Nadir
Having trouble downloading pdf file right now but the crossover I used was troels final crossover on, I believe, page 8 of the file with notch filter consisting of LCR. L being 1.8 mh. The series inductor is .47mh.
I find the midrange quality hard to describe but it seems like it's not totaly flat and depth layering(I dont expect/want artificial depth created by the speakers like I've heard with some) is missing as if there's a phase problem. In fact I kept checking the to see if one tweeter is out of phase with the other.
Thank's Joel
Having trouble downloading pdf file right now but the crossover I used was troels final crossover on, I believe, page 8 of the file with notch filter consisting of LCR. L being 1.8 mh. The series inductor is .47mh.
I find the midrange quality hard to describe but it seems like it's not totaly flat and depth layering(I dont expect/want artificial depth created by the speakers like I've heard with some) is missing as if there's a phase problem. In fact I kept checking the to see if one tweeter is out of phase with the other.
Thank's Joel
Glad to hear you're pleased, Doug. They really are a damned impressive speaker - and I say that despite my misgivings at the beginning of this thread. (I'll be surprised though if you don't end up feeling the need for the notch filter.)
Joel: From your description, you are certainly using the latest Troels-modded xo. It's true that the clones are not the greatest as far as depth layering goes, though image placement in the horizontal plane is better than most. Try pulling them further out into the room for more image depth.
I've just noticed that in Troels' latest v5 PDF there is a mistake. While his crossover schematic in Figure 15 on page 7 shows 1.8 mH for the first woofer series inductor, the biwire layout in Figure 16 on page 8 shows 1.5 mH. This should, of course, be 1.8 also. As Joel says, the notch filter inductor is 1.5 mH, and this is shown correctly in both figures.
Joel: From your description, you are certainly using the latest Troels-modded xo. It's true that the clones are not the greatest as far as depth layering goes, though image placement in the horizontal plane is better than most. Try pulling them further out into the room for more image depth.
I've just noticed that in Troels' latest v5 PDF there is a mistake. While his crossover schematic in Figure 15 on page 7 shows 1.8 mH for the first woofer series inductor, the biwire layout in Figure 16 on page 8 shows 1.5 mH. This should, of course, be 1.8 also. As Joel says, the notch filter inductor is 1.5 mH, and this is shown correctly in both figures.
Has anyone tried using the Notch filter with just the Chinese crossover, not using Jacq's mods?
Also, how significant is the changing of the 5.6 vs the 5ohm resistor as specified in this diagram below when the notch filter is used?
http://www.geocities.com/diyproac25/images/Jacq_Troels.gif
Finally, who is using this design as shown in the image attacheD?
Thanks,
Mike
Also, how significant is the changing of the 5.6 vs the 5ohm resistor as specified in this diagram below when the notch filter is used?
http://www.geocities.com/diyproac25/images/Jacq_Troels.gif
Finally, who is using this design as shown in the image attacheD?
Thanks,
Mike
Last night I added the notch filter to the "Chinese" crossover and am thrilled with the speakers. I was real happy before, but now I'm extatic.
I'm done monkeying with it for a while now. Time to veneer the things in the box and enjoy them. If I want to test a different crossover, I'll build another set of cabinets and try it that way.
I'm done monkeying with it for a while now. Time to veneer the things in the box and enjoy them. If I want to test a different crossover, I'll build another set of cabinets and try it that way.
Hi all,
I am about to embark upon a clone build. Reading this thread has made me re think but having researched this for so long now it seems there is a great conflict of opinion.
I have a pair of R1sc and love the sound as with all Proac. To me they are singly the best speakers this side of panels.
Of those who have made them are they as good as every one says and are they better than other commercially available units at this price? (diy price that is)
If you were going to buy these commercially what price range would they be in?
Welcome any thoughts
Matt
I am about to embark upon a clone build. Reading this thread has made me re think but having researched this for so long now it seems there is a great conflict of opinion.
I have a pair of R1sc and love the sound as with all Proac. To me they are singly the best speakers this side of panels.
