I'm just wondering how audible the IMD of aluminum tweeters' breakup is and what is the usual method of taming this. I noticed in John K's new cheap MTM the breakup above 20k is pretty bad, but no attempt is made to tame it.
Would the usual method be a notch filter or a CR to ground?
BTW I did do a search but got nothing for IM and just couple posts using intermodular so don't kill me
Would the usual method be a notch filter or a CR to ground?
BTW I did do a search but got nothing for IM and just couple posts using intermodular so don't kill me
some things to consider...
you could just buy a better tweeter and save time and money... but if you feel like having a wack at it you can do some physical mods. You can drill out the pole piece and make a large chamfer at the top side. Damp the pole with some fluf and apply some felt to the chamfer. You can take some siliconized acrylic latex calking and paint a thin layer on the tweeter surround (assume that it is also aluminum). You can paint some 3/16 inch diametre dots (4-5)of the same caulking on to the inside surface of the dome (equally spaced) to help break up dome resonances (thin ones so as not to add too much mass). You can apply a good wack of duct seal to the magnet structure. You can tag a small wire from the negitave tweeter terminal to the tweeter top plate (needs to make electrical contact to the top plate). You can mount a 5/16-3/8 inch rod vertically over the tweeter dome (as close as you can) to deflect the dome resonance at the apex. You can make a small felt ring (1/16 inch thick) large enough to fit around the tweeter suspension and just cover over top of the suspension. All of these things should keep you busy for a while and make your tweeter sound a lot better for not much more than your time invested. Oh you can also experiment with a snubber resistor across the tweeter terminals start with about 10x the DCR of the tweeter voice coil value (1/8 or 1/4 watt is lotsdosent matter what kind). Hope that you have fun and like what you hear. Regards Moray James.
you could just buy a better tweeter and save time and money... but if you feel like having a wack at it you can do some physical mods. You can drill out the pole piece and make a large chamfer at the top side. Damp the pole with some fluf and apply some felt to the chamfer. You can take some siliconized acrylic latex calking and paint a thin layer on the tweeter surround (assume that it is also aluminum). You can paint some 3/16 inch diametre dots (4-5)of the same caulking on to the inside surface of the dome (equally spaced) to help break up dome resonances (thin ones so as not to add too much mass). You can apply a good wack of duct seal to the magnet structure. You can tag a small wire from the negitave tweeter terminal to the tweeter top plate (needs to make electrical contact to the top plate). You can mount a 5/16-3/8 inch rod vertically over the tweeter dome (as close as you can) to deflect the dome resonance at the apex. You can make a small felt ring (1/16 inch thick) large enough to fit around the tweeter suspension and just cover over top of the suspension. All of these things should keep you busy for a while and make your tweeter sound a lot better for not much more than your time invested. Oh you can also experiment with a snubber resistor across the tweeter terminals start with about 10x the DCR of the tweeter voice coil value (1/8 or 1/4 watt is lotsdosent matter what kind). Hope that you have fun and like what you hear. Regards Moray James.
augerpro said:
As far as I know this doesn't happen. But, think of it this way - if an ultrasonic breakup caused an anomaly in the audible region, THAT anomaly would show up in the measurements just as surely as the HF one does - and that's what you would want to deal with in your crossover design.
So really the ultrasonic stuff is irrelevant either way - regardless of what caused the problem in the audible region you ignore the stuff you can't hear and work on the stuff you can.
Make sense?
I was thinking along the same lines at first. If it's above 20k I won't hear it. If it causes bump at a lower harmonic i'd see it in a 1/12 smoothed or better FR.
But then I thought would I see it? What if it does show up in FR as a little bump, maybe a dB or two? Most of us wouldn't think anything is wrong with a little bump. And what if that original 20k+ spike is very distorted? The lower harmonic might be very distorted. So 1 or 2 dB distorted bump may be audible I'm thinking.
Just some musings...
