Hi Alan,
The east coast of Australia is very very long. I'm located at the extreme south of the east coast. We have had very hot weather here; no fires but no rain either. Our "once in a hundred years flood/fire/cyclone" cycle now appears to be"once every few years".
.I think I have worked out how to post images. The diyAudio support page has details on how to do it. My post (#941) is (if you can access it) my understanding of your prescription for my crossover detailed at your posts #937 and #938. Forgive me if I have confused "in parallel" with "in series". Have I got it wrong??
Reggie
The east coast of Australia is very very long. I'm located at the extreme south of the east coast. We have had very hot weather here; no fires but no rain either. Our "once in a hundred years flood/fire/cyclone" cycle now appears to be"once every few years".
.I think I have worked out how to post images. The diyAudio support page has details on how to do it. My post (#941) is (if you can access it) my understanding of your prescription for my crossover detailed at your posts #937 and #938. Forgive me if I have confused "in parallel" with "in series". Have I got it wrong??
Reggie
#941 needs a correction ; #943 displays OK , and a comment about ...
G'day Reggie ,
About your schematic in #941 , the positions of the capacitors in the woofer section need to be reversed:
From the junction point of the 3.5mH and 2.2mH inductors connect the 68uF + 1 ohm and the 8.2uF + 8.2 ohm.
From the other end of the 2.2mH at junction to the woofer (+) connect the 68uF + 1.5 ohm -{ this is the Output of the filter }.
The midrange circuit is correct.
The tweeter circuit is correct , presuming the L-pad options are:
1.2 ohm and 15 ohm
1.1 ohm and 18 ohm
Post the schematic again after redrawing it if you want further confirmation.
About the Review in #943 , at bottom of first column where it is stated:
" a trough in the octave centered on 4kHz "
measurement there depends very much on the vertical axis - that is the height of the microphone with respect to the positions of the mid-dome and tweeter.
When measuring , and listening , on the optimum vertical axis that trough will largely fill in.
You should be able to hear the effect by crouching close to one cabinet and with one ear pointed towards it ,
then move up and down so as to listen on all of:
ear at tweeter ; ear about half way between tweeter and mid-dome ; ear at mid-dome.
When listening further away you will need a chair that allows you to have your ears at the height you preferred for the close test.
Listening at heights significantly above the tweeter and significantly below the mid-dome will result in peculiar tonal balances.
The review is interesting in other aspects - thank you for posting it !
Post your listening impressions after you have listened awhile with the new capacitors and L-pad in ,
and then we will consider reduced resistance inductors in the woofer filter.
I refer you also to my Post #925 on Page 93 , in particular to:
(4)- The upper bass problem may be owing to how you have blocked off the ABR/passive radiator.
Is it still in the cabinet , and with some thing fixed over its front ? ,
or , is it removed and something fixed into the open space ,
and if this latter then how thick and stiff is the new fixing and how well sealed so no air gaps around it ?
G'day Reggie ,
About your schematic in #941 , the positions of the capacitors in the woofer section need to be reversed:
From the junction point of the 3.5mH and 2.2mH inductors connect the 68uF + 1 ohm and the 8.2uF + 8.2 ohm.
From the other end of the 2.2mH at junction to the woofer (+) connect the 68uF + 1.5 ohm -{ this is the Output of the filter }.
The midrange circuit is correct.
The tweeter circuit is correct , presuming the L-pad options are:
1.2 ohm and 15 ohm
1.1 ohm and 18 ohm
Post the schematic again after redrawing it if you want further confirmation.
About the Review in #943 , at bottom of first column where it is stated:
" a trough in the octave centered on 4kHz "
measurement there depends very much on the vertical axis - that is the height of the microphone with respect to the positions of the mid-dome and tweeter.
When measuring , and listening , on the optimum vertical axis that trough will largely fill in.
You should be able to hear the effect by crouching close to one cabinet and with one ear pointed towards it ,
then move up and down so as to listen on all of:
ear at tweeter ; ear about half way between tweeter and mid-dome ; ear at mid-dome.
When listening further away you will need a chair that allows you to have your ears at the height you preferred for the close test.
Listening at heights significantly above the tweeter and significantly below the mid-dome will result in peculiar tonal balances.
The review is interesting in other aspects - thank you for posting it !
Post your listening impressions after you have listened awhile with the new capacitors and L-pad in ,
and then we will consider reduced resistance inductors in the woofer filter.
I refer you also to my Post #925 on Page 93 , in particular to:
(4)- The upper bass problem may be owing to how you have blocked off the ABR/passive radiator.
Is it still in the cabinet , and with some thing fixed over its front ? ,
or , is it removed and something fixed into the open space ,
and if this latter then how thick and stiff is the new fixing and how well sealed so no air gaps around it ?
Last edited:
Hello Alan. Thank you for vetting my draft crossover. The above post incorporates your changes (I hope). Unfortunately I cannot purchase anything right now. My wife has left for South America (she said she was going to the shops but I got a text from Manchu Pichu saying she would see me in April) and she has the credit card - I'm not allowed to have one because of my condition. Maybe by then you will have sorted the inductors. Alan, you asked about whether the level of treble was ok for my ears. But that's pretty difficult to assess when one has lost their frequency register. So, I often think that I am missing the full spectrum of sound. My gauge is the sound of cymbals. On something like Michael Frank's album "The Art of Tea" (featuring the great guitarist Larry Carlton) on the track "Eggplant" I can hear everything (so my wife assures me) whereas on something like Wishbone Ash's Album "Argus" at track no 6 "Warrior" about a minute in the music goes quiet with just a guitar playing a soft lead tune. BUT, if I put my ear to the tweeter I can hear the drummer playing a lovely rat a tat tat on the cymbal with the drum stick which I cannot hear from my listening position but my wife can. So, I can only conclude that if a high note is played too softly, is mixed too far back by the engineer or is just too high for my aged hearing then it's not going to be heard by me. As regards the bass I have taken the board off the passive woofer and am happy with what I hear. I think this is also highly dependent on how a record was produced. I think I will leave this at this for now until I hear from you re the inductors. I have one other question for you or other readers who have maybe experimented and that is as regards the foam stuffing inside the Celestion cabinets. Mine is stuffed head to foot with 3 inch yellow foam. Is this correct?
Reggie
Reggie
Corrections to #946 , and replies to #947 , and do we have sba and DennyG reading ?
G'day Reggie ,
In your schematic in #946 , the woofer circuit is now correct ,
however you have omitted the 3.9/4uf + 3.9 ohms in the midrange circuit !
{ The mids circuit was correct in your schematic in #941. }
The L-Pad resistor combinations around the tweeter are NOT correct in #946.
They should be:- 1.2 ohms with 15 ohms , and 1.1 ohms with 18 ohms.
Note to other readers ,
these L-Pad values are only for application with a Hiquphon OWII tweeter , and not for any other tweeters we have discussed to date in this thread.
-
Reggie , if I was to go to Machu Pichu I might not come back !
thus it is fortunate for you that I do not have your credit card.
A while ago my cousin in Australia and I were discussing international payments for components , etc ...
He said that in Australia there are Debit cards available which can be used on-line as substitute for the Credit card information the sellers require.
They are issued by the Banks and Credit Unions and Building Societies there , and are Master Card and Visa Debit cards.
None have annual fees , though some banks apply small monthly fees ,
however my cousin has one with a Credit Union that does not charge any fee for it.
There is an International Currency Transfer fee , and that varies between 2.5% - 3% depending on which Bank or Credit Union.
The good thing about these Debit cards is that one can only spend one's own money ,
because they draw only from what one has in the linked account ,
thus one cannot overspend , and there is no Interest charged because one is using one's own money , not the bank's.
Perhaps you can get one of those for your own use ...
I do not have a copy of Michael Franks' "The Art Of Tea" -{ though I do drink tea ...} , and it is a very long time since I heard it ,
and about the sound I can remember only that it likely had a tidily balanced mix of the type common for such music in that era.
I do have Wishbone Ash - "Argus" - it is one of my favorite albums from way back soon after it was originally released !
... it was a Classic album of its era , and remains such I think !!