Of those who have made them are they as good as every one says and are they better than other commercially available units at this price? (diy price that is)
If you were going to buy these commercially what price range would they be in?
Welcome any thoughts
Matt
Is it just me or do these URL's not work ?
http://members.chello.se/jpo/2_5_clone_measurements_v5.pdf
http://www.geocities.com/diyproac25/images/Jacq_Troels.gif
Can someone post them or email to me ?
Thanks
Andrew
http://members.chello.se/jpo/2_5_clone_measurements_v5.pdf
http://www.geocities.com/diyproac25/images/Jacq_Troels.gif
Can someone post them or email to me ?
Thanks
Andrew
My 2.5 clones have been playing for a year now and impress me as being a bit better than the real ones, using Jaq's mods and the LCR filter. I can say this with much authority as the reason I built them was due to envy of my 2 audio buddies owning the real ones for years. I spent a lot of time listening to their speakers and have compared the two.
I think that we are reaching the limits of the drivers ( which are outstanding at that ) and being obsessively possesed by what have been reduced to very minor flaws in the clones due to the extraordinary efforts of the above mentioned people. Much Thanks!
I say that because one of my buddies let his 2.5's go for the 3.8's.
After much listening to the 3.8's, I can say that they go to the next level, probably due to a better tweeter and more attention to the crossover.
I can also say that if I had to give my clones a ProAc number I would call them 2.8's as they beat the real 2.5's and have a characteristic closer to the 3.8's.
Anyone who builds these speakers will not be dissapointed with their performance. All of the research done with the crossover will allow you to tweak the sound to your own taste, equipment, and room. These are the variables left out of most discussions of the sound of the clones.
I wish I had these speakers a long time ago, I would have saved a lot of money switching amps, preamps and other components trying to get the best sound.
I think that we are reaching the limits of the drivers ( which are outstanding at that ) and being obsessively possesed by what have been reduced to very minor flaws in the clones due to the extraordinary efforts of the above mentioned people. Much Thanks!
I say that because one of my buddies let his 2.5's go for the 3.8's.
After much listening to the 3.8's, I can say that they go to the next level, probably due to a better tweeter and more attention to the crossover.
I can also say that if I had to give my clones a ProAc number I would call them 2.8's as they beat the real 2.5's and have a characteristic closer to the 3.8's.
Anyone who builds these speakers will not be dissapointed with their performance. All of the research done with the crossover will allow you to tweak the sound to your own taste, equipment, and room. These are the variables left out of most discussions of the sound of the clones.
I wish I had these speakers a long time ago, I would have saved a lot of money switching amps, preamps and other components trying to get the best sound.
Thanks for the message. I think, from the postings that the tweeter has changed recently which has caused some problems not apparent with earlier incarnations. I was wondering if recently built systems were just as good.
Re the web links – I have problems with the geocities too, you need 128 bit encryption on your browser I think. The browser on my windows 2000 machine won’t connect but my XP system works fine. For the other one just type the first bit of the location and navigate to the required area.
Matt
Re the web links – I have problems with the geocities too, you need 128 bit encryption on your browser I think. The browser on my windows 2000 machine won’t connect but my XP system works fine. For the other one just type the first bit of the location and navigate to the required area.
Matt
I guess you've found them by now, Matt, but links to Troels Gravesen's last four papers can be found on this page: http://members.chello.se/jpo/.
There are a number of points of interest in these documents:
1. The use of a Scan-Speak 9500 tweeter in place of the 8513. (I haven't tried this personally, but I'd be interested to hear from anyone who has.)
2. The notch filter on the 8535 can be eliminated by the simple addition of a resistor to the capacitor in the LP section. This is a more elegant solution to the 2 KHz bump problem, while preventing a slight "smearing" of the sound produced by the notch filter.
3. Radical surgery to the 8513 tweeter (pole-piece damping, ferrofluid removal and Damar coating of the dome) to eliminate the sibilance problem. It works, but requires some care to prevent damage - this is not for the faint-hearted! However, I did it myself without any problems. Note also some slight amendments to the crossover. Troels now feels this modified 8513 works better in the clone than his earlier experiment with the 9500.
