Headphone/PA with defined Ro

[ricardo]

Ro.gif (6.19 KiB) Viewed 17968 times

The threads on headphone amps prompt me to ask how we can have a defined Ro on an amplifier without wasting too much power.

This is easy for a speaker PA. (Also to modify an existing PA) The Ro isn't exactly given by that formula but more than good enough. It is never necessary to resort to the complicated stuff in http://www.aes.org/e-lib/browse.cfm?elib=6099 which was written by academics with little idea of real speakers.

Bob Carver tells us that Ro=1R goes a long way to emulating the sound of Classic Valve[*] amps. He's one of the few people who have conducted properly controlled Blind Listening Tests of Golden Pinnae amps with the Platinum Pinnae of Stereophile reviewers when he won the Stereophile Challenge to make a transistor amp sound like their Golden Pinnae favourites.

It's easy for speakers cos they are usually floating with separate "hot & cold" wires.

But where variable Ro would be most useful is with headphones. Headphone designers assume a whole lot of different Ro. IEC 61938 specify 5V @ 120R which is a lot to stick in series with a +/- 15V headphone amp. This is where a feedback controlled Ro would be useful.

But headphones have a common (to both L&R) earth return which is the Shield on the usual TRS plug. We can't use this simple circuit which demands separate "cold" connections.

The obvious solution is to have 0R1 on the hot terminal and a differential amp bring this down to earth reference to drive R2. Maybe using THAT1246.

Does anyone have a more elegant solution?

[*] Tuubes to yus Rebel Colonial Yanks

[edit] Add a Ro=120R circuit as IEC61938. Pontificating later on in thread[/edit]

[mediatechnology]

Hmm. Let me think about where I saw something like this recently. I want to say it was an EDN or Electronic Design article or IFD.

What actaully was the Crown Delta/Omega?
We had those where I worked once.

[JR.]

There are many such circuits, and IIRC Sondermeyer at peavey could even make a negative output impedance. to correct for impedance in wiring and connectors.

Yup, the basic ground current sense and application of feedback based on that. Peavey used a current sense in series with the hot and used a small transformer to level shift that down for feedback use, so it would still work with bridged amps.

JR

.

[mediatechnology]

Yeah, I think the Delta-Omega was a negative OP impedance.

ricardo - I think a THAT1240/1246 could be used to measure the floating shunt's current.

[ricardo]

Added a complete HP amp circuit to my original pic with IEC61938 Ro=120R. I use THAT1240 for higher G so R2 is bigger. R2 also sets the gain for the whole amp about 20dB so a log pot volume control with centre detent will be about 0dB.

Having R3, the current sense resistor at the hot end is necessary for phones cos they have a common terminal in TRS plugs.

Swapping the input terminals of the THAT1240 so G is -ve gives -ve Ro.

PRR at http://gilmore2.chem.northwestern.edu/u ... 3&tid=2432 says impedance variation in phones would only amount to 2.5dB response variation from 0R to zillion R. So is this Ro stuff worth doing?

Probably, cos the distortion reduction is worth having. If Ro=Rl, distortion (at your lugholes) around 100Hz is about halved for the major distortions in moving coil and pro-rata as Ro increases down to about 0.1% thd. Means there's little advantage (or effect) with 600R phones but lots with 32R.
___________________________

More difficult is a headphone amp with PRR's 7V 120mA to draw blood with any phones. The 120mA is for 32R iThing phones etc. 5532 only gives 35mA into low loads. You could use THAT1646 driven CM ala Wayne but you'd need a buffer to drive that.

I'm tempted to use a small PA like TDA2040. It's not "1 part per zillion distortion" but as we are using the circuit to reduce total distortion of amp/phones, its the overall result that matters. But those of you with Golden Pinnae 600R phones may want something better in which case 5532 is good.
___________________________

Anyone have recommendations for good/accurate/comfortable phones? The ones I like & trust are long Unobtainium.

[ricardo]

There are many such circuits, and IIRC Sondermeyer at peavey could even make a negative output impedance. to correct for impedance in wiring and connectors.

