interims to: sba & nigel pearson , including about LS3/5a
Hi sba ,
great stuff - your CSD plots , etc ...
Thankyou for posting these !
I see you have some JBL and Altec also - more than I will have time to comment on for a long time,
except to say now that both loudspeakers can be improved, especially the JBL.
--- --- --- --- ---
'ullo Nigel ,
yes, a lot of the sound character of the LS3/5a is owing to its crossover, and in part that tapped inductor.
The purpose of the tapped inductor was to compensate for sample variations amongst the tweeters,
though modern manufactured tweeters should be more consistent/less degree of variation between samples.
For Music reproduction I prefer to not use cored inductors, but some designers insist we cannot hear any detriment result of the core so long as it is suitable size for the intended power dissipation within it.
OK , some people don't hear it, but some people hear differently to others, and the simplistic Text Book engineers will not allow for that ,
because they have not read modern in-depth Audiology text books, and they don't hear what some other people do hear.
I respect the LS3/5a design, and understand what it was intended for:
near-field monitoring of speech for long periods of time, primarily for Radio Broadcast.
It was not designed to portray accurate Timbre of musical instruments, as the BBC had other monitors designed for that.
I can accept that some listeners like its peculiar timbral character,
though mostly what enthusiasts of it like is its Imaging - and that is an artefact of its design, and not actually the truth of the audio signal.
There is PART similar in the Celestion 44 and 66 , which is why I mention this here, because I think Reggie could be hearing that in the treble.
Why are you "too lazy these days" ?
You can be not lazy and help us here:
For a tweeter with 6.5 ohm impedance in the middle of its audio-band ,
with an L-pad of 15 ohms Parallel and 1 ohm Series - how much attenuation in dB ?
I estimate about -2dB or slightly less ... if you have whatever necessary, can you calculate closer for us please ?
Hi sba ,
great stuff - your CSD plots , etc ...
Thankyou for posting these !
I see you have some JBL and Altec also - more than I will have time to comment on for a long time,
except to say now that both loudspeakers can be improved, especially the JBL.
--- --- --- --- ---
'ullo Nigel ,
yes, a lot of the sound character of the LS3/5a is owing to its crossover, and in part that tapped inductor.
The purpose of the tapped inductor was to compensate for sample variations amongst the tweeters,
though modern manufactured tweeters should be more consistent/less degree of variation between samples.
For Music reproduction I prefer to not use cored inductors, but some designers insist we cannot hear any detriment result of the core so long as it is suitable size for the intended power dissipation within it.
OK , some people don't hear it, but some people hear differently to others, and the simplistic Text Book engineers will not allow for that ,
because they have not read modern in-depth Audiology text books, and they don't hear what some other people do hear.
I respect the LS3/5a design, and understand what it was intended for:
near-field monitoring of speech for long periods of time, primarily for Radio Broadcast.
It was not designed to portray accurate Timbre of musical instruments, as the BBC had other monitors designed for that.
I can accept that some listeners like its peculiar timbral character,
though mostly what enthusiasts of it like is its Imaging - and that is an artefact of its design, and not actually the truth of the audio signal.
There is PART similar in the Celestion 44 and 66 , which is why I mention this here, because I think Reggie could be hearing that in the treble.
Why are you "too lazy these days" ?
You can be not lazy and help us here:
For a tweeter with 6.5 ohm impedance in the middle of its audio-band ,
with an L-pad of 15 ohms Parallel and 1 ohm Series - how much attenuation in dB ?
I estimate about -2dB or slightly less ... if you have whatever necessary, can you calculate closer for us please ?
Last edited:
interim to Reggie
Replies:
(1)- read in my comments to Nigel in #901 above: "There is PART similar ... "
The 44 and 66 have their tweeters connected in reverse Polarity to their mid and bass drivers.
This causes a Distance effect in the treble, but not everyone hears it like that, though I suspect you might be.
{ There is at least 2 distinctly different types of human hearing, and perhaps more ... }
Experiment: swap the leads around at the tweeter between + and - terminals ,
then listen again with exactly the same music as before the swap.
Ask you wife to listen before/after also ... but remember, her brain priority allocation may be different to yours in this aspect.
Possible complication: when the Polarity is swapped with a 3rd Order crossover , which this one to tweeter is, the Vertical off-axis lobes swap over ,
thus that lip at the top of the cabinet will reflect treble differently, and may be audibly more unpleasant, and there-by confuse any other audible effects.
You should first stick some soft absorbent material under that lip down to the top of the tweeter flange, and along under the lip to the sides of the cabinet.
Temporary is to use rolled up Wool socks - real Wool , not Polyester because it does not absorb as well.
If you like the sound, then your wife may be familiar with Haberdashery or Crafts shops.
If so, buy some very soft wool felt, or very soft cotton felt if no wool, but NOT a polyester or nylon blend.
Roll it up to optimum thickness to fit to cover all the under side of the lip,
but do not roll it tight - it needs to be loose enough to remain soft to absorb sound.
Stick it in place with double sided tape, or similar that your wife may be able to do with Velcro.
If she wants to sew the roll to hold it together, then sew it only at its ends where it meets the sides of the cabinet.
Do not sew it tight where it is will be anywhere near the tweeter flange.
If the felt is compressed it will not absorb much sound.
-
(2)- read back though this thread to where I posted about solo violin recordings to DennyG.
Do you have any of Bach's Partitas and Sonatas for solo violin ?
If you like such, but don't own recordings, I will list some that are musical more-so than too scholarly technical so that you can enjoy the listening.
It is more difficult to decide about audio balance when listening to things one does not enjoy.
Soprano vocals also, true Soprano, not Mezzo singers who can also reach high notes, because Mezzo voices are richer in lower midrange harmonics even when singing high notes.
{NB: Soprano are Female, not that TV show or Movie about Male Gangsters, or whatever they are ... OK, you likely know that, but I'm getting in before the jokes !}
-
(3)- I'll describe this when we get to modifying the mids of the crossover.
-
(4)- yes, it is possible.
-
(5)- there are a few others, but more-so in posts in Forums than on dedicated web-pages.
-
(6)- these are very large size ! They may not fit on the Celestion board.
Are your Solen caps the 250 volt or 400 volt or 630 volt size ?
I think the Mundorf you have is 400 volt size .. ?
-
(7)- those are intended for Power Supply filtering, and may not be suitable in a crossover, but I do not know ... however look at their size:
55mm x 125mm , that is slightly larger than 2" x 5" !
Also, the Solens are OK there, and I doubt you'll hear much difference in that part of the circuit unless you happen to have very microphonic samples of the Solen.
You can use your Solen 8.2uF cap there later to increase the filtering. I will describe how later.
-
(8)- The Jantzen Superior-Z caps are in limited values.
If you want to use use single caps there, and remember they are physically very large size, you will have to buy a new inductor because 0.14mH will not work at 5kHz with 3.3uF and 10uF caps.
There is the Jantzen 20AWG aircore 0.15mH listed on Speakerbug. That would be suitable,
and not the other Jantzen 0.15mH there because it is too low in DCR. Some resistance is needed there.
If you want to keep the Celestion tweeter inductor you will need to buy 3.6uF and 11uF or 12uF caps.
It is possible to squeeze the filter a bit by slightly changing the effective Impedance, and then use the more common 3.9uF and 10uF values with the Celestion inductor,
but the result will not be quite as flat response through the crossover region.
First decide whether these huge Jantzen caps will fit ... then decide whatever else you would like to do about the above, then post decisions here.
Jantzen Superior-Z cap physical dimensions are here:
http://www.jantzen-audio.com/download/Z-caps series list.pdf
Replies:
(1)- read in my comments to Nigel in #901 above: "There is PART similar ... "
The 44 and 66 have their tweeters connected in reverse Polarity to their mid and bass drivers.
This causes a Distance effect in the treble, but not everyone hears it like that, though I suspect you might be.
{ There is at least 2 distinctly different types of human hearing, and perhaps more ... }
Experiment: swap the leads around at the tweeter between + and - terminals ,
then listen again with exactly the same music as before the swap.
Ask you wife to listen before/after also ... but remember, her brain priority allocation may be different to yours in this aspect.
Possible complication: when the Polarity is swapped with a 3rd Order crossover , which this one to tweeter is, the Vertical off-axis lobes swap over ,
thus that lip at the top of the cabinet will reflect treble differently, and may be audibly more unpleasant, and there-by confuse any other audible effects.
You should first stick some soft absorbent material under that lip down to the top of the tweeter flange, and along under the lip to the sides of the cabinet.
Temporary is to use rolled up Wool socks - real Wool , not Polyester because it does not absorb as well.
If you like the sound, then your wife may be familiar with Haberdashery or Crafts shops.
