1. Home
  2. Electronics Design
  3. Timestep error in LTSpice

Timestep error in LTSpice

Guess we've all been there. Tried to build a current controlled oscillator around a Schmitt inverter in LTSpice and got dinged. "Timestep too small". Again. Fiddled with the usual things like easing off reltol but this time no matter what I do it looks like it won't converge.

Anyway, are the shipped models for digital devices not useful for this 'semi-analog' stuff? If so I'll scoot over to the lab and do it the way I always used to do it, with the solder iron and a pot of coffee.

So I am just looking for a hint at a quick fix or a recommendation along the lines of 'fuggetaboutit'.

Regards, Joerg

formatting link

Reply to
Joerg
Loading thread data ...

sigh........................................

and the Undo button works

martin

Reply to
martin griffith

Hello Martin,

I used the "off" button on the PC and the "on" button on the Weller :-)

Not a great Spice fan, well, maybe a little more than Bob Pease is. It's ok with chip design where I can use foundry parameters and my own models. But most times when I use a "part in a can" I hit the wall.

Regards, Joerg

formatting link

Reply to
Joerg

Found better models than those shipped, on the Yahoo LTSpice group. Helmut had contributed those (thanks!). Now it can do semi-analog stuff, still not all of it but close enough.

Regards, Joerg

formatting link

Reply to
Joerg

Hi Joerg

I would be interested in that too - have been searching the Yahoo group LTspice for you message, but no such luck. Can you point me in the right direction?

Regards

Klaus

Reply to
Klaus Kragelund

Hello Klaus,

It's not my message that fixed the problem but the 74HC library contribution from Helmut. You can find it in the "File" section. Make sure you load both the 74HC.lib file and install the *.asy part (this is done via his posted 74HC.zip file). Then add the ".include 74HC.lib" to your Spice file and you are in business.

The models are still not ideal. For example I can make it switch at VCC/2 which the HC14 Schmitt will not do in real life. But at least it allows me to simulate some basic behavior.

The really correct method would be to obtain full-fledged device-based models and AFAIK Philips offer quite a few. But I didn't have the time, got to wrap up this project by Friday night.

Regards, Joerg

formatting link

Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.