Tweeter Polarity
Having listened to various artists and recordings with my tweeters hooked up normal polarity, I'd have to say I prefer the tweeters hooked up reverse polarity, as they were originally. On some recordings I did not notice much of a difference, while others there was a significant change to the sound being produced. As best as I can describe the change, I felt that the mids had been moved forward and the highs had been pushed back, creating a hard and unbalanced feel to the music. There were a couple recordings where I also felt there was something missing, as if a void in the frequency at which the mid and the tweeter cross over. Overall, I like the sound being reproduced with the tweeters hooked in reverse, it has a warm, natural and balanced feel to it and with some artists just begs to turned up, but never sounds loud.
I am very pleased with the result achieved after changing over to the poly caps. Clear and detailed are words that come to mind and I'm hearing instruments I haven't heard before. But I am still a little unsatisfied with the tweeter, it just does not have the sparkle and clarity that I know other tweeters have and that I remember them to have had. So I have decided to replace it with a modern tweeter and see what kind of result I can achieve.
===============
Alan, hopefully you can help me out with this...
I had previously thought I'd replace my tweeters with the Seas HO737 19TFF1, but having talked with a buddy in the business, he can sell me some Hiquphons at a pretty good price, so I am going to bite the bullet and go with these. My question is which Hiquphon, the OWI or OWII? I know in previous posts you have recommended the OWI, because the sensitivity better matches that of the HF2000. What I wonder about though is, would it not be better to go with the OWII with higher sensitivity to compensate for the higher impedance of the Hiquphon and adjust the sound level with some L pad resistors? Also, it would be my understanding that the Hiquphon will also need the 6.2 caps adjusted, same as the Seas, to compensate for the higher impedance. I have read that the Hiquphon OWI can be used as a straight replacement for the HF2000, but I cannot see how this could be true??
Still learning and tweaking...
Dave
Having listened to various artists and recordings with my tweeters hooked up normal polarity, I'd have to say I prefer the tweeters hooked up reverse polarity, as they were originally. On some recordings I did not notice much of a difference, while others there was a significant change to the sound being produced. As best as I can describe the change, I felt that the mids had been moved forward and the highs had been pushed back, creating a hard and unbalanced feel to the music. There were a couple recordings where I also felt there was something missing, as if a void in the frequency at which the mid and the tweeter cross over. Overall, I like the sound being reproduced with the tweeters hooked in reverse, it has a warm, natural and balanced feel to it and with some artists just begs to turned up, but never sounds loud.
I am very pleased with the result achieved after changing over to the poly caps. Clear and detailed are words that come to mind and I'm hearing instruments I haven't heard before. But I am still a little unsatisfied with the tweeter, it just does not have the sparkle and clarity that I know other tweeters have and that I remember them to have had. So I have decided to replace it with a modern tweeter and see what kind of result I can achieve.
===============
Alan, hopefully you can help me out with this...
I had previously thought I'd replace my tweeters with the Seas HO737 19TFF1, but having talked with a buddy in the business, he can sell me some Hiquphons at a pretty good price, so I am going to bite the bullet and go with these. My question is which Hiquphon, the OWI or OWII? I know in previous posts you have recommended the OWI, because the sensitivity better matches that of the HF2000. What I wonder about though is, would it not be better to go with the OWII with higher sensitivity to compensate for the higher impedance of the Hiquphon and adjust the sound level with some L pad resistors? Also, it would be my understanding that the Hiquphon will also need the 6.2 caps adjusted, same as the Seas, to compensate for the higher impedance. I have read that the Hiquphon OWI can be used as a straight replacement for the HF2000, but I cannot see how this could be true??
Still learning and tweaking...
Dave
Last edited:
On Polarity of tweeters ; on Hiquphon tweeters
OK Doug,
post what you hear when you have listened sufficiently to be happy,
and if you don't hear what Dave did, do not be surprised,
because hearing:- the ear<--->brain connection is different for different people.
Dave,
I'm not surprised by your report about audible effects that you heard,
BUT, I'm going to wait till Doug reports what he hears before I comment further.
***********************************************************
Hiquphon tweeters:-
everything has a Trade-off.
Higher sensitivity means less of something else.
Higher sensitivity is not needed for the 66.
Both versions of the tweeter are going to need resistors or an L-pad to match them to the Celestion crossover.
Using larger resistance values to match OWII sensitivity to the Celestion MD500 mids means you will be hearing more of the audio characteristics of the resistors and less of the audio characteristic of the tweeter.
That does not make sense to me, but you decide for you ...
The only reason I would choose a high sensitivity tweeter and pad it down heavily with resistors is if I needed to use the tweeter with a high power capable mid or mid-bass driver to play at very high volume levels -
{higher than MD500 will safely play without damage to it}.
My experience is that heavily padded dome tweeters only sound OK when driven hard,
and sound a bit anemic when played quietly ... but there could be exceptions ...
*********************************************************
WARNING :- before anyone buys a different tweeter to HF2000,
{except perhaps for Coles CE 4001},
I cannot guarantee the changes to the treble filter of the crossover that I will post will give the best possible performance the other tweeter is capable of.
I can only state that it will give a different audible result to HF2000,
and that this sound will not be worse,
AND, if you are prepared to do some fine-tuning by later substituting some different value resistors and/or capacitors, you may be able to get a better audible result.
That is DIY - we experiment ! -{and we measure when we can}.
I will estimate from the tweeter data the resistors and capacitors that I think will work well,
but to be sure, technical measurements would have to be made on the tweeter and on the MD500.
ALSO, my estimations will include using the existing Inductor that is in the 66 crossover - to keep costs low for you,
BUT, to get the best possible performance of a different tweeter to work with the MD500, a different value inductor may be required.
I can estimate, if necessary, a possible inductor, but same warning applies as for caps and resistors - you may have to try other close values.
I state this before anyone buys an EXPENSIVE tweeter.
Safer would be to spend that money on lowish price SEAS H0737 and to pay someone to measure your MD/MF 500s so that a new, optimum filter can be designed to match both.
Safest seems to be to use a Coles CE4001, because that seems to be close to HF2000 except for Impedance.
CE4001 with a suitable parallel connected resistor will then likely work the same in the 66 crossover as HF2000.
