Strange simulation results

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Post Reply
ilya
Posts: 40
Joined: Tue Dec 29, 2009 3:10 pm
Location: Russia
Contact:

Strange simulation results

Post by ilya » Thu Oct 12, 2017 2:53 pm

I'm playing with a simple DI circuit in the simulator and here's a very strange thing that's happening. I'm following the output of the DI with a basic non-inverting op-amp block. When I use JFET op-amps, everything is fine. But when I put BJT op-amp, the input and output pins show a huge DC offset. I tried another DI topology with FET input, and the simulation behaves the same. Strangely enough, when connecting the generator to the op-amp directly, everything is fine and works as it should.

Any ideas?

EDIT. Damn! Found the culprit - forgot to put a resistor to GND for proper AC coupling. Doh!
Attachments
weirdSim.jpg

User avatar
JR.
Posts: 2025
Joined: Sat May 24, 2008 7:21 pm

Re: Strange simulation results

Post by JR. » Thu Oct 12, 2017 5:05 pm

The JFET shouldn't have worked either.... I grew up before sims....

JR

User avatar
mediatechnology
Posts: 2681
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Strange simulation results

Post by mediatechnology » Fri Oct 13, 2017 8:18 pm

JR. wrote:
Thu Oct 12, 2017 5:05 pm
The JFET shouldn't have worked either.... I grew up before sims....

JR
Yeah, you gotta wonder why the JFET version sim'd OK.
"There does seem to be a pattern where, the more difficult a truth is to stomach, the higher the hysteria levels go internally at Google."
Visit RockmoreLabs on Etsy https://www.etsy.com/shop/RockmoreLabs
http://www.ka-electronics.com

User avatar
JR.
Posts: 2025
Joined: Sat May 24, 2008 7:21 pm

Re: Strange simulation results

Post by JR. » Sat Oct 14, 2017 8:59 am

mediatechnology wrote:
Fri Oct 13, 2017 8:18 pm
JR. wrote:
Thu Oct 12, 2017 5:05 pm
The JFET shouldn't have worked either.... I grew up before sims....

JR
Yeah, you gotta wonder why the JFET version sim'd OK.
Back in the early days, sims were only as good as the models engineers loaded into the software and it was kind of like you had to know the answers to ask it the right questions.

I rolled my own DIY filter design sim, but never found circuit design sims very trustworthy, I'd much rather melt some solder.

JR

User avatar
mediatechnology
Posts: 2681
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Strange simulation results

Post by mediatechnology » Sat Oct 14, 2017 9:45 am

I'd much rather melt some solder.
Me too. Where the rubber hits the road.

My spreadsheet calculators for resistor values, CMRR etc. are kinda handy but I'm not sure I believe them until I confirm it with a scientific calculator and then finally melt the solder.
"There does seem to be a pattern where, the more difficult a truth is to stomach, the higher the hysteria levels go internally at Google."
Visit RockmoreLabs on Etsy https://www.etsy.com/shop/RockmoreLabs
http://www.ka-electronics.com

User avatar
JR.
Posts: 2025
Joined: Sat May 24, 2008 7:21 pm

Re: Strange simulation results

Post by JR. » Sat Oct 14, 2017 11:35 am

My filter software saved me many hours dialing in multipole filters (including embedded pre/de-emphasis networks). With the program (written in basic on my old H-11) I could plug in real cap values and see how close i could get to a target response... before melting solder.

OTOH I had a bad habit of using some semiconductors in unconventional ways with body diodes and intrinsic substrate diodes affecting real outcomes.

JR

User avatar
mediatechnology
Posts: 2681
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Strange simulation results

Post by mediatechnology » Sat Oct 14, 2017 1:56 pm

IIRC one of the things Rosalfonso and I had to accommodate when he simulated the Phantom Menace II is how power supplies were modeled.

The simulator treats them like stiff batteries or 0Ω sources able to sink and source current.
In the real world, with real voltage regulators, the results were entirely different.
"There does seem to be a pattern where, the more difficult a truth is to stomach, the higher the hysteria levels go internally at Google."
Visit RockmoreLabs on Etsy https://www.etsy.com/shop/RockmoreLabs
http://www.ka-electronics.com

Post Reply

Who is online

Users browsing this forum: Baidu [Spider], Bing [Bot] and 6 guests