I have the original vinyl LP plus a later copy as the original got a bit worn ... of course ... and two CDs.
One CD is a direct transfer of the original tapes , plus one extra song that had been originally released only on a 7" single ,
and the other CD is the 2002 "Expanded Edition" , which sounds like a re-mix because a few instruments are placed slightly differently in the stereo field ,
and are at different levels to the original.
It sounds like a Digital re-mix , and possibly the original analog multitrack tape was copied to a digital multitrack before remixing.
This CD has 3 extra tracks - all from a live concert from Memphis , and are songs from the band's earlier albums.
Which version of the CD do you have ?
I am not sure where I have put the first CD , so I listened to only the later CD before replying here.
Its sound is noticeably brighter than the original CD and the LPs , and that is quite nice for the Acoustic Guitar parts ,
though I'm not entirely convinced it is necessarily better for any of the rest.
For the track "Warrior" - you are referring to the section that starts about 1 minute in from the beginning.
The quiet cymbals' pattern there continues behind some vocals , and there are 2 guitars playing ,
and that cymbal pattern continues till about 3 minutes in till the band change rhythm.
That is what I hear in the mix on the "Expanded Edition" CD.
The cymbals are fairly quiet , though I can hear them with no difficulty with the volume level moderate in my medium-small listening room.
After you install the 11uF cap in place of the 8.2uF to the tweeter the lower treble will fill in , and I think you will better hear that cymbal part ,
unless you have the earlier CD and it is mixed differently ... ?
-
FOAM STUFFING :-
sba and DennyG ,
The foam stuffing - 3 inch thick - is it only a single 3" layer against all walls inside your cabinets ,
or is there double thickness or more in some places ?
For the bass response to be even and no major internal resonances inside the cabinet that tall cabinet needs at least one quarter of its height filled.
That can be one eighth at top and one eighth at bottom , or all at bottom behind the ABR/passive radiator.
For balanced control of the ABR it will likely be better if the entire area behind it is filled , firmly , but not stuffed tight.
Behind the woofer needs to be only 3" on both side walls but perhaps thicker than 3" on the back wall , because ideally one quarter of the depth should be filled ,
however if the entire space behind the woofer is filled that may be to obtain the bass damping that Celestion wanted ,
and I would not necessarily change it , but I might experiment with the foam there only on the sides and back to hear how different the sound is.
The 3" should continue on both sides and back to the top , and 3" under the top would be useful.
For Acoustic grade foam to work , it must NOT be stuffed compressed very tight , but only firmly placed so that it cannot move around ,
because if any internal damping material moves like a large floppy area when the woofer is moving , the woofer becomes unevenly Mass Loaded ,
and that detrimentally affects how it works.
The only movement in damping materials should be very small local area movement within the material - too small to see moving.
Damping material needs to vibrate as very small movement , as that is random phase and does not load the woofer at some frequencies differently to at other frequencies.
If the damping material is compressed there is not sufficient space between its fibres for them to vibrate in , nor for sufficient quantity of sound waves to penetrate.
The above also applies to damping material inside sealed mid-range enclosures and inside tweeters behind the dome.
-
INDUCTORS :-
Do the interested persons here want to wind their own inductors , or buy ready wound ?
G'day Reggie ,
In your schematic in #946 , the woofer circuit is now correct ,
however you have omitted the 3.9/4uf + 3.9 ohms in the midrange circuit !
{ The mids circuit was correct in your schematic in #941. }
The L-Pad resistor combinations around the tweeter are NOT correct in #946.
They should be:- 1.2 ohms with 15 ohms , and 1.1 ohms with 18 ohms.
Note to other readers ,
these L-Pad values are only for application with a Hiquphon OWII tweeter , and not for any other tweeters we have discussed to date in this thread.
-
Reggie , if I was to go to Machu Pichu I might not come back !
thus it is fortunate for you that I do not have your credit card.
A while ago my cousin in Australia and I were discussing international payments for components , etc ...
He said that in Australia there are Debit cards available which can be used on-line as substitute for the Credit card information the sellers require.
They are issued by the Banks and Credit Unions and Building Societies there , and are Master Card and Visa Debit cards.
None have annual fees , though some banks apply small monthly fees ,
however my cousin has one with a Credit Union that does not charge any fee for it.
There is an International Currency Transfer fee , and that varies between 2.5% - 3% depending on which Bank or Credit Union.
The good thing about these Debit cards is that one can only spend one's own money ,
because they draw only from what one has in the linked account ,
thus one cannot overspend , and there is no Interest charged because one is using one's own money , not the bank's.
Perhaps you can get one of those for your own use ...
I do not have a copy of Michael Franks' "The Art Of Tea" -{ though I do drink tea ...} , and it is a very long time since I heard it ,
and about the sound I can remember only that it likely had a tidily balanced mix of the type common for such music in that era.
I do have Wishbone Ash - "Argus" - it is one of my favorite albums from way back soon after it was originally released !
... it was a Classic album of its era , and remains such I think !!
I have the original vinyl LP plus a later copy as the original got a bit worn ... of course ... and two CDs.
One CD is a direct transfer of the original tapes , plus one extra song that had been originally released only on a 7" single ,
and the other CD is the 2002 "Expanded Edition" , which sounds like a re-mix because a few instruments are placed slightly differently in the stereo field ,
and are at different levels to the original.
It sounds like a Digital re-mix , and possibly the original analog multitrack tape was copied to a digital multitrack before remixing.
This CD has 3 extra tracks - all from a live concert from Memphis , and are songs from the band's earlier albums.
Which version of the CD do you have ?
I am not sure where I have put the first CD , so I listened to only the later CD before replying here.
Its sound is noticeably brighter than the original CD and the LPs , and that is quite nice for the Acoustic Guitar parts ,
though I'm not entirely convinced it is necessarily better for any of the rest.
For the track "Warrior" - you are referring to the section that starts about 1 minute in from the beginning.
The quiet cymbals' pattern there continues behind some vocals , and there are 2 guitars playing ,
and that cymbal pattern continues till about 3 minutes in till the band change rhythm.
That is what I hear in the mix on the "Expanded Edition" CD.
The cymbals are fairly quiet , though I can hear them with no difficulty with the volume level moderate in my medium-small listening room.
After you install the 11uF cap in place of the 8.2uF to the tweeter the lower treble will fill in , and I think you will better hear that cymbal part ,
unless you have the earlier CD and it is mixed differently ... ?
-
FOAM STUFFING :-
sba and DennyG ,
The foam stuffing - 3 inch thick - is it only a single 3" layer against all walls inside your cabinets ,
or is there double thickness or more in some places ?
For the bass response to be even and no major internal resonances inside the cabinet that tall cabinet needs at least one quarter of its height filled.
That can be one eighth at top and one eighth at bottom , or all at bottom behind the ABR/passive radiator.
For balanced control of the ABR it will likely be better if the entire area behind it is filled , firmly , but not stuffed tight.
Behind the woofer needs to be only 3" on both side walls but perhaps thicker than 3" on the back wall , because ideally one quarter of the depth should be filled ,
however if the entire space behind the woofer is filled that may be to obtain the bass damping that Celestion wanted ,
and I would not necessarily change it , but I might experiment with the foam there only on the sides and back to hear how different the sound is.
The 3" should continue on both sides and back to the top , and 3" under the top would be useful.
For Acoustic grade foam to work , it must NOT be stuffed compressed very tight , but only firmly placed so that it cannot move around ,
because if any internal damping material moves like a large floppy area when the woofer is moving , the woofer becomes unevenly Mass Loaded ,
and that detrimentally affects how it works.
The only movement in damping materials should be very small local area movement within the material - too small to see moving.
Damping material needs to vibrate as very small movement , as that is random phase and does not load the woofer at some frequencies differently to at other frequencies.
If the damping material is compressed there is not sufficient space between its fibres for them to vibrate in , nor for sufficient quantity of sound waves to penetrate.
The above also applies to damping material inside sealed mid-range enclosures and inside tweeters behind the dome.
-
INDUCTORS :-
Do the interested persons here want to wind their own inductors , or buy ready wound ?
