On - Going Quad 909 Performance.

[Hodges]

I have spent quite some time listening to a classical mix on Amazon Music, playing it with my modified loudness control switched in set for 6/12db. It certainly lifts the bass very cleanly and widens the aural spectrum in a way that seem quite natural. However that is at low listening level.


However this morning, I switched it out to "Flat" and upped the volume a little. Compared to when switched in – it sounds so "Raw", as if one is sitting in front of the piano/instruments. Quite impressive really. Some piano notes are so crisp it is like scratching glass, violins full of rawness and guitars like I have the instrument in my hands.


Not sure how long I could listen to it as, as good as it is, it seems like it is unfinished as musical compositions go – before the engineers get to work on it. However, these are popular interpretations of musical compositions so need to be judged less severely than the real articles. The openness of the Spendors continue to amaze me – faultless, with clean bass – but not something modern enthusiasts would hanker after – not pronounced enough, perhaps because of the lack of "Boominess".


I don't think however, I would want conventional tone controls as it would probably mean having to continually adjust bass/treble/filter, for each track. This gizmo leaves things alone in the lower mid range and addresses low notes/bass without emphasizing them too much – it would do if I switched in more but it would be excessive. I think, more out of luck than judgement, it turned out better than I expected.


I will have a week or two listening to the Stax, prior to comparing a pair of HiFi - mans, just to "Get my Eye in". Plus A – B comparison can be done directly with the Quad powering the Stax, while the HiFi's are plugged into the Sugden. However, we have been staying with family at an apartment similar to one we are looking for and it looks like my Spendors may well fit in well - given the size of the lounge. That would suit me down to the ground as I am at a loss as to what could replace them with, within my budget.

Currently my Headmaster Sugden remains redundant as I continue to drive the 909 via a ALPS Pre-amp/potentiometer, with my modified Loudness Control; also switched ALPS, in circuit too. I also have a couple of in - line Switchable Phono 10/15/20db. Attenuators at the input to the Quad. Principally because of the differing sensitivities between the Yamaha CD player, the Zen V2 and the Suggie and Quad. They work well and give me extra protection and piece of mind, knowing I won't overdrive the 909, the Suggie or importantly my Spendors. Our dining room is a bit on the small side and pretty bare as far as soft furnishings go; given everything is packed away, pending our move. Non the less, the brightness this causes is not something that worries me and the Bass doesn't cause the room to boom at the kind of level I play at.

I remain absolutely thrilled/astounded with the Quad 909. What it must be like driving Electrostatics, I would love to know. The grip it exercises over the Spendors is just fabulous. Nowhere is there any looseness in the music, regardless of the instruments, brass, woodwind, piano, stringed or percussion. And it will embarrass any artist that is not up to task with his/her singing or playing. Quite funny on occasions really. I am considering a second 909 as I would like to fit one under each Spendor, into their stands, shortening the loudspeaker cables and driving them with longer phono leads. But that depends on the price and condition I can get one for.

 

[Hodges]

Ok, boring - I know. However, sitting here listening to "Heartstrings" on A.M,, I am still coming to terms with how much more I seem to be getting out of my equipment combination.

Many other amplifiers, including the 303 seem fulsome when a narrow array of instruments are playing. But, it seems now, as though, when a full ensemble plays, power drains from the "Whole" and it appears as if the volume control has been turned down? Illusion, or what? The difference between RMS measurements and Music Power seems to be the only reason I can think of. I.e., the amps power, when shared between a multitude of instruments, can only mean the total power is is distributed between the sum of the instruments, but when integrated, equates to the RMS rating of the amplifier - assuming it is running flat - out?

The reason why this has come up is, the 909, although not at high volume setting, has added weight - a term I used earlier, which in other amps is lacking. I think, it's because of it's power rating; it can sustain the dynamic range and relative loudness of each instrument and thus it sounds that much louder - vis - a - vis Class A amps with their beefy power supplies, even though they may be only 10 to 20 watts rated.

Food for thought - but is a very revealing trait which I can't remember too clearly, if I have experienced before. I discount the times when I have been listening to top notch demo's, I am talking of a domestic setting where one wouldn't play music at ear - piercing levels.

I am in love! My Spendors are sublime. So delightfully silky. Ok new drivers, but fifty years on, they are still riveting to listen to. I can't part with these now.