On the question of the commercial 2.5s, Troels has now heard them and found the tweeter to sound quite different to the unmodified 8513 in the clone (no sibilance!). He concludes that ProAc must have modified the tweeter, hence his own mods to the 8513. (Interestingly, my own local ProAc dealer told me that ProAc has indeed modified the 2.5 over the years, and agreed that early versions did exhibit excessive sibilance.) Troels also reports that the midrange "hardness" referred to earlier in this thread is very much in evidence in the current commercial speaker. (This is certainly at variance with some reviews!)
Troels states that a bi-product of his mods to the 8513 is that integration between tweeter and mid is improved markedly. Personally, I think he's right. Sibilance is not only more natural, but the midrange is less inclined to harden up and the transition from mid to tweeter is less apparent, the effect being an improvement in "presence". It's not perfect, but the sound is much improved. I should add that some of this is also due to Troels' present recommendation of three coatings of Damar to the dust cap of the 8535.
In my opinion, Troels' latest mods to the drivers and crossover produce a much more neutral (and tolerant) speaker than the original clone. On the other hand, I'm sure not everyone will agree. It's certainly a long way from the sound of the original Jacq or Chinese versions. If you like that sound, leave well enough alone, but if the tweeter sibilance and/or midrange hardness is driving you up the wall, or even if you're about to embark on building the clone, give these papers a careful read.
There are a number of points of interest in these documents:
1. The use of a Scan-Speak 9500 tweeter in place of the 8513. (I haven't tried this personally, but I'd be interested to hear from anyone who has.)
2. The notch filter on the 8535 can be eliminated by the simple addition of a resistor to the capacitor in the LP section. This is a more elegant solution to the 2 KHz bump problem, while preventing a slight "smearing" of the sound produced by the notch filter.
3. Radical surgery to the 8513 tweeter (pole-piece damping, ferrofluid removal and Damar coating of the dome) to eliminate the sibilance problem. It works, but requires some care to prevent damage - this is not for the faint-hearted! However, I did it myself without any problems. Note also some slight amendments to the crossover. Troels now feels this modified 8513 works better in the clone than his earlier experiment with the 9500.
On the question of the commercial 2.5s, Troels has now heard them and found the tweeter to sound quite different to the unmodified 8513 in the clone (no sibilance!). He concludes that ProAc must have modified the tweeter, hence his own mods to the 8513. (Interestingly, my own local ProAc dealer told me that ProAc has indeed modified the 2.5 over the years, and agreed that early versions did exhibit excessive sibilance.) Troels also reports that the midrange "hardness" referred to earlier in this thread is very much in evidence in the current commercial speaker. (This is certainly at variance with some reviews!)
Troels states that a bi-product of his mods to the 8513 is that integration between tweeter and mid is improved markedly. Personally, I think he's right. Sibilance is not only more natural, but the midrange is less inclined to harden up and the transition from mid to tweeter is less apparent, the effect being an improvement in "presence". It's not perfect, but the sound is much improved. I should add that some of this is also due to Troels' present recommendation of three coatings of Damar to the dust cap of the 8535.
In my opinion, Troels' latest mods to the drivers and crossover produce a much more neutral (and tolerant) speaker than the original clone. On the other hand, I'm sure not everyone will agree. It's certainly a long way from the sound of the original Jacq or Chinese versions. If you like that sound, leave well enough alone, but if the tweeter sibilance and/or midrange hardness is driving you up the wall, or even if you're about to embark on building the clone, give these papers a careful read.
Nadir, thanks for the comments I have read this thread extensively and am reading through Troles’ papers. My build is underway and I have decided to start with the basic set up and if necessary move towards the modifications.
Your observations and the help I have had from Troles is an inspiration and shows that there's nothing that can’t be achieved with co-operation.
I hope to be able to feed some advice back in as I build these speakers and do so from a total novice point of view. I can only ask that you keep going with this support as it has been a great help to me and I’m sure to others. It is all out there it is just knowing where to find it!
