Bob Cordell said:
G’day Bob.
I agree. However we could extend this argument to parameters other than stability into capacitive loads. How fast does an amplifiers slew rate really need to be? How low the THD? What degree of damping factor at low frequencies is really necessary?
Now, if a designer produces an amplifier with <10ppm THD-20, is that designer necessarily “incompetent” by virtue of having aimed for a performance standard well beyond the necessary?
I’d say no, and I’d say ditto for designing an amplifier to be stable into a highly capacitive (possibly unrealistically so) load connected to the output terminals, with regards to which an output inductor has utility (as dependant on other factors dependant on the design). Should all other performance aspects be simply ignored and the worth amplifier judged by the inclusion of an appropriately sized output inductor – a part which some people assert (with dubious qualification and mostly illogical arguments) necessarily degrades an amplifiers sonic performance?
Cheers,
Glen
AndyC, either Bryston uses a smaller inductor, or 'Stereophile' measures output impedance incorrectly. For example: The 4B measures .13ohms at 20KHz. How can this be? The 3B measures .09ohms., the 9 B measures .07 ohms.
john curl said:
I agree John. Something strange is going on here.
Here's another data point. Bascom H. King's 4B data shows a damping factor of about 40 at 20 kHz, which makes more sense for 2 uH. The resolution on this graph is terrible though. Graph is here, Chart 4
IMO, it's not that trivial to measure a 2uH inductor accurately. My guess is that the inductance value people think they're using is often a ways from what's really there, even for competent commercial manufacturers. It's also easy to be off on very low value output impedance measurements, though the error is usually high, not low. I'd take this stuff with a grain of salt.
Conrad, impedance is impedance. Either 'Stereophiles' measurements are off by about a factor of 2, or almost nobody uses 2uH coils anymore. You can't just 'handwave away' this kind of error.
john curl said:
Can i throw Mr Michaelson in the henhouse ?
=>the fully balanced and choked powersupply KW (aka KW1000) [Sanken 2SA1295/2SC3264] and KW-750 [Sanken SAP15N/SAP15P] power amps.(2 output coils/ch)
john curl said:
Every current SST model Bryston power amplifier schematic shows a 2uH output inductor, as well as all the other power amp series schematic's I bothered to down load. As avaliable here:
http://www.bryston.ca/BrystonSite05/BrystonDocs.html
I trust the accuracy of a manufactures schematics rather than trying to draw a conclusion from differing measured results from differing on-line magazines.
Hi Rodolfo,
I'm not upset by your responses here, you are aiming for the good of everybody and I support you in this.
The more we become involved in trying to sort things out, the more we all make little unitentional mistakes or omissions of detail which we do not know that others seek, on which some ( especially those who have no blind spots ) can pounce with self justified vigour.
Hi teemuk,
Thank you for your input. Would like to discuss more, but presently really short of time, more of which is being wasted having to defend against unproductive argument.
Hi Jan,
>> Inductance withing the loop can affect global loop performance and capabilities in the presence of reactive loads. It can trigger phantom non-linearity events which might arise at some particular instant of a music reproduction performance with a particular loudspeaker, but not others."
Sufficient?
Jan Didden<<
Sufficient ? ???
I have already clearly stated/shown that it is C.doms and stabilisation networks etc which unavoidably make amplifiers *themselves* inductive (reactive) as viewed from their global NFB node/output terminal !!!
This relates to tube based research older than me (before NFB was called NFB), and to SS papers published when useful power amps first started to become readily available. Look up a few *equivalent* circuits.
Also see Bob's post below yours.
Hi Bob,
Like Glen was. I feel that your are picking on one particular amplifier of John's rather than the choke aspect in particular !
I refer to my first sentence here. This has become a 'whipping' topic, and John was not the first to do this.
Hi Tony,
I meant Yes 'I agree with you'.
There is much that I have not yet seen previously discussed.
General ------------------------
Much discussed here takes our eyes of the ball !!!
Surely the point is audiblity of choke Yes/No, and the way in which this should be examined ?
I am not in a position to do any significant audio based work at the moment, but I have asked a friend if he can do so.
