Hi,
What is your opinion about switching the Morel MDT-40 (tweeter) to one of the following to be used with the PEERLESS 832873 driver:
Dayton PT2C-8 Planar Tweeter
HiVi RT1C-A Planar Isodynamic Tweeter
HiVi RT2C-A Planar Isodynamic Tweeter
HiVi RT1L Round Planar Tweeter
What effect might they cause and what’s your view on each?
What is your opinion about switching the Morel MDT-40 (tweeter) to one of the following to be used with the PEERLESS 832873 driver:
Dayton PT2C-8 Planar Tweeter
HiVi RT1C-A Planar Isodynamic Tweeter
HiVi RT2C-A Planar Isodynamic Tweeter
HiVi RT1L Round Planar Tweeter
What effect might they cause and what’s your view on each?
The "HiVi RT1L Round Planar Tweeter" seems very interesting, at 88 dB, the sensitivity matches the woofer. Furthermore it seems to radiate in every direction.
The other 3 tweeters have sensitivity 93-94 dB, so a few resistors must be added to lower the output from these tweeters. Besides, traditional planar tweeters has a reputation of limited horisontal radiation = limited sweetspot.
So, in short I guess the "HiVi RT1L Round Planar Tweeter" could be a very good replacement for the Morel tweeter. Note, that the capacitor should be about 4,4 uF if crossover frequence is kept at 9000 Hz. Enjoy
Link to FR
http://www.swanspeaker.com/product/htm/view.asp?id=226
The other 3 tweeters have sensitivity 93-94 dB, so a few resistors must be added to lower the output from these tweeters. Besides, traditional planar tweeters has a reputation of limited horisontal radiation = limited sweetspot.
So, in short I guess the "HiVi RT1L Round Planar Tweeter" could be a very good replacement for the Morel tweeter. Note, that the capacitor should be about 4,4 uF if crossover frequence is kept at 9000 Hz. Enjoy
Link to FR
http://www.swanspeaker.com/product/htm/view.asp?id=226
had a good look...
and these pencils are very nice.
looking for an inexpensive HT surround speaker (need 4 channels) (sides and rears for a 7.1 or 7.0 system for my daughter's BF).
thinking about subbing some different drivers. We were going to try PVC pipe for an enclosure, and leave the bottom open while elevating it above the ground for a"vent" (more like a TL than anything else). New Onkyo 606 reciever to do the processor/amplifier needs.
stew
and these pencils are very nice.
looking for an inexpensive HT surround speaker (need 4 channels) (sides and rears for a 7.1 or 7.0 system for my daughter's BF).
thinking about subbing some different drivers. We were going to try PVC pipe for an enclosure, and leave the bottom open while elevating it above the ground for a"vent" (more like a TL than anything else). New Onkyo 606 reciever to do the processor/amplifier needs.
stew
Hi Critofur
No, I haven't measured them yet. But has lived with them for some time now. The crossover is just a 2,2 uF for the tweeter. I've done quite some listening tests, and ended up with capacitors from obligato, see http://www.diyhifisupply.com/diyhs_ob_caps.htm.
Why Obligato? Well, I found this shootout: http://www.humblehomemadehifi.com/Cap.html
That guy is very accurate in his description, and I get the same experiences (having used all the Jantzen capacitors in the test).
I have built several other speakers afterwards, but the pencil is still standing in the living room
SPEAKER-CABLE TIP: Check out wires.co.uk. They have a great 2mm Silver-plated copper cable. Find some PTFE-insolation and WHEUUUVV!
No, I haven't measured them yet. But has lived with them for some time now. The crossover is just a 2,2 uF for the tweeter. I've done quite some listening tests, and ended up with capacitors from obligato, see http://www.diyhifisupply.com/diyhs_ob_caps.htm.
Why Obligato? Well, I found this shootout: http://www.humblehomemadehifi.com/Cap.html
That guy is very accurate in his description, and I get the same experiences (having used all the Jantzen capacitors in the test).
I have built several other speakers afterwards, but the pencil is still standing in the living room
SPEAKER-CABLE TIP: Check out wires.co.uk. They have a great 2mm Silver-plated copper cable. Find some PTFE-insolation and WHEUUUVV!