This stuff was very important to me in my previous life. The real advantage is when you start cancelling the DC resistance of the voice coil but it needs to be matched to the speaker and its box. You can do a lot more.

I posted this in another forum.
_________________
> For what it is worth, I think this is impossible. It seems to violate the cause and effect principle. How could the amplifier "know" what the speaker is up to ? Short of sensing the actual motion of the drivers or the actual sound... this has to be sort of by guess and by golly, it seems to me.

It is entirely possible for an "amplifier" to know what the speaker is doing.

It needs to sense the speaker current. An amp which twiddled its Output Z using both current & voltage feedback does this. Speakers act as accurate microphones (sense the actual sound) if operated into Low Z.

See "Loudspeakers as Microphones" - Peter Baxandall special lecture London AES (early 80s, late 70s?)

If operated into High Z, then the voltage at the terminals is a measure of cone velocity.

Both these mechanisms obey superposition & Thevenin so if you're clever, you can look at this while the amp is giving zillion volts and amps to the speaker. But non est tantum facile.

There are several tried & tested methods of using this "controlled output Z" or "current + voltage feedback" or "actual sound & motion feedback" (different descriptions of the same thing) if you incorporate the amplifier design in the speaker. Some of these are in the zanier incarnations of my Powered Integrated Super Sub technology.

The simplest is the negative output R that Fons mentions.

More sophisticated but similar (??!) is ACE technology by Erik Stahl which was used by Audio Pro, Sweden for subs. Unfortunately, since he left, there isn't anyone there who understands it. Anyone have a contact for Erik? Or a clean copy of his original AES preprint?

These methods have the distortion reduction and dynamic overload protection features discussed in Mills & Hawksford. However, they are badly affected by heating of the voice coil.

David Birt (?) did an excellent IoA paper at Windermere where he arranged speaker and amp in a bridge so he could measure and compensate for heating on the fly. Anyone have an email for him?

These are the most elegant methods and they can be analysed from many viewpoints. Some of these viewpoints don't show up the distortion reduction advantages clearly.

I'm contemptous of methods which rely on extra transducers or extra windings (like Mills) or zillion point DSP EQ especially if they don't give ALL the advantages of the elegant methods.

A brute force zillion point approach possible today is measure accurately speaker Z (not that easy) and tailor the output of a High Z amp to suit. This would give some but not all the advantages of the above systems cos it wouldn't "know" what the speaker is up to.
_______________________
Loudspeakers as High-Quality Microphones - Baxandall
Preprint 1593 AES jan80

Synthesis of Loudspeaker Mechanical Parameters by Electrical Means: A new Method for Controlling Low-Frequency Loudspeaker Behaviour - Stahl
Preprint 1381 AES oct78

[mediatechnology]

ricardo: My friend and previous co-worker Rosalfonso Bortoni wrote "Audio Amplifiers Principles, Characteristics, and Applications." His graduate work was with power amplifier output stages and current-driven speakers. As far as I know it has not been translated from Portugese. I think I may have seen it on Amazon. You can contact Ros at THAT Corporation and he may have a translation.

The high-side current shunt and measurement looks really cool. Thank you for posting it. I'm tempted to compare it to "Zero Ohm" drive. If it makes sense for speakers I don't know why it wouldn't for headphones.

How could the amplifier "know" what the speaker is up to ? Short of sensing the actual motion of the drivers or the actual sound... this has to be sort of by guess and by golly, it seems to me.

It is entirely possible for an "amplifier" to know what the speaker is doing.

It needs to sense the speaker current. An amp which twiddled its Output Z using both current & voltage feedback does this. Speakers act as accurate microphones (sense the actual sound) if operated into Low Z.

I have a Panasonic vacuum tube stereo power amp from the 1960s with matching speakers that used the "Motional Feedback System" that had a measurement microphone coaxially mounted to the LF drivers. On the front panel are thumbwheel switches for cut-off and damping. I need to pull and scan that manual for historical reasons.

