I want to make an amp with a Negative Output Resistance, but HOW?? - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 24th May 2004, 04:34 AM   #1
diyAudio Member
 
Join Date: Nov 2003
Location: Brisbane, QLD
Question I want to make an amp with a Negative Output Resistance, but HOW??

Hello,

I'm looking to make an amplifier with a negative output resistance, but I haven't been able to find much information about it, so I'm not really sure where to start. (Some schematics or links to schematics would be good!)

It's been argued that the output resistance generally has very little effect on the sound of a speaker, except to alter the tonal balance slightly, however I don't think that that's entirely correct, and I want to test my hypothesis. A simple experiment can help to illustrate what I mean:

If you tap on the cone of an unplugged box-mounted speaker, you'll find that the cone rings a bit like a drum. This is because the speaker's mass interacts with the air inside the box so that it vibrates at a certain resonant frequency. There is very little to stop the cone from vibrating, only its mechanical damping factor means that the vibrations are eventually absorbed. Short-circuiting the speaker's terminals brings the Qes into the equation, and it's the Q(total) that now determines how quickly the vibrations will be absorbed. Now when you tap the cone, it should produce a dull thud, which is more like tapping against a solid surface, though still not exactly.

In the same way that the speaker can resist forces from the outside (finger tapping), it also works hard - together with the amplifier - at stopping various forces from inside the box when music is being played. Even quite modest listening levels can produce an extremely loud back-wave inside a speaker's box. Imagine a room-full of sound, compacted into into the space of only a few litres (or a small fraction of a litre in the case of tweeters). That's a lot of pressure!

I figured that using an amplifier with a negative output resistance would have an effect that's equivalent to decreasing a speaker's Qe value, and thereby improving its boxed midrange performance. The same thing could be done mechanically by replacing the magnet with a much stronger one, however this gives a skewed frequency response with severely decreased bass, though I'm not 100% sure why. If an amplifier could emulate this electrically, f.r. changes could be equalized if need be. (I'm assuming that the cone movement is pistonic, so paper cones/fabric domes are not allowed.)

I've got an Accuton/Thiel C94 midrange speaker, which has a Qt of 0.18 which is pretty low so it's not that much of a problem, even in a very small box. However, I don't have specs on my C23 tweeter, and I suspect that its Qt is much more average, something like 0.5 perhaps. I think that many hard-domed tweeters would stand to benefit a lot from an amplifier with negative output resistance. The chamber behind the dome is usually tiny (a fraction of a litre is already on the large side), so it probably has a lot of extremely loud hf standing waves when music is played, and if I'm right this would explain why some tweeters have an undesirable metallic sound.

So now the question is: HOW? Normally if I push on the cone of a speaker, the amplifier produces a small current to maintain the correct voltage across the terminals (0V if it's idle). Ie: it wants to be a voltage source, regardless of the current needed to achieve that goal.

How do I make an amplifier deliberately over-compensate by amplifying the "microphonic" currents that are induced in a speaker? Is positive feedback involved? Can it be stable? What are the limitations?

CM
  Reply With Quote
Old 24th May 2004, 05:11 AM   #2
Won is offline Won
diyAudio Member
 
Join Date: Nov 2001
Location: Cambridge, Mass
http://sound.westhost.com/project56.htm

http://sound.westhost.com/z-effects.htm
  Reply With Quote
Old 24th May 2004, 05:25 AM   #3
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
Use a mix of voltage sensing feedback and current sensing feedback in the amplifier. This will allow for almost any output impedance you like. It can be made stable as long as there is a load connected.

An example: http://nds2.ir.nokia.com/downloads/a...isual/AVC1.pdf

Steven
__________________
The Analog Art shows no sign of yielding to the Dodo's fate. The emergence and maturation of monolithic processing finesse has perhaps lagged a bit behind the growth of the Binary Business. But whereas digital precision is forever bounded by bits, there is no limit excepting Universal Hiss to the ultimate accuracy and functional variety of simple analog circuits. - Barry Gilbert, 1973
  Reply With Quote
Old 26th May 2004, 08:57 AM   #4
diyAudio Member
 
Join Date: Nov 2003
Location: Brisbane, QLD
Thanks guys. Something tells me that using negative feedback tends to be a very ineffective and masochistic means to improve the sound of loudspeakers. Although it probably could work, I'd be forever tuning and tweaking the amplifier to cope with a reactive loadspeaker load.


If only there was a success story out there...

Incidentally, some of the comments on the sound.westhost.com/project56.htm site suggested that there's been success with negative output impedance with horn speakers. That makes perfect sense to me, as horn resonances are like just another type of box resonance, and resonances are what I'm trying to eliminate.

