Post filter AND pre filter feedback..!

[Lars Clausen]

Continuing a subject from another thread, about which is better, pre filter feedback (like ZAPpulse) or post filter feedback (like UcD, Icepower etc.). I don't want to risk thread jacking, so therefore this new thread.
It is intended to show that both schemes are possible on almost any Class D platform using a simple switch and a few parts. Below you can see how it's simply implemented on a ZAPpulse 2.3SE. The parts values may differ on other Class D modules, but basicly it's the same circuit.

This way anyone can simply switch over, and compare the benefits of each type of feedback topology. I think that's the most honest way of advocating for a solution. Namely to let people check it out for themselves.

Please note: this is not an official proposition of L C Audio Technology. They have nothing to do with it. It is just a proposition from one DIY'er to another. But it does work!

[Workhorse]

Hi LC,

Why have you Left the LCAT, I thought that you were the man behind them and all, especially the ZAP's.....

regards,
K a n w a r

[JohnW]

Lars,

Looking forward, as I've already worked on both methods and have posted about my subjective results earlier - will be interesting to hear (no pun intended) about other peoples experiences!

John

[SSassen]

Lars,

As I already expressed in the other thread, what's the use of post-filter if you don't use it to allow for load-invariant behaviour of the amplifier? I'm sure lowering THD is a goal, but that isn't something I'd be too concerned about (if you've already got good scores in that department).

Furthermore the values you quoted for the ZapPulse modification are several magnitudes off any usuable value, just simulate a simple hyst. self. osc. with these values and you'll notice this modification does very little, if anything.

Best regards,

Sander Sassen
http://www.hardwareanalysis.com

[Lars Clausen]

Sander: aaaaaa sorry i don't think simulators are that useful. Fun to play with though. But for serious stuff i like to do real life measurements.

With regards to this:

Furthermore the values you quoted for the ZapPulse modification are several magnitudes off any usuable value......

.... i have no idea what you are talking about

I'm sure lowering THD is a goal, but that isn't something I'd be too concerned about (if you've already got good scores in that department).

Exactly, a good filter coil has much lower THD than the rest of the amplifier. So THD would not be a good reason to include the choke in the feedback control loop.

Hi John W:

have posted about my subjective results earlier

Please give me a link, i'd like to see your findings.

Hi Workhorse

L C Audio was bought by an investment firm in 2002 after a chapter 11 bankrupcy. As a part of the deal i made a contract with them to work 3 years at a relatively low salary.
Now i have fulfilled the contract, and then left them in September 2005. Now i work independently and make a lot more than i did before.
However i still like to keep in touch with the audio industry, it's always fun, and i have many good friends here.

Lars

[SSassen]

Lars,

You'll be surprised as to how accurate simulators can be, LTspice has been a valuable tool for me in these past months, I would even go as far as to say that it has saved me months of prototyping, as the simulator showed me beforehand what areas were problematic and what to expect. Atual results measured on prototypes where a lot like those which I'd previously simulated.

If you can combine simulations (and a good understanding thereof) with actual measurements on a prototype you'll see that this speeds up the design process immensely. At least that has been my experience and that of many others, as some of the members here will be happy to explain to you.

As for the filter coil, and the output capacitor, it is a (reactive) combination that will interact with the reactive load that is the loudspeaker and hence represent a complex load to the amplifier that is not corrected for as the feedback takes place before the output filter, you can't correct that unless you apply post-filter feedback. I've done measurements on pre-filter class-D amplifiers that have clearly shown a rise in THD with higher frequencies (much like an ordinary dominant pole amplifier).

It is however perfectly feasible to keep the THD at a reasonably fixed level by applying post filter feedback, as done with UcD, or by applying both pre- and postfilter feedback on a non-UcD amplifier.

If you don't go through the trouble of doing that what exactly are the benefits of class-D? UcD is able to go places where it is difficult for any class-A/AB amplifier to go with its load invariant FR and THD. If other class-D topologies can't do this, where's the improvement? Then they're just a novelty, or a cost-saving solution, not a step forward, just another way of doing the same thing.

Best regards,

Sander Sassen
http://www.hardwareanalysis.com

[Lars Clausen]

It is however perfectly feasible to keep the THD at a reasonably fixed level by applying post filter feedback, as done with UcD, or by applying both pre- and postfilter feedback on a non-UcD amplifier.

If you don't go through the trouble of doing that what exactly are the benefits of class-D?

Well my real life work (not simulators) show that post filter feedback will not actually do much to improve THD.

But on the other hand you can also get a good low THD with pre filter THD (in real life i mean).

So i guess i just dont agree with you on this one.

It's not a matter of 'not going throgh the trouble' of implementing post filter feedback. My experience just tells me that post filter feedback has some inherent limitations that i don't want my amplifiers to have. Limitations that i can't see any way to work around without damaging the sound. As simple as that.

I know it can also be said that pre filter amps have inherent problems, like loss of gain at low impedances at 10-20 kHz. However i find it easy to get around these limitations without damaging the sound. (With a simple RC compensation filter).

UcD is by far not the only post filter feed back modules, there are many of them out there. And they all feature the same characteristics. And none of them are better in sound tests than a good pre filter feedback setup. Sorry.....

Anyway since there is so much prejustice around post filter feedback, and it's so easy to implement, i think my next generation of Class D amplifiers might have both options on board.

[SSassen]

Hi Lars,

Anyway since there is so much prejustice around post filter feedback, and it's so easy to implement, i think my next generation of Class D amplifiers might have both options on board.

There's a BIG difference between implementing post filter feedback and a class-D amplifier having load invariant FR and THD behaviour. When implementing post filter feedback load invariant FR and THD behaviour are not thrown in for free, if that was the case we wouldn't be having this discussion. The real trick, which Bruno managed to pull off with UcD, is to make the performance of the amplifier independant of the load. Whether that's done with pre- or post filter feedback I really don't care, it is the end result that counts, wouldn't you agree?

As for an amp sounding good, I will simply not consider an amplifier that is known to suffer from a (clearly measurable) change in FR or THD with different loads to be a candidate in my listening room. Afterall I'm trying to reproduce the recording as accurately as possible and with the amplifier being such an important contributor I'd like for it to add or substract as little as possible from the recording. Aren't we all just looking for the fictious piece of wire with gain really?

Best regards,

Sander Sassen
http://www.hardwareanalysis.com

[Lars Clausen]

I'm sorry Sander. Now i just think it's plain empty talk from your part.

[km]

hi,

>UcD is by far not the only post filter feed back modules, there are many of them out there. And they all feature the same characteristics. And none of them are better in sound tests than a good pre filter feedback setup. Sorry.....>

my/our experience is the opposite

karsten madsen - cadaudio.dk/digiamps.com