Juspur,
Great looking speaker pair. I was toying with the idea trudging down the path of designing a pair of DIY omni speakers , but between you and Mamboni it appears much or all of the groundwork has already been done. I particularly like the use of quality drivers and your minimalist crossover network. That seems in direct contrast to my Ohm Walsh 5 speakers which don't appear to use the best of drivers, and have the signal passing through -WAY- too many components.
Have you considered building a pair a size or two larger ?
Ohighway
Great looking speaker pair. I was toying with the idea trudging down the path of designing a pair of DIY omni speakers , but between you and Mamboni it appears much or all of the groundwork has already been done. I particularly like the use of quality drivers and your minimalist crossover network. That seems in direct contrast to my Ohm Walsh 5 speakers which don't appear to use the best of drivers, and have the signal passing through -WAY- too many components.
Have you considered building a pair a size or two larger ?
Ohighway
Hi Ohighway
A friend of mine measured the pencil and other drivers mounted the same way. It seems that some of the sound (frequencies above approx. 2,5 kHz) is hitting the back of the magnet - when we're talking about the Peerless I'm using.
As a result, you'll see a rolloff of about 6 db/octave above 2,5 kHz. It's okay - a natural rolloff means that no crossover parts still isn't necessary for the woofer.
However, as a result, I've changed the capacitor for the tweeter to a 3,3 uF from Obligato (see www.diyhifishop.com). So the tweeter now rolls off from 6 kHz with 6 db/octave too.
Now the bad news regarding bigger drivers - if they've got magnets with bigger Ø, they'll absorb more sound (meaning rolloff at lower frequencies).
On the other hand: if you find drivers with magnets with smaller Ø, higher frequencies will be reproduced by the driver without rolloff.
The good news regarding drivers - We've got more drivers to chose from, since a rise in frequency above fx 4 kHz won't be a problem due to the rolloff caused by the magnets fysical Ø.
A friend of mine measured the pencil and other drivers mounted the same way. It seems that some of the sound (frequencies above approx. 2,5 kHz) is hitting the back of the magnet - when we're talking about the Peerless I'm using.
As a result, you'll see a rolloff of about 6 db/octave above 2,5 kHz. It's okay - a natural rolloff means that no crossover parts still isn't necessary for the woofer.
However, as a result, I've changed the capacitor for the tweeter to a 3,3 uF from Obligato (see www.diyhifishop.com). So the tweeter now rolls off from 6 kHz with 6 db/octave too.
Now the bad news regarding bigger drivers - if they've got magnets with bigger Ø, they'll absorb more sound (meaning rolloff at lower frequencies).
On the other hand: if you find drivers with magnets with smaller Ø, higher frequencies will be reproduced by the driver without rolloff.
The good news regarding drivers - We've got more drivers to chose from, since a rise in frequency above fx 4 kHz won't be a problem due to the rolloff caused by the magnets fysical Ø.
Every part of the Walsh 5 driver other than the basket is custom designed specifically for that speaker. I don't think you'll find a pre-made commercial driver that can substitute and give you that kind of low frequency output while still playing smoothly above 2khz (from the BACK side).Ohighway said:
For a smaller speaker like this pencil, yes, there are a few (very few) drivers that work well. Most pre-made drivers simply won't make a decent "walsh type" speaker.
When I went to the Parts Express DIY event and heard lots of different speakers there, I didn't hear any that I thought could still sound good after comparing them to a Ohm Walsh 5 - but, they didn't play them loud enough to hear clearly, so I'm not sure.
I had a pair of Walsh 5 on loan for a couple years, they could play as loud as I could possibly want in a medium size room and still play clean and clear, and low.
When auditioning a bunch of $1,000 + (up to ~ $20,000) speakers, the only ones I liked better than the Walsh 5 were the Revel Salon Ultimas. But, for something like The Matrix soundtrack, the Ultimas fell flat - I don't understand why at all and wish I had a chance to borrow some Ultimas at home for a week or two.
More drivers to choose from? Maybe, but, most drivers are simply crap when you try to use them as a "Walsh" driver.Juspur said:
Also, I bet if you did blind listening tests with different brands of (same value) capacitors, you will not hear a difference?
Rolloff on pencil?
