John wrote, "it includes a Spice model, but of course I don't know how good it is."

I am looking for that SPICE model, but I'm not sure if I am seeing it.? Are you referring to the schematic diagram in Figure 3, in the ResearchGate article?? Or maybe Figure 5 (same article)?? Both circuits are missing considerable information, which I think makes them rather useless.

Now mind you, I have not read either article.? I only skimmed them.

I find it peculiar that the two papers share some identical figures, yet there are no authors in common to both papers.

Andy

Hello,

I saw that paper. This is actually the model I am trying to implement. For the mentioned paper, I don't know how to implement the SDD component in Figure 5. How the equation 2 of that paper can be implemented in the spice model? What are the ports in LTspice? Also, how to use this SRD as a component in other circuits/schematic? These are some of my questions. Thanks.

开云体育

We do not encourage posting links to 3rd party sites, but Research Gate is reliable, so:

\it includes a Spice model, but of course I don't know how good it is.

Could someone tell me how can I model a SRD in LTspice and use that as a component in other circuits? I am new in LTspice and don't have much idea. I am attaching a link to a paper.

?

I am trying to model SRD according to figure 1 based on equation (1) and then design circuits using SRD as component like figure 2 of the paper. Any help would be much appreciated.

Thank you for your response. I have uploaded the pictures of the figures and equation. /g/LTspice/album?id=295221

Please let me know if you can't access the photos. I will really appreciate your help regarding SRD modelling. Thanks.

开云体育

On 22/05/2024 23:06, abid wrote:

Could someone tell me how can I model a SRD in LTspice and use that as a component in other circuits? I am new in LTspice and don't have much idea. I am attaching a link to a paper.

?

I am trying to model SRD according to figure 1 based on equation (1) and then design circuits using SRD as component like figure 2 of the paper. Any help would be much appreciated.

Unfortunately, many of us don't have access to ieeexplore without paying $$, so we can't see the figure 2 you are referring to. If you are able to, it would help if you could transcribe it into an LTspice schematic and upload that, even if you don't have a proper symbol for SRDs. (I do, though.)

I have done a fair bit of modelling of SRDs. I'll try to dig out something for you tomorrow, as it's too late here now.

--
Regards,
Tony

Could someone tell me how can I model a SRD in LTspice and use that as a component in other circuits? I am new in LTspice and don't have much idea. I am attaching a link to a paper.

?

I am trying to model SRD according to figure 1 based on equation (1) and then design circuits using SRD as component like figure 2 of the paper. Any help would be much appreciated.

Thanks everyone,

OK, Andy I'll look into it now. thanks.

john23 asked, "But its for PSPICE.What could be done to convert it properly into LTSPICE?"

Probably nothing.

I might be remembering incorrectly, but I think you were already using that UCC21220 model, before eetech00 made the UCC5304 model.? Most PSpice models are SPICE models, and they work in most SPICE simulators.? That includes LTspice.? ?Chances are it does not need any conversion.? But there are occasional exceptions.

Andy

Hello, I want to test also the suggested UCC21220A.
But its for PSPICE.What could be done to convert it properly into LTSPICE?
Or i just need to treat it like LTSPICE and use the create symbol in the subcircuit?
Thanks.

https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1227885/ucc5304-spice-model

开云体育

Yes. you can't use ^ fro exponentiation in LTspice,, except in Laplace expressions. Use **.

I believe I solved it by using V=sin(pi*20e3*time*time) instead.

The other problem is that you used the wrong formula for a varying frequency.

Yes it seems to be that "sin(pi*20e3*time*time)" would result in a frequency equal to "time".? But that is incorrect, unless the variable "time" is a constant with a time-derivative of zero.? It's the mathematically wrong formula.? Don't get me wrong; you will get a frequency that isn't fixed and changes over time, and maybe that's "good enough" for you, but its frequency is not the value of "time".

Use LTspice's Modulate function instead, and give it a voltage (on its "FM" input pin) that varies linearly with time.? That has the desired result.

Andy

开云体育

I am trying to sweep frequencies to 20K with the behavioral voltage source using V=sin(pi*20e3*time^2). I get no error, yet I don't get the same results as I do using a wav file.

I guess you forgot that "^" means exclusive-OR, not? exponent (except in Laplace).

--

I believe I solved it by using V=sin(pi*20e3*time*time) instead.

Hello,

I am trying to sweep frequencies to 20K with the behavioral voltage source using V=sin(pi*20e3*time^2). I get no error, yet I don't get the same results as I do using a wav file.

OPA855

* OPA855 - Rev. A
* Created by Sean Cashin; 2020-06-04
* Created with Green-Williams-Lis Current Sense Amp Macro-model Architecture
* Copyright 2020 by Texas Instruments Corporation

OPA855 does not meet the bandwidth requirement as I mentioned.
But thank you very much for your support.

On Tue, May 21, 2024 at 03:23 PM, Mirza wrote:

OPA855

* OPA855 - Rev. A
* Created by Sean Cashin; 2020-06-04
* Created with Green-Williams-Lis Current Sense Amp Macro-model Architecture
* Copyright 2020 by Texas Instruments Corporation

On Tue, May 21, 2024 at 09:49 PM, Andy I wrote:

Mirza, I'm assuming that you want an actual purpose-built transimpedance amplifier, one that has everything in it already.? Right?? They are considerably less common than ordinary op-amps, of course.? I wonder what characteristics other than "very high speed" that your TIA needs.

Yeah I want an actual purpose-built transimpedance amplifier like OPA855 from Texas Instruments, LTC6268-10 from Analog Devices. However, these can not meet my bandwidth requirements for gate mode circuit (in my case). Thus, I am looking for spice models from other companies, however, they do not have (or provide) spice model for their high speed TIA.

According to your comments "It is possible to make a generic TIA model, but probably not easy to make one that mimics characteristics of an actual part. ?The answer depends on how closely you need the model to approach the actual part. ?A generic model would lack things like nonlinear behavior. ?Datasheets might not tell you enough about that. ?But that's just my opinion."
Yeah I feel that you are right.

eewiz wrote:

Prior to your suggestion, I did not know that .param statements could do IF().

This opens a whole new world of controllability that I had not previously understood.

It helps here to read the Help page for the .PARAM command.? It is all documented there.

There might be other shortcuts for temporarily disabling the PWL() section for running a simulation, if that is what you are trying to do.? One is to use a parameter for the "high" current level, instead of 1 Amp.? Then change only the parameter assignment.? Another one, though somewhat kludgy, is to right-click on the PWL text next to the current source, and change it from this:
? ? pwl(0 0 48.2m 0 repeat forever (0 0 +20n 1 +11.99998m 1 +20n 0 +11.99998m 0) endrepeat)
to this:
? ? 0 ; pwl(0 0 48.2m 0 repeat forever (0 0 +20n 1 +11.99998m 1 +20n 0 +11.99998m 0) endrepeat)
by adding "0 ;" before the PWL() part.? The ";" character is a comment character.? LTspice ignores everything after it.? So that changes it from a PWL current source to a 0 Amp source, while leaving all the text still in place where it can be restored later.

Andy