If so, buy some very soft wool felt, or very soft cotton felt if no wool, but NOT a polyester or nylon blend.
Roll it up to optimum thickness to fit to cover all the under side of the lip,
but do not roll it tight - it needs to be loose enough to remain soft to absorb sound.
Stick it in place with double sided tape, or similar that your wife may be able to do with Velcro.
If she wants to sew the roll to hold it together, then sew it only at its ends where it meets the sides of the cabinet.
Do not sew it tight where it is will be anywhere near the tweeter flange.
If the felt is compressed it will not absorb much sound.
-
(2)- read back though this thread to where I posted about solo violin recordings to DennyG.
Do you have any of Bach's Partitas and Sonatas for solo violin ?
If you like such, but don't own recordings, I will list some that are musical more-so than too scholarly technical so that you can enjoy the listening.
It is more difficult to decide about audio balance when listening to things one does not enjoy.
Soprano vocals also, true Soprano, not Mezzo singers who can also reach high notes, because Mezzo voices are richer in lower midrange harmonics even when singing high notes.
{NB: Soprano are Female, not that TV show or Movie about Male Gangsters, or whatever they are ... OK, you likely know that, but I'm getting in before the jokes !}
-
(3)- I'll describe this when we get to modifying the mids of the crossover.
-
(4)- yes, it is possible.
-
(5)- there are a few others, but more-so in posts in Forums than on dedicated web-pages.
-
(6)- these are very large size ! They may not fit on the Celestion board.
Are your Solen caps the 250 volt or 400 volt or 630 volt size ?
I think the Mundorf you have is 400 volt size .. ?
-
(7)- those are intended for Power Supply filtering, and may not be suitable in a crossover, but I do not know ... however look at their size:
55mm x 125mm , that is slightly larger than 2" x 5" !
Also, the Solens are OK there, and I doubt you'll hear much difference in that part of the circuit unless you happen to have very microphonic samples of the Solen.
You can use your Solen 8.2uF cap there later to increase the filtering. I will describe how later.
-
(8)- The Jantzen Superior-Z caps are in limited values.
If you want to use use single caps there, and remember they are physically very large size, you will have to buy a new inductor because 0.14mH will not work at 5kHz with 3.3uF and 10uF caps.
There is the Jantzen 20AWG aircore 0.15mH listed on Speakerbug. That would be suitable,
and not the other Jantzen 0.15mH there because it is too low in DCR. Some resistance is needed there.
If you want to keep the Celestion tweeter inductor you will need to buy 3.6uF and 11uF or 12uF caps.
It is possible to squeeze the filter a bit by slightly changing the effective Impedance, and then use the more common 3.9uF and 10uF values with the Celestion inductor,
but the result will not be quite as flat response through the crossover region.
First decide whether these huge Jantzen caps will fit ... then decide whatever else you would like to do about the above, then post decisions here.
Jantzen Superior-Z cap physical dimensions are here:
http://www.jantzen-audio.com/download/Z-caps series list.pdf
Last edited:
The tweeter problem is the typical British solution . A freind made speakers which had additional units to replace lost information in the crossover region . They could reproduce square waves accurately . I loved the speakers . Alas his skills in selling them were zero and they were dammed with faith praise . In my opinion he was reviewed and not his product . I wish I had bought a pair . Even more so than Quad ESL they told truths seldom heard . Not just tonality or low distortion , nor even imaging . Layers never heard before or after . These layers are not supposed to be heard and would annoy some . The speakers worked with modest amplifiers , most surprisingly even the Quad 306 . Amplifier coloration was more oblivious using them I than I ever remembered . Very bright yet not at all fatiguing . Only speaker I ever used which had a genuine 20Hz response .
Orchid Audio PLL1 [English]
Orchid Audio PLL1 [English]
The tweeter problem is the typical British solution . A freind made speakers which had additional units to replace lost information in the crossover region . They could reproduce square waves accurately . I loved the speakers . Alas his skills in selling them were zero and they were dammed with faith praise . In my opinion he was reviewed and not his product . I wish I had bought a pair . Even more so than Quad ESL they told truths seldom heard . Not just tonality or low distortion , nor even imaging . Layers never heard before or after . These layers are not supposed to be heard and would annoy some . The speakers worked with modest amplifiers , most surprisingly even the Quad 306 . Amplifier coloration was more oblivious using them I than I ever remembered . Very bright yet not at all fatiguing . Only speaker I ever used which had a genuine 20Hz response .
Orchid Audio PLL1 [English]
Seems a duplicate !!!!
Orchid Audio PLL1 [English]
Seems a duplicate !!!!
New Crossover
Hello Alan,
Some responses to your questions and suggestions.
1(a). I have played with the tweeter polarity so many times now. The present iteration is that the polarity is reversed at the tweeter terminals. However my pre amp is a valve OTL (a TRAM 2) which the designer, Thorsten Loesch, said reverses polarity so my speaker wires are reversed at the power amps. I have also tried reversing only the midrange. I’ve tried all combinations but cannot truly say I hear a difference. Also, I’m not sure what the polarity is at the crossover itself. Regarding the recessed HF It may be relevant to quote from a review compiled by High Fidelity Magazine on the “Ditton 66” (wood faced) in the late 70’s....”Triangles, bells and such, which have their energy concentrated almost exclusively at the higher frequencies, are reproduced very well, albeit perhaps not quite in perfect loudness balance with the rest of the orchestra. In listening tests the reduced relative high-end response shows up mainly as a lack of sparkle and transient response that, though fairly good, is not up to the best we have heard. On the other hand, that same characteristic lends a sweetness to the sound that is quite appealing. There is no trace of harshness or hiss and yet the highs are indeed there – right out to 16kHz...” Is this relevant?
1(b). I have had wool felt around the tweeter (and midrange) since early last year. I got the idea from some of your earlier posts and from a 2005 article I found on the internet by a David Ralph entitled “Diffraction doesn’t have to be a problem” (www,speakerdesign.net). What I seemed to achieve was a greater focus to the music.
2(a). I will go back and look for the advice you provided to DennyG re listening to violin music.
2(b). I have 3 classical CD’s: (i) Beethoven Concerto for Piano, Cello and Orchestra in C combined with Boccherini’s Concerto for Cello and Strings in G (ii) The Best of Boccherini and (iii) Pegolesi”s Stabat Mater with Gillian Welsh soprano (which I really like) but in general I’m more middle of the road, with some left of centre stuff.
3. Ok
4. Ok
5. Yes, I’ve read a few but nothing as exhaustive as this.
6. Solens 400v. Mcap 400v. As for the size well, the present crossover is in 2 parts, bass and mid/tweeter but my intention is to have the crossovers outside the cabinets (at least initially).
7. If you think it is a waste to get the big Obbligato’s then we will retain the Solens. My concern was with the disparity between the values of the original capacitors (144) and my current ones (136) so your suggestion to use the 8.2uf is welcome.
8. Alan, unless something unusual occurs, I will probably keep these speakers (I’ve had them for 36 years now). So I’m up for spending money on totally renewing the crossovers. That’s new capacitors, coils and resistors. You tell me what I need and I’ll get them.
What I’m hoping to get from this is: (a) improved clarity (which I interpret to be a greater intimacy with and insight into the music, (b) a tighter bass (nb. I’ve closed off the bass reflex speaker with no readily apparent loss of bass (my speakers are near room corners) and (c) maybe a small increase in efficiency (to get the Celestions to sing I have to turn up the volume slightly more than my wife appreciates)(even with the door shut). Quoting again from High Fidelity Magazine (stop me if you’ve heard this before)...”Although the efficiency of the Ditton 66 is about 3 and ½ dB below what we have come to consider average – CBS reports an on-axis sound pressure level of 83 and ¾ dB at 1 meter from a 0.dBW noise level, 250 Hz to 6kHz – it is an easy speaker to drive. The impedance curve is relatively smooth and well contained. Except for a 22.5ohm peak at resonance (49Hz) the load stays between 4 and 10.5 ohms throughout the audio band. We rate the impedance as 4.6 ohms (the minimum reached just above resonance) but over the important midrange the curve averages about 2 ohms higher: low, but not so low as to be dangerous...” and so on (there’s more of this technical stuff).
Several more questions please. Given my desire to try the Morel midrange at some future time (the MF500’s are not going to last forever - they are already 36 + years old and I need the Celestions to last another 20 years (I’m 63) will these new crossovers be compatible?
What is your view of tri wiring. What would have to change on the crossover and how would my push pull valve mono power amps (operating in triode) feel about feeding 3 x 12 gauge wires (Paul Speltz Anti cable). I’ve read all the pro’s and con’s on the internet: some say it’s good some say it’s bad and some say it makes no difference. I think it would be helpful for me in fine tuning the crossovers. I think at the very least I will bi-wire. I won’t be using (banana) plugs but rather some mechanical means of connecting the speaker wire to the crossover.