Your observations and the help I have had from Troles is an inspiration and shows that there's nothing that can’t be achieved with co-operation.
I hope to be able to feed some advice back in as I build these speakers and do so from a total novice point of view. I can only ask that you keep going with this support as it has been a great help to me and I’m sure to others. It is all out there it is just knowing where to find it!
Matt:
If you've read Troels' paper on the tweeter mods, you'll have seen his warning that the latest batch of 8535 bass/mid drivers are heavily coated on the rear of the cone. His measurements of their response indicate that this coating produces an earlier roll-off than previously and will require further modifications to the crossover as a result. Have you examined the rear of your own 8535s?
I think your approach is a sensible one, i.e. starting off with the original crossover and then trying the mods if you feel the need. With the exception of the tweeter surgery, they're all reversible - even the Damar coating of the 8535 dust cap.
I'll be interested to read your impressions of the clone. As you've noted, there is a great diversity of opinion on this speaker. What I've found interesting is that for a number of builders there is an initial sense of excitement and wonder at what they can do, but as time goes on and CD/LP collections are explored more fully, their faults become apparent - and sometimes intolerable. For others, however, they remain everything that could be wished for. It shows that we all perceive sound differently.
If you've read Troels' paper on the tweeter mods, you'll have seen his warning that the latest batch of 8535 bass/mid drivers are heavily coated on the rear of the cone. His measurements of their response indicate that this coating produces an earlier roll-off than previously and will require further modifications to the crossover as a result. Have you examined the rear of your own 8535s?
I think your approach is a sensible one, i.e. starting off with the original crossover and then trying the mods if you feel the need. With the exception of the tweeter surgery, they're all reversible - even the Damar coating of the 8535 dust cap.
I'll be interested to read your impressions of the clone. As you've noted, there is a great diversity of opinion on this speaker. What I've found interesting is that for a number of builders there is an initial sense of excitement and wonder at what they can do, but as time goes on and CD/LP collections are explored more fully, their faults become apparent - and sometimes intolerable. For others, however, they remain everything that could be wished for. It shows that we all perceive sound differently.
Nadir,
I have all the main parts but the drive units sort of hesitating on which tweeter to get. I will go for the 8513 I think so will hopefully be ordering soon. The boxes are almost complete so next is to build the cross over. I will make ‘dope’ the bass/mid with the Damar varnish first and then see if the sound is as bad as you suggest. I am sensitive to sibilance so I’m glad there is this mod available. Failing all of this there is always the 9500.
This is the great thing about DIY. I will hopefully be able to pass on my views in a couple of months when I get the time to complete them. Thanks to every one for their views and Troles with whom I have communicated several times with what must seem stupid questions. But he has been patient. I can’t wait to get them done.
My last thoughts are on internal cable. If the tweeter is ‘hard’ sounding silver coated OFC cable might not be the best option. Any one any views?
I have all the main parts but the drive units sort of hesitating on which tweeter to get. I will go for the 8513 I think so will hopefully be ordering soon. The boxes are almost complete so next is to build the cross over. I will make ‘dope’ the bass/mid with the Damar varnish first and then see if the sound is as bad as you suggest. I am sensitive to sibilance so I’m glad there is this mod available. Failing all of this there is always the 9500.
This is the great thing about DIY. I will hopefully be able to pass on my views in a couple of months when I get the time to complete them. Thanks to every one for their views and Troles with whom I have communicated several times with what must seem stupid questions. But he has been patient. I can’t wait to get them done.
My last thoughts are on internal cable. If the tweeter is ‘hard’ sounding silver coated OFC cable might not be the best option. Any one any views?
Matt:
If you're sensitive to sibilance, I think the unmodified 8513 is going to give you quite a shock! But do listen first before doing the mods.
As far as the 9500 goes, the few who have tried it like the sound, but Troels says the modified 8513 is even better. Don't be scared off by the tweeter mods if you decide they're needed - they're easier than they look. Just take it slowly and carefully.