Hi Alan.
Thank you for actually looking at my diagram. Obviously it was more than I did. A note on credit cards. I got my first - and only - credit card in 1975 and spent my maximum in 1 day (by purchasing a motor-bike for $3000). When I got home I realised the type of person I could become and cut the card up. So now if I haven't got the cash it doesn't get bought. I could get a debit card but I use my wife's instead heh heh.
Michael Frank's "The Art of Tea" is a very beguiling record. It's humorous lyrics and understated playing leads one to dismiss it but.. The final track - Mr Blue - with just piano, guitar and voice and with several complicated changes of tempo is delicious.Anyhoo, I also have the expanded (3 extra tracks) version of the Warrior CD. In common with many other re-releases I find it too bright. Several exceptions to this are the wonderful 40th anniversary editions of Jethro Tull's "Aqualung" and King Crimson's "Islands" and a 2 cd re-release of John Martyn's "Solid Air" (Thick as a Brick was also a 40th anniversary release but something bad must have happened).
Thank you for the note on the filling. Obviously, it's something to experiment with. I might try some pink fibreglass batts (wearing a mask and gloves)(oopps there goes the F/A 18 war planes overhead. A flyover for the Melbourne F1grand-prix on this afternoon) and some synthetic stuffing.
For my money I will probably buy the inductors otherwise they will never be done.
Reggie
Thank you for actually looking at my diagram. Obviously it was more than I did. A note on credit cards. I got my first - and only - credit card in 1975 and spent my maximum in 1 day (by purchasing a motor-bike for $3000). When I got home I realised the type of person I could become and cut the card up. So now if I haven't got the cash it doesn't get bought. I could get a debit card but I use my wife's instead heh heh.
Michael Frank's "The Art of Tea" is a very beguiling record. It's humorous lyrics and understated playing leads one to dismiss it but.. The final track - Mr Blue - with just piano, guitar and voice and with several complicated changes of tempo is delicious.Anyhoo, I also have the expanded (3 extra tracks) version of the Warrior CD. In common with many other re-releases I find it too bright. Several exceptions to this are the wonderful 40th anniversary editions of Jethro Tull's "Aqualung" and King Crimson's "Islands" and a 2 cd re-release of John Martyn's "Solid Air" (Thick as a Brick was also a 40th anniversary release but something bad must have happened).
Thank you for the note on the filling. Obviously, it's something to experiment with. I might try some pink fibreglass batts (wearing a mask and gloves)(oopps there goes the F/A 18 war planes overhead. A flyover for the Melbourne F1grand-prix on this afternoon) and some synthetic stuffing.
For my money I will probably buy the inductors otherwise they will never be done.
Reggie
Responses to posts by Alan
Alan, I've missed some of the recent discussion and suggestions so I've included some comments below under the post #
#925 #931 & #935
DIFFRACTION & FELT ON FRONT SPEAKER PANEL
I think adding the felt might be a very good suggestion Alan. It may help to smooth out the sound a bit. I need to work out what the diagrams you have drawn mean and source some suitable material. It may be a while before I can get to do it but I can source the material in any case. The link to David Ralph's Speaker Pages is appreciated as well as
SAE & Sheet Felt
#937 RESISTORS IN BASS CIRCUIT
---> DennyG - you have effectively 74uF in the central cap's position.
Ideally the 6.2uF cap you connected there needs its own separate resistor.
Use 8.2 ohm or 10 ohm in 1 watt Carbon film from a local shop.
If you have some 1 watt resistors already , then so long as not less than 4.7 ohms or larger than 12 ohms , try those you have.
I think I have (measured values) 66+6 and 67+6 in the central position. What would be affected by adding the 8R-10R to the 6uF's? I assume the resistor should go straight to the common and not to the 1R.
#940
FILE STORAGE FOR DIAGRAMS
The exetel site (it's an ISP) I used to store some files for easy access is a private site for exetel customers. I decided to use it because all the free sites I tried in the past either: deleted the files after a short while; were bought out and closed down by the big boys to crunch the competition or just disappeared. Some ISP's offer free space for file storage so this would be worth checking out if you need to have a reliable place for this sort of thing.
SPEAKER TESTING
When I start testing again I might have to go back to the old system as I haven't had success getting the USB audio interface to do what I want to do.
Note that all the speaker tests were done using the same (left) channel of the amplifier to avoid the inevitable channel differences you mention.
For amp tests I don't have 6.8R but I have two 8 ohm 50W resistors, will these be ok to test the L/R channel freq response? Power level for these tests? Say 5W?
If I remove the crossovers I'll try what you suggest for each channel using dummy R's. Again 5W say?
#948
INTERNAL SPEAKER FOAM
From what I remember the foam is layered on the sides, rear, top and bottom but next time I open them up I'll check. When adding foam what types should be used and what should be avoided? I've various types of absorbent material lying around so what should I avoid?
INDUCTORS
The inductors might be a worthwhile upgrade but what frequencies would be affected? And what inductors are you talking about?
Alan, I've missed some of the recent discussion and suggestions so I've included some comments below under the post #
#925 #931 & #935
DIFFRACTION & FELT ON FRONT SPEAKER PANEL
I think adding the felt might be a very good suggestion Alan. It may help to smooth out the sound a bit. I need to work out what the diagrams you have drawn mean and source some suitable material. It may be a while before I can get to do it but I can source the material in any case. The link to David Ralph's Speaker Pages is appreciated as well as
SAE & Sheet Felt
#937 RESISTORS IN BASS CIRCUIT
---> DennyG - you have effectively 74uF in the central cap's position.
Ideally the 6.2uF cap you connected there needs its own separate resistor.
Use 8.2 ohm or 10 ohm in 1 watt Carbon film from a local shop.
If you have some 1 watt resistors already , then so long as not less than 4.7 ohms or larger than 12 ohms , try those you have.
I think I have (measured values) 66+6 and 67+6 in the central position. What would be affected by adding the 8R-10R to the 6uF's? I assume the resistor should go straight to the common and not to the 1R.
#940
FILE STORAGE FOR DIAGRAMS
The exetel site (it's an ISP) I used to store some files for easy access is a private site for exetel customers. I decided to use it because all the free sites I tried in the past either: deleted the files after a short while; were bought out and closed down by the big boys to crunch the competition or just disappeared. Some ISP's offer free space for file storage so this would be worth checking out if you need to have a reliable place for this sort of thing.
SPEAKER TESTING
When I start testing again I might have to go back to the old system as I haven't had success getting the USB audio interface to do what I want to do.
Note that all the speaker tests were done using the same (left) channel of the amplifier to avoid the inevitable channel differences you mention.
For amp tests I don't have 6.8R but I have two 8 ohm 50W resistors, will these be ok to test the L/R channel freq response? Power level for these tests? Say 5W?
If I remove the crossovers I'll try what you suggest for each channel using dummy R's. Again 5W say?
#948
INTERNAL SPEAKER FOAM
From what I remember the foam is layered on the sides, rear, top and bottom but next time I open them up I'll check. When adding foam what types should be used and what should be avoided? I've various types of absorbent material lying around so what should I avoid?
INDUCTORS
The inductors might be a worthwhile upgrade but what frequencies would be affected? And what inductors are you talking about?
a reply to #950 on Page 95
G'day DennyG ... and reggie , I'm presuming you are still there somewhere ...
Unfortunately it has taken me a while to get back to this , but let's proceed ...
The Link to the felt on baffle measurements was originally posted by reggie ,
thus thanks to him for that.
There will be sellers of Engineering and Technical felts in Australia for felts manufactured there ,
given all the wool that is shorn there !
#937 - Resistors in woofer circuit:
The small capacitor will pass the initial pulse of high frequencies faster than the large capacitor ,
and thus there will be an two resonances - each capacitor causes one with the inductance of the voice-coil of the woofer.
When a larger resistance is connected in Series with the smaller cap than with the larger cap ,
the pulse rise through the smaller cap will be slowed down to closer to the rise time through the larger cap.
Exactly how much resistance is required ?