OK. Enough now.
Regards
Reggie
Hello Alan,
Some responses to your questions and suggestions.
1(a). I have played with the tweeter polarity so many times now. The present iteration is that the polarity is reversed at the tweeter terminals. However my pre amp is a valve OTL (a TRAM 2) which the designer, Thorsten Loesch, said reverses polarity so my speaker wires are reversed at the power amps. I have also tried reversing only the midrange. I’ve tried all combinations but cannot truly say I hear a difference. Also, I’m not sure what the polarity is at the crossover itself. Regarding the recessed HF It may be relevant to quote from a review compiled by High Fidelity Magazine on the “Ditton 66” (wood faced) in the late 70’s....”Triangles, bells and such, which have their energy concentrated almost exclusively at the higher frequencies, are reproduced very well, albeit perhaps not quite in perfect loudness balance with the rest of the orchestra. In listening tests the reduced relative high-end response shows up mainly as a lack of sparkle and transient response that, though fairly good, is not up to the best we have heard. On the other hand, that same characteristic lends a sweetness to the sound that is quite appealing. There is no trace of harshness or hiss and yet the highs are indeed there – right out to 16kHz...” Is this relevant?
1(b). I have had wool felt around the tweeter (and midrange) since early last year. I got the idea from some of your earlier posts and from a 2005 article I found on the internet by a David Ralph entitled “Diffraction doesn’t have to be a problem” (www,speakerdesign.net). What I seemed to achieve was a greater focus to the music.
2(a). I will go back and look for the advice you provided to DennyG re listening to violin music.
2(b). I have 3 classical CD’s: (i) Beethoven Concerto for Piano, Cello and Orchestra in C combined with Boccherini’s Concerto for Cello and Strings in G (ii) The Best of Boccherini and (iii) Pegolesi”s Stabat Mater with Gillian Welsh soprano (which I really like) but in general I’m more middle of the road, with some left of centre stuff.
3. Ok
4. Ok
5. Yes, I’ve read a few but nothing as exhaustive as this.
6. Solens 400v. Mcap 400v. As for the size well, the present crossover is in 2 parts, bass and mid/tweeter but my intention is to have the crossovers outside the cabinets (at least initially).
7. If you think it is a waste to get the big Obbligato’s then we will retain the Solens. My concern was with the disparity between the values of the original capacitors (144) and my current ones (136) so your suggestion to use the 8.2uf is welcome.
8. Alan, unless something unusual occurs, I will probably keep these speakers (I’ve had them for 36 years now). So I’m up for spending money on totally renewing the crossovers. That’s new capacitors, coils and resistors. You tell me what I need and I’ll get them.
What I’m hoping to get from this is: (a) improved clarity (which I interpret to be a greater intimacy with and insight into the music, (b) a tighter bass (nb. I’ve closed off the bass reflex speaker with no readily apparent loss of bass (my speakers are near room corners) and (c) maybe a small increase in efficiency (to get the Celestions to sing I have to turn up the volume slightly more than my wife appreciates)(even with the door shut). Quoting again from High Fidelity Magazine (stop me if you’ve heard this before)...”Although the efficiency of the Ditton 66 is about 3 and ½ dB below what we have come to consider average – CBS reports an on-axis sound pressure level of 83 and ¾ dB at 1 meter from a 0.dBW noise level, 250 Hz to 6kHz – it is an easy speaker to drive. The impedance curve is relatively smooth and well contained. Except for a 22.5ohm peak at resonance (49Hz) the load stays between 4 and 10.5 ohms throughout the audio band. We rate the impedance as 4.6 ohms (the minimum reached just above resonance) but over the important midrange the curve averages about 2 ohms higher: low, but not so low as to be dangerous...” and so on (there’s more of this technical stuff).
Several more questions please. Given my desire to try the Morel midrange at some future time (the MF500’s are not going to last forever - they are already 36 + years old and I need the Celestions to last another 20 years (I’m 63) will these new crossovers be compatible?
What is your view of tri wiring. What would have to change on the crossover and how would my push pull valve mono power amps (operating in triode) feel about feeding 3 x 12 gauge wires (Paul Speltz Anti cable). I’ve read all the pro’s and con’s on the internet: some say it’s good some say it’s bad and some say it makes no difference. I think it would be helpful for me in fine tuning the crossovers. I think at the very least I will bi-wire. I won’t be using (banana) plugs but rather some mechanical means of connecting the speaker wire to the crossover.
OK. Enough now.
Regards
Reggie
If it is not a diversion you might like to see if any of this makes sense . I had the opportunity to buy this company . Alas as someone said you couldn't buy Joe . The company was Joe . An un-respected genius of speaker design . As far as I know not in business now ?
What is a Hi-Fi Loudspeaker? | Royd Audio Loudspeakers
What is a Hi-Fi Loudspeaker? | Royd Audio Loudspeakers
Ditton 66: Midrange Noise
I own a pair of Ditton 44s... every once in a while, one might pick up a scratchy noise from the midranges... if you catch it early, (I was told) that if you take out the midrange, rotate it and then put it back in, the noise might go away...something to do with the weight of the magnet and the way the rear part of the cone sits in it.... one of the other techies in this forum may be able to explain more clearly why it happens... all I can tell you is that I did it and it worked!
DittonFan
I own a pair of Ditton 44s... every once in a while, one might pick up a scratchy noise from the midranges... if you catch it early, (I was told) that if you take out the midrange, rotate it and then put it back in, the noise might go away...something to do with the weight of the magnet and the way the rear part of the cone sits in it.... one of the other techies in this forum may be able to explain more clearly why it happens... all I can tell you is that I did it and it worked!
DittonFan
Good that it worked for you.
I'd guess that the voice coil (on the cone) and the air gap it sits in, in the pole pieces of the magnet assembly have a very tight (too tight) tolerance and the coil is rubbing. If the weight of the magnet and rotating the whole driver fixes it then maybe the chassis of the speaker is subject to distortion (flimsy).
There's many a speaker that "rubs" when you hold the chassis and lightly twist it. Also many a speaker that rubs when the mounting surface isn't flat and the fixing screws twist the frame.
I'd guess that the voice coil (on the cone) and the air gap it sits in, in the pole pieces of the magnet assembly have a very tight (too tight) tolerance and the coil is rubbing. If the weight of the magnet and rotating the whole driver fixes it then maybe the chassis of the speaker is subject to distortion (flimsy).
There's many a speaker that "rubs" when you hold the chassis and lightly twist it. Also many a speaker that rubs when the mounting surface isn't flat and the fixing screws twist the frame.
Given the massive cast frame on the midrange of the 44 its not likely that its due to it being flimsy!
I'll bet Ditton Fan in your case whats happened is over the years gravity has taken its toll and the cone has sagged, this often happens where where the surround is very complaint and/or the cone very heavy. (in the case of the Dittons it'd be compliance) This then causes the voice coil in the air gap to no longer be central and to rub. An extreme example, A friend of mine owns a pair of Chartwell PM400's with massive 12" Bextrene bass units, after many rubbing problems he simply turns the bass units by 1/4 every six months - no more rubbing..
It would be worth you turning the other one too as if one has sagged its likely the other will and if not caught the rubbing can wear through the coil insulation causing it to short..
I'll bet Ditton Fan in your case whats happened is over the years gravity has taken its toll and the cone has sagged, this often happens where where the surround is very complaint and/or the cone very heavy. (in the case of the Dittons it'd be compliance) This then causes the voice coil in the air gap to no longer be central and to rub. An extreme example, A friend of mine owns a pair of Chartwell PM400's with massive 12" Bextrene bass units, after many rubbing problems he simply turns the bass units by 1/4 every six months - no more rubbing..
It would be worth you turning the other one too as if one has sagged its likely the other will and if not caught the rubbing can wear through the coil insulation causing it to short..
Some people rotate Tannoys to extend life . Spendor BC1 drivers also . It also ensures bolts are tight . There was a craze of tweeter cone tightening . I offered to do it for my customers free of charge . It should be done finger tight so as not to deform or overly compress components ( gaskets ) . Spendor always said it was a dilemma whether to optimally tighten the rear cabinet screws or have them all the same . All the same won the day .
Nigel's loudspeakers {from previous page here-of}
'ullo Nigel,
about the Orchid Audio PLL1 ,
I do remember when this loudspeaker was introduced, however I have not heard a pair.
I read the Link you posted.
I agree with what I understand about the design, but as there are no details given for the two "PhaseLink" drivers' crossover sections I cannot comment further , except the limited following:
I understand the difference between Accuracy as regards revealing the actual content of the audio signal, versus euphonic colourations, etc ...