More difficult is a headphone amp with PRR's 7V 120mA to draw blood with any phones. The 120mA is for 32R iThing phones etc. 5532 only gives 35mA into low loads. You could use THAT1646 driven CM ala Wayne but you'd need a buffer to drive that.

You might want to use a buffered THAT1646 using external boost transistors. We have one in another thread.

But those of you with Golden Pinnae 600R phones may want something better in which case 5532 is good.

I don't think there are any 600 Ohm phones left in production. I think the highest I've found is one model at 300 Ohms.

[ricardo]

The high-side current shunt and measurement looks really cool. Thank you for posting it. I'm tempted to compare it to "Zero Ohm" drive. If it makes sense for speakers I don't know why it wouldn't for headphones.

I was hoping someone would come up with an even more elegant circuit using less bits. This is how I did it in Jurassic times for speakers but with separate OPA diff amp.

The 1240 simplifies things and means you can do a supa dupa, controlled Ro, stereo HP amp with 4.5 devices. 5532 to buffer, 2x1646 as main amps, 2x1240 current sensors. That's 4 devices too many IMHO but I can't see how to make it less.

There's actually 3 separate situations.

1 "Normal" speaker amp. Here Ro = 0R is vital cos that's how speakers are designed. 1 part per zillion distortion at the PA terminals bla bla.

2 Powered Integrated Super Stuff. My specialty. You integrate amp & speaker to get something greater than the sum of the parts. Current Drive, voice coil DC cancelling, all come under this. You can get the amp to cancel out some speaker distortions and do loadsa EQ, protection stuff too, both analog & digital. I see some THAT chips that would be useful in this.

3 Headphones. A new field cos I'm not sure H/P are designed for "Zero Ohm" drive. Headphones are less sensitive so there is less electromagnetic damping. The impedance curve tells you how much. If it is flat, then it matters not, response wise, whether R0 = 0 or zillion.

http://www.benchmarkmedia.com/discuss/f ... dphone-amp dun really understand damping. Basically, if the impedance curve is flat, electromagnetic damping is negligible.

If its not flat, that's another story. Certain resonances are reflected in the impedance curve and will be exacerbated by current drive. If they are not reflected, Ro = 0 will not increase the damping at all.

But current drive will reduce many of the distortions inherent in practical moving coil transducers. Difficult to produce hard evidence of this without a B&K or GRAS mike and artificial ear.

You can increase Ro in my H/P amp simply by making R1 = zillion. Then Ro is simply the CM input R of the 1246 at 4k5. Gain structure needs thinking through. Worth trying with the better phones with flat impedance curves. Probably not good for cheap phones with resonances.

> More difficult is a headphone amp with PRR's 7V 120mA to draw blood with any phones. The 120mA is for 32R iThing phones etc. 5532 only gives 35mA into low loads. You could use THAT1646 driven CM ala Wayne but you'd need a buffer to drive that.

You might want to use a buffered THAT1646 using external boost transistors. We have one in another thread.

I saw that but I think your CM driven 1646 in parallel is an elegant sufficiency with 140mA peak. The Golden Pinnae version will have a 5W dropper resistor for the plasma display on a giarnormous H/S so I can claim Class A operation.

[mediatechnology]

Two minor things to overcome using bare 1646s to drive HP:

1) There are internal build-outs of 25R/output. Your defined Ro would need to compensate for deal that.
2) Where do you return feedback?

There almost needs to be on Op Amp ahead of the 1646 with the 1646 only a unity gain power booster.

[ricardo]

Two minor(??!!) things to overcome using bare 1646s to drive HP:

1) There are internal build-outs of 25R/output. Your defined Ro would need to compensate for deal that.
2) Where do you return feedback?

There almost needs to be on Op Amp ahead of the 1646 with the 1646 only a unity gain power booster.

Duu.uh! Was going to say da feedback compensates for da 25R build-outs when you hit me "what feedback?"

Might have to be one of dem BUF things but I've never used them before. (Never used 1646 either but Wayne convinced me they are the dog's bollocks ). Actually as you need a buffer OPA anyway, the 1646 might as well be the BUF. The build-outs save a resistor you should have in place to avoid melting 32R phones.