I have asked for a 5-6uH component to be inserted between a low inductance amplier which does not need a choke, and this to be A/B switched IN/OUT between his loudspeakers whilst monitoring - high energy transients, percussion, sibilance reproduction to see if he can hear a difference in accuracy of detail. Just one channel with the head say 1m away from the loudspeaker and with best ear facing the LS. He could maybe try white noise too.
I will report back.
Cheers ......... Graham.
I'm not upset by your responses here, you are aiming for the good of everybody and I support you in this.
The more we become involved in trying to sort things out, the more we all make little unitentional mistakes or omissions of detail which we do not know that others seek, on which some ( especially those who have no blind spots ) can pounce with self justified vigour.
Hi teemuk,
Thank you for your input. Would like to discuss more, but presently really short of time, more of which is being wasted having to defend against unproductive argument.
Hi Jan,
>> Inductance withing the loop can affect global loop performance and capabilities in the presence of reactive loads. It can trigger phantom non-linearity events which might arise at some particular instant of a music reproduction performance with a particular loudspeaker, but not others."
Sufficient?
Jan Didden<<
Sufficient ? ???
I have already clearly stated/shown that it is C.doms and stabilisation networks etc which unavoidably make amplifiers *themselves* inductive (reactive) as viewed from their global NFB node/output terminal !!!
This relates to tube based research older than me (before NFB was called NFB), and to SS papers published when useful power amps first started to become readily available. Look up a few *equivalent* circuits.
Also see Bob's post below yours.
Hi Bob,
Like Glen was. I feel that your are picking on one particular amplifier of John's rather than the choke aspect in particular !
I refer to my first sentence here. This has become a 'whipping' topic, and John was not the first to do this.
Hi Tony,
I meant Yes 'I agree with you'.
There is much that I have not yet seen previously discussed.
General ------------------------
Much discussed here takes our eyes of the ball !!!
Surely the point is audiblity of choke Yes/No, and the way in which this should be examined ?
I am not in a position to do any significant audio based work at the moment, but I have asked a friend if he can do so.
I have asked for a 5-6uH component to be inserted between a low inductance amplier which does not need a choke, and this to be A/B switched IN/OUT between his loudspeakers whilst monitoring - high energy transients, percussion, sibilance reproduction to see if he can hear a difference in accuracy of detail. Just one channel with the head say 1m away from the loudspeaker and with best ear facing the LS. He could maybe try white noise too.
I will report back.
Cheers ......... Graham.
Graham Maynard said:
Graham,
You mislead us in another post stating that an output inductor inside a fb loop was not the same as speaker cable inductance.
Several of us then mentioned that an output inductor is never inside a fb loop and thus is exactly the same effect as speaker cable inductance.
You said you never said that the inductor would be inside the fb loop.
I corrected you with the post of which you now take a part that is YOUR TEXT and try to mislead us again.
Do you really need this type of manipulation and misleading?
Jan Didden
Hi Jan,
Let's get something clear here.
Your ACCUSATION that I deliberately mislead readers is out of order, and your "Sufficient?" comes over as attempted point scoring in a fruitless and time-wasting argument; as with my fully written up graphical representation not being adequate to satisfy your standards.
I did NOT say about having 'AN INDUCTOR' inside the global loop !!!
Merely 'inductance within a closed NFB loop", in reference to components within the global loop which *generate* that inductance.
I refer you to post#211 above.
You have been following ( and picking on ) my words since 2004, so you really should know by now what I have been writing about, and you should know to question, rather than ACCUSE.
Surely I do not have to repeat myself with every post ?
When I see a C.dom (or components which control gain in a manner not linear with frequency) within an amplifier circuit I immediately visualise the amplifier in a way most people do not; ie. as an equivalent circuit with the series inducance(s) those components generate inside of that closed loop.
I feel that if more amplifier designers viewed amplifiers in this way (with regard to 'inductance within' the global loop) they might actively modify the topologies they presently promote !
Cheers ....... Graham.
Let's get something clear here.
Your ACCUSATION that I deliberately mislead readers is out of order, and your "Sufficient?" comes over as attempted point scoring in a fruitless and time-wasting argument; as with my fully written up graphical representation not being adequate to satisfy your standards.
I did NOT say about having 'AN INDUCTOR' inside the global loop !!!
Merely 'inductance within a closed NFB loop", in reference to components within the global loop which *generate* that inductance.
I refer you to post#211 above.