The reviewer seems to have tried to give an honest appraisal of the abilities of these l'speakers,
however given he stated his preferences to be SET amplifiers and wide-range single-cone driver l'speakers he is not the ideal reviewer for the PLL1.
As you found the PLL1 to work OK with a QUAD 306, I recommend they be heard with the QUAD 909 or the new QUAD Elite QSP power amplifier driving them when in large rooms.
Alternate amplifiers include any of Musical Fidelity's large Class A power amps,
and for valve/tube amp enthusiasts, use any of Tim De Paravicini's EAR PP output type amps.
Paravicini has designed and manufactured amplifiers for driving monitors for recording and mastering studios,
thus if his models are specified to drive the Impedance of the PLL1 I am confident they will be suitable,
albeit within their power range dynamic capabilities.
In a small room the Paravicini designed QUAD II Eighty would likely drive those l'speakers to suitable volume levels ...
... and I think would also drive Celestion 66 in a small to moderate sized room if the Eighty has a 4 ohm Output Tap.
***
Royd/Akroyd :
unfortunately did not gain the large market perception of B&W , Celestion , KEF , Mission , etc ...
thus not worth a lot to buy the company unless one intended to continue with the same or very similar design philosophy as Joe Akroyd.
I have heard a couple of Royd models, and I do agree with his primary design philosophy,
however the Royd sound is not one that everyone likes ... not because it is bad , far from that , but because within the price-budget of many buyers the specific compromises J.A. decided on to keep the basic models affordable caused them to not have the oomph factor that a lot of buyers in that price-region prefer.
I think Royds are very good for portraying fine detail , etc ... and close to accurate thus good for Classical Music and similar , but to get a wider and flatter frequency response more expensive drivers would have to be used,
and similarly to enable high volume replay in large rooms.
I would not buy the company because I do not want to have to make the types of compromises necessary to be able to sell sufficient quantity of l'speakers to be able to stay in business whilst knowing fully that there are people like me who can hear those compromises, AND that there are other listeners who prefer the sounds caused by different compromises and/or different design philosophies.
One cannot please every listener ... it is simpler for me to contribute to DIY about the sound types I can understand, and in the limited amount of time that I have available.
It is a diversion ,
because the Royd sound is not in any way like the Celestion 66 or 44 sound ,
and I think the PLL1 sound is not likely similar to either ... but possibly is somewhere in between the two,
however there will be distinct differences that cannot be changed.
*
It is worth readers here looking at the Links that Nigel posted on Page 91 , because then you may be able to understand further the differences between design philosophies ,
and thus hopefully what can be achieved with Celestion 66 and what cannot.
If you like the basic Celestion 66 or 44 sound , that can be kept , or in some ways improved on in clarity as we have been doing here to date ,
or developed further in a direction to get slightly higher output and bass tightness ... as seems to be the interest of DennyG and reggie in this thread ,
thus I will continue in that direction at the times I have available.
'ullo Nigel,
about the Orchid Audio PLL1 ,
I do remember when this loudspeaker was introduced, however I have not heard a pair.
I read the Link you posted.
I agree with what I understand about the design, but as there are no details given for the two "PhaseLink" drivers' crossover sections I cannot comment further , except the limited following:
I understand the difference between Accuracy as regards revealing the actual content of the audio signal, versus euphonic colourations, etc ...
The reviewer seems to have tried to give an honest appraisal of the abilities of these l'speakers,
however given he stated his preferences to be SET amplifiers and wide-range single-cone driver l'speakers he is not the ideal reviewer for the PLL1.
As you found the PLL1 to work OK with a QUAD 306, I recommend they be heard with the QUAD 909 or the new QUAD Elite QSP power amplifier driving them when in large rooms.
Alternate amplifiers include any of Musical Fidelity's large Class A power amps,
and for valve/tube amp enthusiasts, use any of Tim De Paravicini's EAR PP output type amps.
Paravicini has designed and manufactured amplifiers for driving monitors for recording and mastering studios,
thus if his models are specified to drive the Impedance of the PLL1 I am confident they will be suitable,
albeit within their power range dynamic capabilities.
In a small room the Paravicini designed QUAD II Eighty would likely drive those l'speakers to suitable volume levels ...
... and I think would also drive Celestion 66 in a small to moderate sized room if the Eighty has a 4 ohm Output Tap.
***
Royd/Akroyd :
unfortunately did not gain the large market perception of B&W , Celestion , KEF , Mission , etc ...
thus not worth a lot to buy the company unless one intended to continue with the same or very similar design philosophy as Joe Akroyd.
I have heard a couple of Royd models, and I do agree with his primary design philosophy,
however the Royd sound is not one that everyone likes ... not because it is bad , far from that , but because within the price-budget of many buyers the specific compromises J.A. decided on to keep the basic models affordable caused them to not have the oomph factor that a lot of buyers in that price-region prefer.
I think Royds are very good for portraying fine detail , etc ... and close to accurate thus good for Classical Music and similar , but to get a wider and flatter frequency response more expensive drivers would have to be used,
and similarly to enable high volume replay in large rooms.
I would not buy the company because I do not want to have to make the types of compromises necessary to be able to sell sufficient quantity of l'speakers to be able to stay in business whilst knowing fully that there are people like me who can hear those compromises, AND that there are other listeners who prefer the sounds caused by different compromises and/or different design philosophies.
One cannot please every listener ... it is simpler for me to contribute to DIY about the sound types I can understand, and in the limited amount of time that I have available.
It is a diversion ,
because the Royd sound is not in any way like the Celestion 66 or 44 sound ,
and I think the PLL1 sound is not likely similar to either ... but possibly is somewhere in between the two,
however there will be distinct differences that cannot be changed.
*
It is worth readers here looking at the Links that Nigel posted on Page 91 , because then you may be able to understand further the differences between design philosophies ,
and thus hopefully what can be achieved with Celestion 66 and what cannot.
If you like the basic Celestion 66 or 44 sound , that can be kept , or in some ways improved on in clarity as we have been doing here to date ,
or developed further in a direction to get slightly higher output and bass tightness ... as seems to be the interest of DennyG and reggie in this thread ,
thus I will continue in that direction at the times I have available.
Last edited:
Posts #907 <---> #910 on Page 91
The information in all four posts #907 <---> #910 is useful,
and thank you gentlemen for posting that information.
For Celestion 66 and 44 , and most vintage l'speakers that have not been used for a while thus particually susceptable to cone-sag or similar,
I think it best to do a 180 degree rotation after checking that the drivers are working, and then listen carefully to hear if any extraneous noises.
Later, if all the sound is good, perform the 90 degree rotation after every 6 months or so ...
Of course if capacitors in treble and/or midrange sections have deteriorated those capacitors will have to be replaced before accurate diagnosis of the drivers can be attempted.
The rotating will benefit most cone drivers, and some types of dome drivers,
but whether all types of very small domes such as in Celestion tweeters I do not know.
180 degree rotating would likely benefit Ribbon drivers, BUT be very careful with vintage ribbons:
do NOT cover their face with your hand or any solid object and rapidly pull away,
as the ribbon can break or irrepairably stretch when subject to the force of partial vacuum in front of it and thus pressure behind it.
Hopefully needless to say: do not push on the ribbon's surface , nor drop the driver.
The information in all four posts #907 <---> #910 is useful,
and thank you gentlemen for posting that information.
For Celestion 66 and 44 , and most vintage l'speakers that have not been used for a while thus particually susceptable to cone-sag or similar,
I think it best to do a 180 degree rotation after checking that the drivers are working, and then listen carefully to hear if any extraneous noises.
Later, if all the sound is good, perform the 90 degree rotation after every 6 months or so ...
Of course if capacitors in treble and/or midrange sections have deteriorated those capacitors will have to be replaced before accurate diagnosis of the drivers can be attempted.
The rotating will benefit most cone drivers, and some types of dome drivers,
but whether all types of very small domes such as in Celestion tweeters I do not know.
180 degree rotating would likely benefit Ribbon drivers, BUT be very careful with vintage ribbons:
do NOT cover their face with your hand or any solid object and rapidly pull away,
as the ribbon can break or irrepairably stretch when subject to the force of partial vacuum in front of it and thus pressure behind it.
Hopefully needless to say: do not push on the ribbon's surface , nor drop the driver.
Last edited:
The point about the 66's is I feel they were far more like a a studio speaker than people realize . I speculate that with an active crossover various options could be tried . At the flick of a switch the original sound could be had . I was shock when I heard a pair for the first time . Very accurate would be my description . I had expected overblown bass and enclosed mid-range .