 

[Hodges]

Which brings this full - circle. As I said light - years ago, when I commented upon early Class AB amplifiers and their inadequate power supplies. I am not sure that still is not true of some modern amplifiers either.

"If the aggregate power within the audio envelop comes close to the equivalent RMS rating @ 1khz, there is less headroom left when further musical transients come along". The net effect is the music becomes compressed and that most definitely could be detected during subjective testing. To add to the mix, the issue of loudspeaker efficiency comes into play, as well as the dynamic impedance of the loudspeaker, inclusive of cross-over network, i.e., poor phase response of cross-over networks will result in greater "Selective Losses", in the region where "Return Loss" is poor. Vis - a - vis, "Matching".

From my observations - of late, it is clear to me, a different method of testing amplifier power ratings should be adopted, which would give an indication of the sustainability of dynamics at high mean power levels. I have long felt that power supplies could be involved with poor test results and early statements from within the industry; when Class AB amplifiers were coming into vogue, about the ability of manufacturers to make economies in terms of power supplies and expensive mains transformers, tend to support my thesis.

If this proves to be the case, then there may well be some credence to what Jez says, which sadly he fails to adequately explain, leaving me to believe he has supernatural powers when he bleats on about subjectivity and denies the value of Blind Testing.

Either way, it explains - for me that is, why the Quad 909 sounds superb. The 405, with 100 watt rating, non the less, for this reason alone, must be inferior to the 909, for at least one reason - it's power supplies may not be able to sustain full musical output, because of lesser power supply capacity?

Perhaps Jez can think about this more deeply, be for running - off at the mouth - whoops, I mean Keyboard.

Footnote - for my benefit. If an amplifier is capable of say, 100 watts into 8 ohms, or 200 watts into 4 ohms, if the impedance of the loudspeaker is less than uniformly 8 ohms or 4 ohms respectively, by its very nature, any musical components which appear where the loudspeaker impedance reduces from it's nominal value; hypothetically 8 ohms, then the relative relationship of the music at that frequency, is modified and the dynamics of the music is altered. This is why the cross-over network response is so critical, given, not only has it to separate the respective audio bands, but it must appear as close as possible as a resistive value, to the amplifier, i.e., 8 ohms - thus no angle. Not inviting comment, just recording this for further consideration by me.

Match - return - loss being one and the same thing!!!!!!!!!

 

[Sergeauckland]

I think you're wrong about the power supply capacity. If an amplifier can sustain it's rated output continuously, so 100w 8 ohms, 150w 4 ohms or whatever those numbers are, then the power supply is clearly adequate.

If compared with another higher powered amplifier, there will only be a difference if the lower powered amplifier is clipping. Otherwise, if both amplifers are outputting, say 50 watts, then both are within their capabilities.

I don't see any technical reason for audiophile terms such as 'grip' in relation to amplifiers. They have power capabiliites, current capabilities and output impedance, grip isn't a measurable parameter.

Oh, and haven't we already mentioned that 'return loss' is the wrong term?

S

 

[Arkless Electronics]

To make a "blameless" amplifier is not rocket science... hence everything that really matters "isn't a measurable parameter".

Output power and power supplies is one of the most misunderstood areas of hi fi amongst the non technical such as Hodges... it seems if "they" are not on flights of fancy over enchanted mains fuses they are making up the pseudo technical like "return loss" and "matching".

The vast majority of modern amplifiers with any quality pretensions have way more than adequate PSU's and largely to placate the completely false view that an amplifier should be usable for arc welding even in order to get good results from LS3/5A's!
This is mainly due to snake oil marketing from various companies who realised that by spending an extra £10 on a big heavy transformer they can get away with adding say £100 to the final cost!

Many seem to think that, like the effortless acceleration and unstressed cruising of a car with a 6L V8 engine, an amplifier of huge power and current (eg 200WPC and massive PSU) will give vastly better results when producing 1W into an easy to drive speaker than a 40WPC amplifier will. This is completely untrue and in fact an amplifier with low output impedance and a very stiff PSU but rated at low power (such as the first 20WPC Levinson or indeed my new creation) should give the same "grip", "slam", "authority", "effortlessness" etc etc as a 200WPC when used within its power rating (again you cannot reliably design in any of those favourable subjective parameters as sometimes an amp with huge power and current ability will sound rather weak in all those areas for no measurable reason).
Some of this misunderstanding is obviously due to the majority of budget amps having low output power (15 - 30W ish) to keep the bean counters happy and therefore having small transformers, lowish value smoothing caps and (important but often overlooked) conservative protection circuitry to protect the single pair of the cheapest output transistors they could find and which actually starts to spoil the sound at way below the maximum output. Hence when many people think of the sound of a low power amp they recall what they heard from an el cheapo amp and wrongly assume it sounded like that due to being only 20WPC!