The Damar coating of the 8535 dust cap does not make a dramatic difference, but it's clearly discernable in A-B comparisons and definitely makes the speaker more tolerant of "difficult" recordings.
On internal cable, it depends a bit on where you're placing the crossover. If it's at the base of the enclosure, it may make a slight difference, but if it's directly behind the drivers I doubt you'll hear much change given the short lengths required. I tried various cables over a period of time, and yes, there were very small differences, but the excessively sibilant, almost metallic character of the tweeter didn't really become any more tolerable no matter what I used. Only changes to the driver itself provided a solution that satisfied my ears.
I'll stick my neck out here and say that differences between cables, caps, etc. are a lot less than some would have you believe. Changing the value of a crossover component, sometimes by even a small amount, produces a much more noticeable difference than replacing that component with something costing megabucks, or trying to correct a basic fault with "smoother" cables or whatever.
I think it's only when you come to know the sound of a speaker really well (and that takes time) that you will be able to appreciate the small changes upgraded components and cables can make. So my feeling is that anybody starting this project should buy caps, cables, etc. of good but modest quality and consider high-quality stuff only when they've finalised the design. This is particularly true in the case of the 2.5 clone, given the number of crossover options available. You could easily waste a lot of money along the way. The important things in this case are low-DCR inductors and correct component values throughout.
If you're sensitive to sibilance, I think the unmodified 8513 is going to give you quite a shock! But do listen first before doing the mods.
As far as the 9500 goes, the few who have tried it like the sound, but Troels says the modified 8513 is even better. Don't be scared off by the tweeter mods if you decide they're needed - they're easier than they look. Just take it slowly and carefully.
The Damar coating of the 8535 dust cap does not make a dramatic difference, but it's clearly discernable in A-B comparisons and definitely makes the speaker more tolerant of "difficult" recordings.
On internal cable, it depends a bit on where you're placing the crossover. If it's at the base of the enclosure, it may make a slight difference, but if it's directly behind the drivers I doubt you'll hear much change given the short lengths required. I tried various cables over a period of time, and yes, there were very small differences, but the excessively sibilant, almost metallic character of the tweeter didn't really become any more tolerable no matter what I used. Only changes to the driver itself provided a solution that satisfied my ears.
I'll stick my neck out here and say that differences between cables, caps, etc. are a lot less than some would have you believe. Changing the value of a crossover component, sometimes by even a small amount, produces a much more noticeable difference than replacing that component with something costing megabucks, or trying to correct a basic fault with "smoother" cables or whatever.
I think it's only when you come to know the sound of a speaker really well (and that takes time) that you will be able to appreciate the small changes upgraded components and cables can make. So my feeling is that anybody starting this project should buy caps, cables, etc. of good but modest quality and consider high-quality stuff only when they've finalised the design. This is particularly true in the case of the 2.5 clone, given the number of crossover options available. You could easily waste a lot of money along the way. The important things in this case are low-DCR inductors and correct component values throughout.
I use Aeon (Solen) polypropelene(sp?) capacitors in my crossover and use a higher quality low value 0.1uf metal/tin foil capacitor in parrallel with each capacitor in the tweeter section. This significantly lifts the clarity (10-20+%) although does not resolve the sibilance issue, but helps anyway.
I will just add some comments I made at another forum which will add to the rest of the comments made by others here in this post:
Its been interesting to follow Troel's mods from the notch filter, 9500 tweeter, then back to dropping the notch filter out in place of a simple resistor, and now major surgery to the 8513 tweeter and damar coating.
My clones are now about 2 years old with Jacq's mods and last month I had the opportunity to listen to some commercial 2.5s again after about 3 years since I owned some real ones. Overall the sound was still very close, IMO. The treble was more reserved this time and less inclined to zing. Looks like the site will need some more minor changes to reflect Troel's good work.