CE4001 new is low price, and costs a lot less than ebay sellers are asking for old HF2000s - which may be partially damaged.
It is not likely to be a better or worse sounding tweeter than a perfect condition HF2000,
but it will sound better than a damaged HF2000 !
SEAS H0737 will be audibly better than CE4001, but you will have to be prepared to do some experimenting with crossover components to get the best from it if you are not happy with the audible results of my estimated component values ... and this is what I do myself ...
{I measure what I can, then I calculate what I can, then I estimate from other data whatever else I need to know, then I install and listen,
and if I think I can achieve better I then substitute different value components.}
All 3 filters in the 66 crossover interact, because they are connected in electrical Parallel with each other,
AND in parallel with the Impedances of the 3 active drivers -{woofer; mid-dome; tweeter}.
If any Impedance anywhere in that complete circuit is changed it will cause changes to all other sections' performance -
though to smaller degree in some sections than in others.
My estimations will take all that into account, but there will be at least a small difference in impedance in the treble section -
resulting in small, if any, audible effect, because it is more than an octave below the crossover point of a 3rd order high-pass filter,
and its parallel circuit effect on the mid-range cicuit will be very small.
****************************************************
Dave,
if you're still considering a Hiquphon tweeter after all the above,
has your "buddy in the business" got any additional data - numerical or plots - for OWI, or OWII, that are not on the manufacturer's website ?
I think OWI sound very good, and I would use them in a suitable project.
I do not know if I have ever heard OWII in any loudspeakers.
OK Doug,
post what you hear when you have listened sufficiently to be happy,
and if you don't hear what Dave did, do not be surprised,
because hearing:- the ear<--->brain connection is different for different people.
Dave,
I'm not surprised by your report about audible effects that you heard,
BUT, I'm going to wait till Doug reports what he hears before I comment further.
***********************************************************
Hiquphon tweeters:-
everything has a Trade-off.
Higher sensitivity means less of something else.
Higher sensitivity is not needed for the 66.
Both versions of the tweeter are going to need resistors or an L-pad to match them to the Celestion crossover.
Using larger resistance values to match OWII sensitivity to the Celestion MD500 mids means you will be hearing more of the audio characteristics of the resistors and less of the audio characteristic of the tweeter.
That does not make sense to me, but you decide for you ...
The only reason I would choose a high sensitivity tweeter and pad it down heavily with resistors is if I needed to use the tweeter with a high power capable mid or mid-bass driver to play at very high volume levels -
{higher than MD500 will safely play without damage to it}.
My experience is that heavily padded dome tweeters only sound OK when driven hard,
and sound a bit anemic when played quietly ... but there could be exceptions ...
*********************************************************
WARNING :- before anyone buys a different tweeter to HF2000,
{except perhaps for Coles CE 4001},
I cannot guarantee the changes to the treble filter of the crossover that I will post will give the best possible performance the other tweeter is capable of.
I can only state that it will give a different audible result to HF2000,
and that this sound will not be worse,
AND, if you are prepared to do some fine-tuning by later substituting some different value resistors and/or capacitors, you may be able to get a better audible result.
That is DIY - we experiment ! -{and we measure when we can}.
I will estimate from the tweeter data the resistors and capacitors that I think will work well,
but to be sure, technical measurements would have to be made on the tweeter and on the MD500.
ALSO, my estimations will include using the existing Inductor that is in the 66 crossover - to keep costs low for you,
BUT, to get the best possible performance of a different tweeter to work with the MD500, a different value inductor may be required.
I can estimate, if necessary, a possible inductor, but same warning applies as for caps and resistors - you may have to try other close values.
I state this before anyone buys an EXPENSIVE tweeter.
Safer would be to spend that money on lowish price SEAS H0737 and to pay someone to measure your MD/MF 500s so that a new, optimum filter can be designed to match both.
Safest seems to be to use a Coles CE4001, because that seems to be close to HF2000 except for Impedance.
CE4001 with a suitable parallel connected resistor will then likely work the same in the 66 crossover as HF2000.
CE4001 new is low price, and costs a lot less than ebay sellers are asking for old HF2000s - which may be partially damaged.
It is not likely to be a better or worse sounding tweeter than a perfect condition HF2000,
but it will sound better than a damaged HF2000 !
SEAS H0737 will be audibly better than CE4001, but you will have to be prepared to do some experimenting with crossover components to get the best from it if you are not happy with the audible results of my estimated component values ... and this is what I do myself ...
{I measure what I can, then I calculate what I can, then I estimate from other data whatever else I need to know, then I install and listen,
and if I think I can achieve better I then substitute different value components.}
All 3 filters in the 66 crossover interact, because they are connected in electrical Parallel with each other,
AND in parallel with the Impedances of the 3 active drivers -{woofer; mid-dome; tweeter}.
If any Impedance anywhere in that complete circuit is changed it will cause changes to all other sections' performance -
though to smaller degree in some sections than in others.
My estimations will take all that into account, but there will be at least a small difference in impedance in the treble section -
resulting in small, if any, audible effect, because it is more than an octave below the crossover point of a 3rd order high-pass filter,
and its parallel circuit effect on the mid-range cicuit will be very small.
****************************************************
Dave,
if you're still considering a Hiquphon tweeter after all the above,
has your "buddy in the business" got any additional data - numerical or plots - for OWI, or OWII, that are not on the manufacturer's website ?
I think OWI sound very good, and I would use them in a suitable project.
I do not know if I have ever heard OWII in any loudspeakers.
Last edited:
Alan...
I do understand and realize what you have stated above and in no way am I asking for any guarantee from you or anyone else. I am simply hoping for a starting point and know that it is my choice and up to me to live with or fine tune the result. I do value your expertise as well as that of others.
Given what you have stated regarding the Hiquphons, this was the information I was hoping you could give me, thank you. I am going to go with the OWI, I should be able to get it for slightly more than if I order the Seas. With regard to it being a fairly expensive tweeter, I guess it is, but still less than what some might spend for a night out at the bar, so I guess it's what makes you happy and some new tweeters would make me happy.
Sorry my friend does not have any additional data for the OWI, other that what is on Hiquphons site. He is in the Hi-Fi retail business and has some contacts that are selling or would be willing to sell some drivers they have on hand.