The PLL1 would in someways be a modern 66 . No one makes for the domestic market anything like the 66 now .With the PLL1 I had the designers amp to use , it was excellent . I know another engineer who knows the PLL design . Dave Mate it's designer took great trouble to make the load as restive as possible . Using the Quad 306 was like using a Mini to pull a caravan , it worked ! Other respected amps sounded colouired through the PLL 1's . They show this in measurement terms also ( Naim ) . Not as obvious with other speakers . It was a rare occasion when the Quad was the clearly better choice . What the Quad lacks the speaker had in spades . The designer was shocked . It might be argued that Quad were always right and we have always been listening to speakers mostly . I still find Quad were daft as the extra engineering they needed costs peanuts . The NAD 3020 proves that . It often would drive a difficultly speaker better than 306 or 405 . If 405-2 is used in parallel bridge they will drive mostly anything . To buy a second 405-2 is usually cheap . The 405 should be tried in speaker terminal anti-phase as the 405 phase inverts the signal . A kick drum should kick forward is the usually important reason . One can use a candle flame to verify it if the grills are fixed . Some records are in reverse phase . If so that was the engineers choice and is the sound of the recording .
Royd speakers can sound very neutral in the right system . Measurements were reasonable . I feel what Royd acheived with bass mid-range units is rare . That bass unit coupled with a ribbon tweeter would have been where I would have started with new models . My point would be to explore the potenial of the 66's as you might an E Type Jaguar .
At the pub we all agreed that Celestion should have marketed kits as KEF did . A 15 , 44 and 66 Kit would have been excellent . The 15's and 44's I suspect never were as respected as they deserved to be as they were sold through discount warehouses . KEF also . The kits I think gave KEF the edge as only an idiot would ignore what they had to offer . Many started their interest that way . I am told the Linn Isobarik was 2 x KEF kit 3 . So happy was he that the concept became a product . If you hear Isobaiks and KEF Concerto's it is interesting to note how the sound is a choice and not 100 % the drive units . I think I was told that the Isobarik was only a concept . It advanced but never to the point hoped for , the first 2 weeks of it's life being more or less the pattern for the future . It did gain some neutrality with age .
As to the sound of the Royds . Having been a minor recording engineer of classical music I find the balance of sound from one place to another very different ( tonality ) . What usually is constant is the live sound experience . The Royd speakers and Quad ESL57 and 63 ( 63 preferred ) suggest a live sound . The Royd having the slightly less favorable tonality . The 66 are balanced in tonality like 63's , what they lack " it is live element " . My guess is that it is inside the 66's if the tonality is slightly compromised . Not the treble as that would be a mistake .
The PLL1 would in someways be a modern 66 . No one makes for the domestic market anything like the 66 now .With the PLL1 I had the designers amp to use , it was excellent . I know another engineer who knows the PLL design . Dave Mate it's designer took great trouble to make the load as restive as possible . Using the Quad 306 was like using a Mini to pull a caravan , it worked ! Other respected amps sounded colouired through the PLL 1's . They show this in measurement terms also ( Naim ) . Not as obvious with other speakers . It was a rare occasion when the Quad was the clearly better choice . What the Quad lacks the speaker had in spades . The designer was shocked . It might be argued that Quad were always right and we have always been listening to speakers mostly . I still find Quad were daft as the extra engineering they needed costs peanuts . The NAD 3020 proves that . It often would drive a difficultly speaker better than 306 or 405 . If 405-2 is used in parallel bridge they will drive mostly anything . To buy a second 405-2 is usually cheap . The 405 should be tried in speaker terminal anti-phase as the 405 phase inverts the signal . A kick drum should kick forward is the usually important reason . One can use a candle flame to verify it if the grills are fixed . Some records are in reverse phase . If so that was the engineers choice and is the sound of the recording .
Royd speakers can sound very neutral in the right system . Measurements were reasonable . I feel what Royd acheived with bass mid-range units is rare . That bass unit coupled with a ribbon tweeter would have been where I would have started with new models . My point would be to explore the potenial of the 66's as you might an E Type Jaguar .
At the pub we all agreed that Celestion should have marketed kits as KEF did . A 15 , 44 and 66 Kit would have been excellent . The 15's and 44's I suspect never were as respected as they deserved to be as they were sold through discount warehouses . KEF also . The kits I think gave KEF the edge as only an idiot would ignore what they had to offer . Many started their interest that way . I am told the Linn Isobarik was 2 x KEF kit 3 . So happy was he that the concept became a product . If you hear Isobaiks and KEF Concerto's it is interesting to note how the sound is a choice and not 100 % the drive units . I think I was told that the Isobarik was only a concept . It advanced but never to the point hoped for , the first 2 weeks of it's life being more or less the pattern for the future . It did gain some neutrality with age .
As to the sound of the Royds . Having been a minor recording engineer of classical music I find the balance of sound from one place to another very different ( tonality ) . What usually is constant is the live sound experience . The Royd speakers and Quad ESL57 and 63 ( 63 preferred ) suggest a live sound . The Royd having the slightly less favorable tonality . The 66 are balanced in tonality like 63's , what they lack " it is live element " . My guess is that it is inside the 66's if the tonality is slightly compromised . Not the treble as that would be a mistake .
I think the 66 has two major shortcomings, the baffle and the woofer. Both are fixable, but it would take a considerable amount of DIY time (and money) and it would result in a substantially different speaker.
Excessive diffraction on the baffle shows up early in the impulse response in a series of ripples caused by the depressions (speaker grill channels) and protrusions (woofer chassis, tweeter terminals, mounting brackets, cabinet lip). This shakes up the frequency response and creates quite a few phase shifts.
The mid dome, I think, should be crossed to the woofer about an octave higher, around 1175hz, where the on/off-axis responses are still optimal (and where there’s less risk of damage to the dome). The problem here is that the 66 woofer is smooth only to 600hz or so. Beyond that there’s a broad and audible increase in output, followed by a dip, followed by breakup mode. So not very good behavior for a 12” woofer, and not a good match to the mid dome at 1175hz.
Going active will do nothing to solve the diffraction problem, and may or may not work on the woofer issue. I don’t know much about active, but I suppose that if one crossed at 1200hz AND could use equalization to lower the woofer output between 700-1100, then it might work. But this rules out anyone using tubes.
Another solution is overlaying a new smooth baffle on the 66 and installing a new woofer that can reach 1200 - 1500hz without any issues.. Along with….. new mounts for the mid & tweeter, additional cabinet damping, and a new XO. It wouldn’t be a 66 any longer, but it would probably sound really, really good.
Excessive diffraction on the baffle shows up early in the impulse response in a series of ripples caused by the depressions (speaker grill channels) and protrusions (woofer chassis, tweeter terminals, mounting brackets, cabinet lip). This shakes up the frequency response and creates quite a few phase shifts.
The mid dome, I think, should be crossed to the woofer about an octave higher, around 1175hz, where the on/off-axis responses are still optimal (and where there’s less risk of damage to the dome). The problem here is that the 66 woofer is smooth only to 600hz or so. Beyond that there’s a broad and audible increase in output, followed by a dip, followed by breakup mode. So not very good behavior for a 12” woofer, and not a good match to the mid dome at 1175hz.
Going active will do nothing to solve the diffraction problem, and may or may not work on the woofer issue. I don’t know much about active, but I suppose that if one crossed at 1200hz AND could use equalization to lower the woofer output between 700-1100, then it might work. But this rules out anyone using tubes.
Another solution is overlaying a new smooth baffle on the 66 and installing a new woofer that can reach 1200 - 1500hz without any issues.. Along with….. new mounts for the mid & tweeter, additional cabinet damping, and a new XO. It wouldn’t be a 66 any longer, but it would probably sound really, really good.
Hi, I'm new to this forum. I've been reading this thread for a little now as I wait to receive my beautiful Celestion Ditton 66s. They were recapped in the 90's, but I expect I will need to do it again in the next few years. Any idea how long I can expect them to last under normal load?
It all started when I was searching kijiji for "mac" and came across a MAC 1900. I was a goner from that moment on, about a month ago. It is an earlier 1900 without the impedance switch. I brought it with me to audition the 66s, and I was enthralled. I loved the sounds of the two mated together. I had always wanted a McIntosh, but could never afford one until this 1900 jumped out at me. I had been using an NAD mated with a Bose 8.2 pair I bought in 87, just before my first child was born. I figured that I wouldn't be able to afford audio equipment for a good while. I was right. However, in the last two years I have jumped back in, using speakers I've picked up from garage sales to build my first, and then my second home theatre systems. They sound good, but I love that British sound!
Cheers,
Byron
It all started when I was searching kijiji for "mac" and came across a MAC 1900. I was a goner from that moment on, about a month ago. It is an earlier 1900 without the impedance switch. I brought it with me to audition the 66s, and I was enthralled. I loved the sounds of the two mated together. I had always wanted a McIntosh, but could never afford one until this 1900 jumped out at me. I had been using an NAD mated with a Bose 8.2 pair I bought in 87, just before my first child was born. I figured that I wouldn't be able to afford audio equipment for a good while. I was right. However, in the last two years I have jumped back in, using speakers I've picked up from garage sales to build my first, and then my second home theatre systems. They sound good, but I love that British sound!
