Maybe my question is not very clever.
But what is to think about when designing a special Bass Amp?
Is it same as 'normal' Amplifier?
But what is to think about when designing a special Bass Amp?
Is it same as 'normal' Amplifier?
How do you want the bass to sound?
Moreover, the use of frequency-specific amplifiers is accompanied by a loss of homogeneity and resolution.
The additional components such as plugs, sockets, power supplies... counteracts at best the washed-out sound of multi-way wooden loudspeakers.
...which is probably why so many people are into "active speakers"-)
Moreover, the use of frequency-specific amplifiers is accompanied by a loss of homogeneity and resolution.
The additional components such as plugs, sockets, power supplies... counteracts at best the washed-out sound of multi-way wooden loudspeakers.
...which is probably why so many people are into "active speakers"-)
It can be .....
A few "tweaks" listed below -
1- Since most Subs (bass) will demand more current on a regular basis and/or be lower impedance , fatter tracks , multiple pairs of output
devices and very robust stability are desirable. Triple EF output stages excel here , as well.
2- Compensation ... Since we are just amplifying 20-200hz , high loop gain <1K is what matters. A large Miller cap can be used , 20Khz THD
is an after thought.
3-More robust thermal considerations , Bass is 80% of most music - power wise. OPS can be more conservatively biased. My 2 A/B sub amps
are just 6mV across the .22R emitter resistors (<30ma per device). My amps run VERY cool on idle.
4 - These factors just call for simplicity and reliability in the design - what MOST OEM subs lack these days. (example below).
PS - Behringer used this same amp for ALL its powered PA and commercial subs. Most failures were main filter caps after decades of use.
I used mine for over 3 years - perfect.
A few "tweaks" listed below -
1- Since most Subs (bass) will demand more current on a regular basis and/or be lower impedance , fatter tracks , multiple pairs of output
devices and very robust stability are desirable. Triple EF output stages excel here , as well.
2- Compensation ... Since we are just amplifying 20-200hz , high loop gain <1K is what matters. A large Miller cap can be used , 20Khz THD
is an after thought.
3-More robust thermal considerations , Bass is 80% of most music - power wise. OPS can be more conservatively biased. My 2 A/B sub amps
are just 6mV across the .22R emitter resistors (<30ma per device). My amps run VERY cool on idle.
4 - These factors just call for simplicity and reliability in the design - what MOST OEM subs lack these days. (example below).
PS - Behringer used this same amp for ALL its powered PA and commercial subs. Most failures were main filter caps after decades of use.
I used mine for over 3 years - perfect.
Attachments
What , are you considering Bi-amping or a sub ?
I won't go back to full range amps , I can fine tune the bass amp too well.
Another thing to consider is the advantage of filtering and adjustable phase shift for the separate bass amp.
I am building a 70/115hz sallen-key LPF feeding a ESP project 103 all pass with variable phase shift for my new e-waste sub.
OS
I won't go back to full range amps , I can fine tune the bass amp too well.
Another thing to consider is the advantage of filtering and adjustable phase shift for the separate bass amp.
I am building a 70/115hz sallen-key LPF feeding a ESP project 103 all pass with variable phase shift for my new e-waste sub.
OS
Attachments
I do not intend to make an amplifier.
My question is only theoretical.
The only system I use is my PC side active speakers.
They are Edifier R1600T.
Freq 30-20kHz 2-way: 4" + 19mm tweeter.
But I have big loudspeakers in my stereo. Audio Pro Black Diamond floor standing.
With a NAD C316BEE amplifier.
But I only listen with PC. Playing mp3.
My question is only theoretical.
The only system I use is my PC side active speakers.
They are Edifier R1600T.
Freq 30-20kHz 2-way: 4" + 19mm tweeter.
But I have big loudspeakers in my stereo. Audio Pro Black Diamond floor standing.
With a NAD C316BEE amplifier.
But I only listen with PC. Playing mp3.
Edifier R1600T looks like equivalent of my Thonet & Vander kurbis.
talk about over rated (340w pmpo) with SMPS and a small chip amp.
At least Edifer rates the 1600 at 60W.
What a scam , glad I only paid 70usd for them.
But they do match well with my XT-12 sub and that REAL 120W behringer amp. 2.1 rules.
PS - doubt the edifer's can go down to 30hz. My PSW-10 Polk sub can only do 40hz. Thats why I am
building the XT-12 polk (24hz).
talk about over rated (340w pmpo) with SMPS and a small chip amp.
At least Edifer rates the 1600 at 60W.
What a scam , glad I only paid 70usd for them.
But they do match well with my XT-12 sub and that REAL 120W behringer amp. 2.1 rules.
