Current source amplifiers.

[oon_the_kid]

Hi,

I am planning to build a f2j for my openbafle speaker. Looking to biamp the system with a current source amplifier. One possible way might beto connect a resistor in series and take the feedback from the resistor. Just wondering if anybody tried that before.

Oon

Sent from my MB860 using Tapatalk

[Osvaldo de Banfield]

yes, ok, but why current driven the speaker? This is not good...

[oon_the_kid]

Hi,

The reason is firstwatt f2j is a current drive amplifier intended for fullrange amplifiers. Voltage driven speakers have one weakness, that is the forces acting on the cone is no longer representative of the true signal since the current is no longer proportional to the voltage because of inductance, back emf etc. So the idea of a current amp is to match the f2j.

I am a bit worried about instability though..
Oon


Sent from my MB860 using Tapatalk

[jackinnj]

National Semi's data sheet notwithstanding -- you can turn a chipamp into a Howland current source.

[pacificblue]

Originally Posted by oon_the_kid

Voltage driven speakers have one weakness, that is the forces acting on the cone is no longer representative of the true signal since the current is no longer proportional to the voltage

Most speakers are designed to have a nearly flat frequency response for constant voltage. And they are all current-driven.

Look at the impedance graph of a speaker. Then use Ohm's law (U = I x R) to find out how the voltage will look, when you apply constant current instead of constant voltage. The frequency response will become a scaled image of the impedance curve added to the frequency response for constant voltage.

The amplifier would already be clipping at low volumes, where the impedance is high, e.g. around the speaker's resonant frequency, at very high frequencies, around crossover frequencies, etc., because increased impedance demands increased voltage to achieve a constant current.

Of course the F2J is not actually keeping the current constant. It works more like a shunt regulator, where a constant current is shared among the (MOSFET) transistor and the speaker. It deviates just enough current into the speaker to achieve a certain voltage at the speaker terminals. In the end you still have a voltage controlled speaker.

[Fenris]

Been there, done that a year ago. I've got a LM3886 amp in a classic V-I configuration. For best results, you need to ensure the speaker impedance is as flat as possible and/or use a series crossover. I use a fullrange driver with mine with a zobel to level out the impedance and avoid the complexity of a crossover. Watch out for the impedance spike at resonance frequency, it can cause clipping.

[a.wayne]

Then why constant current ...?

[oon_the_kid]

Dear pacific blue,

Disagree on the voltage control speaker part the F2. What is important the current is proportional to the input voltage. Because the current bleeding by the mosfet is proportional to the input voltage. So in that sense the speaker is current controlled.
There is however a 16 ohm resistor in parallel so at resonance, the amplifier is essentially voltage amplifier.

Oon

Sent from my MB860 using Tapatalk

Sent from my MB860 using Tapatalk

[oon_the_kid]

Dear fenris,

Do you mind sharing a bit more on the details of your circuit. Did you just put the speaker in between the output and the -inverting input. And a 1 ohm between the inverting input and the ground? Or maybe an additional voltage divider?



I actually planned for it for woofer use in an open baffle, so it needs plenty of equalization anyway. The additional boost attacks resonance could come in handy since I need a bit of boost at fs anyway.

I am just thinking the phase relationship between woofer and the fullrange will be better.

How do you like the sound compared to normal voltage amps?

Oon

[pacificblue]

Originally Posted by oon_the_kid

the current bleeding by the mosfet is proportional to the input voltage.

No. If the output current was proportional to the input voltage you would get a frequency response that looks like the impedance curve. The current is proportional to the output voltage divided by the load impedance.

The F2 is a voltage regulator like any other audio power amp. The output voltage is proportional to the input voltage, controlled by voltage feedback through P1, R4, R5 and R6.

The difference to most other amps is that the F2 has less stages. It combines the functions of the typical three stages in one. You find, what Pass calls 'active current source', in most amps' VAS (voltage amplification stage). In the F2 there is simply no additional output (buffer) stage that makes the load for the VAS lighter and more constant.