But then I thought would I see it? What if it does show up in FR as a little bump, maybe a dB or two? Most of us wouldn't think anything is wrong with a little bump. And what if that original 20k+ spike is very distorted? The lower harmonic might be very distorted. So 1 or 2 dB distorted bump may be audible I'm thinking.
Just some musings...
what you hear and what you preceive...
are two different things but they both impact on what you experience.
Sorry to scare you but I did not have a clue as to what you had spent on the tweeter. Still at $300.00 clams you should only have one thing to do with a tweeter...enjoy it. You could find out what part of your tweeter does the high frequency resonancework, it will either be the tip of the dome (doing a little oil can action) or the suspension resonating as a ring radiator. If it is the suspension then you can easily test to see if the interface between the dome and the suspension is behaving properly by applying a thin (3/64 inch wide) strip if cheap black electrical tape(ie. thin thickness 3 thou or so) around the base of the dome just at the point before it reaches the suspension. This can be applied with light finger pressure no need to press it down firmly. Give the tweeter a listen and if it sounds better then you can simply peel off the tape and apply a thin band of the siliconized caulk with a fine art brush in place of the tape. If it is the tip of your dome that is doing your high frequency duty then you can apply a little layer of the same tape to the top edge of the suspension half round all the way around the dome. Have a listen and if it sounds better then simply apply a layer of the caulk to the suspension as mentioned before directly to the suspension. You can also experiment with placing a couple of 1/16 inch wide by 3/16 inch long strips of PVC electrical tape on the very top of your dome in a cross configuration and listen. If the tweeter sounds better then you can apply a dot about 3/16 inch diametre to the tip of the dome. All of this is easy fast and reversable if you dont like what you have done as the tape application is your removable trial damper, easy on and easy off with no worries. If your dome tip is your hi freq. radiator then the felt overlay covering (but not touching the suspension half roll) will be a good idea (in addition to any damping layer) to reduce the impact of any noise generated by the suspension that you don't want to hear anyway. Try the deflector in front of your dome tip (in vertical orientation) also you can grab a round bic pen or a crayon and tape it in place with some scotch tape to see what you think. The rod can be as thick a 3/8 inch. The cool thing is with this trick is that the top end will actually be smoother and more extended than without it. The only thing that the rod will interfear with (in a good way) is to deflect the dome's oil can resonance on the very tip of the dome toward your side walls and delay its arrival long enough for your brain to recognize it as a reflection (reduced in level from the trip and its reflections) and so dismiss it as part of the rest of the direct radiation from the dome that just shoots on past the rod on its way to you. You still get the high frequency air from the tweeter but it will sound sweeter smoother and surprisingly more extended. You can't make the dome resonance go away (nor do you want to) and you need it as this is your high frequency wide dispersion energy source but you can time delay it and eat up a little level so that it intigrates with the dome at lower frequencies. Remember that a one inch dome starts to get in its own way at about 10KHz. That's why either the dome tip or the suspension is designed to resonate for you at high frequency and a small area so you get the dispersion. The only tweeters that can behave as a true piston at 20 KHz. are plasma's, ribbons and strip ESL's with very thin film diaphragms 1-3 micron thick. All domes really (by design) give you wide dispersion high frequency through making some physically small part of the system resonate up high. It's kind of like the door bell, rings when you need (expect it) and then goes away when you don't. Sounds worse to hear it discribed this way than it actually is to listen to but if you get used to top notch ribbon you will have a hard time going back to a dome. You just cannot physically (practically speaking) make a dome or a cone small enough to do the job at 20 KHz (with wide dispersion).
Try the snubber resistor you can play with the values up and down 10 then 5 ohms to dial in the value but like I said 10 X DCR is a good starting point. This is really simple to do and costs nothing and as always if you do not like the effect simply take it off. Hope that you do some experiments and have some fun listening to your handy work. Regards Moray James. PS: if you are still too freaked out to touch uour spendy tweeters you can try all this on a $5.00 tweeter to acess the impact and get a feel for what these things do. PPS: don't forget the duct seal and you can also make a butterfly gasket out of a single layer of plumbers skived teflon tape (the thin white stuff not the thicker yellow gas fitters tape) and place it inbetween the butterfly and the magnet assembly as well as between the top side of the butterfly assembly and the tweeter cover plate. This will insure that there are no little buzzes anywhere and it will compress down to almost nothing so no worries about the position of the coil in the gap.
are two different things but they both impact on what you experience.