I do not know , and to determine one would need to measure the pulse rise times through each cap ,
but we do not need to be that precise here , because part of the pulse is already slowed by the 3.5mH inductor ,
and more is slowed by the 2.2mH inductor , thus we need only to slow the initial one a little more than the following one.
One estimate way to calculate is to use the RC product , that is make Ra x Ca = Rb x Cb .
Thus , 1 ohm x 66uF = 11 ohms x 6uF ,
but before any Theory knowledgeable readers correct me here , I will state that that formula is for equal Time Constants ,
and there is no simple formula for equal Pulse Rise Times , however the Time Constants' formula is useful for an Estimate to get close.
Because the pulse are already slowed by the inductors the second resistor does not need to be as large as 11 ohms.
A resistor that is between about half the calculated value and the full value will suffice.
How much difference will be heard in this circuit I do not know ,
however because of the several significant resonances in the woofer above the crossover point -{ as shown in sba's CSD plots }-
we do not want to add to those by introducing another double resonance , but rather we want to reduce all the resonances ,
thus why I'm recommending adding the extra resistors.
You could listen immediately before installing another resistor , and again immediately after to hear if any better clarity in the midrange ,
because those woofers do reproduce a significant amount of midrange that needs to be adequately filtered out.
Yes , connect the additional resistor to the Common , and not to the top of the 1 ohm.
Note also :
Twelve 6uF caps each with its own 12 ohm Series resistor is equal to a single 72uF cap in Series with 1 ohm , if all caps and resistors are of the same types.
#940 - File Storage - take note reggie !
I see that some of sba's attachments are no longer available , likely as result of this ,
however sufficient of his that are directly useful to the 66 are still there , thus in a different storage , perhaps his own ..?
Speaker Testing :
Good , all done through the same channel of the amp.
Do this again for your two HF2000 tweeters for Frequency Response and Impedance ,
and 2 passes for each so as to double-check.
We need only Relative measurements , as that will show the differences between the two tweeters ,
so do not not worry too much about trying to get absolute true-to-life results ,
though as close as you can get is useful - see again my #890 on Page 89.
8 ohm/50 watt is fine for the Amplifier tests , and you can use the 8 ohm resistors as substitute loads in the crossover when the mid-domes are disconnected.
Use 4.7 ohm in either 5 watt or 10 watt to substitute for the tweeters and woofers in the crossover.
For all tests apply a signal level no larger than cause the resistors to get a little more than warm , but not too hot to touch.
With the 50 watters that will result in the amp being tested at about 20 watts per channel ,
presuming these 50 watters are heat-sinked , or don't require such ..?
Note:
some types of Power resistors will only dissipate in the upper wattage of their ratings when mounted on an external heat sink.
Such resistors will only dissipate about 10% of their rating when not heat-sinked.
All resistors will only dissipate their Maximum power rating if force cooled to their rated temperature -
- see the complete Data Sheets for any resistor type - there is a De-rating Plot to apply to the power for when the temperature of the resistor is higher ,
and when the Maximum temperature is reached the resistor will not dissipate any power ... usually it will have started burning before that temperature is reached.
Which model microphone did you buy with that "Cross-spectrum" system you bought Denny ?
The Behringer ECM8000 or Dayton EMM-6 or UMM-6 ?
Internal Foam :
These "various types of absorbent material" you have Denny , what is each ?
Some types of plastic foam do not absorb sound well , even some types that look like the Egg-Crate types of Acoustic Foam , but are only shock absorbing packaging foam ,
{ and beware , some unscrupulous sellers advertise these packaging foams as Acoustic Foam , because they are lower cost to buy , and look similar ,
however they feel different to touch/press if one has experience of genuine acoustic foam }.
If the bass response of your 66s is currently acceptable to you , then perhaps little or no better can be easily achieved.
I would experiment with filling inside a Ditton 44 , and various other sealed box loudspeakers ,
but not inside a well-balanced Passive Radiator/ABR loudspeaker , unless its interior foam has visibly deteriorated.
I introduced this topic about the fill in case there has been extra fill added by some-one to reggie's 66s.
Inductors:
I started an option for further modification of the 66 crossover in #865 on Page 87 , and more specifically about the inductors in #869.
The bandwidth affected would be:
the lowest bass frequencies from the ABR would be tighter , because the woofer would then be able to control the ABR faster ,
the mid and upper bass from the woofer would be noticeably louder and faster with less power from the amplifier ,
as also would be the lower midrange , though to slightly lesser degree.
Inductors to achieve the above will need to have no greater than half the DC resistance of the Celestion inductors , and preferably lower.
That requires thicker wire , thus will be larger size , thus will need a larger area board to mount on.
You can use the Celestion board for the midrange and tweeter sections of the crossover , and a new board for the woofer filter ,
or install all sections on a larger board.
Minimum gauge for the woofer inductors has to be 16AWG , though 15AWG and 14AWG will cause noticeably better results ,
as will 12AWG , etc ... though 12AWG and thicker are very large and quite expensive ... expensive to buy the wire ,
and more if you buy ready-made plus cost of shipping such heavy items.
Note: the lower the AWG number , the thicker the wire.
Celestion's Midrange and Tweeter inductors in the 66 are good , and do not need to be replaced.
I will do a separate post about the inductors when I compile a list of sizes and relative prices , etc ... and fairly soon I hope.
G'day DennyG ... and reggie , I'm presuming you are still there somewhere ...
Unfortunately it has taken me a while to get back to this , but let's proceed ...
The Link to the felt on baffle measurements was originally posted by reggie ,
thus thanks to him for that.
There will be sellers of Engineering and Technical felts in Australia for felts manufactured there ,
given all the wool that is shorn there !
#937 - Resistors in woofer circuit:
The small capacitor will pass the initial pulse of high frequencies faster than the large capacitor ,
and thus there will be an two resonances - each capacitor causes one with the inductance of the voice-coil of the woofer.
When a larger resistance is connected in Series with the smaller cap than with the larger cap ,
the pulse rise through the smaller cap will be slowed down to closer to the rise time through the larger cap.
Exactly how much resistance is required ?
I do not know , and to determine one would need to measure the pulse rise times through each cap ,
but we do not need to be that precise here , because part of the pulse is already slowed by the 3.5mH inductor ,
and more is slowed by the 2.2mH inductor , thus we need only to slow the initial one a little more than the following one.
One estimate way to calculate is to use the RC product , that is make Ra x Ca = Rb x Cb .
Thus , 1 ohm x 66uF = 11 ohms x 6uF ,
but before any Theory knowledgeable readers correct me here , I will state that that formula is for equal Time Constants ,
and there is no simple formula for equal Pulse Rise Times , however the Time Constants' formula is useful for an Estimate to get close.
Because the pulse are already slowed by the inductors the second resistor does not need to be as large as 11 ohms.
A resistor that is between about half the calculated value and the full value will suffice.
How much difference will be heard in this circuit I do not know ,
however because of the several significant resonances in the woofer above the crossover point -{ as shown in sba's CSD plots }-
we do not want to add to those by introducing another double resonance , but rather we want to reduce all the resonances ,
thus why I'm recommending adding the extra resistors.
You could listen immediately before installing another resistor , and again immediately after to hear if any better clarity in the midrange ,
because those woofers do reproduce a significant amount of midrange that needs to be adequately filtered out.
Yes , connect the additional resistor to the Common , and not to the top of the 1 ohm.
Note also :
Twelve 6uF caps each with its own 12 ohm Series resistor is equal to a single 72uF cap in Series with 1 ohm , if all caps and resistors are of the same types.
#940 - File Storage - take note reggie !
I see that some of sba's attachments are no longer available , likely as result of this ,
however sufficient of his that are directly useful to the 66 are still there , thus in a different storage , perhaps his own ..?
Speaker Testing :
Good , all done through the same channel of the amp.
Do this again for your two HF2000 tweeters for Frequency Response and Impedance ,
and 2 passes for each so as to double-check.
We need only Relative measurements , as that will show the differences between the two tweeters ,
so do not not worry too much about trying to get absolute true-to-life results ,
though as close as you can get is useful - see again my #890 on Page 89.