You have been following ( and picking on ) my words since 2004, so you really should know by now what I have been writing about, and you should know to question, rather than ACCUSE.
Surely I do not have to repeat myself with every post ?
When I see a C.dom (or components which control gain in a manner not linear with frequency) within an amplifier circuit I immediately visualise the amplifier in a way most people do not; ie. as an equivalent circuit with the series inducance(s) those components generate inside of that closed loop.
I feel that if more amplifier designers viewed amplifiers in this way (with regard to 'inductance within' the global loop) they might actively modify the topologies they presently promote !
Cheers ....... Graham.
Graham Maynard said:
Bonsai says (and I paraphrase):
"Any 2uH output inductor effect is swamped by speaker cable inductance, xover inductance, etc. "
You, dear Graham, say (and I paraphrase):
"Inductance inside a nfb loop is not the same as cable inductance outside the nfb loop"
Please explain how we should interprete then your statement? Should we consider it as totally unrelated to Bonsai's statement? If yes, what was the purpose of your reply, if any?
I have never attacked you as a person. I have no issues with Mr Graham Maynard at all.
I do have issues with your technical statements and your refusal to answer members concerns if they don't suit you.
I do have issues with your insistence to built a whole framework of bogus theories as based on a single input signal that is irrelevant to audio. I do have issues with you refusing to repeat your tests with relevant audio-band signals. I think this is intellectually unfair and misleading.
I will continue to point this out.
Jan Didden
I just found this tread.
John Curl and all who have contributed ... many thanks for your help to folks like myself who like to build our own amps.
I stopped using an output coil 20 years ago. In the 1980's I built a Stan Curtis pair of 60 watt mono blocks. Later stan mentioned that the sound may improve without the output coil. When I removed the coil at that time I heard little difference. I was then using kef 105's which are large three way speakers.
I later changed my speakers to full range direct connected 5" speakers with just a little assistance from a tweeter and a sub base. Then I went back and tried all my pre and power amps with the new speakers. The new speakers bring much more insite. This time around I could here the difference of including or not including the output coil.
More concrete tests from others would also help. After all it is about enjoyable sound quality.
Don
John Curl and all who have contributed ... many thanks for your help to folks like myself who like to build our own amps.
I stopped using an output coil 20 years ago. In the 1980's I built a Stan Curtis pair of 60 watt mono blocks. Later stan mentioned that the sound may improve without the output coil. When I removed the coil at that time I heard little difference. I was then using kef 105's which are large three way speakers.
I later changed my speakers to full range direct connected 5" speakers with just a little assistance from a tweeter and a sub base. Then I went back and tried all my pre and power amps with the new speakers. The new speakers bring much more insite. This time around I could here the difference of including or not including the output coil.
More concrete tests from others would also help. After all it is about enjoyable sound quality.
Don
Hi
The difference was one of clarity. more detail and clearer notes. Most obvious on complex pasages in the mid and treble. The differences between notes and silences becomes more obvious. I guess we can never remove all distortion from electronically projected music but this did help.
It was the same type of difference I heard when I first changed speakers from the Kef 105's to a high quality direct connected 5" speaker - but to a lesser extent. ( For info my speakers are connected to the power amp by 1 meter thick cable )
Don
The difference was one of clarity. more detail and clearer notes. Most obvious on complex pasages in the mid and treble. The differences between notes and silences becomes more obvious. I guess we can never remove all distortion from electronically projected music but this did help.
It was the same type of difference I heard when I first changed speakers from the Kef 105's to a high quality direct connected 5" speaker - but to a lesser extent. ( For info my speakers are connected to the power amp by 1 meter thick cable )
Don
Hi
I do not recall the value. However I seem to remember it was about 20 to 30 turns of wire in a 1" dia coil around a wire wound resistor.
Don
I do not recall the value. However I seem to remember it was about 20 to 30 turns of wire in a 1" dia coil around a wire wound resistor.
Don
Hi Don,
I wonder was this the same amplifier with and then without a choke such that you can categorically state it was the choke alone causing the change in reproduction, and not some amp/choke combination ?
Was this an improvement in defition/clarity and silence behind notes/voices (this silence as with good tube amps) such that you would never willingly want to go back to listening via a choke again ?
