Spice Model doesn't work, is there a fix?

Hi RobertV ,

Will look into this model problem. When you say DEMO, do you mean the jig from LTspice, or the circuits from the LT3081 web page. If the latter, which circuit:

https://www.analog.com/en/products/lt3081.html

If you would like try the device itself, there are demo boards available at the bottom of that page.

Thanks,

mike

Hi RobertV ,

RobertV said:

The DEMO is fine at 5V, but when using 10-25V the output stays at 5.8V or so independent of the programing resistor.

I cannot seem to reproduce this. Are you using the latest model? Can you upload your schematic with the components set to values where you see the behavior.

Thanks.

mike

I'm guessing one of the "other forums" you refer to is this discussion on eevblog. Like I mentioned there, I was unable to replicate any of the issues others were reporting. Others mentioned that the simulation might be affected by SOA limiting, though I have not observed this.

If you want help troubleshooting this, you should attach your schematic and report the version of LTspice you're running (and make sure your component models are up to date as well).

The DEMO is the LT3081_TA01A zip file. It works fine when running version 24.0.9 but when I run 26.0.2 I get the odd behavior. I simply changed the V1 source to 25V and changed the R3 to 400K (20V output), changed Rload to 200 Ohms (100mA)

Okay, I tried this on my home machine which has 26.0.2 installed. Initially I was able to observe the same behavior, Vout stuck around 5.7-5.9V. Then I updated the component library (Tools > Update Components). After that, things work as expected.

Looking at the Model Change Log, I see:

09/01/26 LT3081 Fixed output range.

So looks like the fix just went in yesterday.