8 ohm/50 watt is fine for the Amplifier tests , and you can use the 8 ohm resistors as substitute loads in the crossover when the mid-domes are disconnected.
Use 4.7 ohm in either 5 watt or 10 watt to substitute for the tweeters and woofers in the crossover.
For all tests apply a signal level no larger than cause the resistors to get a little more than warm , but not too hot to touch.
With the 50 watters that will result in the amp being tested at about 20 watts per channel ,
presuming these 50 watters are heat-sinked , or don't require such ..?
Note:
some types of Power resistors will only dissipate in the upper wattage of their ratings when mounted on an external heat sink.
Such resistors will only dissipate about 10% of their rating when not heat-sinked.
All resistors will only dissipate their Maximum power rating if force cooled to their rated temperature -
- see the complete Data Sheets for any resistor type - there is a De-rating Plot to apply to the power for when the temperature of the resistor is higher ,
and when the Maximum temperature is reached the resistor will not dissipate any power ... usually it will have started burning before that temperature is reached.
Which model microphone did you buy with that "Cross-spectrum" system you bought Denny ?
The Behringer ECM8000 or Dayton EMM-6 or UMM-6 ?
Internal Foam :
These "various types of absorbent material" you have Denny , what is each ?
Some types of plastic foam do not absorb sound well , even some types that look like the Egg-Crate types of Acoustic Foam , but are only shock absorbing packaging foam ,
{ and beware , some unscrupulous sellers advertise these packaging foams as Acoustic Foam , because they are lower cost to buy , and look similar ,
however they feel different to touch/press if one has experience of genuine acoustic foam }.
If the bass response of your 66s is currently acceptable to you , then perhaps little or no better can be easily achieved.
I would experiment with filling inside a Ditton 44 , and various other sealed box loudspeakers ,
but not inside a well-balanced Passive Radiator/ABR loudspeaker , unless its interior foam has visibly deteriorated.
I introduced this topic about the fill in case there has been extra fill added by some-one to reggie's 66s.
Inductors:
I started an option for further modification of the 66 crossover in #865 on Page 87 , and more specifically about the inductors in #869.
The bandwidth affected would be:
the lowest bass frequencies from the ABR would be tighter , because the woofer would then be able to control the ABR faster ,
the mid and upper bass from the woofer would be noticeably louder and faster with less power from the amplifier ,
as also would be the lower midrange , though to slightly lesser degree.
Inductors to achieve the above will need to have no greater than half the DC resistance of the Celestion inductors , and preferably lower.
That requires thicker wire , thus will be larger size , thus will need a larger area board to mount on.
You can use the Celestion board for the midrange and tweeter sections of the crossover , and a new board for the woofer filter ,
or install all sections on a larger board.
Minimum gauge for the woofer inductors has to be 16AWG , though 15AWG and 14AWG will cause noticeably better results ,
as will 12AWG , etc ... though 12AWG and thicker are very large and quite expensive ... expensive to buy the wire ,
and more if you buy ready-made plus cost of shipping such heavy items.
Note: the lower the AWG number , the thicker the wire.
Celestion's Midrange and Tweeter inductors in the 66 are good , and do not need to be replaced.
I will do a separate post about the inductors when I compile a list of sizes and relative prices , etc ... and fairly soon I hope.
Last edited:
LTSpice simulation for midrange crossover
DennyG ,
as you have LTSpiceIV , etc ... there is something useful you can simulate with it for the 66.
I have done some calculations from data in the Impedance and Phase plots posted by sba and Wayne Swan.
We can look at the upper crossover filter section applied to the mid-domes.
The results will be accurate ONLY at 5kHz - the nominal crossover frequency - but close to accurate at little on both sides of 5kHz ,
however for about 1 octave on either side we will also be able to see the general trend of the slope well enough for that to be useful also.
To do this:
(1)- we need a Reference Plot.
This can be a horizontal straight line near the top of the scale.
It will represent the output of the mid-dome with no filter attached.
If you have to enter any component values to get that , then enter as follows:
R = 8.97 ohm + L = 0.179mH for the voice-coil impedance , that is the total Load at 5kHz which we need to measure the output across.
If you have to reduce the number of significant digits , then use 9.0 ohm + 0.18mH.
Both R and L are needed for the speaker load.
(2)- next do a plot with the filter attached as follows:
C in = 24uF + 1.8 ohm ; L in = 2.2mH + 1.3 ohm ; L out = 0.34mH + 0.6 ohm ; C out = 3.9uF + 3.9 ohm.
Plot the output across the the Load R + L as in (1) above.
Plot from 2.5kHz to 10kHz at least , though you can start from 1kHz or 2kHz if you like.
Use a Vertical scale in dB so that we can read the exact value at 5kHz.
This plot should all be a little below the Reference plot straight line when both are combined onto one double graph ,
but it may go a little above that line around 2.5kHz , because the filter is resonant and thus it amplifies the output.
The purpose of the 1.8 ohm input resistor is to reduce the amplification , however that value is an estimate ...
the actual resistance needed to reduce the amplification may be different , and you will not be able to find it with this data which is accurate at only 5kHz ,
but do not worry about that - we will get sufficient result to be useful.
(3)- next do a plot with 2 different component values :
change L out to 0.42mH + 0.6 ohm , and C out to 4.7uF + 3.9 ohm.
Leave all other values the same as for (2).
Post a combined graph with both (2) and (3) on it so that we can see the difference in dB at 5kHz.
Ideally this combination plot should have the Reference Plot (1) in it also ,
but if you can only combine two plots , then do a (1) + (2) and a (2) + (3).
Yes , I have changed two components for (3) , and Scientific Method is to change only one thing , but this is to make it less time-consuming for you.
If you want to you can plot (3) with the 0.42mH change only , using the original 3.9uF + 3.9 ohm ,
and then do a plot (4) with 0.42mH and 4.7uF + 3.9 ohm , however that is not necessary because I can interpolate and guesstimate the interim step.
If you want to you can do other plots with changes to the 3.9 ohm resistor ,
reducing it to successively 3.3 , 2.7 , 2.2 , 1.8 , all using the same output cap of 3.9uF ,
but this may not show anything significant that is useful within the limits of what an LTSpice can validly present. I do not know.
You would have to do at least three plots to see if any difference ,
because any two resistor values could by co-incidence produce almost exactly the same result if they happen to be either side of the optimum resistance.
All the above resistances will likely cause different dB at 5kHz , but it is in the region between 5kHz to 10kHz that the optimum resistance needs to be found for.
I will explain the results after I see them.
The above will not show anything useful at the lower crossover point of 500 Hz , because the Load impedance there is very different to that at 5kHz ,
and as sba's Impedance plots have shown a lot of sample variation around 500 Hz , any one I calculated for would be a bit different to all the others ,
however if you want to see a typical example of what is happening at 500 Hz with the filter attached ,
then I will calculate a Load combination for an Average sample of the mid-dome ,
so request here if you would like to see such.
DennyG ,
as you have LTSpiceIV , etc ... there is something useful you can simulate with it for the 66.
I have done some calculations from data in the Impedance and Phase plots posted by sba and Wayne Swan.
We can look at the upper crossover filter section applied to the mid-domes.
The results will be accurate ONLY at 5kHz - the nominal crossover frequency - but close to accurate at little on both sides of 5kHz ,
however for about 1 octave on either side we will also be able to see the general trend of the slope well enough for that to be useful also.
To do this:
(1)- we need a Reference Plot.
This can be a horizontal straight line near the top of the scale.
It will represent the output of the mid-dome with no filter attached.
If you have to enter any component values to get that , then enter as follows:
R = 8.97 ohm + L = 0.179mH for the voice-coil impedance , that is the total Load at 5kHz which we need to measure the output across.
If you have to reduce the number of significant digits , then use 9.0 ohm + 0.18mH.
Both R and L are needed for the speaker load.
(2)- next do a plot with the filter attached as follows:
C in = 24uF + 1.8 ohm ; L in = 2.2mH + 1.3 ohm ; L out = 0.34mH + 0.6 ohm ; C out = 3.9uF + 3.9 ohm.