Many with reasonably efficient speakers (>87dB/W lets say), in an average smallish UK living room would be absolutely stunned at the volume and control that even a 10WPC amp can give when it has been built up to a quality rather than down to a price!

 

[Hodges]

I think you're wrong about the power supply capacity. If an amplifier can sustain it's rated output continuously, so 100w 8 ohms, 150w 4 ohms or whatever those numbers are, then the power supply is clearly adequate.

If compared with another higher powered amplifier, there will only be a difference if the lower powered amplifier is clipping. Otherwise, if both amplifers are outputting, say 50 watts, then both are within their capabilities.

I don't see any technical reason for audiophile terms such as 'grip' in relation to amplifiers. They have power capabiliites, current capabilities and output impedance, grip isn't a measurable parameter.

Oh, and haven't we already mentioned that 'return loss' is the wrong term?

S

Thank you for that - please note Arkless - the operative words are "I think you're wrong".

However, back to your reply. I will try to explain my reasoning more clearly as I believe it has been mis-understood. It might have been clearer if I had stated; were the 405 power supplies fitted to the 909.

Since the issue of telemetry has been raised, if one takes an example such as MCVFT: Multi-channel Voice Frequency telephony, the power in a single channel is given as 0dbm.

However, if four channels are used, each with a power level of 0dbm., the composite power is 4 times, or +6dbm. So any drive (audio) is 6db up on the base figure for a single channel. If an amplifier was being driven to just below clipping with a single channel, to increase the drive to four channel, each at 0dbm, the amplifier would saturate - be over driven. So to achieve a level, just below clipping, each channel drive, needs to be reduced by 6db.

Should the composite signal be audio, within the range 20hz. to 20khz, Power Spectral Density; PSD, must be taken into account. The total PSD must not exceed the power of a single frequency, to achieve an output, just below clipping, If the drive level (audio) fails to take this into account, any transient above would take the amplifier beyond the point of clipping.

Returning to the 909, as max power rating is given as 140 WPC, at normal listening levels, there is enough reserve (Headroom) to cope with such transients - not limitless however. This is a function of both the Linear Power Capacity of the amplifier and the Power Supplies ability to sustain supplies at 140 WPC.

Simply put - if the 405's supplies were fitted to the 909, this probably would not be the case. "In both cases, drive level to achieve full output are identical."

I believe, historically, the Americans and, or the Japanese introduced the IHF Standard for "Music Power", which in effect is the PSD needed to achieve full output.

It may well be that you feel the term "Return loss" is inappropriate, however, I do not. There are many examples where Matching is critical and by using return loss; to express matching, quoting in dB's simplifies issues with respect levels regardless of internal impedances. However, both the 405 and the 909, and all Quad CD amplifiers in between and after, utilize an Impedance Bridges to determine the "Error Signal", to enable the principle to work. The error is inclusive of the output load - i.e. loudspeakers. In designing this, to balance the bridge, xC = xL, and where that occurs, it would have been determined by an Impedance Bridge, where the match would have been expressed in db. return loss. 20db. return loss equates to a match of 100:1. So regardless of whether 50 ohms or 600 ohm transmission lines or any other impedance used, a 20db. figure would be a vey good match.

I am also forming the opinion that, the Impedance Bridge; as it senses the load on the amplifier, could additionally function by way of a very short time - constant. In this, I am starting to consider it, not only in terms of error signal, but error signal amplitude. In this it may well be that it acts as a very fast "AGC" system, which counters changes in Frequency Selective Load Impedance.

 

[Hodges]

To make a "blameless" amplifier is not rocket science... hence everything that really matters "isn't a measurable parameter".

Output power and power supplies is one of the most misunderstood areas of hi fi amongst the non technical such as Hodges... it seems if "they" are not on flights of fancy over enchanted mains fuses they are making up the pseudo technical like "return loss" and "matching".