Sorry to scare you but I did not have a clue as to what you had spent on the tweeter. Still at $300.00 clams you should only have one thing to do with a tweeter...enjoy it. You could find out what part of your tweeter does the high frequency resonancework, it will either be the tip of the dome (doing a little oil can action) or the suspension resonating as a ring radiator. If it is the suspension then you can easily test to see if the interface between the dome and the suspension is behaving properly by applying a thin (3/64 inch wide) strip if cheap black electrical tape(ie. thin thickness 3 thou or so) around the base of the dome just at the point before it reaches the suspension. This can be applied with light finger pressure no need to press it down firmly. Give the tweeter a listen and if it sounds better then you can simply peel off the tape and apply a thin band of the siliconized caulk with a fine art brush in place of the tape. If it is the tip of your dome that is doing your high frequency duty then you can apply a little layer of the same tape to the top edge of the suspension half round all the way around the dome. Have a listen and if it sounds better then simply apply a layer of the caulk to the suspension as mentioned before directly to the suspension. You can also experiment with placing a couple of 1/16 inch wide by 3/16 inch long strips of PVC electrical tape on the very top of your dome in a cross configuration and listen. If the tweeter sounds better then you can apply a dot about 3/16 inch diametre to the tip of the dome. All of this is easy fast and reversable if you dont like what you have done as the tape application is your removable trial damper, easy on and easy off with no worries. If your dome tip is your hi freq. radiator then the felt overlay covering (but not touching the suspension half roll) will be a good idea (in addition to any damping layer) to reduce the impact of any noise generated by the suspension that you don't want to hear anyway. Try the deflector in front of your dome tip (in vertical orientation) also you can grab a round bic pen or a crayon and tape it in place with some scotch tape to see what you think. The rod can be as thick a 3/8 inch. The cool thing is with this trick is that the top end will actually be smoother and more extended than without it. The only thing that the rod will interfear with (in a good way) is to deflect the dome's oil can resonance on the very tip of the dome toward your side walls and delay its arrival long enough for your brain to recognize it as a reflection (reduced in level from the trip and its reflections) and so dismiss it as part of the rest of the direct radiation from the dome that just shoots on past the rod on its way to you. You still get the high frequency air from the tweeter but it will sound sweeter smoother and surprisingly more extended. You can't make the dome resonance go away (nor do you want to) and you need it as this is your high frequency wide dispersion energy source but you can time delay it and eat up a little level so that it intigrates with the dome at lower frequencies. Remember that a one inch dome starts to get in its own way at about 10KHz. That's why either the dome tip or the suspension is designed to resonate for you at high frequency and a small area so you get the dispersion. The only tweeters that can behave as a true piston at 20 KHz. are plasma's, ribbons and strip ESL's with very thin film diaphragms 1-3 micron thick. All domes really (by design) give you wide dispersion high frequency through making some physically small part of the system resonate up high. It's kind of like the door bell, rings when you need (expect it) and then goes away when you don't. Sounds worse to hear it discribed this way than it actually is to listen to but if you get used to top notch ribbon you will have a hard time going back to a dome. You just cannot physically (practically speaking) make a dome or a cone small enough to do the job at 20 KHz (with wide dispersion).
Try the snubber resistor you can play with the values up and down 10 then 5 ohms to dial in the value but like I said 10 X DCR is a good starting point. This is really simple to do and costs nothing and as always if you do not like the effect simply take it off. Hope that you do some experiments and have some fun listening to your handy work. Regards Moray James. PS: if you are still too freaked out to touch uour spendy tweeters you can try all this on a $5.00 tweeter to acess the impact and get a feel for what these things do. PPS: don't forget the duct seal and you can also make a butterfly gasket out of a single layer of plumbers skived teflon tape (the thin white stuff not the thicker yellow gas fitters tape) and place it inbetween the butterfly and the magnet assembly as well as between the top side of the butterfly assembly and the tweeter cover plate. This will insure that there are no little buzzes anywhere and it will compress down to almost nothing so no worries about the position of the coil in the gap.