Plot the output across the the Load R + L as in (1) above.
Plot from 2.5kHz to 10kHz at least , though you can start from 1kHz or 2kHz if you like.
Use a Vertical scale in dB so that we can read the exact value at 5kHz.
This plot should all be a little below the Reference plot straight line when both are combined onto one double graph ,
but it may go a little above that line around 2.5kHz , because the filter is resonant and thus it amplifies the output.
The purpose of the 1.8 ohm input resistor is to reduce the amplification , however that value is an estimate ...
the actual resistance needed to reduce the amplification may be different , and you will not be able to find it with this data which is accurate at only 5kHz ,
but do not worry about that - we will get sufficient result to be useful.
(3)- next do a plot with 2 different component values :
change L out to 0.42mH + 0.6 ohm , and C out to 4.7uF + 3.9 ohm.
Leave all other values the same as for (2).
Post a combined graph with both (2) and (3) on it so that we can see the difference in dB at 5kHz.
Ideally this combination plot should have the Reference Plot (1) in it also ,
but if you can only combine two plots , then do a (1) + (2) and a (2) + (3).
Yes , I have changed two components for (3) , and Scientific Method is to change only one thing , but this is to make it less time-consuming for you.
If you want to you can plot (3) with the 0.42mH change only , using the original 3.9uF + 3.9 ohm ,
and then do a plot (4) with 0.42mH and 4.7uF + 3.9 ohm , however that is not necessary because I can interpolate and guesstimate the interim step.
If you want to you can do other plots with changes to the 3.9 ohm resistor ,
reducing it to successively 3.3 , 2.7 , 2.2 , 1.8 , all using the same output cap of 3.9uF ,
but this may not show anything significant that is useful within the limits of what an LTSpice can validly present. I do not know.
You would have to do at least three plots to see if any difference ,
because any two resistor values could by co-incidence produce almost exactly the same result if they happen to be either side of the optimum resistance.
All the above resistances will likely cause different dB at 5kHz , but it is in the region between 5kHz to 10kHz that the optimum resistance needs to be found for.
I will explain the results after I see them.
The above will not show anything useful at the lower crossover point of 500 Hz , because the Load impedance there is very different to that at 5kHz ,
and as sba's Impedance plots have shown a lot of sample variation around 500 Hz , any one I calculated for would be a bit different to all the others ,
however if you want to see a typical example of what is happening at 500 Hz with the filter attached ,
then I will calculate a Load combination for an Average sample of the mid-dome ,
so request here if you would like to see such.
Last edited:
I was just browsing Canuck Audio Mart and found the following. I don't know the fellow who is selling them, but wow! 3 Pairs of Ditton 66s for sale in British Columbia?
Celestion Ditton - Canuck Audio Mart
Maybe someone on here might be interested, although they aren't cheap.
Celestion Ditton - Canuck Audio Mart
Maybe someone on here might be interested, although they aren't cheap.
Hello Alan
I'm about to go on an extended trip so haven't time to go through your last posts in detail. Also haven't been able to get back to any of this speaker stuff for some time. Hope to get back to it in a couple of months.
Felt
I did fine some local felt suppliers that look promising:
Felt Sales • Engineering
Engineering Felt/Industrial Felt | JJ Davies
Australian Felt Specialists Pty Ltd
Microphone
The calibrated mic. I purchased was the EMM-6 Premium+
Cross·Spectrum - Calibrated Dayton Audio EMM-6 Microphones for Sale
Foam
As you say Alan, the 66's sound fine with the amount of foam thay have. The condition of the old foam looks ok to me. I have dismantled a couple of mattress pads (basically unused, from Ikea) so I can reuse the materials in some furniture restoration and they have foam (flat), foam (egg shell) latex (egg shell), polyester (blanket), wool (blanket). I thought about using some of this stuff when speaker testing to dampen reflections.
Bass Inductor
The changes you predict to the frequency response with the lower impedance inductor in-circuit may be helpful to the overall sound and probably reinforce the region that was vastly improved when the ESR resistors were added. This may help with some of the (badly?) recorded/mixed material that can sound a bit 'wooden' on these speakers. The 66's, being very open sounding, will sound superb on some material but pretty ordinary on this other stuff.
Spice simulation
If someone else can't do as you suggest then I'll tackle this when I get back next month!
I'm about to go on an extended trip so haven't time to go through your last posts in detail. Also haven't been able to get back to any of this speaker stuff for some time. Hope to get back to it in a couple of months.
Felt
I did fine some local felt suppliers that look promising:
Felt Sales • Engineering
Engineering Felt/Industrial Felt | JJ Davies
Australian Felt Specialists Pty Ltd
Microphone
The calibrated mic. I purchased was the EMM-6 Premium+
Cross·Spectrum - Calibrated Dayton Audio EMM-6 Microphones for Sale
Foam
As you say Alan, the 66's sound fine with the amount of foam thay have. The condition of the old foam looks ok to me. I have dismantled a couple of mattress pads (basically unused, from Ikea) so I can reuse the materials in some furniture restoration and they have foam (flat), foam (egg shell) latex (egg shell), polyester (blanket), wool (blanket). I thought about using some of this stuff when speaker testing to dampen reflections.
Bass Inductor
The changes you predict to the frequency response with the lower impedance inductor in-circuit may be helpful to the overall sound and probably reinforce the region that was vastly improved when the ESR resistors were added. This may help with some of the (badly?) recorded/mixed material that can sound a bit 'wooden' on these speakers. The 66's, being very open sounding, will sound superb on some material but pretty ordinary on this other stuff.
Spice simulation
If someone else can't do as you suggest then I'll tackle this when I get back next month!
A reply to #954 , + other comments about Celestion 66
Hello DennyG , and who-ever else is interested in what he posted above ,
FELT
The first and second Suppliers you posted Links to seem to be selling felts from the same German manufacturer ,
thus interested Australians can buy from the one that has outlets in the State they live in.
The third seller seems to be one who sells large quantity to trade users , thus may not sell a single roll or package of self-adhesive pieces to hobbyists.
Of the first and second , their :-
Engineering Felt Grade A and Fine White Felt are both too dense to absorb sufficient sound on baffle or inside a speaker cabinet ,
though can be put to other sound related purposes not applicable here.
Engineering Felt Grade B and Saddle Felt are both more suitable for the sound absorption we are trying to achieve.
Specialty Felt --> Medical Felt seems to be the most suitable , as is the least dense , thus will absorb upper midrange and treble well ,
however it is not listed whether that is Pure Wool or a Composite , but such may not matter.
Only problem is that it is pure White colour , thus if not visually acceptable to look at you will have to dye it - carefully , see my earlier post about that.
Best option from these sellers for the baffle seems to be the Saddle Felt ,
as I see it is available in Brown and Black colours , so choose from the photos.
The Medical Felt could be used to replace the plastic foam inside the cabinet if the foam has deteriorated ,
but at likely too high in price compared to likely Australian grown and made wool felt batts from another local seller.
MICROPHONE
From the frequency response plots , the most useful way to use the Dayton EMM-6 mic for Accuracy is to use it at 90 degrees angle ,
that is point it vertically upwards , and have its top edge directly in line with the centre of the driver that is to be measured.
I recommend very slightly lowering it , or very slightly tilting it towards the driver , so that the angle is about 85 degrees ,
as this will likely give a more even response in some parts of the spectrum , even though that does not show on the Dayton plot.
Note , the above gives a result with respect to a flat response only to 10kHz , however that is sufficient for measuring the mid-domes and woofers ,
and sufficient for measuring tweeters to an Octave or more higher in frequency than the crossover points usually used.
The mic could then be used at 0 degrees or 45 degrees to make RELATIVE plots between two or more tweeters
to evaluate any amount of DIFFERENCE between samples to about 18kHz , and that is sufficient.
FOAM
The foam and latex you have taken from the mattress will not likely be much good for sound absorption ,
however the two "egg shell" types can be used at a distance in the room to assist with reflecting sound away from going directly back to the microphone.
For larger reflecting , roll up the Flat foam into about 6 inch or larger diameter , and place it so as to reflect sound away from its source.