The vast majority of modern amplifiers with any quality pretensions have way more than adequate PSU's and largely to placate the completely false view that an amplifier should be usable for arc welding even in order to get good results from LS3/5A's!
This is mainly due to snake oil marketing from various companies who realised that by spending an extra £10 on a big heavy transformer they can get away with adding say £100 to the final cost!

Many seem to think that, like the effortless acceleration and unstressed cruising of a car with a 6L V8 engine, an amplifier of huge power and current (eg 200WPC and massive PSU) will give vastly better results when producing 1W into an easy to drive speaker than a 40WPC amplifier will. This is completely untrue and in fact an amplifier with low output impedance and a very stiff PSU but rated at low power (such as the first 20WPC Levinson or indeed my new creation) should give the same "grip", "slam", "authority", "effortlessness" etc etc as a 200WPC when used within its power rating (again you cannot reliably design in any of those favourable subjective parameters as sometimes an amp with huge power and current ability will sound rather weak in all those areas for no measurable reason).
Some of this misunderstanding is obviously due to the majority of budget amps having low output power (15 - 30W ish) to keep the bean counters happy and therefore having small transformers, lowish value smoothing caps and (important but often overlooked) conservative protection circuitry to protect the single pair of the cheapest output transistors they could find and which actually starts to spoil the sound at way below the maximum output. Hence when many people think of the sound of a low power amp they recall what they heard from an el cheapo amp and wrongly assume it sounded like that due to being only 20WP

Many with reasonably efficient speakers (>87dB/W lets say), in an average smallish UK living room would be absolutely stunned at the volume and control that even a 10WPC amp can give when it has been built up to a quality rather than down to a price!

What a load of obtuse drivel, designed to further advance you self proclaimed understanding of amplifier design, to wit, your "World - Beating Designs" should be accepted, without any peer - checking as to your claims.

Put up, or shut up! This is a debate, of which opinions may and will differ, according to, not just the understanding of any author, but his or her ability to make their point, in a manner which other parties understand. Just cut the flannel about fuses, blue, red, or green wire and your insistence that everyone other than your good self is conning the public. Even were that to be true, it keeps my taxes down by employing people.

Can you not grasp the fact that no one is interested in your designs, insofar as they know nothing about them, what they look like or how they work? Or the fact that, if any were sold, they would have to come back to you for repair, as only you understand the way your particular brand of "Snake - oil" works?

Does your vanity know any bounds. If you have a point to make - make it! Without shooting off at a tangent every time as a means of being offensive to everyone who deigns to disagree with you. Even those who don't!

Give us some Maths, or are your designs inclusive of components found lying on the bench, the use of which you know nothing about, beyond the observation that it is revolutionary in the way you have chosen to employ them?

 

[Sergeauckland]

Thank you for that - please note Arkless - the operative words are "I think you're wrong".

However, back to your reply. I will try to explain my reasoning more clearly as I believe it has been mis-understood. It might have been clearer if I had stated; were the 405 power supplies fitted to the 909.

Since the issue of telemetry has been raised, if one takes an example such as MCVFT: Multi-channel Voice Frequency telephony, the power in a single channel is given as 0dbm.

However, if four channels are used, each with a power level of 0dbm., the composite power is 4 times, or +6dbm. So any drive (audio) is 6db up on the base figure for a single channel. If an amplifier was being driven to just below clipping with a single channel, to increase the drive to four channel, each at 0dbm, the amplifier would saturate - be over driven. So to achieve a level, just below clipping, each channel drive, needs to be reduced by 6db.

Should the composite signal be audio, within the range 20hz. to 20khz, Power Spectral Density; PSD, must be taken into account. The total PSD must not exceed the power of a single frequency, to achieve an output, just below clipping, If the drive level (audio) fails to take this into account, any transient above would take the amplifier beyond the point of clipping.

Returning to the 909, as max power rating is given as 140 WPC, at normal listening levels, there is enough reserve (Headroom) to cope with such transients - not limitless however. This is a function of both the Linear Power Capacity of the amplifier and the Power Supplies ability to sustain supplies at 140 WPC.

Simply put - if the 405's supplies were fitted to the 909, this probably would not be the case. "In both cases, drive level to achieve full output are identical."

I believe, historically, the Americans and, or the Japanese introduced the IHF Standard for "Music Power", which in effect is the PSD needed to achieve full output.