Lech
  Reply With Quote
Old 26th May 2004, 09:06 AM   #5
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
Quote:
That makes perfect sense to me, as horn resonances are like just another type of box resonance, and resonances are what I'm trying to eliminate.
With negative output impedance you can lower Qes (and Qtc subsequently) but you can't tame box resonances etc.

Regards

Charles
  Reply With Quote
Old 26th May 2004, 09:34 AM   #6
diyAudio Member
 
Join Date: Nov 2003
Location: Brisbane, QLD
Sorry about my wording. Anyone could get a large chisel, a block of granite, and make themselves a speaker box with completely solid walls. By box resonances I meant echoes, resonances and such-like in the internal air volume. How those waves are absorbed depends on their wavelength, and since they vary between >15m and <15mm, it's a pretty safe bet that no mechanical damping efforts will eliminate all of 'em. I'm hoping that I can make the speaker's cone electrically emulate a solid wall too.
  Reply With Quote
Old 26th May 2004, 09:41 AM   #7
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
Quote:
electrically emulate a solid wall too
The problem may be that your cone isn't a solid piston (not even an Accuton's one). I.e. the compensation signal applied to the voice-coil might not be the one in fact needed to cancel the acoustic signal going through the cone from behind.

IMO you are better off by designing the internal box shape accordingly to get the lowest achievable standing waves and reflections.

Just one more thing: Way above fs it is even better to drive speakers by a current source than a voltage source (let alone a negative output impedance) because it lowers THD and improves transient response.


Regards

Charles
  Reply With Quote
Old 26th May 2004, 11:02 AM   #8
diyAudio Member
 
Join Date: Nov 2003
Location: Brisbane, QLD
Quote:
Originally posted by phase_accurate


The problem may be that your cone isn't a solid piston (not even an Accuton's one). I.e. the compensation signal applied to the voice-coil might not be the one in fact needed to cancel the acoustic signal going through the cone from behind.
Yeah, I had that on my mind a bit earlier on. I'm assuming that at approx 5kHz (at the peak of the first break-up mode), the edge of the cone will lag by 180 degrees compared to the middle. Presumably the sound would cancel out, but not when considering all the physical offsets and probable constructive interference. I'm guessing that the phase shift gradually gets smaller at lower frequencies, down to a relatively benign <90 degrees below 2.5kHz. However, I can't be certain of how linear that phase shift is, and perhaps the electrical correction would only be valid up to a lower frequency such as 1.5kHz? If that's a reasonable limit, then it would still be usable, and better than nothing. For the C23 tweeter the limit could be 10kHz, just so it's more than an octave below resonance.

Quote:
...IMO you are better off by designing the internal box shape accordingly to get the lowest achievable standing waves and reflections...
Been there, done that. It's too much effort. I always need a bigger and bigger box to reduce the amplitude of internal waves, with no end in sight. This has led me to search for a better way. Perhaps I can achieve the boxless clarity of dipoles, in a box?

Quote:
...Way above fs it is even better to drive speakers by a current source than a voltage source (let alone a negative output impedance) because it lowers THD...
Btw why does it lower THD? I think there are inherent similarities between the required topologies for a current-source amp and negative output impedance, so I'll certainly look into it.


Another thought simmering away: maybe I could apply a similar concept to a headphone amp? Here the problem is reversed, because I would want to eliminate resonances on the inside by allowing sounds to be transmitted to the outside. If I can straighten out my thoughts on this, it would be a much easier step than designing for multiple loudspeakers, and multiple high-power channels.

CM
  Reply With Quote
Old 26th May 2004, 11:26 AM   #9
moamps is offline moamps  Croatia
diyAudio Member
 
Join Date: Oct 2002
Location: Croatia
Quote:
Originally posted by CeramicMan

If only there was a success story out there...
Maybe here you can find something usefull.
Gainclone with negative output impedance

Regards
  Reply With Quote
Old 26th May 2004, 11:32 AM   #10
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
Quote:
Originally posted by CeramicMan
Btw why does it lower THD? I think there are inherent similarities between the required topologies for a current-source amp and negative output impedance, so I'll certainly look into it.
A current source amp *negative* feedbacks a voltage proportional to load current.

A negative output impedance amp *positive* feedbacks a voltage proportional to load current.

A practical NOI amp has normal negative voltage feedback too, and this must be dominant for the system to be stable.
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
PSPICE and negative resistance rtarbell Software Tools 2 16th May 2009 08:34 PM
NPN negative collector resistance?? darkfenriz Parts 7 16th September 2007 05:45 PM
Two port negative resistance element rtarbell Parts 9 18th March 2006 09:27 AM
TDA 7294 -negative outlet resistance ?!! PAVLIN Analogue Source 5 21st November 2002 07:19 PM
Negative resistance better than current source for MOSFET follower amp? Circlotron Solid State 16 6th October 2002 12:54 PM


New To Site? Need Help?

All times are GMT. The time now is 01:53 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