Cheers,
Byron
For those of you here who might be contemplating replacing your old & tired electrolytic caps, but are also concerned that that process might alter the voicing of your speakers, you might want to try polyester caps. It appears that at least some of them share the same ESR as electrolytics. Here's a link to a post on another site. The poster is Carlspeak ( I think this person is speakerdoctor on this forum). He's done some very interesting studies on caps using a modified WT-2 setup--
Effect of film dielectric on capacitor ESR - Mods, Tweaks, and Upgrades to the Classics - The Classic Speaker Pages Discussion Forums
**********************************
Very nice! I use a MAC 4100 with mine.
Effect of film dielectric on capacitor ESR - Mods, Tweaks, and Upgrades to the Classics - The Classic Speaker Pages Discussion Forums
**********************************
Very nice! I use a MAC 4100 with mine.QUAD amplfiers ; Linn Isobaric loudspeakers .
Here is something for Nigel , and whoever else may be interested.
I did not buy a QUAD 520f at the time , and now QUAD has a better amp ...
and maybe I will build a better ... if I ever get time !
The below about QUAD is taken from another web-site:
The Quad 510 and 520 were the first professional current dumping amplifiers. Produced from 1984 until 1995.
Production numbers:
Quad 510 Mono 19" Rack Amplifier(Pro), produced from 1984-1989 numbers: 3200
Quad 510 Mono 19" Rack Amplifier(Neutrik)Pro), produced from 1989-1995 numbers: 600
Quad 520 Stereo 19 " Rack Amplifier(Pro), produced from 1984-1986 numbers: 4000
Quad 520f Stereo 19 " Rack Amplifier(Pro), produced from 1986-1989 numbers: 7500
Quad 520f Stereo 19 " Rack Amplifier(Neutrik)Pro), produced from 1989-1995 numbers: 3000
Introduction:
The 520f with Pcb 12828-8 was based on the final circuit of the 606 and 306 which were introduced in 1986. Prior to the Pcb 12828-8 there was also a 12282-7 version with a slightly different circuit diagram (mayby there are more variants?). It shows how Quad solved the problems around the current limitter circuit and the H+ rail for the class A amplifier. Still not the best solution, but a fare and good working compromise. The 520 differ on two major aspects from a 520f, a double conventional power supply instead of the power supply with the regulated earth to prevent any DC current through the loudspeaker, and discrete build current sources in the input circuit. The 606 design was based on single integrated current sources. The 510 never got the “f” suffix but was after 1986 or so also based on the new circuit (and modules).
The 520f had standard unbalanced inputs, balanced high quality input modules could be ordered separately or factory installed. The model type suffix was then changed in “M”. There was also an “S” model where no input volume controls where installed. In the first series of 520f the inrush current limiter was an option, from serial 6500 this was implemented in the factory. When buying second hand; be alert, the suffix on the front for all models is the same;”f”. There was also a special version for the BBC; 521f. Here the type number on the front was changed. The 521 had all the options, including balanced input volume controls and a provision to separate the mains earth from the chassis. To make it simple; Quad also built 520’s to special order. Recently I saw units with the number 522 on the front for sale on eBay, as far as I know, these were also build for the BBC.
Because of the pro use, it is very difficult to buy an unmolested 510 or 520(f). Only a few found their way to domestic applications or received TLC in pro environments. But from a technical (electronic) viewpoint most units can be restored to pristine order.
The 510 and 520 families have double output transistors in stead of the triples in a 606. This limits the output power, but in practice you will not notice the difference when playing music. The 520 ranges of amplifiers are thermal protected, so if you wish, the current limiters can be changed to allow a higher peak output current with low impedance loads. The 510 is not thermal protected, althoug it is mentioned in the manuals.
If you can chose: buy a 520f. The 510 is also very good, but is single channel and primarily designed for driving special loads.
Current limiters:
Quad struggled with the design of the limiters from the first 405 schematic. The key problem is the position of the upper dumper driver. Ideal this driver, which is a part of the class A stage, should be connected to H+ (like the rest of the class A stage). In most schematics, including some of the the 606, the voltage of the emitter of the driver is modulated by the voltage across the current sensing resistor in the dumper circuit. The final solution from Quad is to connect the class A stage to this point. And in this way trusted on the very good Psu ripple rejection ratio of the class A stage. But in this case the ripple is not only mains borne, but also ”musical”. Why not implemented it the “right” way?
*** *** *** *** ***
Linn Isobaric :
A major difference between the Linn design and the KEF Concerto was that Linn used the two bass drivers in acoustic series
- that is , one is inside the cabinet , directly behind the one visible on the front baffle.
This causes a doubling of moving mass , and thus a lower in-box resonant frequency for the bass.
The Linn would play bass clearly to lower in frequency than any KEF model using the same B139 driver.
This Isobaric loading allows bass response that is truer to the original signal than the type of bass that is achieved from vented cabinets and passive radiator/ABR cabinets -{such as Celestion 66 is}.
If the Celestion bass drivers were available in 8 ohm options we could remove the ABR , block off its aperature ,
install a second 12" driver behind the visible one , and achieve tighter bass response -{if optimally done}.
Using two Celestion 4 ohm drivers connected in electrical Series will not produce as tight bass as two 8 ohm drivers connected in electrical Parallel.
A second 12" active driver mounted in place of the ABR will not play bass to lower in frequency , but will only play the existing bass louder.
We can achieve part of that with lower DCR inductors in the bass filter ... which I will get to.
I prefer now to not continue with side-issues in discussion , even though such are interesting ,
because there is so much back-log for me to get through from posts that are directly related to practical ways to restore or improve the 66 ,
but in keeping with its characteristic sound.
A Celestion 66 can never be got to sound like a QUAD ESL63 , not like any type of panel-diaphragm/no-cabinet loudspeaker.
Here is something for Nigel , and whoever else may be interested.
I did not buy a QUAD 520f at the time , and now QUAD has a better amp ...
and maybe I will build a better ... if I ever get time !
The below about QUAD is taken from another web-site:
The Quad 510 and 520 were the first professional current dumping amplifiers. Produced from 1984 until 1995.
Production numbers:
Quad 510 Mono 19" Rack Amplifier(Pro), produced from 1984-1989 numbers: 3200
Quad 510 Mono 19" Rack Amplifier(Neutrik)Pro), produced from 1989-1995 numbers: 600
Quad 520 Stereo 19 " Rack Amplifier(Pro), produced from 1984-1986 numbers: 4000
Quad 520f Stereo 19 " Rack Amplifier(Pro), produced from 1986-1989 numbers: 7500
Quad 520f Stereo 19 " Rack Amplifier(Neutrik)Pro), produced from 1989-1995 numbers: 3000
Introduction:
The 520f with Pcb 12828-8 was based on the final circuit of the 606 and 306 which were introduced in 1986. Prior to the Pcb 12828-8 there was also a 12282-7 version with a slightly different circuit diagram (mayby there are more variants?). It shows how Quad solved the problems around the current limitter circuit and the H+ rail for the class A amplifier. Still not the best solution, but a fare and good working compromise. The 520 differ on two major aspects from a 520f, a double conventional power supply instead of the power supply with the regulated earth to prevent any DC current through the loudspeaker, and discrete build current sources in the input circuit. The 606 design was based on single integrated current sources. The 510 never got the “f” suffix but was after 1986 or so also based on the new circuit (and modules).
The 520f had standard unbalanced inputs, balanced high quality input modules could be ordered separately or factory installed. The model type suffix was then changed in “M”. There was also an “S” model where no input volume controls where installed. In the first series of 520f the inrush current limiter was an option, from serial 6500 this was implemented in the factory. When buying second hand; be alert, the suffix on the front for all models is the same;”f”. There was also a special version for the BBC; 521f. Here the type number on the front was changed. The 521 had all the options, including balanced input volume controls and a provision to separate the mains earth from the chassis. To make it simple; Quad also built 520’s to special order. Recently I saw units with the number 522 on the front for sale on eBay, as far as I know, these were also build for the BBC.
Because of the pro use, it is very difficult to buy an unmolested 510 or 520(f). Only a few found their way to domestic applications or received TLC in pro environments. But from a technical (electronic) viewpoint most units can be restored to pristine order.
The 510 and 520 families have double output transistors in stead of the triples in a 606. This limits the output power, but in practice you will not notice the difference when playing music. The 520 ranges of amplifiers are thermal protected, so if you wish, the current limiters can be changed to allow a higher peak output current with low impedance loads. The 510 is not thermal protected, althoug it is mentioned in the manuals.