There is a step level of quality in the equipment being used as a composite system necessary before you can hear this difference, and frankly, I would suggest that anyone who has tried but not yet observed it on their own equipment still has some way to go !
Cheers ......... Graham.
I wonder was this the same amplifier with and then without a choke such that you can categorically state it was the choke alone causing the change in reproduction, and not some amp/choke combination ?
Was this an improvement in defition/clarity and silence behind notes/voices (this silence as with good tube amps) such that you would never willingly want to go back to listening via a choke again ?
There is a step level of quality in the equipment being used as a composite system necessary before you can hear this difference, and frankly, I would suggest that anyone who has tried but not yet observed it on their own equipment still has some way to go !
Cheers ......... Graham.
Bob, do you really think that I have made a synthetic 2uH inductor? I feel like the caveman when asked questions by the lady psychriatrist. WHAT? 
Hi guys, what a peaceful thread, let's bring in some roses 😀
Executive summary :
- John Curl listens to amp with output inductor
- John Curl listens to amp without output inductor
- John Curl notes difference in sonics and wonders why this is so
- John Curl make the (wrong) choice to trust human nature and post his findings in the forum to trigger some collaborative group thinking effort
- Legions come and post 10 pages saying Curl is so full of cerumen he can't hear.
So, what now ? Here are the things that come to my mind when John Curl (who IMHO Knows What He's Doing) says that the output inductor affects the sound...
- RF : If you want to keep RF out, you use a RF specified lowpass filter, or common mode choke, not just some inductor, which may act like an AM antenna (esp. air cores), worsens common mode RF, injects RF into ground, makes the two terminals have a different RF impedance, may turn your cable into a tuned 1/4-wave antenna, paint your CDs edges green while you sleep, etc.
- Inductors are known to emit unspecified electromagnetic fields into your enclosure, and to form magnetic circuits with anything ferrous onboard.
- Did someone stick a network analyzer in there ?
- Does it sound different with the output inductor on the amp PCB, or at the speaker terminals at the other end of the wire ? (if yes, then the problem is not the inductance, but the parasitics)
- Inductors wound around resistors make a nice transformer of unspecified properties with the resistor itself, which is a metal film cut in SPIRAL (ie. a coil) on a ceramic cylinder. The steel endcaps will also saturate and add distortion. Winding it on capacitors like Dynaco did is plain stupid : it just adds shorted turns.
- Didn't someone stick a network analyzer in there, dammit ?
- Does it mess with the class-AB crossover distortion and switching characteristics ?
- Does it vibrate and thus inject EMF into the circuit ? (air core inductors have mechanical resonances...)
- etc.
Executive summary :
- John Curl listens to amp with output inductor
- John Curl listens to amp without output inductor
- John Curl notes difference in sonics and wonders why this is so
- John Curl make the (wrong) choice to trust human nature and post his findings in the forum to trigger some collaborative group thinking effort
- Legions come and post 10 pages saying Curl is so full of cerumen he can't hear.
So, what now ? Here are the things that come to my mind when John Curl (who IMHO Knows What He's Doing) says that the output inductor affects the sound...
- RF : If you want to keep RF out, you use a RF specified lowpass filter, or common mode choke, not just some inductor, which may act like an AM antenna (esp. air cores), worsens common mode RF, injects RF into ground, makes the two terminals have a different RF impedance, may turn your cable into a tuned 1/4-wave antenna, paint your CDs edges green while you sleep, etc.
- Inductors are known to emit unspecified electromagnetic fields into your enclosure, and to form magnetic circuits with anything ferrous onboard.
- Did someone stick a network analyzer in there ?
- Does it sound different with the output inductor on the amp PCB, or at the speaker terminals at the other end of the wire ? (if yes, then the problem is not the inductance, but the parasitics)
- Inductors wound around resistors make a nice transformer of unspecified properties with the resistor itself, which is a metal film cut in SPIRAL (ie. a coil) on a ceramic cylinder. The steel endcaps will also saturate and add distortion. Winding it on capacitors like Dynaco did is plain stupid : it just adds shorted turns.
- Didn't someone stick a network analyzer in there, dammit ?
- Does it mess with the class-AB crossover distortion and switching characteristics ?
- Does it vibrate and thus inject EMF into the circuit ? (air core inductors have mechanical resonances...)
- etc.
- Status
- Not open for further replies.