Nasty breakups
Agree with "hear" end "perceive":
In 2004 Focal changed from its classic Titanium dome (with heavy breakups just above 20kHz) to a Alu-Magnesium Alloy, where the breakups occur at higher frequencies, and less pronounced - the result is that the tweeter sound less sharp.
Ther is an excellent material out there which will resolve your problems, it's a Varnish based on natural ingredients, and specially designed for the purpose. It's called "Sound Varnish D" (there is also a Sound Varnish without D, which is for paper cones and softdomes), it's made by Jürgen Ultée (a highly regarded engineer) and avilable here:
http://www.lautsprechershop.de/ see section "Damping".
It's expensive: 28 Euro (about 35 $), but with 10ml you can coat lots a tweeters or (alu cone) woofers. Also suitable for Glassfibre cones.
cheers
- dan
Agree with "hear" end "perceive":
In 2004 Focal changed from its classic Titanium dome (with heavy breakups just above 20kHz) to a Alu-Magnesium Alloy, where the breakups occur at higher frequencies, and less pronounced - the result is that the tweeter sound less sharp.
Ther is an excellent material out there which will resolve your problems, it's a Varnish based on natural ingredients, and specially designed for the purpose. It's called "Sound Varnish D" (there is also a Sound Varnish without D, which is for paper cones and softdomes), it's made by Jürgen Ultée (a highly regarded engineer) and avilable here:
http://www.lautsprechershop.de/ see section "Damping".
It's expensive: 28 Euro (about 35 $), but with 10ml you can coat lots a tweeters or (alu cone) woofers. Also suitable for Glassfibre cones.
cheers
- dan
Attachments
Moray I may try some of that. The 9800 has a little plastic thing over the center of the dome. I suppose this is to disperse the break up energy in way similar to what you have mentioned. Jon Marsh has a mod he does to those tweeters, I'll have try that thread again and see what he was trying to accomplish.
Alfetta87: thanks for the tip. I think Focal makes something like that too.
Alfetta87: thanks for the tip. I think Focal makes something like that too.
Re: Nasty breakups
Generally it's best to have the breakup mode at 40KHz or higher. Damping needs to be applied with care. make sure there is some way you can get it off in case you don't like it. For testing purposes, I have used Colgate toothpaste (not the gel type but the type that dries to feel like chalk).
The little mask type thing normally in the middle of a metal dome tweeter is used to suppress the effects of the breakup mode. If you take it off, you will measure a more pronounced peak in the top range. This type of design however cuts the resolution, openess, and image precision capabilities of the tweeter.
Alfetta87 said:
Generally it's best to have the breakup mode at 40KHz or higher. Damping needs to be applied with care. make sure there is some way you can get it off in case you don't like it. For testing purposes, I have used Colgate toothpaste (not the gel type but the type that dries to feel like chalk).
augerpro said:
The little mask type thing normally in the middle of a metal dome tweeter is used to suppress the effects of the breakup mode. If you take it off, you will measure a more pronounced peak in the top range. This type of design however cuts the resolution, openess, and image precision capabilities of the tweeter.
Re: Re: Nasty breakups
The Sound Varnish is not just a damping coating, it has other capabilities of for example shifting the breakup frequency to higher frequencies, where they disturb less. Damping only would mean that you damp also in lower bands than the ultrasonic - which comes then with a drop of efficiency. Believe me, try it and you'll surprised.
To take it of, balsamic turpentine (NOT chemical) removes it with ease.