Dave
I do understand and realize what you have stated above and in no way am I asking for any guarantee from you or anyone else. I am simply hoping for a starting point and know that it is my choice and up to me to live with or fine tune the result. I do value your expertise as well as that of others.
Given what you have stated regarding the Hiquphons, this was the information I was hoping you could give me, thank you. I am going to go with the OWI, I should be able to get it for slightly more than if I order the Seas. With regard to it being a fairly expensive tweeter, I guess it is, but still less than what some might spend for a night out at the bar, so I guess it's what makes you happy and some new tweeters would make me happy.
Sorry my friend does not have any additional data for the OWI, other that what is on Hiquphons site. He is in the Hi-Fi retail business and has some contacts that are selling or would be willing to sell some drivers they have on hand.
Dave
more about the tweeter options ...
Hi Dave, ... and whoever else following this ...
I have some more information about the SEAS H/H0586:-
originally it was designated 19TFF.
Later SEAS changed the designation to 19TDF.
The reason is apparently because its surround is a polymer/plastic and the dome is fabric.
The FF designation for the new H0737/19TFF1 is because the dome and surround are BOTH fabric.
Thus for H/H0586, the 19TDF and 19TFF Datasheets are the same,
BUT not the H0737/19TFF1 which has its own Datasheet.
I do not know which dome & surround option of the SEAS T19 tweeter would be better.
Both will be slightly different, perhaps audibly so.
Possibly each is technically as good as the other,
thus decide by sound ... or by price.
***************************
Hiquphon are now making their OW series tweeters available in 4 screw holes and No screw holes versions, as well as their original 3 screw holes version,
thus for HF2000 cabinet holes fitting be sure to order the 3 Screw Holes OW1.
Also, look at the data on:-
Hiquphon 1st natural choice in 3/4" soft dome tweeters - 20mm soft dome tweeter type OWI, OWII, OW3, OW4 .... and more. Designed and produced in Denmark since 1983.
and check that the holes' spacing is the same as the HF2000 - both the screw holes AND the centre cut-out diameter to fit the tweeter in.
There is no 30degrees off-axis frequency response plot shown on the Hiquphon web-site,
but as it is a 19mm dome with a very similar front plate curved exit to SEAS, it will have very similar off-axis response ...
and that response looks good .
OWII looks to be designed for crossover at about 4.5kHz <---> 5kHz to a high sensitivity mid-driver,
or with a non-standard crossover filter slope if crossing at a lower frequency.
Really I think OW1 or one of the above-mentioned SEAS tweeters are the more suitable for Celestion 66.
BUT, as OW1 has a different Fs - at a lower frequency than:-
HF2000 ; Coles CE4001 ; and both SEAS tweeters,
it presents a different type of Impedance to the crossover filter in the frequency band which the filter is rolling off the treble response in,
thus BOTH capacitors may have to be changed to different uF values,
even if I can calculate a suitable resistive network to allow use of the original Celestion inductor ...
and I think I can, and will post about that next time.
Hi Dave, ... and whoever else following this ...
I have some more information about the SEAS H/H0586:-
originally it was designated 19TFF.
Later SEAS changed the designation to 19TDF.
The reason is apparently because its surround is a polymer/plastic and the dome is fabric.
The FF designation for the new H0737/19TFF1 is because the dome and surround are BOTH fabric.
Thus for H/H0586, the 19TDF and 19TFF Datasheets are the same,
BUT not the H0737/19TFF1 which has its own Datasheet.
I do not know which dome & surround option of the SEAS T19 tweeter would be better.
Both will be slightly different, perhaps audibly so.
Possibly each is technically as good as the other,
thus decide by sound ... or by price.
***************************
Hiquphon are now making their OW series tweeters available in 4 screw holes and No screw holes versions, as well as their original 3 screw holes version,
thus for HF2000 cabinet holes fitting be sure to order the 3 Screw Holes OW1.
Also, look at the data on:-
Hiquphon 1st natural choice in 3/4" soft dome tweeters - 20mm soft dome tweeter type OWI, OWII, OW3, OW4 .... and more. Designed and produced in Denmark since 1983.
and check that the holes' spacing is the same as the HF2000 - both the screw holes AND the centre cut-out diameter to fit the tweeter in.
There is no 30degrees off-axis frequency response plot shown on the Hiquphon web-site,
but as it is a 19mm dome with a very similar front plate curved exit to SEAS, it will have very similar off-axis response ...
and that response looks good .
OWII looks to be designed for crossover at about 4.5kHz <---> 5kHz to a high sensitivity mid-driver,
or with a non-standard crossover filter slope if crossing at a lower frequency.
Really I think OW1 or one of the above-mentioned SEAS tweeters are the more suitable for Celestion 66.
BUT, as OW1 has a different Fs - at a lower frequency than:-
HF2000 ; Coles CE4001 ; and both SEAS tweeters,
it presents a different type of Impedance to the crossover filter in the frequency band which the filter is rolling off the treble response in,
thus BOTH capacitors may have to be changed to different uF values,
even if I can calculate a suitable resistive network to allow use of the original Celestion inductor ...
and I think I can, and will post about that next time.
Alan...
Thanks for all the above info.
The screw holes, cutouts, etc. on the HF2000 do not match any of the Seas or Hiquphon tweeters, but the speaker baffle can be modified to accomodate. The HF2000 doesn't have mounting holes in the flange, Celestion uses a type of screw clamp to hold the tweeter in place. The cutout for both the Seas and the Hiquphon magnets will have to be enlarged slightly. If using the Seas, a slot in the cutout is also needed for the tweeter connectors. With either brands of tweeter, the recess in the speaker cabinet for the tweeter is larger, so there will be a gap around the flange of the tweeter.
Reference for others who may be interested in replacing tweeters...
- the recess for the tweeter flange is 102mm dia.
- the cutout is 62mm dia., somewhat ovaled to 84mm to accomodate the tweeter connectors
Hope all that made sense, the bottom line is the Seas and Hiquphon tweeters can be used with some slight modification, but some may not like the end result cosmetically.
Dave
Thanks for all the above info.