If you can chose: buy a 520f. The 510 is also very good, but is single channel and primarily designed for driving special loads.
Current limiters:
Quad struggled with the design of the limiters from the first 405 schematic. The key problem is the position of the upper dumper driver. Ideal this driver, which is a part of the class A stage, should be connected to H+ (like the rest of the class A stage). In most schematics, including some of the the 606, the voltage of the emitter of the driver is modulated by the voltage across the current sensing resistor in the dumper circuit. The final solution from Quad is to connect the class A stage to this point. And in this way trusted on the very good Psu ripple rejection ratio of the class A stage. But in this case the ripple is not only mains borne, but also ”musical”. Why not implemented it the “right” way?
*** *** *** *** ***
Linn Isobaric :
A major difference between the Linn design and the KEF Concerto was that Linn used the two bass drivers in acoustic series
- that is , one is inside the cabinet , directly behind the one visible on the front baffle.
This causes a doubling of moving mass , and thus a lower in-box resonant frequency for the bass.
The Linn would play bass clearly to lower in frequency than any KEF model using the same B139 driver.
This Isobaric loading allows bass response that is truer to the original signal than the type of bass that is achieved from vented cabinets and passive radiator/ABR cabinets -{such as Celestion 66 is}.
If the Celestion bass drivers were available in 8 ohm options we could remove the ABR , block off its aperature ,
install a second 12" driver behind the visible one , and achieve tighter bass response -{if optimally done}.
Using two Celestion 4 ohm drivers connected in electrical Series will not produce as tight bass as two 8 ohm drivers connected in electrical Parallel.
A second 12" active driver mounted in place of the ABR will not play bass to lower in frequency , but will only play the existing bass louder.
We can achieve part of that with lower DCR inductors in the bass filter ... which I will get to.
I prefer now to not continue with side-issues in discussion , even though such are interesting ,
because there is so much back-log for me to get through from posts that are directly related to practical ways to restore or improve the 66 ,
but in keeping with its characteristic sound.
A Celestion 66 can never be got to sound like a QUAD ESL63 , not like any type of panel-diaphragm/no-cabinet loudspeaker.
continuing with some of the current backlog/s
Hi sba , and others interested ,
you have discovered correctly about the diffraction !
Part of it can be ameliorated ... and I'll get to that in conjunction with a related matter at some time.
Also you are correct about the Celestion woofer , and indeed your long-ago measurements showed that ,
and that is why it does not work well with the HF1300s in the 25 model , because the crossover there will be above 1500Hz and probably above 2000Hz.
{I have forgotten the specification data for the 25.}
Optimum crossover needs to be below that approx 750Hz region resonant mode , and that is well below the ragged behavior higher in the woofer's midrange.
About the MF500 and MD500 - they could be crossed over higher than the notional 500Hz
{ and from your plots they probably cross somewhere between 550Hz-600Hz , and there will be differences owing to the age-related deterioration of some of the domes }
BUT they cannot be crossed too much higher or the sound will deteriorate in the midrange.
Bandwidth for a midrange driver needs to be no less than 2 octaves , and preferably a little wider , because when less the sound becomes quite Honky or squawky resonant ,
particually when with a Passive crossover , because the capacitors and inductors form a resonant circuit , and that is coupled into the driver.
We have been trying to minimise the resonance via the ESR simulating resistors , particually the one in series with the 4uF cap.
Basically, resistance in Series with the Parallel elements of a bandpass filter { that midrange filter } are anti-resonance resistors.
There is about 1.3 ohms in the 2.2mH inductor , which is probably sufficient , and we don't want less there , especially when trying various values between 1.8 ohm <--> 3.9 ohm in series with the 4uF capacitor.
The upper frequency limit of the mid-domes is about 5kHz , thus for 2 octaves they cannot be used below 1.25kHz , but even with supression resistors there will be more audible honk/squawk if only 2 octaves than with the current slightly greater than 3 octaves.
Audibly , I think 3 octaves is the minimum , and some experts prefer 1 Decade ... which here would be 500Hz <--> 5kHz ... which co-incidentally or purposely is the Celestion specification !
We can sneak the crossover up to about 600Hz , but I doubt that better can be achieved ,
AND , for those domes their low end roll-off is part of the crossover.
The 24uF cap + 2.2mH inductor combine with the driver's own roll-off to form a 4th Order/24dB per octave slope filter , thus if one wants to cross higher the driver's own response will assist less and an additional filter component would have to be added ... and that would upset the .34mH + 4uF filter at the other end , because all the passive components interact ... so you would be opening a can of worms if altering too far with the existing bandpass filter.
Celestion already sneaked it up to likely as high as it may work when changing from 30uF to 24uF , thus I doubt we can do a lot better unless substantial re-design ... and to do that will require a lot more measurements to be made.
*** *** ***
Your Link to the Polyester capacitors is interesting - thankyou ,
however these will only cause part of the original sound , because Polyester caps though having greater ESR and Dielectric Absorption than Polypropylene caps still have a lot less of both than the old Electrolytic caps had , and even than most modern electrolytic caps have.
A few modern electrolytic caps are designed to have very low ESR , and these were specifically for Switch-Mode Power Supply use -{though clever engineers do apply them elsewhere},
however they still have poor Dielectric Absorption -{which is not a problem in SMPS application}- but if one does not want the clarity possible via a good Polypropylene cap , nor the limited life of an Electrolytic cap , then do try Polyester caps.
*** *** ***
Please have a look at Post #896 on Page 90 .
Can you advise DennyG about a suitable USB audio interface ?
If so , I can post about another useful measurement he can do.
*** *** ***
Hi TechnoDweeb/Byron ,
to assist to answer your question , can you post photos of the crossovers showing all the capacitors currently in there ?
If no photos possible , then tell us , are they Electrolytic or what ?
Post all the Letters and Digits that are printed on the capacitors.
*** *** ***
G'day Reggie ,
to answer two more of your questions:
I have read the web-site for the Paul Speltz Anti Cable.
This is an excellent idea , and similar to what I have stated in several places.
Basically it is the Dielectric - the Insulation - of the cable that causes some of the audible differences between different cable designs.
I am not convinced that his design for Interconnect Cable will work in all applications ... have you tried it ? ... and if so to what audible result ?
If you are using that as l'speaker cable the do use it also inside the enclosures -{I recommended similar in a post some time ago}
- but do not throw away the old Celestion internal wire , because that is good wire to solder to if such is ever needed , and better than some modern low-priced wire.
If you intend to replace the inductors in the bass filter later when I get to the low DCR matter , then I recommend you make a second crossover board for those then ,
and mount the new to come caps for the bass filter there-on also , and wire that board separately to the output ,
and that could be fully Bi-Wire back to the amplifier.
I doubt there will be a lot of audible benefit to splitting the midrange and treble filters and fully Tri-Wiring back to the amplifier , but you can if you want to ,
and you may hear some improvement , or perhaps your wife will ... but how much cable does she want covering the floor ... and , who cleans the floor !
*** *** ***
I have to go now , but hope to be back before too long , and will continue with the back-log then.
Hi sba , and others interested ,
you have discovered correctly about the diffraction !
Part of it can be ameliorated ... and I'll get to that in conjunction with a related matter at some time.
Also you are correct about the Celestion woofer , and indeed your long-ago measurements showed that ,
and that is why it does not work well with the HF1300s in the 25 model , because the crossover there will be above 1500Hz and probably above 2000Hz.
{I have forgotten the specification data for the 25.}
Optimum crossover needs to be below that approx 750Hz region resonant mode , and that is well below the ragged behavior higher in the woofer's midrange.
About the MF500 and MD500 - they could be crossed over higher than the notional 500Hz
{ and from your plots they probably cross somewhere between 550Hz-600Hz , and there will be differences owing to the age-related deterioration of some of the domes }
BUT they cannot be crossed too much higher or the sound will deteriorate in the midrange.
Bandwidth for a midrange driver needs to be no less than 2 octaves , and preferably a little wider , because when less the sound becomes quite Honky or squawky resonant ,
particually when with a Passive crossover , because the capacitors and inductors form a resonant circuit , and that is coupled into the driver.
We have been trying to minimise the resonance via the ESR simulating resistors , particually the one in series with the 4uF cap.
Basically, resistance in Series with the Parallel elements of a bandpass filter { that midrange filter } are anti-resonance resistors.
There is about 1.3 ohms in the 2.2mH inductor , which is probably sufficient , and we don't want less there , especially when trying various values between 1.8 ohm <--> 3.9 ohm in series with the 4uF capacitor.
The upper frequency limit of the mid-domes is about 5kHz , thus for 2 octaves they cannot be used below 1.25kHz , but even with supression resistors there will be more audible honk/squawk if only 2 octaves than with the current slightly greater than 3 octaves.
