IRF610 & switchercad strangeness

[Rescue Toaster]

Been fiddling around with a two-stage DC amp, and when I use IRF's provided model in switchercad, I get REALLY strange results when I do AC analysis. If I do an AC analysis, I get magic currents appearing from nowhere (from INSIDE the IRF610?) flowing through the load.

The biggest problem is when I do something like the D1's volume control, with the shunt across a differential circuit, then the AC analysis goes nuts (output goes UP as shunt resistance goes down = wtf), generating amps of current through the load, yet if I do a DC op point with DC bias across the input, it works fine. (My circuit has no capacitors, so the results should be identical.)

I guess switchercad just doesn't handle differential circuits well at all. It routinely fails on transient analysis and such.

Anyone else had problem doing balanced/diff circuits like pass's stuff in switchercad? I was so pleased with it, and it handles the huge node numbers I have with cascoded current sources, that student pspice won't do.

[metalman]

I've modelled a few of the Pass balanced designs, mostly preamps, using IR's spice model for teh IRF 610 and never encountered problems like what you're describing. Unfortunately I don't have any ideas as to what is going wrong.

[jcx]

you could show us your circuit - change your .asc filename to have .txt extension and you can attach it to a post here

the irf610 model doesn't seem to be included in the distribution lib, maybe your model has compatability problems, you can include the model on your .asc schematic by copying its text into a .inc statement

you have to register to become a member of the yahoo LtSpice user's group but it is a very good resource, you may find your problem addressed in the files or past messages and the level of Spice expertise on the forum is high, the program author incorporates bug fixes into near weekly new releases

[jcx]

updated LtSpice recently?

Tools>Sync Release

from last weeks 2.17h release: [in Changelog.txt]

"04/27/06 Corrected an error monitoring some MOSFET terminal currents."

[Rescue Toaster]

Hey thanks, I'm back on 2.17e, so I'll give it a shot.

Hopefully it improves the transient analysis too. I get a lot of them that cruise along at like 10 nanoseconds per second, ie like 5 years to complete 1/10th of a second of a sinewave.

[GRollins]

I gave up on simulators years ago. There are just too many circumstances where they fall apart.
A great example can be found in the Aleph-X thread. I posted the circuit. People started simulating. Hordes of people told me the circuit wouldn't work based on their simulations. I just smiled, knowing all along that I had a working prototype sitting at my elbow.
If I had simulated the Aleph-X, it would have told me that it wouldn't work, and I never would have followed up on the idea. Fortunately, I had already quit using them.
People tell you that they save time. I disagree. There's no time savings in being lied to. Build the circuit and let reality tell you whether it works or not.
Reality beats theory seven days a week.

Grey

[jh6you]

Originally posted by GRollins

. . . simulations.

I understand the simulation as math, where limitations and assumptions are introduced and mixed up. So that if the limitations and the assumptions have a certain distance from the reality, the simulation might be less useful.

[keantoken]

People use simulators usually because they don't have the resources to just build every prototype they come up with, or they just don't want to waste time with a circuit. However, some experienced engineers don't use them because they have learned how to use their brain as one (good for them). These resources are-

#1: good source of broken equipment
#2: lots of friends
#3: lots of money
#4: own a store
These are the resources I am familiar with, but there are more.

If you are one of the lucky people who have some of these resources, awesome for you, but as for me and others, simulators are more efficient if you know what your doing. jh6you, I think you are a very experienced person, I will not insult another's knowledge unless I have a reason to, and this, is not the reason of my post.

[Babowana]

jh6you, I think you are a very experienced person, I will not insult another's knowledge unless I have a reason to, and this, is not the reason of my post.

Hi Albert Einstein,

I know him . . . jh6you . . . He left diyAudio. He said diyAudio is a very good place, but he thought that he showed his ignorance, mistakes, non-senses and pretendings about electronics too much to the diyAudio members. He seemed to have felt ashamed of those things. I encouraged him saying your decision was very wise.

As far as I know, he is an okay structural engineer, who understands structural things quite okay. He often complains his young engineers as they rely too much on the advanced structural analysis tools, without understanding the structures. He agrees that the tools are good, very useful and absolutely necessary, but after understanding the structures. Hmm . . . maybe, he is right . . . Umm . . . surely, he is not an EE.

Even though you have a good reason to, avoid insulting other knowledges . . . Very often their voices are gathering and loud enough to make you . . . mmm . . . moreover, diyAudio keeps a strict moderator team.

Use indirect and gentle comments . . . I believe you know how . . . Long live in diyAudio!

See you