Spice simulation

[KSTR]

Here we go...

rename the attachment to opsup.sub and store it in the LTspice \lib\sub folder for easy access (no paths requiered)

- place a command .lib opsup.sub in your sim
- use the opamp2 symbol for your opamps and change the value field to opsup or opsupcl.

You may specifiy the usual Aol (default: 1e6) and Gbw (default: 50MegHz) parameters for both plus there are additional ones:
opsup:
- Iq (defines the quiescent current, default is 10mA)

opsupcl (cl stands for clamped -- clamps the output voltage to the supply levels)
- Iq (defines the quiescent current, default is 10mA)
- Rout (defines output resistance, default is 20R)

You can specify those by editing spiceline or spiceline2 in the attribute editor dialog (right-click on the opamp)

If you want to make those permanently visible then copy and edit the opamp2.asy symbol.

- Klaus

[darkfenriz]

Thanks, I'll try that

[Rafael L]

Originally posted by jcx
Hi John,

I believe I could implement Bob's "PIM" measurement in LtSpice

See this to measure TIM

[aparatusonitus]

If anyone have spicemodels for these transistors or know where i can find them, please let me know.

[unclejed613]

Originally posted by Rafael.luc


See this to measure TIM

i've been experimenting with a program called "Spectrum Lab" available as freeware here:
http://freenet-homepage.de/dl4yhf/spectra1.html

it has a DSP function block that you can create custom signals and filters with, and it seems like it would be simple to implement an IM analyzer with it.

[Rafael L]

Thanks unclejed

This software is powerful

[tiefbassuebertr]

Very large threat. But I miss by this intense discussion one thing: The verification how far the simulation results to real life measurements actually differ
If a member the measured values of his amp posted here, I could check by simulation of my software to see how far the simulated results differ (preferably simple circuit - at best a pure power amp device).
What do you like about this?

[Rafael L]

Hi

I wonder if someone did simulation of IMD in LTspice?

http://www.diyaudio.com/forums/showt...71#post1349571

I'm trying to do, but I do not know if is correct(Attached)!

Thanks

[unclejed613]

that looks about right. you may want to set your minimum timestep to about 1us or so and make sure you have data compression turned off. also, look up some of the standard tone pairs used for IMD measurements, as there are some like SMPTE using 60hz+7khz in a 4:1 ratio and measuring the sidebands of the 7khz tone, or ITU-R which uses 19khz+20khz in a 1:1 ratio and measuring any 1khz component that turns up. if you have heard test signals such as the two-tone test used as a warning signal in the Emergency Alert System in the US, it's usually 900hz+1khz in a 1:1 ratio, and any 100hz (and sometimes the 1900hz) component is measured. such a two-tone IMD test is primarily used for radio equipment with a limited audio bandwidth (2khz to 5khz).

[Rafael L]

Originally posted by unclejed613
that looks about right. you may want to set your minimum timestep to about 1us or so and make sure you have data compression turned off. also

Yes I'm doing it, and disabling the compression with the command:
.options plotwinsize=0

Originally posted by unclejed613
look up some of the standard tone pairs used for IMD measurements, as there are some like SMPTE using 60hz+7khz in a 4:1 ratio and measuring the sidebands of the 7khz tone, or ITU-R which uses 19khz+20khz in a 1:1 ratio and measuring any 1khz component that turns up. if you have heard test signals such as the two-tone test used as a warning signal in the Emergency Alert System in the US, it's usually 900hz+1khz in a 1:1 ratio, and any 100hz (and sometimes the 1900hz) component is measured. such a two-tone IMD test is primarily used for radio equipment with a limited audio bandwidth (2khz to 5khz). [/B]

I read that in LTspice that is only as possible to measure in frequencies next.

this website is IMD measurements (Graf-6)

http://www.littlefishbicycles.com/poweramp/

I want to know is how to analyze the graphic, would be the amplitude of residual?
-seems that there is any way to get the value of IMD in %