If you want to actually perform this experiement, you'll have to first invent a loudspeaker design that meets the requirements! A much harder nut to crack. And while you are at it, look into the room accoustics as well. Even if you have such a perfect amplifier, you will have no way to verify it (given the stated goal) until these other "minor" problems are resolved.
Very wise and very true indeed
. At least I know my goal, and with my current experiment data, I see a point of light in the dark alley, means it is not impossible. I dont know how far is the light from where I am standing now, but maybe if I walk and walk, I will get to that source of the light.I will not get down if I cannot achieve this. My "teachers" here (like Nelson Pass, John Curl, etc ) have spend their life with audio amps and I believe these guys have reach the limit of what can be done with audio power amp (with current technology).
Maybe a "perfect" amp is never should be made. If it existed, all people in the world will buy only 1 amp. This will make the world (of audio) stop turning.
I notice how the consumer(s) likes to change amps, changes speakers etc. They are looking for the "perfect" gear and in the process, they buy this, buy that....., this makes all the power amp companies in the world can live, the employees can feed their families. This "journey" of the consumer in looking for the perfect gear should be rolling and rolling, never stop because the "perfect gear" does not existed.
I also have the same clue. Prefer amps with flat openloop response, up to 20khz. The only way to achieve this is not using feedback at all or using small feedback.
Ah..... It is happening in low frequency too. Makes the problem more difficult now
The curve have "S" shape. does all phase curve is like this? Always shift phase in lower and upper frequencies?
A view of negative feedback is to consider its action as “implicit inversion”
In feedback theory the feedback attenuation is called Beta, the closed loop gain of a feedback amplifier is then –1/Beta (or 1/Beta +1) in the high loop gain region, the attenuation of the feedback network has been “inverterted” to a gain – the accuracy of the inversion is determined by the loop gain (or Bode’s T, = loop transmission), which is a function of frequency
The same is true for a active integrator circuit, the feedback network transfer function from op amp output to op amp negative input is a differentiator, the negative feedback “inverts” this to give a good approximation of an integrator in the region where loop gain is high
Viewing negative feedback as implicit inversion also works for the op amp’s transfer function itself, if you have a flat gain (constant Beta) feedback network, you can see the integral characteristic of the op amp’s forward gain is “inverted” to a differentiating input across the op amp’s +,- inputs (easy in spice, put a pulse into a op amp constant gain feedback amplifier and look at the input error voltage)
In feedback theory the feedback attenuation is called Beta, the closed loop gain of a feedback amplifier is then –1/Beta (or 1/Beta +1) in the high loop gain region, the attenuation of the feedback network has been “inverterted” to a gain – the accuracy of the inversion is determined by the loop gain (or Bode’s T, = loop transmission), which is a function of frequency
The same is true for a active integrator circuit, the feedback network transfer function from op amp output to op amp negative input is a differentiator, the negative feedback “inverts” this to give a good approximation of an integrator in the region where loop gain is high
Viewing negative feedback as implicit inversion also works for the op amp’s transfer function itself, if you have a flat gain (constant Beta) feedback network, you can see the integral characteristic of the op amp’s forward gain is “inverted” to a differentiating input across the op amp’s +,- inputs (easy in spice, put a pulse into a op amp constant gain feedback amplifier and look at the input error voltage)
Two solutions.
1] Remove the feedback cap and use pure DC coupling with optional DC servo for offset correction, and use a 30Hz high pass 24dB/octave filter using an opamp at input to get rid of subsonic frequencies.
2] Simply use the amplifier in Inverting mode and get rid of high value cap at feedback.
Does any body tried its hands and ears to build and listen to an Inverting amplifier, Acc to my experience with inverting amplifiers they yield 35 % more Damping factor than same amp having Non-inverting mode.
When I first construct this inverting amp and while listening to Bass-content the speaker Cone was having very less vibration and the bass was something to be heard of , full of punch, whereas listening same amp in Non-inverting mode [with cap & also without Dc servo but DC coupled feedback ] results in much lower speaker damping , lot of cone movement and bass was less punchy.
I didnot figure out exactly that why the inverting mode has more damping Maybe there are other reasons involved in it. but i repeated this test with several other amps and the result was same.
correct me , If i were wrong in some way.
regards,
Kanwar
I remember there is an explenation for comparison of inverting and non inverting topology sound in "chip amps" thread, but I don't remember where it is.
I look at it a simple way. This is for bipolar differential. The current in bipolars is 100% function of VBE. Unfortunately it is not linear relation between VBE and current (Ic).
In non-inverting topology, both base(s) of left and right differential are modulated, the left by input signal and the right by feedback signal. The common point is the emitors (which is attached to CCS)
In inverting topology, only 1 base (the left) is modulated by both input and feedback signal, the right base is usually steady at ground point. There is no (or much less) VBE modulation in right base.
So, in inverting topology, the "non linearity" only happens in one transistor, the left transistor. Makes sense if the inverting topology has better sound, because only 1 transistor has modulated VBE. While in non-inverting amp, there are 2 VBE's being modulated.
I look at it a simple way. This is for bipolar differential. The current in bipolars is 100% function of VBE. Unfortunately it is not linear relation between VBE and current (Ic).
In non-inverting topology, both base(s) of left and right differential are modulated, the left by input signal and the right by feedback signal. The common point is the emitors (which is attached to CCS)
In inverting topology, only 1 base (the left) is modulated by both input and feedback signal, the right base is usually steady at ground point. There is no (or much less) VBE modulation in right base.
So, in inverting topology, the "non linearity" only happens in one transistor, the left transistor. Makes sense if the inverting topology has better sound, because only 1 transistor has modulated VBE. While in non-inverting amp, there are 2 VBE's being modulated.
"1] Remove the feedback cap and use pure DC coupling with optional DC servo for offset correction, and use a 30Hz high pass 24dB/octave filter using an opamp at input to get rid of subsonic frequencies." - Kanwar
Try an amp design that doesn't need a servo or coupling C's - like mine are DC-100KHz , Gain 29dB! Any bass rolloff necessary should be tailored in the preamp for the particular source i.e. CD none needed, Phono -3dB 20Hz (RIAA) continuing at 12dB/octave or greater.
Try an amp design that doesn't need a servo or coupling C's - like mine are DC-100KHz , Gain 29dB! Any bass rolloff necessary should be tailored in the preamp for the particular source i.e. CD none needed, Phono -3dB 20Hz (RIAA) continuing at 12dB/octave or greater.
Mr Evil said:
30Hz is subsonic now?!
In Regular technicl and theoratical context frequencies below 20Hz are said to SUBsonic, but In context to environment in Public addressing and professional audio gear people usually refer frequencies below which are inaudible or barely audible at low end is known as subsonic, just said because of the influence from the field operators of sound
amplifierguru said:
Hi Mr. Greg,
All our ampsare designed as Dc coupled with Full Power bandwidth ranging from DC to 250KHZ gain 36dB using mosfets, but we have to use an input coupling capacitor to avoid any external Dc signal entering the amp and tends to change its offset and create problems.
What are your views regarding inverting and non-inverting mode of amplifiers.
sincere regards,
Kanwar
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.