The screw holes, cutouts, etc. on the HF2000 do not match any of the Seas or Hiquphon tweeters, but the speaker baffle can be modified to accomodate. The HF2000 doesn't have mounting holes in the flange, Celestion uses a type of screw clamp to hold the tweeter in place. The cutout for both the Seas and the Hiquphon magnets will have to be enlarged slightly. If using the Seas, a slot in the cutout is also needed for the tweeter connectors. With either brands of tweeter, the recess in the speaker cabinet for the tweeter is larger, so there will be a gap around the flange of the tweeter.
Reference for others who may be interested in replacing tweeters...
- the recess for the tweeter flange is 102mm dia.
- the cutout is 62mm dia., somewhat ovaled to 84mm to accomodate the tweeter connectors
Hope all that made sense, the bottom line is the Seas and Hiquphon tweeters can be used with some slight modification, but some may not like the end result cosmetically.
Dave
Just listened to 80 mins of Mozart. Tweets on +to+,-to-. Think I did hear them functioning. But then remebered my hearing test numbers. Both ears require 60 to 85 db to hear from about 5000hz & up. Have decided to just leave everything as is & enjoy what I can & not fuss with it anymore. Might put tweeets on reverse polarity , run that CD again & see if I notice any difference. Thanks Alan & everyone else for a winters time of learning & doing.
Doug
Doug
KCD66SMO
I have really enjoyed reading all this thread. I don't want to start an arguement or upset anyone but felt I had to comment.
I have 5 pairs of Celestion 66's and am fortunate in that I have never had the need to open them up or fiddle about inside.
I do however admire all you clever people that know how to fix things. I can swap drivers and tighten things up but that's about it!
It is my quote about specifications I copied from the old Celestion website and comments that often get added to Celestion 66 listings.
I recall a few years back Celestion 44's (especially tatty ones) were very cheap to buy (a lot cheaper than now) and therefore provided a good source of spares especially if one needed a replacement HF2000.
What I don't understand is why anyone would replace a faulty tweeter with anything other than another HF2000? Second hand one's are not that expensive when you compare the cost to how much 66's are worth. Adding any other tweeter is likely to reduce both the value and one's listening pleasure IMHO.
I'd only buy some with different tweeters if they were so cheap that they were still cheap after factoring in the cost of buying a pair of secondhand HF2000. And even then I would be worried what else was wrong. So they would have to be really cheap! Say less than £200.
Yes upgrade the wiring and crossovers if you need to but only replace the drivers with original parts. Else what you are left with isn't really a Celestion 66.
KCD66SMO - Keep Celestion Ditton Studio Monitors Original - you know it makes sense!
I have really enjoyed reading all this thread. I don't want to start an arguement or upset anyone but felt I had to comment.
I have 5 pairs of Celestion 66's and am fortunate in that I have never had the need to open them up or fiddle about inside.
I do however admire all you clever people that know how to fix things. I can swap drivers and tighten things up but that's about it!
It is my quote about specifications I copied from the old Celestion website and comments that often get added to Celestion 66 listings.
I recall a few years back Celestion 44's (especially tatty ones) were very cheap to buy (a lot cheaper than now) and therefore provided a good source of spares especially if one needed a replacement HF2000.
What I don't understand is why anyone would replace a faulty tweeter with anything other than another HF2000? Second hand one's are not that expensive when you compare the cost to how much 66's are worth. Adding any other tweeter is likely to reduce both the value and one's listening pleasure IMHO.
I'd only buy some with different tweeters if they were so cheap that they were still cheap after factoring in the cost of buying a pair of secondhand HF2000. And even then I would be worried what else was wrong. So they would have to be really cheap! Say less than £200.
Yes upgrade the wiring and crossovers if you need to but only replace the drivers with original parts. Else what you are left with isn't really a Celestion 66.
KCD66SMO - Keep Celestion Ditton Studio Monitors Original - you know it makes sense!
Replies soon
Hello - damaltor ; moermusic ; ittf -
I am currently very short of time.
I will reply to each of your posts as soon as I have time available to reply comprehensively.
For the interim period:-
Dave,
look at the data for Visaton KE 25 SC.
It requires a 50mm x 80mm cutout, thus may be preferable for you if you are going to cut more around the HF2000 cutout,
BUT its Flange is stated as 104.5mm, thus is larger than HF2000.
Look also at data for Monacor tweeters - I have not had time to study them all.
If you are happy to cut for Hiquphon OWI, we can proceed with that one,
BUT you will have to change either BOTH capacitors OR one capacitor and the inductor.
Doug,
do not give up yet.
Remove the 0.5ohm resistor from between the 6.2uF cap and the tweeter and connect the tweeter directly and listen to that.
I will explain why next time.
ittf,
basically would I agree with you about reducing value -{if cutting into the cabinet to fit a larger size}-
and keeping HF2000 if they are working well,
but for damaged HF2000 cases an alternate tweeter seems preferable because the prices asked by ebay sellers for old HF2000 are excessive,
particually when a Coles CE 4001/K can be got to fit the cutout, and is easily adaptable to the existing crossover, and is low price - at least in the UK -
and it has a similar level of audio quality to HF2000.
Hello - damaltor ; moermusic ; ittf -
I am currently very short of time.
I will reply to each of your posts as soon as I have time available to reply comprehensively.
For the interim period:-
Dave,
look at the data for Visaton KE 25 SC.
It requires a 50mm x 80mm cutout, thus may be preferable for you if you are going to cut more around the HF2000 cutout,
BUT its Flange is stated as 104.5mm, thus is larger than HF2000.
Look also at data for Monacor tweeters - I have not had time to study them all.
If you are happy to cut for Hiquphon OWI, we can proceed with that one,
BUT you will have to change either BOTH capacitors OR one capacitor and the inductor.
Doug,
do not give up yet.
Remove the 0.5ohm resistor from between the 6.2uF cap and the tweeter and connect the tweeter directly and listen to that.
I will explain why next time.
ittf,
basically would I agree with you about reducing value -{if cutting into the cabinet to fit a larger size}-
and keeping HF2000 if they are working well,
but for damaged HF2000 cases an alternate tweeter seems preferable because the prices asked by ebay sellers for old HF2000 are excessive,
particually when a Coles CE 4001/K can be got to fit the cutout, and is easily adaptable to the existing crossover, and is low price - at least in the UK -
and it has a similar level of audio quality to HF2000.
Last edited:
Hi Alan...
