Page 5 of 10

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier: Pre

Posted: Thu Sep 18, 2014 10:12 am
by mediatechnology
My patented circuit that sounded surprisingly good for a simple diode clipper was to add a pair of anti-parallel diodes across just the boost leg of the bass section of a Baxandal tone control. The diodes were out of the circuit for most normal signal levels, but if/when the operator was commanding a lot of bass boost "and" there was significant bass signal present, the diodes would clamp just that bass part of the signal at +/- a diode drop. Since this bass leg of the Baxandal tone control is naturally LPF by the circuit the HF distortion from diode clipping is reduced and then masked by the normal HF audio signal being passed cleanly by the treble side. Since the fixed instal products were combination mixer-amps the gain structure was fixed and I could index the bass diode clamping to specific output power levels.

While I would not call this a hifi circuit it was surprisingly good sounding and very effective, for that application all for the cost of a couple diodes.
Very cool, yet simple concept.
Definitely KISS.
I can believe that it sounded good too.

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier: Pre

Posted: Thu Sep 18, 2014 6:54 pm
by ricardo
JR. wrote:My patented circuit that sounded surprisingly good for a simple diode clipper was to add a pair of anti-parallel diodes across just the boost leg of the bass section of a Baxandal tone control. The diodes were out of the circuit for most normal signal levels, but if/when the operator was commanding a lot of bass boost "and" there was significant bass signal present, the diodes would clamp just that bass part of the signal at +/- a diode drop. Since this bass leg of the Baxandal tone control is naturally LPF by the circuit the HF distortion from diode clipping is reduced and then masked by the normal HF audio signal being passed cleanly by the treble side. Since the fixed instal products were combination mixer-amps the gain structure was fixed and I could index the bass diode clamping to specific output power levels.

While I would not call this a hifi circuit it was surprisingly good sounding and very effective, for that application all for the cost of a couple diodes.
I like this too.

It is similar in philosophy if not implementation, to the most successful incarnation of my Powered Integrated Super Sub tech. I used a different evil clipper that didn't need any extra bits cos .. as you know .. I get anal over even 2 extra resistors :mrgreen:

The clipping was indexed to the overload of the speaker in its ported box ... and yes, it was HiFi cos it sounded better both on music as well as the dinosaur footsteps to several HiFi competitors.

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier: Pre

Posted: Sat Sep 20, 2014 11:24 am
by JR.
To change this back on topic, I have invested some of my beer o'clock time reviewing my old class AB bias circuit and it does not seem of any utility here.

Ideally a class A output stage where both outputs are independently biased up to always have positive current flow and never cut off. Back decades ago I recall thinking about this and IIRC I used some steering diodes in front of the pre-drivers to drive the up and down separately so they could be biased up class A. Not very elegant and I never melted any solder on this.

Getting back to Wayne's version of class A, I still lean toward investigating a way to modulate the class A current with the audio signal so it can be more efficient at low levels while delivering class A at full output. This class A modulation would also be open loop, predicting the current draw based on voltage and expected load impedance.

JR

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier: Pre

Posted: Sat Sep 20, 2014 12:35 pm
by mediatechnology
I still lean toward investigating a way to modulate the class A current with the audio signal so it can be more efficient at low levels while delivering class A at full output. This class A modulation would also be open loop, predicting the current draw based on voltage and expected load impedance.
I think it's doable.
It could also be labeled a "feed-forward" bias modulator.
It also seems like the AC component would be differential (at the output bases) and would self-cancel somewhat.

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Sun Sep 21, 2014 9:44 am
by JR.
Thinking about this I am reminded of an old amp design I did with a boosted PS rail. So similar to modulating class A current to be enough, I had to modulate the rail voltage to be high enough and before the amp output voltage got there. As I recall my trick PS rail was a bit of a PIA. having to be faster than than signal slew, but if too fast became another new source of distortion. I ended up using some HF pre-emphasis in the PS side chain but recall a lot of bench tweaking to realize both good performance and good efficiency. Driving the rails too high, too much, pissed away a lot of my efficiency benefit. Note: efficiency is important to power amp design because power supply capacity and heat sink dissipation are direct drivers of amplifier cost.

