Why not?
CURRENT
and it needs to be between strict limits for the regulating action of the choke input filter to work.
If the current goes low then the voltage rises as it starts to approach the the value given out by a capacitor input filter (1.414* instead of 0.9*)
ClassA average draw is OK, but the rail currents move towards zero and 2*Iq as output current increases. This will demand that the capacitors smooth out the dips enough that the choke NEVER sees anything as low as zero current.
CURRENT
and it needs to be between strict limits for the regulating action of the choke input filter to work.
If the current goes low then the voltage rises as it starts to approach the the value given out by a capacitor input filter (1.414* instead of 0.9*)
ClassA average draw is OK, but the rail currents move towards zero and 2*Iq as output current increases. This will demand that the capacitors smooth out the dips enough that the choke NEVER sees anything as low as zero current.
It needs to be designed properly. The inductor should be well above the critical inductance for the average (!) current but this is not too difficult in a Class A circuit.
Correct, this could happen in failure mode, so you should choose capacitors with a sufficiently high voltage rating.
In other words you still need sufficient capacitance but this is the same with a capacitor input filter. It is not an issue that the inductor sees a zero current as long as it is only for a short period of time. See also this post: http://www.diyaudio.com/forums/showthread.php?postid=658596#post658596 : For a properly dimensioned LC filter the time constant will be lower than the lowest audio frequency. Calculate an example, enter it in PSUDII and use stepped load, this will give you an indication of how much the voltage will rise or dip. Worst case would be a 16 Hz signal (half a cycle, i.e. ca. 30 msec) and the actual inductance being the bare minimum critical inductance.
> did you mean a pair of bridged amplifiers from the SAME supply?
No. I meant :
http://www.diyaudio.com/forums/showthread.php?s=&postid=658958&highlight=xed#post658958
Patrick
No. I meant :
http://www.diyaudio.com/forums/showthread.php?s=&postid=658958&highlight=xed#post658958
Patrick
Hi,
My name is Patrick. You may wish to address me so.
> is that the one you referred to as Xed?
Yes.
> Can you explain the difference?
Bridged amps from the same supply is like what JLH showed in Fig 4 of his postscript in 1970. You convert a single-ended input signal to differential (+/-) and drive two individual amplifiers which are totally independent on each other, other than a common ground, and optionally but not necessarily the same power supply.
An X'ed amplifier (as patented by Nelson Pass, but also published by Kaneda amongst others) uses a diff pair frontend to link the two halves of the bridge, and thus interacting to cancel distortion better, according to Nelson. There are plenty of explanation of the X circuit on the Pass forum or at his websites, as well as the patent, which describe the principle much better than I can do here with reasonable effort. It only works with negative feedback, however.
I have built different versions of at least 3 variants of the X topology, including the Aleph-X, JLH-X as linked, and a MOSFET version of the JLH-X (also previous published on the Pass Forum PLH thread). They all work very well, with very constant current draw even loaded fully by a dummy speaker at maximum swing.
Patrick
My name is Patrick. You may wish to address me so.
> is that the one you referred to as Xed?
Yes.
> Can you explain the difference?
Bridged amps from the same supply is like what JLH showed in Fig 4 of his postscript in 1970. You convert a single-ended input signal to differential (+/-) and drive two individual amplifiers which are totally independent on each other, other than a common ground, and optionally but not necessarily the same power supply.
An X'ed amplifier (as patented by Nelson Pass, but also published by Kaneda amongst others) uses a diff pair frontend to link the two halves of the bridge, and thus interacting to cancel distortion better, according to Nelson. There are plenty of explanation of the X circuit on the Pass forum or at his websites, as well as the patent, which describe the principle much better than I can do here with reasonable effort. It only works with negative feedback, however.
I have built different versions of at least 3 variants of the X topology, including the Aleph-X, JLH-X as linked, and a MOSFET version of the JLH-X (also previous published on the Pass Forum PLH thread). They all work very well, with very constant current draw even loaded fully by a dummy speaker at maximum swing.
Patrick
Hi Euvl,
thanks for that explanation of the differences.
What I can see is that an Xed is a bridged amplifier designed to be integrated with the feedback crossed over and in this case a shared supply.
These types of bridged amps do indeed have virtually constant current draw and would be ideal for a choke input filter.
thanks for that explanation of the differences.
What I can see is that an Xed is a bridged amplifier designed to be integrated with the feedback crossed over and in this case a shared supply.
These types of bridged amps do indeed have virtually constant current draw and would be ideal for a choke input filter.
> These types of bridged amps do indeed have virtually constant current draw and would be ideal for a choke input filter.
And have all even harmonics largely cancelled (if you have perfectly matched transistors).
Whether that is a good thing is a matter of taste. For me no question yes.
Patrick
And have all even harmonics largely cancelled (if you have perfectly matched transistors).
Whether that is a good thing is a matter of taste. For me no question yes.
Patrick
Hi Andrew, EUVL
An inductor filter will be OK when used on a Class A amp. even when the current goes to zero (which it will only do so if one side cuts off, and the amp would be clipping) if there is a large capacitor as well (after the inductor). This will serve to "fill in" the missing load.
So it does not seem necessary to use an X-config. nor bridge, actually.
As a large capacitor is already part of the PSU it still seems worth considering an inductor filter!
The transformer voltage needs to be increased by root 2 times 1.1, but the current requirement can reduce proportionaly. A minimum load can always be provided by a dummy-shunt transistor. This could be normally off but cut in if the voltage increases (such as during test when an amp may be disconnected).
cheers
John
An inductor filter will be OK when used on a Class A amp. even when the current goes to zero (which it will only do so if one side cuts off, and the amp would be clipping) if there is a large capacitor as well (after the inductor). This will serve to "fill in" the missing load.
So it does not seem necessary to use an X-config. nor bridge, actually.
As a large capacitor is already part of the PSU it still seems worth considering an inductor filter!
The transformer voltage needs to be increased by root 2 times 1.1, but the current requirement can reduce proportionaly. A minimum load can always be provided by a dummy-shunt transistor. This could be normally off but cut in if the voltage increases (such as during test when an amp may be disconnected).
cheers
John
Interestingly, Viola Audio Labs still uses choke input filters in the power supplies of their pre- and power amps. Quote: “The Bravo uses a choke input filter type power supply with a generously rated 2kVA mains transformer. Four 80,000μF filter capacitors are located in the power amplifier chassis to provide optimum power supply decoupling. The Bravo's choke input filter design power supply produces far less Electro-Magnetic Interference than the capacitor input designs most commonly used in power amplifiers. “
In the past, Cello audio used choke input filters and I believe an earlier model of an Ayre power amplifier also had them. I have a couple of pictures of the power supply of the Viola Bravo power amplifier but for some reason I can’t upload them.
In the past, Cello audio used choke input filters and I believe an earlier model of an Ayre power amplifier also had them. I have a couple of pictures of the power supply of the Viola Bravo power amplifier but for some reason I can’t upload them.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.