Hi,
Since power consumption regulations are enforced on TV manufactures, I believe it is only a matter of time before similar requirements are put on audio equipment.
This will heavily influence the solutions that manufacturers can pick and choose from – and will bring up a lot of interesting technical challenges.
Variable rails (Class G like) is looking more and more interesting to bring down idle losses across the whole audio amplifier field – maybe using the latest SMPS technology solutions can be found to provide the audio quality we all would like while maintaining a reasonable power consumption level.
Best regards
Jens
Since power consumption regulations are enforced on TV manufactures, I believe it is only a matter of time before similar requirements are put on audio equipment.
This will heavily influence the solutions that manufacturers can pick and choose from – and will bring up a lot of interesting technical challenges.
Variable rails (Class G like) is looking more and more interesting to bring down idle losses across the whole audio amplifier field – maybe using the latest SMPS technology solutions can be found to provide the audio quality we all would like while maintaining a reasonable power consumption level.
Best regards
Jens
Member
Joined 2009
Paid Member
Class G will be used only by those hanging on to what they know. For the rest, it's all going to be Class D, especially for consumer applications. Hopefully, the law makers will carve out space for other options for the low volume boutique crowd. If you think about it, this will have an impact on the kinds of parts that will be manufactured in high quantities (i.e. readily available and inexpensive) verses those that are not (think tubes, LatFETs, ClassAB chip-amps, whopping great power transformers etc.). The silver lining is that there will be some amazing switching FETs available that will do double duty for us in our non-switching DIY projects (think SiC, GaN as well as good ole Silicon).
Last edited:
To bring Class D idle consumption down, something must be done...
I'm testing a Class D amp using GaN (on silicon) right now.... 50V switch in 4ns :-D
The problem with GaN is that the great switching speeds means that normal packages for power mosfets will be useless due to high inductance levels, so they will have to be pin-less packages like e.g. BGA or similar. - not DIY grade components.
\\\Jens
Member
Joined 2009
Paid Member
I dare say that advanced Class D will not be suitable for the casual DIY'er. But we will still be building our DIY Class A and Class AB amplifiers using NOS transistors and tubes, plus taking advantage of new switching FETs that do still have legs on 'em. And DIY'ers are a resourceful bunch, they'll find a way 
Hi Bigun,
Yes I was working on this with Hugh from Aspen but unfortunately it never got to see its way into a commercial product as health issues intervened. I've done much work on it since then to get the best out of it, and it's an astounding open loop buffer which is like listening to class A but without the heat and inefficiency. The other feature is it literally allows you to roll input stages since it's out of the loop .
Colin
Yes I was working on this with Hugh from Aspen but unfortunately it never got to see its way into a commercial product as health issues intervened
. I've done much work on it since then to get the best out of it, and it's an astounding open loop buffer which is like listening to class A but without the heat and inefficiency. The other feature is it literally allows you to roll input stages since it's out of the loop . Colin
I will be really surprised if class A/AB is banned.
If you pareto energy wasters, there are far more worse things - think heating, airconditioning, cars etc. A PC chews up 250W while a class AB amp (even a big one) is probably no more than 40~50W on average.
But, lawmakers for the most part are ********, so I suspect if they can find something to pass a law on, they will, no matter what the misery is that they will bring to bona fide music lovers . . . .
If you pareto energy wasters, there are far more worse things - think heating, airconditioning, cars etc. A PC chews up 250W while a class AB amp (even a big one) is probably no more than 40~50W on average.
But, lawmakers for the most part are ********, so I suspect if they can find something to pass a law on, they will, no matter what the misery is that they will bring to bona fide music lovers . . . .
Objective testing
This point actually grabbed my interest. I disagree with the statement that zeners and LEDs "mask" the signal because of the noise increase, so I will run a series of tests with regards to this.
I've got a pro-grade microphone amplifier (Behringer MIC200, s/n >120 db, 26-60 db adjustable gain). Attached schematic will be connected to its input and then I measure the noise floor, having zener, LED, diode, resistor, battery, etc. at the place of D1, adjusting the CCS quiescent current to, let say, 2 mA for each setup, without touching the gain. Test circuit will be powered from the regulated PSU +/- 24V DC.
What do you think? Will it make sense?
Cheers,
Valery
P.S. I can also sent the same signal to the input and publish the output spectrum for each option... would be fun to see if there is some difference visible
This point actually grabbed my interest. I disagree with the statement that zeners and LEDs "mask" the signal because of the noise increase, so I will run a series of tests with regards to this.
I've got a pro-grade microphone amplifier (Behringer MIC200, s/n >120 db, 26-60 db adjustable gain). Attached schematic will be connected to its input and then I measure the noise floor, having zener, LED, diode, resistor, battery, etc. at the place of D1, adjusting the CCS quiescent current to, let say, 2 mA for each setup, without touching the gain. Test circuit will be powered from the regulated PSU +/- 24V DC.
What do you think? Will it make sense?
Cheers,
Valery
P.S. I can also sent the same signal to the input and publish the output spectrum for each option... would be fun to see if there is some difference visible
Attachments
Last edited:
Sure it will! Please, do it, Sir
Doubt that
. We'll see...EDIT: What about splitting this off into a new Thread?
Member
Joined 2009
Paid Member
It would be a shame if it stopped there so I hope you have the chance to pursue this one day.
I would like to make a tube + buffer and currently my plans are to use a complementary CFP buffer (think Nelson Pass F5 whiteout gain).
Does the new D Self book cover this ?
Last edited:
I have. Have you done any measurements?
Ummm... you don't see the logical problem here?
And have any been built and measured?
Oh yes I can. As I've explained in APAD6, symmetrical slew limits are the last thing you want. And I am still waiting to see any convincing explanation of why CFA could be an improvement.
Now this I take exception to. I defy you to name any other detailed account of power amplifiers that gives more hard facts and less hand-waving. And of course APAD6 contains real measurements of real amplifiers rather than mere simulations.Give us facts Doug, and not opinions
It's worth bearing in mind that treating Early voltage as a single figure, and so assuming a straight line, is just an approximation. See Figure 7.13 in APAD6.
There has been precious little published on this.
Anybody got a good test setup for measuring Early voltage without a curve-tracer?
Not unless you put a big cap across the Zener, or use RC filtering. (see upthread)
This is standard practice with Zeners.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.