In absolute perfection terms I can agree with these comments. Call me biased, but if you apply the same test to a broad range of other speakers many would not even do as well. I've had about 20-30 speakers in my house, heard 2 times plus many more and the clones still standout among 10 of the best in my experience, given their minor faults. I've heard the 8513 in several other speakers and the general zingy nature is about the same, in some applications padded down more. The speculation about the clones as its generally healthy discussion. I've had a small army of people thru the house and most love the clones within a given context. Some who are unhappy compared them to totally different classes of speakers so the comparison needs to be more reasonable.
As a beginners project or anyone who is coming from mid hifi and looking for something better it is certainly worth it. Those who want absolute cleanliness and perfection should look elsewhere because even the commercial 2.5 does not meet these objectives. Troels latest statement on new 8535s having different coating is a little worrying. Its also very hard to standardise the various comments from builders around the world when people are using slightly different xo parts, stuffing, and damping etc on top of different driver batch variations and bad amp matches. I have heard 4 different clones built in my local area and each was a little different but in general very similar, none were failures and all four builders/owners are very happy. They would have had to spend for new speakers 2-4 times more to get anything near as good. Show me a speaker that is the same thing to all men and women ?.
The original group who set out 2 years ago to make the clone site never intended it to be this widely accepted (see JPO site and the various ClonAc shrine sites).
I will just add some comments I made at another forum which will add to the rest of the comments made by others here in this post:
Its been interesting to follow Troel's mods from the notch filter, 9500 tweeter, then back to dropping the notch filter out in place of a simple resistor, and now major surgery to the 8513 tweeter and damar coating.
My clones are now about 2 years old with Jacq's mods and last month I had the opportunity to listen to some commercial 2.5s again after about 3 years since I owned some real ones. Overall the sound was still very close, IMO. The treble was more reserved this time and less inclined to zing. Looks like the site will need some more minor changes to reflect Troel's good work.
In absolute perfection terms I can agree with these comments. Call me biased, but if you apply the same test to a broad range of other speakers many would not even do as well. I've had about 20-30 speakers in my house, heard 2 times plus many more and the clones still standout among 10 of the best in my experience, given their minor faults. I've heard the 8513 in several other speakers and the general zingy nature is about the same, in some applications padded down more. The speculation about the clones as its generally healthy discussion. I've had a small army of people thru the house and most love the clones within a given context. Some who are unhappy compared them to totally different classes of speakers so the comparison needs to be more reasonable.
As a beginners project or anyone who is coming from mid hifi and looking for something better it is certainly worth it. Those who want absolute cleanliness and perfection should look elsewhere because even the commercial 2.5 does not meet these objectives. Troels latest statement on new 8535s having different coating is a little worrying. Its also very hard to standardise the various comments from builders around the world when people are using slightly different xo parts, stuffing, and damping etc on top of different driver batch variations and bad amp matches. I have heard 4 different clones built in my local area and each was a little different but in general very similar, none were failures and all four builders/owners are very happy. They would have had to spend for new speakers 2-4 times more to get anything near as good. Show me a speaker that is the same thing to all men and women ?.
The original group who set out 2 years ago to make the clone site never intended it to be this widely accepted (see JPO site and the various ClonAc shrine sites).
H i all, this all makes very interesting reading, seems there has been significant effort by troels to eliminate the "bump"@ 2 kHz which is blamed for the sibilance problem 
but this IMHO is totally misguided as the real culprit is the bump in the tweeter response see FR graph# 5 & IS LOCATEd between 4- 5 kHz as I found out to my chagrin after much modd ing etc, of the x/over of an IPl tL 3 I was building for a friend last year , took me an equally long time to locate the problem until I pinned it down to a small rise in the tweeter response as it begins to roll in so if you all want to go ahead and try the notch filter I can tell you only that it will make a difference but the sibilance will still be there untill you eliminate that little plateau between 4-5 k which looks so inoccous that it is easy to overlook and dismiss out of hand as being insignificant, but be warned! ignore it at your peril as the sibilance will still be there unless you can eliminate that little plateau in the x/over, my solution was to go to a series x/over just as troels has done
cheers to all and have fun
TCThis is getting interesting. Qwad, Troles has made some major mods to the xover and now to the tweeter. Are you suggesting that the sibilance could be eliminated by simply modifying the xover?
