John Curl's Blowtorch preamplifier

Status
Not open for further replies.
Jump to Latest
#7,281
Zoebel & RFI

Christer said:
Sorry for quoting myself, but I just wonder if this got lost in the noise in the thread or if everyone finds it so trivial and boring that isn't interesting?

Anyhow, Sigurd has, perhaps unknowingly, commented on some of it in another thread recently and seems to agree that an output Zobel filter is a good idea.
http://www.diyaudio.com/forums/showthread.php?postid=1551357#post1551357
Click to expand...

Hi Christer,

I think it's certainly not trivial. Perhaps boring, because output coils were ad nausea discussed in a dedicated thread:
http://www.diyaudio.com/forums/showthread.php?postid=1202201#post1202201

Anyhow, a Zoebel filter is a good idea, for two reasons:
1. To block RFI from outside.
2. To avoid instabilities in amps with (high) NFB from the output stage.

One problem is that some people claim they can hear the (detrimental) effect of a coil, that is, if larger than say 1uH.

Another problem with Zoebel networks is that the de facto standard configuration, blindly copied by almost everyone, is seldom optimal. See: E.M. Cherry, "Ironing out distortion", EW+WW, Jan. 1995, pp.19-20, fig. 9a and 9b.

Cheers,
Edmond.
 
#7,282
Re: Zoebel & RFI

Edmond Stuart said:


Anyhow, a Zoebel filter is a good idea, for two reasons:
1. To block RFI from outside.
2. To avoid instabilities in amps with (high) NFB from the output stage.
Click to expand...


Yes, power amps usually have a Zobel filter for stability reasons, but that one is usually placed before the coil, although Leach recommends placing it at the output connector (although for entirely different reasons). For RFI reasons, it seems better to have it att the connector for the same reason as we discussed for inputs. However, does that work as well for stability? Perhaps there should be two Zobels, one before the coild and one at the connector?

Then of course, I was not only considering power amps, but also line level amps etc. where we usually do not have a coil or Zobel filter. There it seem reasonable to add a Zobel at the connector.


One problem is that some people claim they can hear the (detrimental) effect of a coil, that is, if larger than say 1uH.
Click to expand...

Yes, and some don't like the sound of Zobels and burn their amps instead. At least serious designers who avoid coils and/or Zobels usually care about coping with stability in other ways.



What about my other question, whether it matters if the input connector filter is a voltage divider type or a shunt type (ie. a Zobel)? My thinking was that since the input impedance of the amp is usually at least many kOhms, it shouldn't matter for HF termination which method we use. However, then I realised that maybe the wires from connector to PCB should also be considered as transmission lines so the actual amp input impedance matters too, in which case the voltage divider is probably better. (Maybe this is more of a theoretical question, since an extra 50 Ohm series input impedande usually doesn't matter).
 
#7,283
mlloyd1 said:
lineup:
i am not one of the experts, but a question for you if i may:
did you build real prototypes and test those? i would be concerned that you are tweaking for spice models differences otherwise .
Click to expand...

Tell this to E Stuart, PMA, jcx etc. etc. the list is loooong ....
There are a whole bunch of others here is doing more spice works than me.
I have not heard anyone complain about that. To them.

And we are getting more and more day by day, using this wonderful tool.
Even Nelson Pass uses spice. But he avoids to mention it.
John Curl wishes he knew more how to use spice!

Spice is a great tool for explorations, man 😉
It has also been witnessed by those guys,
that the real thing is not too far from a good analyse.
Figures will not be at same level.
But the tendencies we can see in a good spice circuit
will more than often be exactly the same in a real amplifier.
-----------------------------------------------------------------------------

😎
No offense, but
I am more interested to hear John Curl / Scott Wurcer / Bob Cordell comments on my issue brought up.
They would know, very well ,what I am addressing.
About complementary JFET input stages & the adjustment of
😎

regars to mlloyd1
Lineup
 
#7,284
Fellow engineers, let's get real. If Spice works for your needs, and you have 'mastered' it, OK. Then use it and be productive.
Spice is not absolutely important for audio design. We have done without it for decades. However, the future is yours. IF you find a new way to do something better with Spice, that will be very good. Let us see what comes out, but please remember what has already been done without Spice, before announcing a new 'breakthrough'.
 
#7,285
Re: Zoebel & RFI

Edmond Stuart said:
......a Zobel filter is a good idea, for two reasons:
1. To block RFI from outside.
2. To avoid instabilities in amps with (high) NFB from the output stage.

One problem is that some people claim they can hear the (detrimental) effect of a coil, that is, if larger than say 1uH.

Another problem with Zobel networks is that the de facto standard configuration, blindly copied by almost everyone, is seldom optimal. See: E.M. Cherry, "Ironing out distortion", EW+WW, Jan. 1995, pp.19-20, fig. 9a and 9b.
Click to expand...

previous posting on this topic.
http://www.diyaudio.com/forums/showthread.php?postid=1475396#post1475396

and the excel spreadsheet based on Dr Cherry's formulae.
http://www.diyaudio.com/forums/attachment.php?s=&postid=1475396&stamp=1207315043
Sorry this is off the pre topic
 
#7,287
A coil I use is one made out of thick silver wire wrapped in unbleached cotton and all this is then epoxied.

It gets fairly large as the attached image below shows.


Sigurd
 

Attachments

  • dsc07013-2.jpg
    dsc07013-2.jpg
    54.2 KB · Views: 627
#7,288
Re: Re: Re: Re: Re: Re: RFI

Bob Cordell said:
One thing that I like to do is right at the RCA connector where the signal comes in, run it through a series 49.9 ohm resistor and then a shunt 100 pF capacitor to the connector ground so that the interconnect is effectively terminated in 50 ohms at very high frequencies. This, of course, is a compromise, as not all interconnects have a characteristic impedance of 50 ohms, but I think it is better than nothing, and it forms a first line of defense at the very high frequencies where things get difficult to manage.

Cheers,
Bob
Click to expand...

Would there be a reason for 100p?
Why not 220p, 330p, or even 1000p while we're at it?

Thanks for the input (termination) 😉
 
#7,292
output coils

john curl said:
And of course, I, as well as many other noted audio designers, here, on this website, have found that NO coil is the best coil. Your design may demand it, but we have 'weaned off' from this 'bandaid', after many decades of using it.
Click to expand...

Hi John,

I don't like coils either, but as long as you don't know what kind of 'rogue' speakers some people will use, I prefer to stay on the safe side.
Moreover, are there other (or better) means to keep outside RFI away (picked up from the speaker cables) without a coil and a capacitor?

Cheers,
Edmond
 
Status
Not open for further replies.
Top Bottom