Hi Juspur, a few questions/comments about the measurements?
In making these comments I am assuming the measurements were made in the "horizontal Walsh mode" ... I would really like to know what angle your friend measured from.
The basic premise: the magnet/basket "shadow" is probably NOT the reason for the roll-off you are seeing at 2500Hz.
It is probably caused by measuring at an effective angle of about 60 degrees off axis relative to the back of the woofer cone. Please read on for further clarification.
1) Even with a "small" size profile, the incident angle of the microphone to the speaker cone would have an effect on the frequency response. Even a 1" tweeter shows substantial changes in FR vs angle, why not a cone that has a 2" backside?
(Yes it´s a 5" driver, but in Walsh mode you are looking at about a 2" radiating "dimension" on the cone, for al 360 degrees.)
2) It looks like the Peerless has a "cone angle" of about 30 degrees relative to the horizontal listening plane when in "Walsh mode"
3) if you want a measurement that compares reliably to the frontal "on axis mode", you would perhaps place the mic at an angle which is perpendicular to the middle of the back surface of the cone.
This would be about 60 degrees UP from the horizontal listening plane!
4) If you DID measure from the "horizontal"... then it was actually measuring the 60 deg off axis relative to the back of the cone
Now, what if we compare this theoretical "horizontal Walsh mode" response to the 60 degrees off axis of the front of the driver?
Let´s check the 60 degree FR on the Tymphany site at http://tymphany.com/files/products/pdf/832873.pdf
Hey, look! at 60 degrees off axis, the driver shows a roll-off right around 2500Hz... much like what your friend measured off the back of the cone.
sounds very similar to the results you quoted from your friend´s measurements, yes?
OK, some guesses and conclusions:
A) The roll off at 60 degrees measuring angle may be fairly close between front and back of cone. No way of knowing for sure without comparing at effectively identical angles, but your comments would indicate they (maybe) are the same?
B) The basket/magnet may cause some reflective peaks and dips in the response, but I don´t think there is any substantial filtering of high frequencies due to this.
C) It would be GREAT if your friend could measure at angles of perpendicular to the back of the cone, and also 30 deg from perpendicular, to see if the curves are fairly similar to the published front cone measurements.
SUGGESTIONS:
D) To get an even better idea of the "actual" roll-off characteristics of the back of the cone, you would want to average the response curves at Perpendicular, +/-15, +/- 30, +/- 45, +/- 60, +/- 75 and +/- 90 degree angles from the perpendicular. This would give you something closer to the true average response. And, according to Floyd Toole, the average response is highly meaningful.
E) If option E is too complicated, try measuring at 45 deg off of perpendicular and maybe at 30 deg off of perpendicular.
In practice, freq response at these angles usually comes ifairly close to the averaged output.
For a design such as this one, if I could only measure at one angle, I would probably go with 45 as being the one to use.
F) So, if you could get your friend to run a measurement at 15 degrees ABOVE the horizontal, and another at 30 deg above, the curves from those points would probably get you much closer to the "optimized" curve to use in your design.
AND.... Before you volunteer me to do this, please bear in mind I travel 90% of the time and have no access to measuring equipment.
So how about it? Any chance your friend could look at this?
Thanks for posting a fun and effective design.
Best regards, Jack
Hi Juspur, a few questions/comments about the measurements?
In making these comments I am assuming the measurements were made in the "horizontal Walsh mode" ... I would really like to know what angle your friend measured from.
The basic premise: the magnet/basket "shadow" is probably NOT the reason for the roll-off you are seeing at 2500Hz.
It is probably caused by measuring at an effective angle of about 60 degrees off axis relative to the back of the woofer cone. Please read on for further clarification.
1) Even with a "small" size profile, the incident angle of the microphone to the speaker cone would have an effect on the frequency response. Even a 1" tweeter shows substantial changes in FR vs angle, why not a cone that has a 2" backside?
(Yes it´s a 5" driver, but in Walsh mode you are looking at about a 2" radiating "dimension" on the cone, for al 360 degrees.)
2) It looks like the Peerless has a "cone angle" of about 30 degrees relative to the horizontal listening plane when in "Walsh mode"
3) if you want a measurement that compares reliably to the frontal "on axis mode", you would perhaps place the mic at an angle which is perpendicular to the middle of the back surface of the cone.
