Hi.i am using multisim 10 for my simulations.could anyone tell me how to plot loop gain to see if my amp will be stable or not??
what we need to investigate is open loop gain or loop gain in an amplifiers stability?
what we need to investigate is open loop gain or loop gain in an amplifiers stability?
I believe the consensus is loopgain.
Look at the example Audioamp in the examples for LT spice for circuit setup.
Then dB(V(AC1+)/V(AC1-) ) and P(-V(AC1+)/V(AC1-)). This is for pspice but you can rearrange the syntax for multisim.
David.
Look at the example Audioamp in the examples for LT spice for circuit setup.
Then dB(V(AC1+)/V(AC1-) ) and P(-V(AC1+)/V(AC1-)). This is for pspice but you can rearrange the syntax for multisim.
David.
examples loopgain.asc and loopgain2.asc in ltspice are used to plot the open loop gain.
example audioamp.asc is refered as to look the gain and phase margins.
which one is better to be used?
if we have a 30db gain amplifier ,can we plot the open loop gain based on the formula on loopgain example,then go at 30db at the open loop gain graph and look at the phase margin corresponding there??
example audioamp.asc is refered as to look the gain and phase margins.
which one is better to be used?
if we have a 30db gain amplifier ,can we plot the open loop gain based on the formula on loopgain example,then go at 30db at the open loop gain graph and look at the phase margin corresponding there??
Really hope I do not make some angry and cause bad speak with this post...
Just read the whole the whole thread. First off good luck. Second, maybe you should read the thread http://www.diyaudio.com/forums/solid-state/154070-watt-sucking-fireball-series.html for some insight. Third, speakers are not loads, they are sources of signal. The left channel drives the right channel which produces a signal and applies it directly to the output and inverting input of the feedback amplifier. This changes the way the amp behaves considerably over the simple notion of a passive load. Fourth, output and driver transistors has lousy S21 parameter whereas the load is very well reflected back to the the last gain stage. It is the stability and ability of the last gain stage to drive the reflected load at the output. Fifth, experience has shown if the amplifier needs a Zobel or output inductor to be stable with the speaker load it does not have enough phase margin. Seventh, I have had almost no luck using simulations with power amps though others claim to. The problem came in with the S21 parameter not properly handled. Eighth, bandwidth and slew rate are not the same thing and almost unrelated. The Miller cap usually effects slew rate but not bandwidth. Ninth, slew rate only needs to be high enough to provide full power output to 25kHz unless ultra sonic power is needed for some reason.
In 35 years of working on and with amplifiers my experience says about 98% are somewhat unstable and are therefore easy to cause oscillation. There are many sources of these oscillations but almost every amp does. Of course mine will not. Phase margin and gain margin under no load are not the same as under a bad load with the result being most amplifiers will oscillate during a portion of the wave into the bad load like a speaker with a complex musical signal. I have successfully used as much as 100dB feedback at 10kHz which achieves very low distortion but is not without some transient issues sonically. Overall pretty good. A good amp will be unconditionally stable or blow the line power fuse when abused with a bad load or something like that. Driving a speaker that is active and reactive is very much like an automotive situation of pulling a heavy trailer. A light weight sedan car does not do nearly as well as a heavy duty truck connected to the trailer. Same for power amps, the amp should not loose its cool behavior when driving the bad load. I test amps as shown here: http://www.diyaudio.com/forums/solid-state/161155-solid-state-testing-tug-war.html The amps ability to perform as a virtual ground for the other channel is a very good method to determine real world stability. Ideal of course is the virtual ground amp will have no signal.
Lots of good luck. Truly do appreciate your wanting to do it yourself. Having tested over 1000 amplifiers I can surely see why and how about anyone who is diligent could do better than most all the off the shelf stuff.
Just read the whole the whole thread. First off good luck. Second, maybe you should read the thread http://www.diyaudio.com/forums/solid-state/154070-watt-sucking-fireball-series.html for some insight. Third, speakers are not loads, they are sources of signal. The left channel drives the right channel which produces a signal and applies it directly to the output and inverting input of the feedback amplifier. This changes the way the amp behaves considerably over the simple notion of a passive load. Fourth, output and driver transistors has lousy S21 parameter whereas the load is very well reflected back to the the last gain stage. It is the stability and ability of the last gain stage to drive the reflected load at the output. Fifth, experience has shown if the amplifier needs a Zobel or output inductor to be stable with the speaker load it does not have enough phase margin. Seventh, I have had almost no luck using simulations with power amps though others claim to. The problem came in with the S21 parameter not properly handled. Eighth, bandwidth and slew rate are not the same thing and almost unrelated. The Miller cap usually effects slew rate but not bandwidth. Ninth, slew rate only needs to be high enough to provide full power output to 25kHz unless ultra sonic power is needed for some reason.
In 35 years of working on and with amplifiers my experience says about 98% are somewhat unstable and are therefore easy to cause oscillation. There are many sources of these oscillations but almost every amp does. Of course mine will not. Phase margin and gain margin under no load are not the same as under a bad load with the result being most amplifiers will oscillate during a portion of the wave into the bad load like a speaker with a complex musical signal. I have successfully used as much as 100dB feedback at 10kHz which achieves very low distortion but is not without some transient issues sonically. Overall pretty good. A good amp will be unconditionally stable or blow the line power fuse when abused with a bad load or something like that. Driving a speaker that is active and reactive is very much like an automotive situation of pulling a heavy trailer. A light weight sedan car does not do nearly as well as a heavy duty truck connected to the trailer. Same for power amps, the amp should not loose its cool behavior when driving the bad load. I test amps as shown here: http://www.diyaudio.com/forums/solid-state/161155-solid-state-testing-tug-war.html The amps ability to perform as a virtual ground for the other channel is a very good method to determine real world stability. Ideal of course is the virtual ground amp will have no signal.
Lots of good luck. Truly do appreciate your wanting to do it yourself. Having tested over 1000 amplifiers I can surely see why and how about anyone who is diligent could do better than most all the off the shelf stuff.
Hi.to which frequency should i take into account the curve of an amplifier simulation?
I mean to say if i am simulating open loop gain or loop gain and i am using output transistors with a transition frequency of 30Mhz should i take into account the results of the curve above 30Mhz ?
for example if there is a gain peak at 100Mhz but until 30Mhz the curve is ok ,should i worry about the gain peak at 100Mhz which is out of the transistor's Ft?
I mean to say if i am simulating open loop gain or loop gain and i am using output transistors with a transition frequency of 30Mhz should i take into account the results of the curve above 30Mhz ?
for example if there is a gain peak at 100Mhz but until 30Mhz the curve is ok ,should i worry about the gain peak at 100Mhz which is out of the transistor's Ft?
Thanks for such a great post
Hi, everyone, I am a new here and just begin my adventure with audio ampifier from ground recently. Thank you for such a great post here. It really give me tons of information about how to get the correct loop gain simulation. See pictures of my recent design. Maybe it's over compensated as a result of following the open loop gain way. I am glad to be here.....
Hi, everyone, I am a new here and just begin my adventure with audio ampifier from ground recently. Thank you for such a great post here. It really give me tons of information about how to get the correct loop gain simulation. See pictures of my recent design. Maybe it's over compensated as a result of following the open loop gain way. I am glad to be here.....
Attachments
karajanlee- you have it very right and no I do not think your amp is necessarily over compensated. I give it an excellent design award. Believe you will find this amp to be exceedingly clear and smooth unlike the other 99% of what you will ever experience. I prefer push pull complementary output though.
Good luck, SUM
Good luck, SUM
- Status
- Not open for further replies.