don't know the provenance of the models in his collection, but I know it includes lots of Russian devices that many of us probably can't obtain. A minority users have stated they use them. Good luck to them. But when it comes to fielding questions related to models they have got from there but have forgotten they did so, what are we supposed to do?
toggle quoted message
Show quoted text
Collections of models uploaded to the group's files section are welcome. Even better when they have some sort of traceability or QA. People can use them in the usual 3rd party mode, with the usual proviso of caveat emptor. The ideal situation is that ADI have dozens of testers validating thousands of models that we can trust. That's not the reality. I recently uploaded a MOSFET model that I developed from the datasheet. I provided a report that compared the model's output to the datasheet. People can make up their own minds about that. It took me about a day to do it, and I have a set of prepared jigs and spreadsheets that have evolved over the years. On that basis it would take me over 5 years to validate 1875 BJT models if I didn't have anything else to do. But... we don't need 1875 BJTs. Most likely, many of them are not even available any more. Many devices are so similar that in 98% of cases, they are completely interchangeable. We don't need 10 devices that are essentially the same. But some are not. There are few alternatives to the Onsemi ring emitter high power audio BJTs, especially the ThermalTrak ones, for example. Sadly, they aren't in the standard library. You can download them from Onsemi. You can also download a sub-set of those (and others) from Bob Cordell's website, that he produced to address supposed deficiencies in the models from Onsemi - you can get those . Bob is a highly respected audio designer, but as far as I know neither he nor Onsemi have published any quality control documents on the efficacy of said models. Instead, there are usually disclaimers (from manufacturers) to not rely on them. Say what? Some people often say you can't trust simulations -? that you must build and test every circuit. Maybe they're right on that basis. I guess when Intel introduce a new processor with 10 billion transistors, they have previously made a dozen or so cut'n'strap breadboards to prove it works. As far as I can tell, ADI are not validating discrete devices for the standard.xx libraries. Instead there are periodic announcements of xxx devices kindly contributed by XYZ Semiconductor Corporation. But when we check (I mean "I"), these are not always the same as those directly available from the original producer's website. In the case of the JFETs, we must therefore conclude they have been mangled by ADI in the process of translation to the standard library file format. As far as I'm concerned, I would rather have 100 models I can trust, rather than a 1000 that I can't. At the moment, it seems, we don't have either. --
Regards, Tony On 10/07/2023 00:05, John Woodgate wrote:
|
|
Re: Warning: Multiple definitions of model ...
开云体育Some valid points, but I think the proposal is not 'either/or'. I see no objection to putting the expanded files in a new subfolder of Files, as long as it has a warning about how to use it and not use it. Not everyone has Tony's skills
in verifying a model. Some of us need more support. ======================================================================================
Best wishes John Woodgate OOO-Own Opinions Only Rayleigh, Essex UK I hear, and I forget. I see, and I remember. I do, and I understand. Xunzi (340 - 245 BC) On 2023-07-09 22:02, Tony Casey wrote:
|
|
Re: Warning: Multiple definitions of model ...
I'm not against a repository of models - we already have one in the group 's Files section. But I am against trying usurp the standard libraries. Most people coming to this group, at least initially, will have only the installation files. Once people have learnt to deal with 3rd party models, they can use what they want. We have had endless issues over the years with people adding extra models to the standard libraries and then wondering why no one can run their problematic schematics, when they believe they have uploaded everything necessary. Many of us of us are not satisfied with ADI's chaotic recent efforts with the standard libraries, but the solution is not additional chaos. Although this is a big group, it still only represents a fraction of the LTspice user base. We may not like the current situation, but we need to concentrate on enabling people to use LTspice despite the current issues, by freeing them from the initial dependency on the standard libraries. ADI's current preoccupation is making sure that their own example schematics run with what is what is currently in the standard libraries. That is where they see the critical value-added issues. People complaining about multiple BC848s are way down in the noise, because they are probably not buying lots of ADI's premium products. For heaven's sake, even the multiple BC848s issue is trivial compared to the dozens of JFETs in that library that have massively faulty models. If anything, I think emphasis ought to be on educating users to validate whatever models they are using. Quality trumps quantity. What's the value in having 1875 BJT models that have little to zero provenance? Have you validated any of them? German Ergueta is right when he says that validation is a time consuming issue. I'd like to think that's because he has lots of people validating things. He probably does - just not the standard libraries. Could we do a better job? Possibly. But who's going to do the heavy lifting?? And not get paid for doing it. And would it have a global impact? No. Regards, Tony? On 9 Jul 2023 20:21, eewiz <eewiz@...> wrote:
|
|
Re: Warning: Multiple definitions of model ...
开云体育It sounds a very good idea, but
keeping the same filenames could cause untold confusion. Is
there any way of making the names different?? The only thing I
can think of is making them all upper-case, i.e. STANDARD.BJT
etc., which humans see as different but LTspice doesn't. ======================================================================================
Best wishes John Woodgate OOO-Own Opinions Only Rayleigh, Essex UK I hear, and I forget. I see, and I remember. I do, and I understand. Xunzi (340 - 245 BC) On 2023-07-09 19:21, eewiz wrote:
|
|
Re: Warning: Multiple definitions of model ...
开云体育Hello All, Members could maintain a set of the standard.xxx files for use by the group. I've been doing that for many years now. If the group retained a set of maintained files in an easy to reach repository similar to /files/temp, group members could downloaded them when needed to replace the LTspice versions that appear after an LTspice update. I have a set of standard.xxx files that I have maintained across LTspice updates since LTspice IV was all the rage. ?????????????????????? ? ?? # OF DEVICES FILE??????????????? MAINTAINED? LTspice standard.bjt?? ? ? ?? 1875 ?????? ?? 306 standard.dio???????? 2575 ?????? ?? 928 standard.jft ? ?????? 1306??????????? 140 standard.mos????? 1710????????? 1207 The library files could be kept in a single zip that could be easily dropped onto the \cmp directory to clobber the undesirable LTspice official standard.xxx files after an LTspice update. Of course there are no guaranties of model quality from anyone including the author of LTspice but, I much prefer having 1875 BJT models to choose from as opposed to the 306 model provided by AD. The paradigm could then be changed from "missing BJT model, be sure to include your model in your working directory" to "Missing BJT model, be sure to download the set of standard.xxx files from the group." Then group members could leave the substandard standard.xxx files to suffer the rest of the world. If I knew anything about PC programming, I would write something to move the maintained files out of the way, call LTspice into update mode, combine any new devices into the maintained files and finally overwrite the downloaded LTspice files. But, I don't. I write embedded system code in assembler and C, where I get to roll all my own code. I have no idea how to deal with calling a zillion pieces of somebody else's code to piece together a PC program. So I just maintain my own copies of the standard.xxx files that I process in a text editor. Back in the 90's I used to easily accomplished such tasks with dBase/Clipper but that capability disappeared for me when M$ abandoned 16-bit software altogether. All for now
|
|
Re: Warning: Multiple definitions of model ...
开云体育Output is engraved on stone
tablets. ======================================================================================
Best wishes John Woodgate OOO-Own Opinions Only Rayleigh, Essex UK I hear, and I forget. I see, and I remember. I do, and I understand. Xunzi (340 - 245 BC) On 2023-07-09 16:07, cander shelter via
groups.io wrote:
|
John Woodgate
Tony Casey
Christoph
eewiz