John Curl's Blowtorch preamplifier part II

[john curl]

Studies, are not everything. Success it.

[syn08]

Originally Posted by john curl

Studies, are not everything. Success it.

I have not seen any evidence that special RF ingress protection is or ever was the key toward a successful audio product. To me, such special protections are just another attempt "to make a difference" on a market where differentiating from your competitor is a critical factor. Otherwise said, just another sales pitch, good for the Stereophile readers.

Do you have such evidence?

[ikoflexer]

Originally Posted by syn08

I am concerned, within reason. I don't think potential RF ingress issues require any special or outstanding measures. To my experience, common sense practices like good grounding, proper shielding, smart wiring (reducing loops), filtering, decoupling, ground loops (or lack, thereof), RF input filters, etc... are addressing 99.99% of the potential issues. No need to get paranoic about, and shielding audio gear like a RF transmitter, designing power supplies for tenths of MHz rejection, etc... is a waste of time, energy and money. Of course, there are always pathological cases like people living in the proximity of RF transmitters, but I don't think they should set a design target.

For example, while I absolutely agree that toroidal transformers are allowing some (?) RF ingress through the winding capacitive coupling, I have never seen any shred of evidence or credible study about the impact (and superiority, thereof) of (e.g.) R-core or even split bobbin transformers on the audio performance. Moreover, if such impact exists, the transformer is not necessary the evil part. To me, (e.g.) building a gain stage with virtually zero PSRR, then complaining about the transformer/power supply allowing RF ingress, is plain wrong design.

I appreciate your response.

Here's something John might like. A quote from someone's presentation:
"The AudioFool Viewpoint
• The Myth: “We can hear stuff above 20 kHz”
• Reality: Some distortion mechanisms DO
produce audible artifacts from ultrasonic
signals, but we hear the problems, not the
signals!
• Intermodulation distortion (40 kHz – 30
kHz = 10 kHz)
• Slew rate limiting within electronics of
ultrasonic output of a mic (or of square
waves from a test generator)"

The rest can be found here
http://audiosystemsgroup.com/Ferrites-Ham.pdf

It all sounds quite reasonable to a noob like me. What do you guys think?

[ikoflexer]

OK, by the same fellow, named Jim Brown, who gave a presentation at the 2008 Fall Convention of the Audio Engineering Society in San Francisco.

http://audiosystemsgroup.com/AES-RFI-SF08.pdf

Seems like a worthwhile presentation to have attended.

[anatech]

Hi Kannan,

now we have no answers to this - what frequency max and what amplitude of this range can cause problems in a line amplifier?

No one can give you any numbers for this off the cuff. There are so many variables that have an effect. The one thing that is true though, many (as in most) consumer electronics are built with very substandard power supplies. So the current state of the art in audio is very poor and very basic. Why else would a PS Audio power line conditioner (actually a power re-generator) actually make any difference to a product plugged into it?

If you read syn08's post, copied here:

I don't think potential RF ingress issues require any special or outstanding measures. To my experience, common sense practices like good grounding, proper shielding, smart wiring (reducing loops), filtering, decoupling, ground loops (or lack, thereof), RF input filters, etc... are addressing 99.99% of the potential issues.

He has pointed out the basic issues. I agree completely here. Just take basic precautions and use common sense. Too often, I believe that a power supply is just "tacked on" without too much thought. Same goes for thermal issues.

The susceptibility of the circuits that comprise the equipment will determine how much injected noise is acceptable.

Can we assume that 3 mm Aluminium enclosure would adequately shield a line amplifier from radiating RFI?

Not alone. Never forget the other cables running through the wall. Even a display may allow enough noise to enter to cause problems. That enclosure will attenuate the offending signals some, but we're back to my last statement above this quote.

Never assume anything.

I also assume that PS line should get to the rule so defined for RF noise( If John can give us some guidelines on Freq/amplitude guidelines for PSU noise - will be nice?)

Again, there is no easy answer for this. John could probably answer for circuits he designed and built (the build is another very important factor here). The way I approach this, and probably many others, is to follow good design practices and layout guidelines. Once you have those down, you just build that way. If you need more shielding because your noise is too high, then you need to investigate and find out why. Also, Aluminum will not block electromagnetic disturbances. So your wiring could easily couple to a noise source. May the games begin.

Look at any good piece of test equipment for clues. What you will see is generally far more attention paid to shielding and power supplies. If you then have a peek at some consumer audio equipment, you should be able to see clearly what the differences are.

Last point. Noise reduction from interference is a money game also. How far do you go? How much money is spent on the pretty box? How far down your noise floor is depends on how much money and research goes into the design (= money). Everything is a trade-off.

-Chris

[anatech]

Hi ikoflexer,
Jim Brown is a smart cookie. His stuff is well worth reading and he has a wealth of information on his site.

I'd recommend you start reading. Remember to keep a perspective on what effects they are talking about. His concepts are solid and based in reality from what I can recall.

-Chris

[john curl]

To minimize confusion here, I have found virtually everything is important in serious audio design. RF contamination is just one thing.
I would like to point out that of the 3 stages: We are now at stage 2.
This is: 'It exists, but it is not important'
IF you want to read anything further on AC line contamination, I recommend getting and reading the IBM article from the IEEE from Ikoflexer. Thanks Ike!

[PMA]

Yesterday I forgot to show the measurement extended to 1GHz. We can see FM broadcast, TV broadcast and GSM in the PSU output spectrum. Of course, the very high frequencies are coupled from the air.

[janneman]

Originally Posted by ikoflexer

OK, by the same fellow, named Jim Brown, who gave a presentation at the 2008 Fall Convention of the Audio Engineering Society in San Francisco.

http://audiosystemsgroup.com/AES-RFI-SF08.pdf

Seems like a worthwhile presentation to have attended.

I did attend. While it was worthwhile as a refresher on these issues, I don't think there was anything new presented.

jd

[syn08]

Originally Posted by PMA

Yesterday I forgot to show the measurement extended to 1GHz. We can see FM broadcast, TV broadcast and GSM in the PSU output spectrum. Of course, the very high frequencies are coupled from the air.

I can see from your measurement that GSM, TV broadcast, FM broadcast, etc... do exist.

And what about the spectra from 10-20GHz? There are Ku band satellite DTH TV broadcasts there, high power transponders with 27 MHz bandwidth, carrying digital signals. Those are much powerful than the next GSM tower! Not to mention the evil Ka band at some 54GHz.

So? Should we get tin foil helmets? Should we encase our amps in Faraday cages?