Like yor Colgate tip though, but since we only brush our teeth with Sensodyne, what would I tell my wife? "Honey, can you buy some Colgate for my tweeters?" she is already surprised when I use milled bone dust with Damar Varnish as a coating...
soongsc said:
The Sound Varnish is not just a damping coating, it has other capabilities of for example shifting the breakup frequency to higher frequencies, where they disturb less. Damping only would mean that you damp also in lower bands than the ultrasonic - which comes then with a drop of efficiency. Believe me, try it and you'll surprised.
To take it of, balsamic turpentine (NOT chemical) removes it with ease.
Like yor Colgate tip though, but since we only brush our teeth with Sensodyne, what would I tell my wife? "Honey, can you buy some Colgate for my tweeters?" she is already surprised when I use milled bone dust with Damar Varnish as a coating...
Re: Re: Re: Nasty breakups
I'm not sure what Sound Varnish consists of, but in general, in order to shift the breakup frequency to a higher frequency, the overall stiffness must be increased. And in order to do that without having significant impact on ifficiency and bandwidth, the youngs modulus/weight ratio needs to be higher than the original material. Not really something a varnish normally has. Another way to do some experiment is to apply some things (Sensodyne if you prefer) using a cotton swab but in randum locations to break up the breakup modes. Either way, be prepared to lose around 2db~3db efficiency.
Alfetta87 said:
I'm not sure what Sound Varnish consists of, but in general, in order to shift the breakup frequency to a higher frequency, the overall stiffness must be increased. And in order to do that without having significant impact on ifficiency and bandwidth, the youngs modulus/weight ratio needs to be higher than the original material. Not really something a varnish normally has. Another way to do some experiment is to apply some things (Sensodyne if you prefer) using a cotton swab but in randum locations to break up the breakup modes. Either way, be prepared to lose around 2db~3db efficiency.
Tweeter tweaks...
your wife may like the idea of tweaking your tweeters.... the Sensodyne may also eliminate those sharp spikey twinges you get when you tweeters squeek. Sorry about that but... I do like the paste idea. If you ever play with PVA adhesives you can try micing in a little Borax powder (try this with your varnish too) as it will cause the material to cross link and have greater stretch qualities but don't add too much. A little PVA mixed into your toothpast will give it a little stretch and not only make it stick better to the dome but also stay stuck together better as forces in a tweeter are as high as you will ever encounter in any speaker. You can play with Borax and white glue (PVA) small quantities of Borax will enhance the strength and bonding qualities and if you add enough it will become a thick rubbery goop.
Regarding damping Vs stiffining of a dome I agree that a varnish will not stiffen a dome but it will damp (some) the lesser harmonic resonances but as for the fundamental resonance at the top end well the Q of that one (where ever it is) will not go away with a dab dob or blob of anything you can stick on to it. You need to deal with the fundamentals with structural reinforcements which are best designed into the dome when you make it (ie. ribs or dimples or multiple profiles to the dome). I would be surplised to see a 2-3 db loss in efficiency (across the band) with the kind of modifications that we have been discussing unless you were right on the hairy edge of the driver's motor strength Vs the diaphragm's weight. I have had some interesting results using ring and disk shaped Neo magnets fitted in bucking action on the top of the tweeters pole piece.
As far as messing with expensive drivers, well some will and most will not and only you can make the choice. That said if a driver is almost what you want and you can milk out of it more of what you do want that works for me regardless of the unit price. The only reason not to do mods would be if you plan to resell the unit. In an industry that is so controlled by marketing numbers specifications and graphs these can and often do make the difference between a marketable product and one that you can't even give away. I have developed loudspeakers for commercial products and have had the use of some very sofisticated measurement equipment. This can be both a tremondous benefit as well as a potential detroment. You have to make sure that you do not simply follow the measurements alone. The human ear must at all times be the device to which final decissions are left to. Marketing aside I have built drivers that measured great and sounded like crap and vice versa. One would like to develope any product so that not only does it sound good but that it measures well too. That however is not always the case and we both know which kind of driver we would rather listen to for the long haul.