PS - doubt the edifer's can go down to 30hz. My PSW-10 Polk sub can only do 40hz. Thats why I am
building the XT-12 polk (24hz).
My suggestion:
A best solution for professional applications!
Then I would still use same poweramps. Same wires. Same psu, or I would connect the psus (maybe per little fuse) to prevent different audible characters.
Most "cheap" professional amps are good pp-circuits, comparable figured in #3 - better than the most fat, expensive HigEnd-amps of well known companies.
A best solution for professional applications!
Then I would still use same poweramps. Same wires. Same psu, or I would connect the psus (maybe per little fuse) to prevent different audible characters.
Most "cheap" professional amps are good pp-circuits, comparable figured in #3 - better than the most fat, expensive HigEnd-amps of well known companies.
The amplifier I'm designing is going to be an integrated part of an active speaker. Not built-in, but sensitive to the woofer specs, so changing the woofer might involve some re-tuning.
Preliminary specs:
Totem-pole (not sure if that's what it's called) stacked N-channel output stage with a current sensing resistor in the middle. The top MOSFET divides its time between being a voltage follower at low frequencies (driven by a low-pass filtered input signal) and a CCS at higher frequencies. The lower MOSFET is open-drain and modulates the speaker current in the higher frequencies.
The mid-point is going to be about 300Hz or so, based on the woofer's minimum impedance between the bass peak and where it starts climbing again due to voice coil impedance. A pair of 12" woofers are now waiting to be driven to 1kHz+, without the usual either-or compromise of crisp bass and shoddy mid-range OR mediocre bass with smooth extended mid-range.
Someone challenged me in a thread on current drive a while ago, to design a mixed-mode amplifier without negative feedback. A gentle 1-order or 2-order ladder filter can be included in some global feedback, for an amplifier to regulate voltage in the bass, and push current in the treble, but man is it finicky... I know it can be half-done with an RC filter, but the slope is too gradual to keep the response flat for the bass-hump and fix the mid-range distortion at the same time.
Preliminary specs:
Totem-pole (not sure if that's what it's called) stacked N-channel output stage with a current sensing resistor in the middle. The top MOSFET divides its time between being a voltage follower at low frequencies (driven by a low-pass filtered input signal) and a CCS at higher frequencies. The lower MOSFET is open-drain and modulates the speaker current in the higher frequencies.
The mid-point is going to be about 300Hz or so, based on the woofer's minimum impedance between the bass peak and where it starts climbing again due to voice coil impedance. A pair of 12" woofers are now waiting to be driven to 1kHz+, without the usual either-or compromise of crisp bass and shoddy mid-range OR mediocre bass with smooth extended mid-range.
Someone challenged me in a thread on current drive a while ago, to design a mixed-mode amplifier without negative feedback. A gentle 1-order or 2-order ladder filter can be included in some global feedback, for an amplifier to regulate voltage in the bass, and push current in the treble, but man is it finicky... I know it can be half-done with an RC filter, but the slope is too gradual to keep the response flat for the bass-hump and fix the mid-range distortion at the same time.
30Hz = fantasy. At least if we are talking -3dB level.
The woofer is only 116mm. ~4.5inch. The box is not big. It is however a bass-reflex.
I will check in XT-12 Polk
I recently bought some P3A boards from ESP that I intend to use in an active system. I’m not familiar with project 103.
What is an e-waste sub?
I think it means energy waste .... ?
That subwoofer Polk XT-12 is at 100W
That subwoofer Polk XT-12 is at 100W
Project 103 - Subwoofer Phase Controller
That amplifier NX400 looks simple enough for my taste.
Still it has some interesting details.
The drivers are 2SC4793 + 2SA1837 = very good transistors.
Thanks! I'm gonna need it! At this stage it's really tempting to remember the KISS principle and go with single-ended class A. My 2-order feedback filter produces a high-pass shelf, which is good, except for an unwanted peak where the feedback switches phase. But if I do it with 2 transistors, I may be able to use filters on both sides for better summing.
I'm also keen to try out some old tricks like winding ~1mH air-core inductors instead of the whole CCS, but I might leave that for the tweeter.
Aside:
Disadvantage with unbalanced half-wave amplifier amps, as shown in #3, AP-12, and #11, NX400: one half-wave "sees" the components of the "bias" control, which modulate this one half-wave additionally with their noises, i.e. considerably differently. And this modulation is also not independent of the working states.
Disadvantage with unbalanced half-wave amplifier amps, as shown in #3, AP-12, and #11, NX400: one half-wave "sees" the components of the "bias" control, which modulate this one half-wave additionally with their noises, i.e. considerably differently. And this modulation is also not independent of the working states.
Last edited:
One that ends up in the landfill 366 days after it’s purchased. Those Polk subwoofer amplifiers have a 100% fail rate.