Audibly , I think 3 octaves is the minimum , and some experts prefer 1 Decade ... which here would be 500Hz <--> 5kHz ... which co-incidentally or purposely is the Celestion specification !
We can sneak the crossover up to about 600Hz , but I doubt that better can be achieved ,
AND , for those domes their low end roll-off is part of the crossover.
The 24uF cap + 2.2mH inductor combine with the driver's own roll-off to form a 4th Order/24dB per octave slope filter , thus if one wants to cross higher the driver's own response will assist less and an additional filter component would have to be added ... and that would upset the .34mH + 4uF filter at the other end , because all the passive components interact ... so you would be opening a can of worms if altering too far with the existing bandpass filter.
Celestion already sneaked it up to likely as high as it may work when changing from 30uF to 24uF , thus I doubt we can do a lot better unless substantial re-design ... and to do that will require a lot more measurements to be made.
*** *** ***
Your Link to the Polyester capacitors is interesting - thankyou ,
however these will only cause part of the original sound , because Polyester caps though having greater ESR and Dielectric Absorption than Polypropylene caps still have a lot less of both than the old Electrolytic caps had , and even than most modern electrolytic caps have.
A few modern electrolytic caps are designed to have very low ESR , and these were specifically for Switch-Mode Power Supply use -{though clever engineers do apply them elsewhere},
however they still have poor Dielectric Absorption -{which is not a problem in SMPS application}- but if one does not want the clarity possible via a good Polypropylene cap , nor the limited life of an Electrolytic cap , then do try Polyester caps.
*** *** ***
Please have a look at Post #896 on Page 90 .
Can you advise DennyG about a suitable USB audio interface ?
If so , I can post about another useful measurement he can do.
*** *** ***
Hi TechnoDweeb/Byron ,
to assist to answer your question , can you post photos of the crossovers showing all the capacitors currently in there ?
If no photos possible , then tell us , are they Electrolytic or what ?
Post all the Letters and Digits that are printed on the capacitors.
*** *** ***
G'day Reggie ,
to answer two more of your questions:
I have read the web-site for the Paul Speltz Anti Cable.
This is an excellent idea , and similar to what I have stated in several places.
Basically it is the Dielectric - the Insulation - of the cable that causes some of the audible differences between different cable designs.
I am not convinced that his design for Interconnect Cable will work in all applications ... have you tried it ? ... and if so to what audible result ?
If you are using that as l'speaker cable the do use it also inside the enclosures -{I recommended similar in a post some time ago}
- but do not throw away the old Celestion internal wire , because that is good wire to solder to if such is ever needed , and better than some modern low-priced wire.
If you intend to replace the inductors in the bass filter later when I get to the low DCR matter , then I recommend you make a second crossover board for those then ,
and mount the new to come caps for the bass filter there-on also , and wire that board separately to the output ,
and that could be fully Bi-Wire back to the amplifier.
I doubt there will be a lot of audible benefit to splitting the midrange and treble filters and fully Tri-Wiring back to the amplifier , but you can if you want to ,
and you may hear some improvement , or perhaps your wife will ... but how much cable does she want covering the floor ... and , who cleans the floor !
*** *** ***
I have to go now , but hope to be back before too long , and will continue with the back-log then.
Last edited:
When I had a shop I often thought of selling the Quad professional series . I never got around to it . I always said if I win the lottery I would reopen the shop and do it right . I would set aside a sum of money to loose and see how many years I could keep going . The point of the shop to be a museum that sells things . Aga cookers were mostly that for years and made where the industrial revolution started .
Isobariks contrived not to show the best of KEF . The Concerto being better balanced . I am told the up firing drive units were to cancel vibrational traits like a V twin engine . I saw it analyzed years ago , it seemed a reasonable idea .
My friend a man of 98 years showed me his speakers surrounded in carpet . He said surely we put the damping in the wrong place ? We want to stop reflections from side walls ? Classic stereo might be more to do with that than the writings of Biblical proportions ? Make the reflected wave from the side go anywhere except into our ears at -20dB , -40 dB would be a better figure ?
The friend also had broom sticks and weights to brace the speaker to the wall ( corner ) . The speaker on a slight tilt .
This was my friend . I doubt anyone knew him as a mathematician ?
https://www.google.co.uk/search?q=n...rGKWM0AWnloDgCA&ved=0CCgQsAQ&biw=1280&bih=885
Isobariks contrived not to show the best of KEF . The Concerto being better balanced . I am told the up firing drive units were to cancel vibrational traits like a V twin engine . I saw it analyzed years ago , it seemed a reasonable idea .
My friend a man of 98 years showed me his speakers surrounded in carpet . He said surely we put the damping in the wrong place ? We want to stop reflections from side walls ? Classic stereo might be more to do with that than the writings of Biblical proportions ? Make the reflected wave from the side go anywhere except into our ears at -20dB , -40 dB would be a better figure ?
The friend also had broom sticks and weights to brace the speaker to the wall ( corner ) . The speaker on a slight tilt .
This was my friend . I doubt anyone knew him as a mathematician ?
https://www.google.co.uk/search?q=n...rGKWM0AWnloDgCA&ved=0CCgQsAQ&biw=1280&bih=885
corrections and additions to my #917 and 918 above
In #917 where about the B139s in the Linn Isobaric I stated:
"This causes a doubling of moving mass , and thus a lower in-box resonant frequency for the bass."
I should have stated:
This causes a doubling of moving mass but without doubling the moving surface area , and thus there is a lower in-box resonant frequency for the bass.
***
In #918 in comment following sba's #916 the obvious thing that I didn't fully describe include is:
The ESR of some modern Electrolytic caps may be as low , or almost as low , as that of Polyester caps ,
but none of these caps have ESR as high as the type of Electro caps used in the Celestion 66 , 44 , 25 , 15 , etc ...
The Polyesters' ESR , whilst not as low as in Polypropylene caps , is very low ,
and thus the resistors I have been recommending will still need to be installed if one wants to get close to the original sound with Polyester caps.
If modern Electrolytic caps are installed there will likely also be some resistance needed , unless a low quality modern Electro cap.
***
'ullo Nigel ,
that could be part of the reason for the upwards firing , top of cabinet installed drivers in the Linn Isobaric ,
but I think it was also intended so as to spread more of the midrange and treble around the room , plus to compensate for the lower efficiency of the KEF B110 and T27A versus the B139 ,
and thereby present the full frequency spectrum all at the same level in the room ... though the drivers' arrangement does not permit accurate transient response.
Imaging and purity of tone were not a priority for Linn , but Dynamic Range and some aspects of colouration were.
Linn Isobarics did not sound like Neutral reproducers to me , but they did present the dynamics' scale of of a wide dynamic range recording to greater degree than the KEF Concerto.
I think that Linn Isobarics are only accurate tonality in the low bass.
About that QUAD amplifiers article , I thought the bits that would interest you most are the parts about the Current Limiters and the Class A stage ,
given your earlier comment about the QUAD 405 versus the original NAD 3020.
In #917 where about the B139s in the Linn Isobaric I stated:
"This causes a doubling of moving mass , and thus a lower in-box resonant frequency for the bass."
I should have stated:
This causes a doubling of moving mass but without doubling the moving surface area , and thus there is a lower in-box resonant frequency for the bass.
***
In #918 in comment following sba's #916 the obvious thing that I didn't fully describe include is:
The ESR of some modern Electrolytic caps may be as low , or almost as low , as that of Polyester caps ,
but none of these caps have ESR as high as the type of Electro caps used in the Celestion 66 , 44 , 25 , 15 , etc ...
The Polyesters' ESR , whilst not as low as in Polypropylene caps , is very low ,
and thus the resistors I have been recommending will still need to be installed if one wants to get close to the original sound with Polyester caps.
If modern Electrolytic caps are installed there will likely also be some resistance needed , unless a low quality modern Electro cap.
***
'ullo Nigel ,
that could be part of the reason for the upwards firing , top of cabinet installed drivers in the Linn Isobaric ,
but I think it was also intended so as to spread more of the midrange and treble around the room , plus to compensate for the lower efficiency of the KEF B110 and T27A versus the B139 ,
and thereby present the full frequency spectrum all at the same level in the room ... though the drivers' arrangement does not permit accurate transient response.
Imaging and purity of tone were not a priority for Linn , but Dynamic Range and some aspects of colouration were.
Linn Isobarics did not sound like Neutral reproducers to me , but they did present the dynamics' scale of of a wide dynamic range recording to greater degree than the KEF Concerto.
I think that Linn Isobarics are only accurate tonality in the low bass.
About that QUAD amplifiers article , I thought the bits that would interest you most are the parts about the Current Limiters and the Class A stage ,
given your earlier comment about the QUAD 405 versus the original NAD 3020.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.