I would still go for the Hiquphon OWI. The cutout required is 69.5mm, so I will only need to enlarge it slightly. The flange on the Hiq will fit within the recess on the Celestion baffle, so I will not be altering the apperance of the speaker, other than the smaller flange size of the Hiq.
Resale value is not really a consideration, I've owned these speakers since 1980 and have no intension of selling them. Better sounding tweeter is what I'm after and feel that the Hiquphon will give me that. Worse case... if the Hiq doesn't sound any better than the HF2000, I put the HF2000 back in and resell the Hiq.
Dave
I would still go for the Hiquphon OWI. The cutout required is 69.5mm, so I will only need to enlarge it slightly. The flange on the Hiq will fit within the recess on the Celestion baffle, so I will not be altering the apperance of the speaker, other than the smaller flange size of the Hiq.
Resale value is not really a consideration, I've owned these speakers since 1980 and have no intension of selling them. Better sounding tweeter is what I'm after and feel that the Hiquphon will give me that. Worse case... if the Hiq doesn't sound any better than the HF2000, I put the HF2000 back in and resell the Hiq.
Dave
any audible differences ? , etc ...
Hello again ittf,
Some additional reply to yours -{my initial reply in #449 on the previous page to this}.
You are indeed fortunate to have obtained 5 pairs of 66s ... {before the prices got to ridiculous levels ?}.
Thankyou for listing those Specifications - I think the ones I found in another Celestion 66 thread in diyaudio are the ones you are referring to.
Is there still an address for the old Celestion web-site, or anywhere where old Celestion Hi-Fi speakers' data can be found ?
200 quid may be a reasonable threshold for you, but for some buyers that is a LOT of money,
thus why I previously posted about the SEAS and Coles tweeters - both quite a lot cheaper !
Unfortunately for some 66 owners used HF2000 is either too expensive now,
or, no likelihood of a non-faulty pair in the Country where they live.
Look again at the Impedance plots that sba posted for his old drivers a few months ago in this thread.
Of his 4 tweeters, the Yellow sample has a fault - likely an air leak through its diaphragm or surround or mounting plate,
and both the Blue and the Purple seems to have a smaller, similar fault,
plus the Purple has either a weak magnet or voice-coil out of alignment.
Only the Orange sample seems to be in good condition.
I do not know the audibility of these faults,
and as most are within the roll-off of a 3rd Order filter they may not be very audible,
except for the upper range faults in the Yellow sample indicated by the bumps in its Impedance plot.
Given one out of four looks OK, that is NOT good odds for buying old HF2000,
particually if price is more than double that of Coles CE 4001/K,
unless one simply wants to preserve the visual authenticity.
What I am trying to achieve in this thread is modifications which keep the basic character of the original 66 sound whilst using new parts.
Other than the above, I agree with your KCD66SMO !
Please Post and tell us any differences you can hear between any of your sample, even if only very small.
Hello again ittf,
Some additional reply to yours -{my initial reply in #449 on the previous page to this}.
You are indeed fortunate to have obtained 5 pairs of 66s ... {before the prices got to ridiculous levels ?}.
Thankyou for listing those Specifications - I think the ones I found in another Celestion 66 thread in diyaudio are the ones you are referring to.
Is there still an address for the old Celestion web-site, or anywhere where old Celestion Hi-Fi speakers' data can be found ?
200 quid may be a reasonable threshold for you, but for some buyers that is a LOT of money,
thus why I previously posted about the SEAS and Coles tweeters - both quite a lot cheaper !
Unfortunately for some 66 owners used HF2000 is either too expensive now,
or, no likelihood of a non-faulty pair in the Country where they live.
Look again at the Impedance plots that sba posted for his old drivers a few months ago in this thread.
Of his 4 tweeters, the Yellow sample has a fault - likely an air leak through its diaphragm or surround or mounting plate,
and both the Blue and the Purple seems to have a smaller, similar fault,
plus the Purple has either a weak magnet or voice-coil out of alignment.
Only the Orange sample seems to be in good condition.
I do not know the audibility of these faults,
and as most are within the roll-off of a 3rd Order filter they may not be very audible,
except for the upper range faults in the Yellow sample indicated by the bumps in its Impedance plot.
Given one out of four looks OK, that is NOT good odds for buying old HF2000,
particually if price is more than double that of Coles CE 4001/K,
unless one simply wants to preserve the visual authenticity.
What I am trying to achieve in this thread is modifications which keep the basic character of the original 66 sound whilst using new parts.
Other than the above, I agree with your KCD66SMO !
Please Post and tell us any differences you can hear between any of your sample, even if only very small.
Last edited:
relative output levels ?
OK Dave,
here we go ...
First, I need to know the relative output levels that YOU are hearing from the samples of the drivers in YOUR 66s.
{It seems there may be sample variations between drivers which include at least small differences in outputs.}
Listening to the upgraded crossover 66s,
are the Tweeters' levels well matched to the levels from the mid-domes,
or slightly too loud or quiet,
or more than a little too different ?
Please describe as best you can.
And, with regard to the mid-dome sound:- is it Balanced OK between uppper and lower mids,
or does one region of its audio spectrum dominate over the other,
and if so, to what degree ?
Also, does the lower mids' levels from the mid-domes seem to match the lower mids' levels from the woofers around the lower crossover region ?
It will not be perfect here, owing to a resonance seemingly in the woofer's cone,
which dloper has reported hearing,
but describe any other anomolies that you hear there between both drivers.
I need to know this to be able to calculate how much resitive attenuation the Hiquphon tweeter needs.
It will have higher Sensitivity than HF2000, but perhaps not a lot higher.
OK Dave,
here we go ...
First, I need to know the relative output levels that YOU are hearing from the samples of the drivers in YOUR 66s.
{It seems there may be sample variations between drivers which include at least small differences in outputs.}
Listening to the upgraded crossover 66s,
are the Tweeters' levels well matched to the levels from the mid-domes,
or slightly too loud or quiet,
or more than a little too different ?
Please describe as best you can.
And, with regard to the mid-dome sound:- is it Balanced OK between uppper and lower mids,
or does one region of its audio spectrum dominate over the other,
and if so, to what degree ?
Also, does the lower mids' levels from the mid-domes seem to match the lower mids' levels from the woofers around the lower crossover region ?
It will not be perfect here, owing to a resonance seemingly in the woofer's cone,
which dloper has reported hearing,
but describe any other anomolies that you hear there between both drivers.