The Wool blanket will be the best to place on floor and ceiling below and above the mic and loudspeaker ,
and with the polyester blanket behind the wool if not sufficient wool to fold for multi-thickness.
Fold loosely for on floor.
Hang about 6 inches to 1 foot from ceiling , and curved hanging will be better than pulled tight flat.
Thank you for posting the above , because some crossover reconditioners do not know this from other sources ,
and may not believe what I have been posting about the use of Resistors.
One has to hear it to really understand why I am recommending this.
Most of the experienced designers who work for the Brand Name loudspeaker manufacturers will not tell DIYers these things ,
and some do not know it them selves , thus why unpleasant sounding loudspeakers that use otherwise good quality modern components.
Most Parts Sellers in the Industry do not know these things.
The above includes reasons why the Ditton 66 was so popular with some types of listeners when it was new ,
and similarly for many other classic designs then , albeit different designs were preferred by different listeners.
Hello TechnoDweeb , if you are reading the above !
SPICE SIMULATION
There is no hurry for this.
It is for use by DIYers following this thread , and who will have had to be patient to get most of the information to date.
If something is useful , it is worth waiting for , as we are not in Emergency circumstances here , I hope ...
When you get time , and get a useful measurements' procedure worked out Denny ,
there will be other uses you can put the Spice to , such as designing a crossover for your Bespoke loudspeakers with those other drivers.
I will be back with a post about alternate Inductors , so that reggie can get started ,
and whoever else may be interested.
Hello DennyG , and who-ever else is interested in what he posted above ,
FELT
The first and second Suppliers you posted Links to seem to be selling felts from the same German manufacturer ,
thus interested Australians can buy from the one that has outlets in the State they live in.
The third seller seems to be one who sells large quantity to trade users , thus may not sell a single roll or package of self-adhesive pieces to hobbyists.
Of the first and second , their :-
Engineering Felt Grade A and Fine White Felt are both too dense to absorb sufficient sound on baffle or inside a speaker cabinet ,
though can be put to other sound related purposes not applicable here.
Engineering Felt Grade B and Saddle Felt are both more suitable for the sound absorption we are trying to achieve.
Specialty Felt --> Medical Felt seems to be the most suitable , as is the least dense , thus will absorb upper midrange and treble well ,
however it is not listed whether that is Pure Wool or a Composite , but such may not matter.
Only problem is that it is pure White colour , thus if not visually acceptable to look at you will have to dye it - carefully , see my earlier post about that.
Best option from these sellers for the baffle seems to be the Saddle Felt ,
as I see it is available in Brown and Black colours , so choose from the photos.
The Medical Felt could be used to replace the plastic foam inside the cabinet if the foam has deteriorated ,
but at likely too high in price compared to likely Australian grown and made wool felt batts from another local seller.
MICROPHONE
From the frequency response plots , the most useful way to use the Dayton EMM-6 mic for Accuracy is to use it at 90 degrees angle ,
that is point it vertically upwards , and have its top edge directly in line with the centre of the driver that is to be measured.
I recommend very slightly lowering it , or very slightly tilting it towards the driver , so that the angle is about 85 degrees ,
as this will likely give a more even response in some parts of the spectrum , even though that does not show on the Dayton plot.
Note , the above gives a result with respect to a flat response only to 10kHz , however that is sufficient for measuring the mid-domes and woofers ,
and sufficient for measuring tweeters to an Octave or more higher in frequency than the crossover points usually used.
The mic could then be used at 0 degrees or 45 degrees to make RELATIVE plots between two or more tweeters
to evaluate any amount of DIFFERENCE between samples to about 18kHz , and that is sufficient.
FOAM
The foam and latex you have taken from the mattress will not likely be much good for sound absorption ,
however the two "egg shell" types can be used at a distance in the room to assist with reflecting sound away from going directly back to the microphone.
For larger reflecting , roll up the Flat foam into about 6 inch or larger diameter , and place it so as to reflect sound away from its source.
The Wool blanket will be the best to place on floor and ceiling below and above the mic and loudspeaker ,
and with the polyester blanket behind the wool if not sufficient wool to fold for multi-thickness.
Fold loosely for on floor.
Hang about 6 inches to 1 foot from ceiling , and curved hanging will be better than pulled tight flat.
Thank you for posting the above , because some crossover reconditioners do not know this from other sources ,
and may not believe what I have been posting about the use of Resistors.
One has to hear it to really understand why I am recommending this.
Most of the experienced designers who work for the Brand Name loudspeaker manufacturers will not tell DIYers these things ,
and some do not know it them selves , thus why unpleasant sounding loudspeakers that use otherwise good quality modern components.
Most Parts Sellers in the Industry do not know these things.
The above includes reasons why the Ditton 66 was so popular with some types of listeners when it was new ,
and similarly for many other classic designs then , albeit different designs were preferred by different listeners.
Hello TechnoDweeb , if you are reading the above !
SPICE SIMULATION
There is no hurry for this.
It is for use by DIYers following this thread , and who will have had to be patient to get most of the information to date.
If something is useful , it is worth waiting for , as we are not in Emergency circumstances here , I hope ...
When you get time , and get a useful measurements' procedure worked out Denny ,
there will be other uses you can put the Spice to , such as designing a crossover for your Bespoke loudspeakers with those other drivers.
I will be back with a post about alternate Inductors , so that reggie can get started ,
and whoever else may be interested.
Last edited:
Schematics , and Photos of Layout of Crossover
Peter ,
almost everything you need to know is in this thread.
Yes , the thread is very long , and much of it is not related to your specific questions ,
however if you have not read all of it , then look at these posts:
Page 85 , #845 , has the Schematic of the crossover for a 66 with the original Celestion tweeter.
You can see there where the recommended resistors are connected to the capacitors.
{The capacitors in that schematic show the summed total of the original Parallel connected capacitances for the tweeter in the Mk2 66.}
Above that in #844 is a Schematic that shows two Parallel connected pairs of capacitors in the bass section.
These sum to the total required 72uF in each position.
Note how a single resistor is connected to each pair of caps.
{ This schematic is also for a different tweeter , so ignore the tweeter filter section there if you still have the original Celestion tweeters.}
Page 66 , #659 , has a Schematic that shows two alternate capacitors that can be used in the bass section ,
for if you want to use a single cap in each position only instead of Parallel pairs of caps to sum to 72uF.
It is not necessary to use exactly 72uF , but when using close to 72 values it is recommended to put the larger and smaller in the particular circuit positions shown in the diagram.
If their positions are reversed the filter will not work as well.
This single caps' option has been used by at least 2 participants of this thread , and both reported good audible results.
There is no exact 72uF cap available now , and nearest available is one brand's 70uF cap which is a very large physical size ,
so I have recommended the moderate sized 68uF and 75uF caps as easier to fit onto the board.
Page 43 , #429 , has 2 photos which show a re-conditioned crossover that has those 68uF and 75uF in the bass section - the 2 Black capacitors.
The Blue capacitors shown there are in the midrange and tweeter filters.
See particually in the 2nd photo how the resistors are connected.
{ Those are Mills MRA-5 , miniature 5 watt wire-wound resistors.}
If you want to know why resistors are needed with the new capacitors ,
then go to Page 21 and read my corrections' Post #206.
There is more on that subject in a few of my posts later in the thread.
You will see in the Schematics above that a different value of resistance has been used in two circuit locations in one schematic than in another.
Those are currently subject to user experiment , however I will recommend specific values for you to start with.
There is NO single supplier in the UK that stocks all the components in the optimum types I have been recommending.
If you will only buy from one supplier then you will have to compromise
and accept either lesser or better caps than you actually need for the tweeter and midrange ,
and physically very large 70uF caps for the woofer , and lesser quality resistors , and which are also physically larger than the Mills MRA-5 resistors.
Mills MRA-5 resistors and better than you need caps for tweeter and midrange can be bought from another UK seller ,
but not caps for the woofer , { unless you will accept physically very large , higher priced caps that will have to be connected as Parallel pairs }.
The nearest seller where all the caps can be bought , in moderate physical sizes and sufficiently good audio quality , is in Germany.