Since you are already comfortable with full class A, perhaps don't go too crazy starving the class A current at low level, which will make getting there in time for transients easier.


JR

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Sun Sep 21, 2014 9:20 pm
by JR.
OK, this took a couple days but I think I have a BOM that even Ricardo would approve.

Typical class A/B amps often use a Vbe multiplier to bias the output stage,,, Instead of a single Vbe multiplier, use one NPN vbe multiplier for the top NPN driver and power device, then a second PNP Vbe multiplier to bias the negative PNP driver and power device. Set the Vbe multipliers for a comfortable (say 100-250mA or more) idle class A current.

Now for the modulated Class A mod... add a resistor and diode from each Vbe multiplier base to ground. With the diode in the right direction for each bias network, a positive output voltage will increase the class A current for just the top driver/output, and a negative voltage swing will increase the negative driver/output class A current.

So i figure about $0.10-$0.15 worth of parts and we get a modulated class A. 8-)

JR

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Mon Sep 29, 2014 1:28 pm
by mediatechnology
OK, this took a couple days but I think I have a BOM that even Ricardo would approve.

Typical class A/B amps often use a Vbe multiplier to bias the output stage,,, Instead of a single Vbe multiplier, use one NPN vbe multiplier for the top NPN driver and power device, then a second PNP Vbe multiplier to bias the negative PNP driver and power device. Set the Vbe multipliers for a comfortable (say 100-250mA or more) idle class A current.

Now for the modulated Class A mod... add a resistor and diode from each Vbe multiplier base to ground. With the diode in the right direction for each bias network, a positive output voltage will increase the class A current for just the top driver/output, and a negative voltage swing will increase the negative driver/output class A current.

So i figure about $0.10-$0.15 worth of parts and we get a modulated class A. 8-)
Very cool. Bootstrapped bias.
The design shown with the OPA2134 and external transistors has the independent bias circuitry that might easily lend itself to that.

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Mon Sep 29, 2014 2:03 pm
by JR.
FWIW I have never searched this, so would not be surprised to learn that somebody else has used this. Often when I search my better ideas I learn that the ancients beat me to it.

I hate to admit that I am almost tempted to open up my living room amp (built in early '70s) to look at modifying it... nah... I haven't even turned it on for years.

JR

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Mon Oct 06, 2014 10:44 am
by JR.
My overnight engineering team has found a flaw in my simple design. If the class A current is dialed in for say 4 ohms, but no speaker load is actually connected to the output, the opposing driver is asked to sink the full output current across the worst case 2x rail voltage. This is a huge thermal dissipation load, and could cause secondary breakdown issues for the power devices.

Back to the drawing board. :o

JR

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Posted: Tue Oct 07, 2014 8:25 pm
by JR.
Untitled.gif
Untitled.gif (10.31 KiB) Viewed 17325 times
OK I am back with a class AB variant but this one does not let the power transistors ever cut off, so no crossover distortion related to the power devices. The class AB transition occurs in fast small geometry pre-driver transistors.

This also uses my favorite 3 transistor LTP I guess I need to call these a "long tail triple"

When the input drive voltage swings opposite polarity, the third transistor keeps the power device turned on at the class A bias current, so the output never cuts off. This class A bias current is set by the voltage drop in the resistor coming from the current sense resistor.. So 1 mA (.5V/510) times 51 ohms is 51mV. Divided by .22 is about 200-225 mA class A. Hot for a small heatsink amp, but should be thermally stable

This should be a lot lower distortion than my old auto-bias approach.


I need to get a life... :lol:

JR

[edit] I figured how to do it with two less resistors. Instead of the resistor from base to emitter of the current sense transistor to form a current source which develops the reference for class A current voltage drop. I could just put a resistor in series with the emitter sized so that at equilibrium I have a happy class A bias level. The former approach is more stable, while the later approach would starve the class A bias a little at high output where it clearly is not needed. [/edit]