John Curl's Blowtorch preamplifier part II

[john curl]

Personally, I do not understand why you say such crazy things, Wavebourn.
Perhaps, you do not understand the definition of: NO COMPROMISE
Perhaps a Skoda is your sort of practical 'compromise' and a Porsche is my sort of 'no compromise' design. It is not an impossible design, but a design with a minimum of performance compromises.
Are you trying to impugn my design reputation? This is what you appear to do constantly, and I think that it is to 'sell' yourself to perspective clients, if that is possible. It won't work, and I suspect that this sort of attitude has hurt your job prospects in Silicon Valley as well. They want team players as well as good engineers.
I would normally keep this to myself, but I think that you should know it.

[VivaVee]

Originally Posted by andy_c

Bob, I have a stupid question. How does one wind an air-core toroid? Or more specifically, isn't there a problem removing the form? I'm thinking of this in comparison with the solenoid case where the form might be, say, a dowel which you can just slide out when done winding.

Try a non-magnetic non-conductive former.

I use 10mm diameterTeflon tubing. Then vacuum impregnate the construction with varnish to stop it 'singing'.

[PMA]

Originally Posted by john curl

Skoda

They produce Moskwitch, Zaporozhets, Volga, ZIS, ZIM, UAZ, Ziguli and similar in the USSR

Not Skodas, Skoda has been a part of Volkswagen group for about 18 years now.

[Bob Cordell]

Originally Posted by syn08

Then never touch Class D with NFB. This is what it takes: http://eprints.nottingham.ac.uk/501/1/final.pdf

P.S. I have spend only 10 minutes reading this work, so I don't claim understanding more than the basic plot. I don't plan spending more time on it

Hi syn08,

Thanks for posting this. It looks like a great reference and an excellent read on a very important aspect of class D amplifiers. When and where was it published?

Thanks,
Bob

[Bob Cordell]

Originally Posted by john curl

It never ceases to amaze me, as to why people bring their most fun topic of what they are doing to this thread.

Now, taking a deep breath, I will attempt to remind everyone as to why this thread has grown so 'popular' (in a sick sort of way) and my original intentions of contributing to this thread.

Now, where to go from here? Back to discussing the subtleties of quality preamp design, or continue the aimless diversions that just confuse the issue?

Hi John,

This is indeed a popular thread. It is the "watercooler" thread in this forum. It has touched on many topics of interest to people here, certainly not just the topic of preamplifier design.

Are you complaining about the recent conversations about class D amplification?

I plead guilty to bringing that up, but it was in the context of PIM. The follow-up discussion that occurred seems to reflect some interest on the part of the participants in class D.

Cheers,
Bob

[Bob Cordell]

Originally Posted by andy_c

Bob, I have a stupid question. How does one wind an air-core toroid? Or more specifically, isn't there a problem removing the form? I'm thinking of this in comparison with the solenoid case where the form might be, say, a dowel which you can just slide out when done winding.

Hi Andy,

Very good question. I first used air core toroids for the output network of my Super Gain Clone amplifier, wishing to minimize possible magnetic field interactions from the coil. It was just a fun thing, and I really don't know whether it made much difference.

I made the coil by winding the turns on a pencil, removing the pencil and then forming the solenoid structure into a toroid by gently bending it by hand.

Cheers,
Bob

[andy_c]

Originally Posted by ostripper

I have only seen this in a PC SMPS , only a few turns , but it was toroidial (wrapped around a plastic former) . Was PE or a teflon- like plastic.

Originally Posted by VivaVee

Try a non-magnetic non-conductive former.

I use 10mm diameterTeflon tubing. Then vacuum impregnate the construction with varnish to stop it 'singing'.

Originally Posted by Bob Cordell

I first used air core toroids for the output network of my Super Gain Clone amplifier, wishing to minimize possible magnetic field interactions from the coil. It was just a fun thing, and I really don't know whether it made much difference.

I made the coil by winding the turns on a pencil, removing the pencil and then forming the solenoid structure into a toroid by gently bending it by hand.

Thanks very much guys.

Love this new multi-quoting feature too .

[Bob Cordell]

Originally Posted by Wavebourn

Hi Bob;

as I asked before;

when eliminating source of PIM we replaced the filter choke by an air core one, then surround the whole thingy by feedback, do we create PIM as the result, since the amp is not ideally linear?

What is better:
to leave an inductor that creates PIM on the place, then reduce PIM using feedback (as you said a negative feedback reduces pre-existing PIM);
or to replace it by an air core one, then generate PIM applying a negative feedback?

Hi Wavebourn,

I may have been unclear in what I was trying to convey in my original post relating the the generation of PIM in class D amplifiers.

As we have noted in many PIM discussions, PIM can be created in two ways (at least). First, it can be created by negative feedback that surrounds an amplifier that does not otherwise have PIM. This is what Otala, myself and Gilbert discussed the most.

The second way that PIM can be generated is often by passive components whose values change as a function of signal voltage or current. In a linear amplifier a good example of this is variation in the collector base capacitance of the VAS.

In the class D example I was citing primarily the second type of PIM caused by passive components, even in the absence of negative feedback. This was simply due to the output filter inductance varying as a function of the signal current. This can happen in class D amplifiers that have no negative feedback,

There are also class D amplifiers that have negative feedback that is either tapped off ahead of or after the output filter. Those may have PIM created by the negative feedback process as well.

For those where the feedback is tapped off ahead of the output filter, those amplifiers will still be left with the undiminished PIM created in the output filter. This may often be larger than any PIM that is created by the feedback process (that is another point of comparison I was making in the original post).

For those amplifiers that tap the feedback from the far side of the output filter, the negative feedback enclosing the output filter will reduce distortions introduced by the filter, including PIM.

The best solution is to use the highest-quality output inductor in the first place, possibly even an air core inductor. Then negative feedback should be applied around the output filter if possible. Although the use of a high quality inductor will reduce or eliminate the PIM and other filter distortions, it is still desirable to have feedback enclose the output filter to reduce output impedance at high frequencies, retaining damping factor and making the amplifier frequency response less load-dependent. The challenge is to be able to employ a meaningful amount of negative feedback while retaining stability.

Self-oscillating class D modulators are a whole 'nother story.

Cheers,
Bob

[john curl]

Hi PMA, I didn't live in the USSR, so I only know Skoda, from the old days. I have seen Russian autos, but I could not know them by name. Stlll, if Skoda does what I think it does, today, then it is still a pretty good example of a good 'compromised' car. Let us compare it to an upscale Mercedes, for example.

[PMA]

No problem, just mentioning that Skoda is a Czech, not Russian car