John Curl's Blowtorch preamplifier part II - Page 3052 - diyAudio
Go Back   Home > Forums > Member Areas > The Lounge

The Lounge A place to talk about almost anything but politics and religion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th November 2012, 03:44 PM   #30511
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Quote:
Originally Posted by DF96 View Post
Thanks for providing the link. As you say, it illustrates what I have been saying. On the left side there is the equation for propagation constant, which may answer Esperado's question.


Be careful. At lower frequencies the line may have greater attenuation because a greater proportion of the series impedance is made of resistance. As I have said, your method and my method, if applied fully and correctly, must provide the same result as your impulses synthesise my audio waveform. Any deviation will be due to different approximations being used. The net result will be a small time delay and a small audio phase shift.
Chapter 5 of my reference contains the derivation in the Belden white paper.
__________________
Silence is so accurate.
  Reply With Quote
Old 30th November 2012, 03:45 PM   #30512
diyAudio Member
 
Joshua_G's Avatar
 
Join Date: Nov 2008
Location: Small village, Israel
Quote:
Originally Posted by PMA View Post
Listening tests performed are the evidence to me. I do not care we do not have an appropriate measuring method yet.
Indeed, this is also my approach.
  Reply With Quote
Old 30th November 2012, 03:49 PM   #30513
diyAudio Member
 
jneutron's Avatar
 
Join Date: Jan 2004
Location: away
Quote:
Originally Posted by DF96 View Post
Be careful. At lower frequencies the line may have greater attenuation because a greater proportion of the series impedance is made of resistance.
Absolutely concur. A #12awg will have net 2 milliohms per foot, so for a ten foot run, 100 transits is still only 2 ohms.

As the signal attenuates, the steps to final value become smaller, so more steps will be required. That's why I said before that it will cause a longer tail.

In the end of course, the final value across the simple 10 foot length will follow ohms law, but it takes time to get there. More time than my simplistic LC analysis would indicate.

jn
  Reply With Quote
Old 30th November 2012, 04:31 PM   #30514
diyAudio Member
 
jneutron's Avatar
 
Join Date: Jan 2004
Location: away
Quote:
Originally Posted by PMA View Post
My efforts is always to get as little HF contamination at the power amplifier output as possible. Measured in a whole audio chain. Wideband measurement and reduction of HF contamination by circuit design, shielding, grounding, wiring, PSU design, everything that matters. Reduction and minimization of HF contamination goes hand in hand with resulting "sound quality".
If you notice, my test is just a basic 20 to 20k stimulus of ground current. I had a spare dynaco 400 gathering dust in the basement, so decided to put it to use. I used a 4 ohm series resistor because I didn't like to have a pure inductance loading the amp. That tends to increase output stage dissipation as it goes hard 4 quadrant.

I used an open gap magnet to stimulate the ground loop for several reasons.
1. If I inserted a source into the ground loop, I alter the loop resistance. The purpose of the test is to do it in situ.
2. Using the gapped magnet allows me to just move the flux generator into the loop without disconnecting either PC's or IC's. The alternative, a toroid for example, would work just as well but requires disconnect. There have been anecdotal accounts of insertion contact problems, so I needed to remove that particular confounder.
3. I had one available...

I use a current viewing toroid for the same reasons as the loop magnetic drive. A non insertion capability of reading the drive current. Smaller is better of course, as it will slightly change the loop inductance as will the drive magnetics, but hey, something's gotta give.

The reason for the current view is to establish constant drive current. My intent is to establish a test protocol using standards, say 100 milliamps sine 20 to 20k, with a look towards either input referred or output referred coupling..swept.

It would be very good to do this while driving the poweramp hard as well, as there is also coupling between input supply currents from the line, supply dc line coupling to ground, and even mundane things like where the star connection for the input pair is and how it couples.

So watching the output spectra during a sweep no signal, or looking for a specific driven frequency buried within music.

The net end result? Step 1 in eliminating ground loop sensitivities of the equipment.

ah, where's my manners...
The pic on the left shows the couplings. Pink is the physical loop in the source component which will generate a voltage as a result of the ground current Blue is the same in the amp. edit(well, the origonal jpegs were pink and blue..) The IC error signal is the voltage generated by the rate of change of flux within the overall loop minus the flux created by the ground current which results.