Al.M I think you have answered most of my thoughts on this. Anticipating what I can expect. I am concerned that the clone will be worse than my R1SC which are an absolute dream. The only better speakers to my ears have been more expensive Proacs or panels.
That many are happy with the clone is encouraging. The sibilance thing is a worry but it has to be in context. There are shortcomings with everything and I’ve not yet heard anything that is perfect. So Hopefully, Nadir, you are being hyper critical and looking for perfection. However the suggestion there is some work to be done in the 4-5KHz range is intriguing.
The only problem is that I have no idea how these things are sorted. Nadir seems to be some one who has some insight into this. How can this area be modified? Perhaps Troles has some ideas.
Personally I intend to go ahead with my build and start to learn about speaker design and try and sort this out. The thought of having a Proac which I can’t afford is quite an inspiration.
Thanks all for you comments they keep me going!!
Al.M I think you have answered most of my thoughts on this. Anticipating what I can expect. I am concerned that the clone will be worse than my R1SC which are an absolute dream. The only better speakers to my ears have been more expensive Proacs or panels.
That many are happy with the clone is encouraging. The sibilance thing is a worry but it has to be in context. There are shortcomings with everything and I’ve not yet heard anything that is perfect. So Hopefully, Nadir, you are being hyper critical and looking for perfection. However the suggestion there is some work to be done in the 4-5KHz range is intriguing.
The only problem is that I have no idea how these things are sorted. Nadir seems to be some one who has some insight into this. How can this area be modified? Perhaps Troles has some ideas.
Personally I intend to go ahead with my build and start to learn about speaker design and try and sort this out. The thought of having a Proac which I can’t afford is quite an inspiration.
Thanks all for you comments they keep me going!!
2.5 clone
Well, there certainly are a number of things to clarify here.
QWAD: yes, a lot of things were tried to get rid of the 2 kHz bump and eventually you can take the basic crossover (Jacq) and apply a 3.3 ohm resistor in series with the capacitor in the bass section. That's all and the bump is gone. If you further increase the tweeter series resistor to 8.2 ohm you basically have a flat frequency response all over.
The 2 kHz bump was never blamed for giving sibilance and couldn't give sibilance as such as we are here talking 5-10 kHz region. The 2 kHz bump creates a presence effect that can be seductive on certain recordings but most troublesome on particularly vocal recordings.
The minor bump at 5 kHz seen in some graphs was a regrettable artefact of microhone setup and tweaking in this area didn't bring anything usefull around. Except some better understanding of "sibilance".
As seen from the graphs in my "The Final 2.5 clone" we have a decent flat response also in this range. Nadir and I spent a lot of mails discussing the "sibilance" problem and in the strict sense of the word the tweeter is not sibilant. We have sibilance when we have a general too high tweeter level, that is from 4-5 kHz and upwards. This can easily be corrected by the tweeter series resistor. What we agreed on was that the tweeter was distorted in some obscure way that I cannot account for. It doesn't show up om harmonic distortion which I can measure by the CLIO system. But by comparing the 8513 tweeter to the HIQUPHON OWI tweeter, which is basically identical to the 8513, it turned out that the 8513 tweeter is far from being uncolored and the heavy coating of the OWI dome clearly pointed to the direction of doing the same thing to the 8513 tweeter. And so far three people have reported back with good results.
I never went for a series crossover? I tried the ALDO series crossover and found it to give a deep dip at 2 kHz and it needed some serious work to perform correctly.
I have strayed with the parallel crossover all the time.
Matt:
As said above sibilance is generally a matter of level and can be corrected.
If you are happy with the SC1s you will probably also like the unmodified 8513 presentation from the 2.5 clone.
I'm sure I don't offend Nadir when I say that we are being very critical on the 8513 tweeter. Hyper critical? Well, we want the best from our recordings and other speakers can do better, so why not? And if a 10$ bottle of resin can do the job, it's a go!
I suggest you start with a 8.2 ohm resistor to the tweeter and listen carefully and make up your mind about the performance.