I need to know this to be able to calculate how much resitive attenuation the Hiquphon tweeter needs.
It will have higher Sensitivity than HF2000, but perhaps not a lot higher.
Alan...
to answer your questions...
"are the Tweeters' levels well matched to the levels from the mid-domes"
I feel the level of the tweeter is a little too quiet, the mid-dome does seem to dominate. Also of note, I am not using an ESR resistor (.5R) on the tweeter cap.
"mid-dome sound:- is it Balanced OK between uppper and lower mids"
I would say it is balanced OK, with a slight egde going to the upper mids.
"does the lower mids' levels from the mid-domes seem to match the lower mids' levels from the woofers"
I think the relationship between the mid-dome and woofer is pretty good, there is nothing here that jumps out at or bothers me.
Dave
to answer your questions...
"are the Tweeters' levels well matched to the levels from the mid-domes"
I feel the level of the tweeter is a little too quiet, the mid-dome does seem to dominate. Also of note, I am not using an ESR resistor (.5R) on the tweeter cap.
"mid-dome sound:- is it Balanced OK between uppper and lower mids"
I would say it is balanced OK, with a slight egde going to the upper mids.
"does the lower mids' levels from the mid-domes seem to match the lower mids' levels from the woofers"
I think the relationship between the mid-dome and woofer is pretty good, there is nothing here that jumps out at or bothers me.
Dave
which components' change option ?
The 0.5 ohm to the tweeters was not intended to simulate ESR in this application.
It was intended to keep the sound from the tweeters in the same audio characteristic as is from the mid-domes and woofers to maintain the authentic Celestion 66 sound.
It will not be identical, and there are two 66 treble sounds,
as result of the different types of caps in the original version and the MkII version, but it will be closer than with no small value resistor there.
Also, various diy experimenters have reported prefering sound from various tweeters with about 0.5ohm in Series in other systems.
It seems to depend on the type of driver used for midrange, and/or the type of crossover filter ... and that is my experience.
With no resistor there you should be hearing a fairly fast or sharp treble character, but as you report the level to be a little too quiet wrt the upper mids I think completely "authentic" Celestion 66 sound is not your listening priority ... and that is not a problem.
I think it best now to try only a small attenuation of the OW1 tweeter,
and then decide whether to increase or decrease its level a very little further.
Ideally I think it best to keep the Total Impedance of the treble circuit close to the same as the original, so as to not change its loading on the midrange circuit,
BUT to do that will require reducing the inductor to .13mH.
If you do that you can keep the 3.9uF cap, and only change the output cap to larger ... to about 11uF.
Look on Solen's website at the 18 AWG aircore inductors for the dimensions of their 0.13mH to see if it will neatly fit on your PCB.
Alternate is to increase the Impedance of the treble circuit a LITTLE,
and that should not cause a major loading problem that would affect the midrange sound.
For this option you keep the Celestion inductor, and change both caps ...
to about 3.6uF and 11uF.
As you are buying a very good sound quality tweeter, consider that to obtain the best possible from it,
it will benefit from better audio caps than Solen.
Solens are not bad, but are slightly microphonic ... and this may vary a little between samples,
and Hiquphon tweeters are transparent enough to clearly hear the differences between cap brands via.
ClarityCap PX series are audibly better ... in my experience ... but you will have to buy from USA, and they have no 3.6uF nor 11uF,
thus you would have to parallel connect 1.8uF pair and 5.6uF pair, and that is possible on your PCB because it has a Parallel pair of holes for both of those circuit positions.
Sonic Craft Sonicaps are a bit more expensive, but are available in 3.6 and 11uF.
You can experiment with cheaper caps first - to hear if this change to crossover works to your liking or if other cap changes may be necessary -
and for that option I see that Parts Connexion now have AXON True Caps in 250 volt series and in both 3.6 and 11uF, and are reduced in price to lower than the Solen PA series, thus if you want to experiment at low price, but still a reasonable quality cap, then buy two of each of those AXON{AEON} caps - 3.6 and 11uF.
Decide about the Inductor first though.
If you decide to buy caps from Parts Conn. then buy also two of the 1R Mills MRA-5 plus two of 2R2 MRA-5 for a fine-tuning option to balance treble to upper mids.
Keep your 0.5R MRA-5s, as you may need to use those for a later tweek.
For the Parallel leg resistors for the tweeters you will need at least a 5 watt rated wirewound resistor, but it does NOT have to be an expensive non-inductive type.
The wirewounds Solen have are 10 watt, and would be suitable, though are 48mm long - 2 inches - thus you will have to decide where to fit those:-
(1) - each could be soldered across the output terminals where the wires to the tweeter are connected, or
(2) - each could be soldered across the input terminals on the back of the tweeter itself, but if the resistors are slightly microphonic the vibration there may cause audible detriment of the sound, or
(3) - look in a local electronic parts shop for small Tag Strips. A type with 2 or 3 terminals will be sufficient.
One can be attached via screws to the cabinet beside the crossover PCB and wires or components' leads used to make the connections.
The 5 watt wirewound resistors sold in electronic parts shops will be OK to use for these parallels at least till you get the fine-tuning of resistance decided,
or if buying from Farnell buy the 6 watt or 9 watt Welwyn W Series wirewounds.
Ideally buy two of 18 ohm for the first option to listen with, but if no 18 is available, then 16 ohm is OK, and 15 ohm will suffice.
Buy two of 27 ohm at the same time for the second listening option, because these resistors are much cheaper than Shipping cost is to buy two additional resistors later.
{If no 27 available, then 25 or 24 ohm will suffice, but 22 ohm will be getting a bit too low.}
BUT, if you decide to change the Inductors, then DON'T buy any resistors yet, instead Post here and I'll calculate suitable values.
I'm guessing you will prefer to keep the inductors original, but you don't have to.
I don't think I have forgotten anything, but if it seems to you that I have, then Post here before you buy anything and I will reply.
The 0.5 ohm to the tweeters was not intended to simulate ESR in this application.
It was intended to keep the sound from the tweeters in the same audio characteristic as is from the mid-domes and woofers to maintain the authentic Celestion 66 sound.
It will not be identical, and there are two 66 treble sounds,
as result of the different types of caps in the original version and the MkII version, but it will be closer than with no small value resistor there.