The pic on the right is one method of taking the ground loop current and bypassing it's effects. It is based on the principle that a cylindrical shell of current has no internal field (the center black region of my avatar is the field map of a dual coax cable.) So I take the input ground reference wire, and pass it through that no field region within the current carrying conductor down to the reference node.

jn
Attached Images
File Type: jpg ground loop.jpg (59.5 KB, 170 views)
File Type: jpg input noise isolation.jpg (142.9 KB, 173 views)

Last edited by jneutron; 30th November 2012 at 04:44 PM.
  Reply With Quote
Old 30th November 2012, 05:33 PM   #30515
diyAudio Member
 
Esperado's Avatar
 
Join Date: Jul 2011
Location: France
Quote:
Originally Posted by john curl View Post
JN are you telling me that the difference between quality and average wire is the tendency toward ground loops? '-)
How do-you define "average" and "quality ?
__________________
Ultimate Protection and more.The Only Source of Knowledge is experience, everything else is just information” ©A. Einstein
  Reply With Quote
Old 30th November 2012, 05:39 PM   #30516
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
John N., thanks for the description of your test. In my case, I have everything in class II (so there is no galavanic ground loop, though loop is still closed through stray and parasitic capacitances).
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html
  Reply With Quote
Old 30th November 2012, 05:58 PM   #30517
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
this is intended to address the claim that speaker impedance variation between channels - due to the speakers parameters varying with different operating point could cause "large" us delay "modulation" - due in part to cable Z

used 12' (repeated sections) of Pass 18 awg zip cord model from http://www.passdiy.com/pdf/spkrcabl.pdf

amp modeled with 1 uH isolation L

speaker 4 Ohm nominal @1, step definitions swap in C reactance to give ~ |4 Ohm| 1 kHz Z with ~75 degree phase angle @2

sorry about cursor window over schematic - but you should get the idea - the differential group delay between the loads is ~ 3.7us @1 kHz

to test cable Z role you could parameterize the cable Z - not thinking of that ahead of time - I instead raised the amp, speaker model Z 4x - with the result that the 1 kHz group delay difference dropped to ~900 ns

this basically should give the same result as dropping cable Z 4x with the same amp, speaker models


so it does look like single digit uS delays at audio are possible with zip cord, low Z speaker - and lower cable Z would reduce the effect


need to parameterize the cable - but a quick check shows that for the same L,C but 4 mOhm per section the group delay difference reduced to 1.2us @ 1 kHz so this low at least the effect seems to be dominated by cable series R
Attached Images
File Type: png spkr_cbl.PNG (85.2 KB, 169 views)
Attached Files
File Type: asc spkr_cbl2.asc (6.7 KB, 3 views)

Last edited by jcx; 30th November 2012 at 06:17 PM.
  Reply With Quote
Old 30th November 2012, 06:19 PM   #30518
diyAudio Member
 
jneutron's Avatar
 
Join Date: Jan 2004
Location: away
Quote:
Originally Posted by jcx View Post
this is intended to address the claim that speaker impedance variation between channels - due to the speakers parameters varying with different operating point could cause "large" us delay "modulation" - due in part to cable Z

used 12' (repeated sections) of Pass 18 awg zip cord model from http://www.passdiy.com/pdf/spkrcabl.pdf

amp modeled with 1 uH isolation L

speaker 4 Ohm nominal @1, step definitions swap in C reactance to give ~ |4 Ohm| 1 kHz Z with ~75 degree phase angle @2

sorry about cursor window over schematic - but you should get the idea - the differential group delay between the loads is ~ 3.7us @1 kHz

to test cable Z role you could parameterize the cable Z - not thinking of that ahead of time - I instead raised the amp, speaker model Z 4x - with the result that the 1 kHz group delay difference dropped to ~900 ns

this basically should give the same result as dropping cable Z 4x with the same amp, speaker models


so it does look like single digit uS delays at audio are possible with zip cord, low Z speaker - and lower cable Z would reduce the effect


need to parameterize the cable - but a quick check shows that for the same L,C but 4 mOhm per section the group delay difference reduced to 1.2us @ 1 kHz so this low at least the effect seems to be dominated by cable series R
Very nice indeed. The assuption that it scales is a good one.

If you match cable to load, will the group delay drop to 0? If so, the model doesn't include the physical length. I'm not aware of how to include the actual physical length into the model.

What would be a good range of speaker impedances to verify? Is 1 to 60 reasonable?

jn
  Reply With Quote
Old 30th November 2012, 06:25 PM   #30519
diyAudio Member
 
jneutron's Avatar
 
Join Date: Jan 2004
Location: away
Quote:
Originally Posted by PMA View Post
John N., thanks for the description of your test. In my case, I have everything in class II (so there is no galavanic ground loop, though loop is still closed through stray and parasitic capacitances).
Is your amplifier a stereo input with unbalanced rca's?

If so, do you have it setup so that the shield return currents do not share?

jn
  Reply With Quote
Old 30th November 2012, 06:38 PM   #30520
PMA is offline PMA  Europe
diyAudio Member
 
PMA's Avatar
 
Join Date: Apr 2002
Location: Prague
Stereo input, RCA SE, 2 completely separated channels, 2 transformers, 2 power supplies.

You can see the amp at bottom right corner.
Attached Images
File Type: jpg studio esos.JPG (142.6 KB, 162 views)
__________________
Pavel Macura
http://web.telecom.cz/macura/audiopage.html

Last edited by PMA; 30th November 2012 at 06:43 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 10:48 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2