This would be about 60 degrees UP from the horizontal listening plane!
4) If you DID measure from the "horizontal"... then it was actually measuring the 60 deg off axis relative to the back of the cone
Now, what if we compare this theoretical "horizontal Walsh mode" response to the 60 degrees off axis of the front of the driver?
Let´s check the 60 degree FR on the Tymphany site at http://tymphany.com/files/products/pdf/832873.pdf
Hey, look! at 60 degrees off axis, the driver shows a roll-off right around 2500Hz... much like what your friend measured off the back of the cone.
sounds very similar to the results you quoted from your friend´s measurements, yes?
OK, some guesses and conclusions:
A) The roll off at 60 degrees measuring angle may be fairly close between front and back of cone. No way of knowing for sure without comparing at effectively identical angles, but your comments would indicate they (maybe) are the same?
B) The basket/magnet may cause some reflective peaks and dips in the response, but I don´t think there is any substantial filtering of high frequencies due to this.
C) It would be GREAT if your friend could measure at angles of perpendicular to the back of the cone, and also 30 deg from perpendicular, to see if the curves are fairly similar to the published front cone measurements.
SUGGESTIONS:
D) To get an even better idea of the "actual" roll-off characteristics of the back of the cone, you would want to average the response curves at Perpendicular, +/-15, +/- 30, +/- 45, +/- 60, +/- 75 and +/- 90 degree angles from the perpendicular. This would give you something closer to the true average response. And, according to Floyd Toole, the average response is highly meaningful.
E) If option E is too complicated, try measuring at 45 deg off of perpendicular and maybe at 30 deg off of perpendicular.
In practice, freq response at these angles usually comes ifairly close to the averaged output.
For a design such as this one, if I could only measure at one angle, I would probably go with 45 as being the one to use.
F) So, if you could get your friend to run a measurement at 15 degrees ABOVE the horizontal, and another at 30 deg above, the curves from those points would probably get you much closer to the "optimized" curve to use in your design.
AND.... Before you volunteer me to do this, please bear in mind I travel 90% of the time and have no access to measuring equipment.
So how about it? Any chance your friend could look at this?
Thanks for posting a fun and effective design.
Best regards, Jack
Hi Jack
I'm getting the point. If I measure at another angle, I'll get another response.
Hmmm... now there's the problem: With ordinary box-speakers, you simply measure on axis (which often is related to desired listening height, typically 90-95 cm above ground).
But - the way the pencil is built, we actually measure at (what Jack correctly points out) about 60 degrees off axis (when we're sitting at typical listening position). But that's how we're placed, related to the driver. It's a bit difficult to explain. Can you catch it?
So the big question is: do we want to measure the frequency, seen from listening position, or an odd angle related to the driver?
As said before, I understand what you mean - and yes - I'll try to get my friend to measure the speakers as you describe (a little patience necessary, however)
I plan to build another identical set of enclosures with the driver + tweeter mounted traditionally (with same x-over too). Could be fun with a direct comparison/measuring.
Have a nice day/afternoon/night
I'm getting the point. If I measure at another angle, I'll get another response.
Hmmm... now there's the problem: With ordinary box-speakers, you simply measure on axis (which often is related to desired listening height, typically 90-95 cm above ground).
But - the way the pencil is built, we actually measure at (what Jack correctly points out) about 60 degrees off axis (when we're sitting at typical listening position). But that's how we're placed, related to the driver. It's a bit difficult to explain. Can you catch it?
So the big question is: do we want to measure the frequency, seen from listening position, or an odd angle related to the driver?
As said before, I understand what you mean - and yes - I'll try to get my friend to measure the speakers as you describe (a little patience necessary, however)
I plan to build another identical set of enclosures with the driver + tweeter mounted traditionally (with same x-over too). Could be fun with a direct comparison/measuring.
Have a nice day/afternoon/night
Juspur said:
Well, if the brands only includes Mundorf Silver and Obligato, I'm in it for a beer
I would include whichever cheap brand non-electrolytic capacitor that is available in the same size.