The 2.5 clone is a great learning project because the drivers are by all means not perfect. The 8535 bass has a hard time reproducing the midrange up to 3 kHz. I guess that's the price to be paid for the incredible bass performance of a small 6½" driver. And the 8513 is not the greatest tweeter around. In that case we would probably have to pay twice as much (like the OWI).
AlM:
English is not my native tongue, but I like your word "zingy" and I think I know what you mean. If I'm correct, this is just what you eliminate by using the 9500, OWI or coated 8513 tweeters.
Regards and happy tweaking!
Well, there certainly are a number of things to clarify here.
QWAD: yes, a lot of things were tried to get rid of the 2 kHz bump and eventually you can take the basic crossover (Jacq) and apply a 3.3 ohm resistor in series with the capacitor in the bass section. That's all and the bump is gone. If you further increase the tweeter series resistor to 8.2 ohm you basically have a flat frequency response all over.
The 2 kHz bump was never blamed for giving sibilance and couldn't give sibilance as such as we are here talking 5-10 kHz region. The 2 kHz bump creates a presence effect that can be seductive on certain recordings but most troublesome on particularly vocal recordings.
The minor bump at 5 kHz seen in some graphs was a regrettable artefact of microhone setup and tweaking in this area didn't bring anything usefull around. Except some better understanding of "sibilance".
As seen from the graphs in my "The Final 2.5 clone" we have a decent flat response also in this range. Nadir and I spent a lot of mails discussing the "sibilance" problem and in the strict sense of the word the tweeter is not sibilant. We have sibilance when we have a general too high tweeter level, that is from 4-5 kHz and upwards. This can easily be corrected by the tweeter series resistor. What we agreed on was that the tweeter was distorted in some obscure way that I cannot account for. It doesn't show up om harmonic distortion which I can measure by the CLIO system. But by comparing the 8513 tweeter to the HIQUPHON OWI tweeter, which is basically identical to the 8513, it turned out that the 8513 tweeter is far from being uncolored and the heavy coating of the OWI dome clearly pointed to the direction of doing the same thing to the 8513 tweeter. And so far three people have reported back with good results.
I never went for a series crossover? I tried the ALDO series crossover and found it to give a deep dip at 2 kHz and it needed some serious work to perform correctly.
I have strayed with the parallel crossover all the time.
Matt:
As said above sibilance is generally a matter of level and can be corrected.
If you are happy with the SC1s you will probably also like the unmodified 8513 presentation from the 2.5 clone.
I'm sure I don't offend Nadir when I say that we are being very critical on the 8513 tweeter. Hyper critical? Well, we want the best from our recordings and other speakers can do better, so why not? And if a 10$ bottle of resin can do the job, it's a go!
I suggest you start with a 8.2 ohm resistor to the tweeter and listen carefully and make up your mind about the performance.
The 2.5 clone is a great learning project because the drivers are by all means not perfect. The 8535 bass has a hard time reproducing the midrange up to 3 kHz. I guess that's the price to be paid for the incredible bass performance of a small 6½" driver. And the 8513 is not the greatest tweeter around. In that case we would probably have to pay twice as much (like the OWI).
AlM:
English is not my native tongue, but I like your word "zingy" and I think I know what you mean. If I'm correct, this is just what you eliminate by using the 9500, OWI or coated 8513 tweeters.
Regards and happy tweaking!
Just another small comment on the sibilance question. As Troels says, it really appears that there's some obscure distortion in the 8513. The point about the sound of s's and t's with the unmodified tweeter is not so much that they sound excessive, but rather the way they "ring out" in an almost metallic way ("zingy" is a pretty good description). If they were simply excessive, we could just increase attenuation on the tweeter as Troels says, and that would be that. Padding the tweeter down certainly reduces their level, but they still manage to jump out of the mix - even with the tweeter level down to the point of dullness. They even seem to follow you into the next room! Given that the response is quite flat, this has to be caused by distortion, whether we're able to measure it or not. If it doesn't bother you, fine; but if it does, I think Troels' solution fixes the problem very effectively.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.