Postage to UK is 18 euro , though they might reduce that if that package weighs less than 1 Kilo { 2.2lb } ,
but even if they will not , the total price including 18 euro Postage is still less than you can buy in UK as the good audio quality caps in UK cost a lot more.
The German seller does not have Mills MRA-5 resistors , however they do have a sufficiently good quality alternate , which is physically a bit larger but not huge size.
They have a web-page Order form , and accept payment via PayPal.
I bought a large number of capacitors from them - everything proceeded well and all arrived.
Think about all the above , then post here your decision and I will continue about whichever Sellers.
In your "newbie" post in the other thread you mention a Celestion 66 thread that has 51 pages.
Please post here the exact Title , and/or user-name of who started it , of that thread so readers here can find it.
Peter ,
almost everything you need to know is in this thread.
Yes , the thread is very long , and much of it is not related to your specific questions ,
however if you have not read all of it , then look at these posts:
Page 85 , #845 , has the Schematic of the crossover for a 66 with the original Celestion tweeter.
You can see there where the recommended resistors are connected to the capacitors.
{The capacitors in that schematic show the summed total of the original Parallel connected capacitances for the tweeter in the Mk2 66.}
Above that in #844 is a Schematic that shows two Parallel connected pairs of capacitors in the bass section.
These sum to the total required 72uF in each position.
Note how a single resistor is connected to each pair of caps.
{ This schematic is also for a different tweeter , so ignore the tweeter filter section there if you still have the original Celestion tweeters.}
Page 66 , #659 , has a Schematic that shows two alternate capacitors that can be used in the bass section ,
for if you want to use a single cap in each position only instead of Parallel pairs of caps to sum to 72uF.
It is not necessary to use exactly 72uF , but when using close to 72 values it is recommended to put the larger and smaller in the particular circuit positions shown in the diagram.
If their positions are reversed the filter will not work as well.
This single caps' option has been used by at least 2 participants of this thread , and both reported good audible results.
There is no exact 72uF cap available now , and nearest available is one brand's 70uF cap which is a very large physical size ,
so I have recommended the moderate sized 68uF and 75uF caps as easier to fit onto the board.
Page 43 , #429 , has 2 photos which show a re-conditioned crossover that has those 68uF and 75uF in the bass section - the 2 Black capacitors.
The Blue capacitors shown there are in the midrange and tweeter filters.
See particually in the 2nd photo how the resistors are connected.
{ Those are Mills MRA-5 , miniature 5 watt wire-wound resistors.}
If you want to know why resistors are needed with the new capacitors ,
then go to Page 21 and read my corrections' Post #206.
There is more on that subject in a few of my posts later in the thread.
You will see in the Schematics above that a different value of resistance has been used in two circuit locations in one schematic than in another.
Those are currently subject to user experiment , however I will recommend specific values for you to start with.
There is NO single supplier in the UK that stocks all the components in the optimum types I have been recommending.
If you will only buy from one supplier then you will have to compromise
and accept either lesser or better caps than you actually need for the tweeter and midrange ,
and physically very large 70uF caps for the woofer , and lesser quality resistors , and which are also physically larger than the Mills MRA-5 resistors.
Mills MRA-5 resistors and better than you need caps for tweeter and midrange can be bought from another UK seller ,
but not caps for the woofer , { unless you will accept physically very large , higher priced caps that will have to be connected as Parallel pairs }.
The nearest seller where all the caps can be bought , in moderate physical sizes and sufficiently good audio quality , is in Germany.
Postage to UK is 18 euro , though they might reduce that if that package weighs less than 1 Kilo { 2.2lb } ,
but even if they will not , the total price including 18 euro Postage is still less than you can buy in UK as the good audio quality caps in UK cost a lot more.
The German seller does not have Mills MRA-5 resistors , however they do have a sufficiently good quality alternate , which is physically a bit larger but not huge size.
They have a web-page Order form , and accept payment via PayPal.
I bought a large number of capacitors from them - everything proceeded well and all arrived.
Think about all the above , then post here your decision and I will continue about whichever Sellers.
In your "newbie" post in the other thread you mention a Celestion 66 thread that has 51 pages.
Please post here the exact Title , and/or user-name of who started it , of that thread so readers here can find it.
Last edited:
Celestion Ditton 66
Regards
Pete
Alan, thanks for your reply. I have printed it off for my friend to look at. He is the electronics expert, not me, i'm sorry to say, just a humble mechanic. My friend does any work of this kind. When I saw the post with all that info, I was hoping to show him, so he could upgrade my Ditton 66's. I did not think he would want to wade through all those pages, and thought maybe I could get a rough idea of what I needed. (My foolishness) In my newbie post I did mention 50 pages, but it was this site, I was on about, somehow missed the rest of post as I see now there is a lot more than 50. I will be looking to upgrade my Dittons soon, so will let you know what my friend advises me to do and what to buy, then hopefully you may be able to supply me with the details of the suppilers for me to purchase. Thanks again for you great reply, very useful indeed.
Regards
Pete
able to measure the capacitance of the Capacitors ?
'ullo Pete ,
I forgot to include in my last post , so will now here , albeit now relevant to your electronics' minded friend.
Ask him , does he have a Capacitance Meter or a Capacitance measurement function in his Mulitmeter ?
And , if yes , then ask him what is the largest amount of Capacitance the meter will measure ?
This may be relevant if you decide to use 2 or 3 intermediate capacitances in Parallel to sum to the 72uF required for the 2 positions in each woofer filter ,
and particually so if you decide to buy from the all parts from one seller in Germany , given the two reasonable options from there.
He may suggest various ways of combining capacitors in Parallel to sum to 72uF , including without need to measure anything ,
however as you originally asked for the "best" way , then that is what I will advise ,
and of the other ways - those I have posted about earlier in this thread ,
and also in parts of the thread titled:
Crossover nightmare!!!!!!!
started by:
lorienblack
which is about the Celestion Ditton 44 , a loudspeaker that has the same woofer and tweeter filter sections as the 66 ,
and a very similar midrange filter in its crossover.
The upgrade principles are the same for both the 44 and 66 crossovers.
I described the case of Parallel capacitors a little differently there , as there were different questions and different responses to my posts.
About your self-description as "a humble mechanic" , there is nothing detrimental about being "humble" ,
and an "electronics expert" would do well to be likewise ,
{ and of course I acknowledge you may have been making a joke there }.
Can you solder well , or will your friend be doing that for you ?
Reply to both the above matters in the same post when you reply about your/his buying decision ,
or when you ask more questions if he suggests to you some alternates.
'ullo Pete ,
I forgot to include in my last post , so will now here , albeit now relevant to your electronics' minded friend.
Ask him , does he have a Capacitance Meter or a Capacitance measurement function in his Mulitmeter ?
And , if yes , then ask him what is the largest amount of Capacitance the meter will measure ?
This may be relevant if you decide to use 2 or 3 intermediate capacitances in Parallel to sum to the 72uF required for the 2 positions in each woofer filter ,
and particually so if you decide to buy from the all parts from one seller in Germany , given the two reasonable options from there.
He may suggest various ways of combining capacitors in Parallel to sum to 72uF , including without need to measure anything ,
however as you originally asked for the "best" way , then that is what I will advise ,
and of the other ways - those I have posted about earlier in this thread ,
and also in parts of the thread titled:
Crossover nightmare!!!!!!!
started by:
lorienblack
which is about the Celestion Ditton 44 , a loudspeaker that has the same woofer and tweeter filter sections as the 66 ,
and a very similar midrange filter in its crossover.
The upgrade principles are the same for both the 44 and 66 crossovers.
I described the case of Parallel capacitors a little differently there , as there were different questions and different responses to my posts.
About your self-description as "a humble mechanic" , there is nothing detrimental about being "humble" ,
and an "electronics expert" would do well to be likewise ,
{ and of course I acknowledge you may have been making a joke there }.
Can you solder well , or will your friend be doing that for you ?
Reply to both the above matters in the same post when you reply about your/his buying decision ,
or when you ask more questions if he suggests to you some alternates.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.