It may well be that you feel the term "Return loss" is inappropriate, however, I do not. There are many examples where Matching is critical and by using return loss; to express matching, quoting in dB's simplifies issues with respect levels regardless of internal impedances. However, both the 405 and the 909, and all Quad CD amplifiers in between and after, utilize an Impedance Bridges to determine the "Error Signal", to enable the principle to work. The error is inclusive of the output load - i.e. loudspeakers. In designing this, to balance the bridge, xC = xL, and where that occurs, it would have been determined by an Impedance Bridge, where the match would have been expressed in db. return loss. 20db. return loss equates to a match of 100:1. So regardless of whether 50 ohms or 600 ohm transmission lines or any other impedance used, a 20db. figure would be a vey good match.

I am also forming the opinion that, the Impedance Bridge; as it senses the load on the amplifier, could additionally function by way of a very short time - constant. In this, I am starting to consider it, not only in terms of error signal, but error signal amplitude. In this it may well be that it acts as a very fast "AGC" system, which counters changes in Frequency Selective Load Impedance.

An amplifier provides a maximum peak-to-peak voltage into a particular load. It doesn't matter if it's a single frequency, a wideband noise signal or music. The maximum peak to peak voltage is the same.

Consequently, if you have one frequency, say 1kHz, you can get, 79v P-P into 8 ohms for a 100w amplifier. (28.3v RMS x 2.8) If you have two frequencies, not phase-locked, the waveform will be higher on its peaks than a single frequency, so the power at each frequency has to be reduced to be accommodated in the maximum P-P. None of this affects the power supply, as the power supply only has to provide its DC voltage. The power supply has no knowledge of what the amplifier is doing, how many frequencies it's providing, or at what level. It just provides volts at whatever amps the amplifier requires, up to its maximum rating. A conventional sagging supply, i.e. one without active regulation, will change its volts depending on current drawn, which is why most conventional class AB amplifiers have what's sometimes called dynamic headroom, sometimes called Music Power, on music peaks that's some 10-20% higher than the amplifier's continuous power depending on the duration of the peaks.

There is no interaction (to the limits of the amplifier's linearity, i.e. distortion characteristics) in an amplifier between 1kHz and, say, 10kHz outputs up to the amplifier's peak-peak level output. If you doubt this, look up the Superposition Principle, which briefly states that for any linear system, (which an amplifier is to a very high level) there is no interaction between frequencies.
If the superposition principle wasn't true, all amplifiers would have massive intermodulation distortion, and the only way that can happen is if they are hugely non-linear, which is clearly not the case.

As to the way a Quad Current Dumping amplifier works, there were some very good articles written in Wireless World at the time of the 405's release. There was, and I suppose still is, misunderstanding of the feed-forward concept. There is no attempt at impedance matching with the load,. the bridge balance takes place entirely within the amplifier, regardless of the external load. That was one of the beauties of it, that the external load had no effect on the balance. There were limitations of course, partially corrected in the 405.2 and the 405 or 405.2 was never suited to extreme loads.

S.

 

[Arkless Electronics]

What a load of obtuse drivel, designed to further advance you self proclaimed understanding of amplifier design, to wit, your "World - Beating Designs" should be accepted, without any peer - checking as to your claims.

Put up, or shut up! This is a debate, of which opinions may and will differ, according to, not just the understanding of any author, but his or her ability to make their point, in a manner which other parties understand. Just cut the flannel about fuses, blue, red, or green wire and your insistence that everyone other than your good self is conning the public. Even were that to be true, it keeps my taxes down by employing people.

Can you not grasp the fact that no one is interested in your designs, insofar as they know nothing about them, what they look like or how they work? Or the fact that, if any were sold, they would have to come back to you for repair, as only you understand the way your particular brand of "Snake - oil" works?

Does your vanity know any bounds. If you have a point to make - make it! Without shooting off at a tangent every time as a means of being offensive to everyone who deigns to disagree with you. Even those who don't!

Give us some Maths, or are your designs inclusive of components found lying on the bench, the use of which you know nothing about, beyond the observation that it is revolutionary in the way you have chosen to employ them?

What a deluded gob shite you really are, and definitely not an engineer. Find a few technical terms on Wikipedia did you? and can't even get them right. You don't know your arse from your elbow and such a complete dick that you now try to tell us about misunderstandings on current dumping and where the articles are to be found after being utterly clueless (still are) and yourself finding out about them from myself and Serge!