Also, various diy experimenters have reported prefering sound from various tweeters with about 0.5ohm in Series in other systems.
It seems to depend on the type of driver used for midrange, and/or the type of crossover filter ... and that is my experience.
With no resistor there you should be hearing a fairly fast or sharp treble character, but as you report the level to be a little too quiet wrt the upper mids I think completely "authentic" Celestion 66 sound is not your listening priority ... and that is not a problem.
I think it best now to try only a small attenuation of the OW1 tweeter,
and then decide whether to increase or decrease its level a very little further.
Ideally I think it best to keep the Total Impedance of the treble circuit close to the same as the original, so as to not change its loading on the midrange circuit,
BUT to do that will require reducing the inductor to .13mH.
If you do that you can keep the 3.9uF cap, and only change the output cap to larger ... to about 11uF.
Look on Solen's website at the 18 AWG aircore inductors for the dimensions of their 0.13mH to see if it will neatly fit on your PCB.
Alternate is to increase the Impedance of the treble circuit a LITTLE,
and that should not cause a major loading problem that would affect the midrange sound.
For this option you keep the Celestion inductor, and change both caps ...
to about 3.6uF and 11uF.
As you are buying a very good sound quality tweeter, consider that to obtain the best possible from it,
it will benefit from better audio caps than Solen.
Solens are not bad, but are slightly microphonic ... and this may vary a little between samples,
and Hiquphon tweeters are transparent enough to clearly hear the differences between cap brands via.
ClarityCap PX series are audibly better ... in my experience ... but you will have to buy from USA, and they have no 3.6uF nor 11uF,
thus you would have to parallel connect 1.8uF pair and 5.6uF pair, and that is possible on your PCB because it has a Parallel pair of holes for both of those circuit positions.
Sonic Craft Sonicaps are a bit more expensive, but are available in 3.6 and 11uF.
You can experiment with cheaper caps first - to hear if this change to crossover works to your liking or if other cap changes may be necessary -
and for that option I see that Parts Connexion now have AXON True Caps in 250 volt series and in both 3.6 and 11uF, and are reduced in price to lower than the Solen PA series, thus if you want to experiment at low price, but still a reasonable quality cap, then buy two of each of those AXON{AEON} caps - 3.6 and 11uF.
Decide about the Inductor first though.
If you decide to buy caps from Parts Conn. then buy also two of the 1R Mills MRA-5 plus two of 2R2 MRA-5 for a fine-tuning option to balance treble to upper mids.
Keep your 0.5R MRA-5s, as you may need to use those for a later tweek.
For the Parallel leg resistors for the tweeters you will need at least a 5 watt rated wirewound resistor, but it does NOT have to be an expensive non-inductive type.
The wirewounds Solen have are 10 watt, and would be suitable, though are 48mm long - 2 inches - thus you will have to decide where to fit those:-
(1) - each could be soldered across the output terminals where the wires to the tweeter are connected, or
(2) - each could be soldered across the input terminals on the back of the tweeter itself, but if the resistors are slightly microphonic the vibration there may cause audible detriment of the sound, or
(3) - look in a local electronic parts shop for small Tag Strips. A type with 2 or 3 terminals will be sufficient.
One can be attached via screws to the cabinet beside the crossover PCB and wires or components' leads used to make the connections.
The 5 watt wirewound resistors sold in electronic parts shops will be OK to use for these parallels at least till you get the fine-tuning of resistance decided,
or if buying from Farnell buy the 6 watt or 9 watt Welwyn W Series wirewounds.
Ideally buy two of 18 ohm for the first option to listen with, but if no 18 is available, then 16 ohm is OK, and 15 ohm will suffice.
Buy two of 27 ohm at the same time for the second listening option, because these resistors are much cheaper than Shipping cost is to buy two additional resistors later.
{If no 27 available, then 25 or 24 ohm will suffice, but 22 ohm will be getting a bit too low.}
BUT, if you decide to change the Inductors, then DON'T buy any resistors yet, instead Post here and I'll calculate suitable values.
I'm guessing you will prefer to keep the inductors original, but you don't have to.
I don't think I have forgotten anything, but if it seems to you that I have, then Post here before you buy anything and I will reply.
Last edited:
where in the midrange ?
Hi Doug,
First read my Post #452.
Do you want to reduce the entire midrange level, or only part of its audio spectrum ?
Focus on the lower mids where the sound crosses over to the woofer.
Is the level from the 500 dome too loud -{or perhaps even slightly too quiet}- in that portion of the audio spectrum,
or OK in proportion to that from the woofer ?
Try to ignore the Bass, but listen to the lower-midrange portion of the audio spectrum which is coming from both the 500 dome and from the woofer.
{This will be easier to determine with Classical Music than with Radio-friendly Pop.}
Next, focus on the upper mids from the 500 dome,
which in your particular case will be sounding like the lower treble region of the audio spectrum.
Is this part of the mids in OK proportion/balance with the lower mids also from the 500 dome,
or does this lower treble/upper mids dominate over the lower mids ?
-{BOTH spectrum regions from the mid-dome, not wrt the tweeter because I now know that is not sufficiently audible to you}.
Listen with several different recordings, because any single one may not have an optimum audio-spectrum balance itself.
Post as many details as you like, and I'll try to address them all.
Hi Doug,
First read my Post #452.
Do you want to reduce the entire midrange level, or only part of its audio spectrum ?
Focus on the lower mids where the sound crosses over to the woofer.
Is the level from the 500 dome too loud -{or perhaps even slightly too quiet}- in that portion of the audio spectrum,
or OK in proportion to that from the woofer ?
Try to ignore the Bass, but listen to the lower-midrange portion of the audio spectrum which is coming from both the 500 dome and from the woofer.
{This will be easier to determine with Classical Music than with Radio-friendly Pop.}
Next, focus on the upper mids from the 500 dome,
which in your particular case will be sounding like the lower treble region of the audio spectrum.
Is this part of the mids in OK proportion/balance with the lower mids also from the 500 dome,
or does this lower treble/upper mids dominate over the lower mids ?
-{BOTH spectrum regions from the mid-dome, not wrt the tweeter because I now know that is not sufficiently audible to you}.