I think that for the most part that the kinds of tweaks that are being discussed here will have a fairly subtle impact upon the FR of the driver and are not severe enough to loose any sleep over especially when one considers the sweeping changes that are often made with the push of a loudness button or the turn of a tone control. Plus or minus a couple of db seems huge when comparing drivers but in the context of room deviations that are typical they are but a drop in the bucket.
You will have found that with better designed drivers where greater attention to detail has been excercised that there is generally less for you to tweak as most of that work was done at the factory. Bear in mind that the R&D guys and designers most often do not make the final decission of what will or will not go to market and that "good enough for right now" is a common reason for why some products sound the way they do. Regards Moray James.
PS: large flake graphite particals will add considerable stiffness yet at the same time self damps as the structure permits energy to be dissapated. Better to have a structure that goes gracefully into a mild resonance sooner than you would like than one that gets you to where you want to be but then abruptly goes to hell on you.
your wife may like the idea of tweaking your tweeters.... the Sensodyne may also eliminate those sharp spikey twinges you get when you tweeters squeek. Sorry about that but... I do like the paste idea. If you ever play with PVA adhesives you can try micing in a little Borax powder (try this with your varnish too) as it will cause the material to cross link and have greater stretch qualities but don't add too much. A little PVA mixed into your toothpast will give it a little stretch and not only make it stick better to the dome but also stay stuck together better as forces in a tweeter are as high as you will ever encounter in any speaker. You can play with Borax and white glue (PVA) small quantities of Borax will enhance the strength and bonding qualities and if you add enough it will become a thick rubbery goop.
Regarding damping Vs stiffining of a dome I agree that a varnish will not stiffen a dome but it will damp (some) the lesser harmonic resonances but as for the fundamental resonance at the top end well the Q of that one (where ever it is) will not go away with a dab dob or blob of anything you can stick on to it. You need to deal with the fundamentals with structural reinforcements which are best designed into the dome when you make it (ie. ribs or dimples or multiple profiles to the dome). I would be surplised to see a 2-3 db loss in efficiency (across the band) with the kind of modifications that we have been discussing unless you were right on the hairy edge of the driver's motor strength Vs the diaphragm's weight. I have had some interesting results using ring and disk shaped Neo magnets fitted in bucking action on the top of the tweeters pole piece.
As far as messing with expensive drivers, well some will and most will not and only you can make the choice. That said if a driver is almost what you want and you can milk out of it more of what you do want that works for me regardless of the unit price. The only reason not to do mods would be if you plan to resell the unit. In an industry that is so controlled by marketing numbers specifications and graphs these can and often do make the difference between a marketable product and one that you can't even give away. I have developed loudspeakers for commercial products and have had the use of some very sofisticated measurement equipment. This can be both a tremondous benefit as well as a potential detroment. You have to make sure that you do not simply follow the measurements alone. The human ear must at all times be the device to which final decissions are left to. Marketing aside I have built drivers that measured great and sounded like crap and vice versa. One would like to develope any product so that not only does it sound good but that it measures well too. That however is not always the case and we both know which kind of driver we would rather listen to for the long haul.
I think that for the most part that the kinds of tweaks that are being discussed here will have a fairly subtle impact upon the FR of the driver and are not severe enough to loose any sleep over especially when one considers the sweeping changes that are often made with the push of a loudness button or the turn of a tone control. Plus or minus a couple of db seems huge when comparing drivers but in the context of room deviations that are typical they are but a drop in the bucket.
You will have found that with better designed drivers where greater attention to detail has been excercised that there is generally less for you to tweak as most of that work was done at the factory. Bear in mind that the R&D guys and designers most often do not make the final decission of what will or will not go to market and that "good enough for right now" is a common reason for why some products sound the way they do. Regards Moray James.
PS: large flake graphite particals will add considerable stiffness yet at the same time self damps as the structure permits energy to be dissapated. Better to have a structure that goes gracefully into a mild resonance sooner than you would like than one that gets you to where you want to be but then abruptly goes to hell on you.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.