Need functional SPICE models for common op amps

Hi Jeffrey,

many of the TI's models are optimized for TINA-TI. So why not using TINA-TI? I'm familiar with both simulators and can say that TINA-TI is way easier to handle than LTSpice.

And I want to add a warning: Simulations are a valuable tool. But they never tell the whole truth. This is especially true for the elder models. So don't rely too much on them but carry out very thorough testings with the real world circuits.

Kai

Hi Jeffrey,

for me the LM833 TINA-TI Spice model file works in LTSpice:

LM833.zip

See also this thread:

https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1034332/pga281-ltspice-simulation-error

Kai

Well then are there any other parts in this range that you DO have PSPICE models for? How about MC33078, NE5532, LM358? I couldn't find models for any of them either. I DID find a model for OPA1678 but it doesn't do me any good since delivery is at best 35 weeks from any major distributor. This nonsense about "you must use TINA-TI or we cannot support you" is indefensible because as you could see if you had even bothered reading my very first sentence you don't support that EITHER. As far as LM833 are you actually claiming you are getting that miserable encrypted HSPICE model to work in TINA-TI? Is that your new "marketing campaign", instead of developing superior products you just develop an entire alternate "SPICE ecosystem" that breaks the tool that everyone else knows and trusts so nobody can really tell how bad YOUR products are? From what I'm seeing so far it wouldn't surprise me in the least! Your products are getting like Google support of Android phones, you get two years from model release and after that we're no longer supporting software updates to keep the malware away, after that if you want to actually continue to use our phones then buy a new one you cheapskate! Please "stay in your own lane" and provide usable SPICE models for your products on ALL major simulation platforms, or I will have NO REASON to buy your products.

I'm using LTSpice XVII, you're on the previous release. I can't tell from what you posted where you got your .lib or .sub model for the LM833 because it's not in the .zip file you linked which is just a schematic. It's possible an earlier release was part of the "standard library" in the previous release but it doesn't matter, it's not the same as the one I found that was identified as snlm202.zip. I'm not an expert on which SPICE formats the current version is capable of supporting, but I know they they do try to support PSPICE models where possible. (You can tell from the error log that in the copy I was using there was an obvious bug, several of the nodes were disconnected from each other internally, the model wouldn't even converge!)

Hey Jeffrey,

Because LTspice is not our software, we don't have any internal insight into the program, or much familiarity with it. We are not even able to install LTspice on our computers. Many of our legacy parts do not have models or have limited models, but our development efforts are focused on our new products. The spice models we release on products from at least the last 10 years or so are unencrypted and are able to be used across many simulators. 

Best,
Jerry

Jeffrey,

Jeffrey Lawton said:

I'm using LTSpice XVII, you're on the previous release. I can't tell from what you posted where you got your .lib or .sub model for the LM833 because it's not in the .zip file you linked which is just a schematic.

you have to import the LIB file first, as discussed in the link I gave above:

https://e2e.ti.com/support/amplifiers-group/amplifiers/f/amplifiers-forum/1034332/pga281-ltspice-simulation-error

Kai

Right, in the case of the LM833 that's the encrypted file LM833.LIB inside snlm202.zip.

Jeffrey Lawton said:

Well then are there any other parts in this range that you DO have PSPICE models for? How about MC33078, NE5532, LM358? I couldn't find models for any of them either.

You will find many of them in TINA-TI, in the folder "Spice Macros".

Kai

I have TINA-9, I don't see a directory "Spice Macros", in the directory "SPICELIB" I was able to find a file NSC.LIB (presumably from the Nat Semi merger) which contains as a subset a .MODEL file for LM358, that may help (that is however a very low-slew-rate part, at the low end of where my interest lies), but none of the others are there.

That's for the NE5534 which is externally compensated. I had asked about the NE5532 which is not, nowadays it's a much more common device (although quite old and somewhat eclipsed by the unavailable OPA1678). And the format of your response suggests that your TINA tool is usable by me, while I have explained several times in this post that it is NOT and specifically why that is (missing the ability to correctly model an asymmetrically wound transformer from the specification of the winding inductances and the coupling constant, it only has a model for symmetric transformers and "mutual inductance" whatever that is which is useless) although the "politically expedient response" here in customer "support" is to simply ignore all complaints about legitimately missing functionality in TINA and just "assume" it's all nothing more than a "customer preference issue" which will go away eventually. (Do you have any influence over DesignSoft to get them to close out issues? Apparently they never resolve anything either.) I was able to ferret my way through the many TINA support files and find that the NE5534 subcircuit file was located in the SPICELIB subdirectory in the TI.LIB file, which is the information I WOULD HAVE had the need for if that information had been of use to me, NOT "showing me where it is in the GUI" in a tool that I cannot use. I would be willing (but not eager) to use TINA if the open item with DesignSoft were closed, but for now my entire project is simulated on LTSpice and I haven't found any discrepancies yet. So far you found me a link for an OPA2990, I managed to find one for an RC4580 and an LM358 since this started, I still don't have a functional LM833 subcircuit that will simulate correctly on LTSpice XVII, the simulation failures I'm getting running the subcircuit file from snlm202.zip (the only file I have available) suggest there are some severe flaws in that model, from the prefix this was clearly originally a NatSemi part so you folks own this one.

That "24 hour response" from DesignSoft is now over 72 hours and counting.

OK I'll bite, what purpose does a .tsm file serve? Here's the "best guesses" according to Google (NONE of which is exaggerated):

TwinCat System Manager (a file of instructions for a PLC)

TurboAssembler for OS/2

TRSI Sound Monitor song file

Time-Structured Merge Tree

Toad Query Builder File

(Of course LTSpice XVII has no idea what such a file is for, I'm sure Engelhardt himself doesn't know.)

So yes I suppose if you're intent on going off and completely redefining what SPICE even consists of in terms of file content and extensions, you probably CAN create models in an incompatible format that appear to "break" other highly accurate and successful tools. Here's the thing though, you might THINK this fools some people, but not NEARLY enough, this whole charade is pretty much common knowledge by now, I mean this was over six years ago:

https://forum.allaboutcircuits.com/threads/how-to-include-ltspice-simulation-folder.117134/ 

(and note the models for that example that were found THAT WORKED came from an alternate source!)

If you think this is a "great plan to influence people and sell semiconductors" then you've made a serious tactical error!

If you don't intend to provide a LM833 SPICE model USABLE BY ME then yes go ahead and close the ticket in defeat - serves you right!

Hello Jeffrey,

Unfortunately it looks like we are unable to further assist you. I wish we could have been of more help. Best of luck with your design.