Listen with several different recordings, because any single one may not have an optimum audio-spectrum balance itself.
Post as many details as you like, and I'll try to address them all.
Last edited:
Alan...
THANK YOU very much for all your help and time spent on this thread and especially for your recommendation regarding the OWI.
The wife and I are leaving for vacation in a couple weeks and we will be gone till the middle of august. So I am going to put this tweeter project on hold till then.
I haven't ordered the OWI tweeters yet, I'll do so when we get back, along with the recommended caps and resistors. I think I'll try with changing the caps first and see how she sounds.
Thanks again
Dave
THANK YOU very much for all your help and time spent on this thread and especially for your recommendation regarding the OWI.
The wife and I are leaving for vacation in a couple weeks and we will be gone till the middle of august. So I am going to put this tweeter project on hold till then.
I haven't ordered the OWI tweeters yet, I'll do so when we get back, along with the recommended caps and resistors. I think I'll try with changing the caps first and see how she sounds.
Thanks again
Dave
Hi Alan - Finally got round to replacing my 80uF caps with the 70(ish) Solen Caps in the manner you suggested.
First impressions are mixed. The bass/mid thickening I heard before is certainly reduced. Human voice, piano and acoustic guitar don't have the same lower register exaggeration.
There seems to be an overall increase in clarity in the mid range, but I think I'm noticing the upper mid more now. Soprano voices and any female singing or brass instruments in that sort of range can seem a little 'hard' on occasion, so I'm using a little more mid attenuation, a little in excess of the 1 ohm mark I made on my l-pads.
Bass seems as well extended as before, and bass clarity seems improved, yet there is a change I can't quite put my finger on. I don't know if it's in the upper bass or a slight attenuation in the deepest registers. Early days yet though.
First impressions are mixed. The bass/mid thickening I heard before is certainly reduced. Human voice, piano and acoustic guitar don't have the same lower register exaggeration.
There seems to be an overall increase in clarity in the mid range, but I think I'm noticing the upper mid more now. Soprano voices and any female singing or brass instruments in that sort of range can seem a little 'hard' on occasion, so I'm using a little more mid attenuation, a little in excess of the 1 ohm mark I made on my l-pads.
Bass seems as well extended as before, and bass clarity seems improved, yet there is a change I can't quite put my finger on. I don't know if it's in the upper bass or a slight attenuation in the deepest registers. Early days yet though.
as others have heard also
Hi Stuart,
I'm not surprised that you are hearing the greater clarity along with less exaggeration of the lower registers in the manner/instruments you describe.
That has been my experience, and also that of other listeners.
I am not surprised that you are now hearing the upper-mids in the manner that you have described,
as you will likely have read in recent posts that 1, perhaps 2, other listeners here are hearing also.
This can be addressed, but please remind me:-
(1) - do you have MD500 or MF500 mid-domes ?
(2) - what value of resistance, if any, do you have in Series with the 3.9uF{4uF} cap that is in Parallel with the mid-dome ?
Also, report any other audible aspects you have noticed if more by time of your reply ...
eg: if decided about where in the bass register, etc ... because we want the bass to complement the mids.
*******************************************************************************
Hi Dave,
I hope you and your wife have a good vacation.
When you return, check this Thread again before you buy the resistors, because the 2R2 MRA-5 may be reduced to slightly lower by then after experimments by others interested here -{may or may not - remember, this requires substituting/listening experience to determine}.
I'm short of time today, thus have to go now, but I'll read any updated Posts ASAP.
Hi Stuart,
I'm not surprised that you are hearing the greater clarity along with less exaggeration of the lower registers in the manner/instruments you describe.
That has been my experience, and also that of other listeners.
I am not surprised that you are now hearing the upper-mids in the manner that you have described,
as you will likely have read in recent posts that 1, perhaps 2, other listeners here are hearing also.
This can be addressed, but please remind me:-
(1) - do you have MD500 or MF500 mid-domes ?
(2) - what value of resistance, if any, do you have in Series with the 3.9uF{4uF} cap that is in Parallel with the mid-dome ?
Also, report any other audible aspects you have noticed if more by time of your reply ...
eg: if decided about where in the bass register, etc ... because we want the bass to complement the mids.
*******************************************************************************
Hi Dave,
I hope you and your wife have a good vacation.
When you return, check this Thread again before you buy the resistors, because the 2R2 MRA-5 may be reduced to slightly lower by then after experimments by others interested here -{may or may not - remember, this requires substituting/listening experience to determine}.
I'm short of time today, thus have to go now, but I'll read any updated Posts ASAP.
Last edited:
Hi Alan - My mid drivers are MD500s, and there's a 4 ohm resistor in series with the 4 uF cap.
I'd definitely say the upper mid can be a little hard, in the way I described in my last post. I may try swapping the tweeters back to reverse polarity to see if this has an effect?
As for the bass, on further listening, bass notes are clear and easy to follow but are either a bit less extended or a little attenuated in the lowest registers. I can't say which, but my wine doesn't wobble now!
Overall, the speakers have a lighter, more delicate sound. I mean this in a positive sense - they have some of the qualities I enjoyed in my Contour 1.8s (sold on Ebay for a good price by the way).
If the upper mid was a little more restrained, and the lower bass had 'more' I'd be well pleased.
I'm still thinking about building new cabinets too, but I suspect I'd be better leaving the L-pads in place for a while once I've done that so I can decide on a restistor value in place of the L-pads. Perhaps I could do away with them altogether if the upper mid can be tamed.
.
I'd definitely say the upper mid can be a little hard, in the way I described in my last post. I may try swapping the tweeters back to reverse polarity to see if this has an effect?
As for the bass, on further listening, bass notes are clear and easy to follow but are either a bit less extended or a little attenuated in the lowest registers. I can't say which, but my wine doesn't wobble now!
Overall, the speakers have a lighter, more delicate sound. I mean this in a positive sense - they have some of the qualities I enjoyed in my Contour 1.8s (sold on Ebay for a good price by the way).
If the upper mid was a little more restrained, and the lower bass had 'more' I'd be well pleased.
I'm still thinking about building new cabinets too, but I suspect I'd be better leaving the L-pads in place for a while once I've done that so I can decide on a restistor value in place of the L-pads. Perhaps I could do away with them altogether if the upper mid can be tamed.
.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.