I have a mix of 1uF, 2.2uF, 3.3uF, 3.9uF, 4.7uF, 10uF, and 20uF that I mix and match to get whatever size I need.
I've never known of a case where someone could actually (consistantly) hear a difference (without knowing which one they're listening to, or whether or not they've actually been switched (blind test)).
But knowing they've been changed, is, of course, possible to have a powerful bias even if there isn't actually a difference.
More Freq response considerations
Hi Juspur,
just a few more comments about the "design axis" for measurements.
Dr Floyd Toole performed much research which showed that for natural sounding reproduction the on axis response must be very flat, AND of equal importance, to design for an "average omnidirectional" response that is within about 2 dB of the on axis.
This is important for real room performance because for our ears to hear it as "real", the sound waves that are reflecting in the room need to be very coherent with the on axis response.
In other words, it is important to have the "average power response" and the on axis response both as close to flat as possible.
HBTW, Dr Toole´s research work got him appointed as the head of design for the Harman Group. The designs which most conform to his recomendations are the ones from Revel, and they are the subject of constant rave reviews. His conclusions are right on target.
For those of us that do not have anechoic test chambers and the like, it is well to note that in most cases, the frequency response of the driver at about 45degrees off axis compares fairly closely to the "average hemispherical power response", so if I had only one frequency response measurement to work with, i would select the 45 degree as the one which is most likely to predict how neutral the driver will be.
When measured from the front side of the cone at 45deg the Peerless unit you have selected performs very well, very flat up to about 2.8KHz and then about a 6dB/octave rolloff up to 5KHz where it transitions to a faster roll-off.
If the backside perpendicular response is anywhere close to this, we probably have a serious winner here!
Other comments:
My prediction is you will find that the "direct radiator" version of the Pencil will sound very forward, not nearly as good as the Pencil.
Why? Well, if you use these drivers "unfiltered" and mounted in a regular box they sound very good off axis... BUT they can sound very aggressive, shouty and forward on-axis.
This is because the front baffle creates a gradual 6dB on axis lift.
This is not theory, I already built a couple of these using the same Peerless drivers... and unfiltered they sound very aggressive when front baffle loaded in a small profile box.
To compensate for this many designers use baffle lift compensation or add in parallel a 2nd driver which only operates in the bass range, but this usually throws the "Dr Toole" parameters out of balance.
Another option is to use two of these in "bi-pole" configuration ... which I also tried, and it sounds superb off axis and in the room, but it still sounds too forward on axis and on overall balance.
(I ended up applying about a 2 dB shelf filter from about 700Hz using a Behringer DigitalEqualizer, and it sounds very, vey good, but this requires much more complexity... the Behringer sounds OK in digital mode, but the analog sections sounds bad... so i use it in digital mode only, with an external DAC... and yes, my system now sounds really good...)
but what i really want is to use it in minimalist configuration, no filters, no Behringer, no DAC... just a direct triode amplifier... so, how to do this?
Well, that's where your Pencil idea is just the ticket.
In the case of the Pencil, having the driver mounted in "Walsh mode" brilliantly solves the baffle lift effect.
Yes, the baffle lift still exists, but now instead of having it pointed at the listener, it´s pointing up at the ceiling!
And as I mentioned, the Peerless sounds really great unfiltered when listening "off axis".
Having it mounted in Walsh mode probably brings much more freedom from that dreaded baffle lift "shoutiness". It really fits the "DrToole" criteria much better than the direct front radiation mode does.
Given my prior experimentation with this driver, I can predict that it would sound extremely balanced and natural in the Walsh mode, provided the basket/magnet assembly doesn´t create too many nasty aberrations in the response.
So, I can't wait to get home and find out.... AND, unfortunately, that will have to wait until Christmas. I would love to do this testing myself, but I am traveling almost all of the time and am far away from my system and my test equipment.
As you can see, I have more than a passive interest in knowing if the back of the cone response is close to that of the front !!
If it IS, identical, then we have a truly excellent driver for mounting in "Walsh mode".
Also, to placate all you Walsh purists, I harbor no illusions that this is really close to a Walsh driver, it really is not the same thing.
However, for this application, the way Juspur integrated it with a tweeter, no filters on the mid-woofer, it is a brilliant idea.
So, Juspur, having come this far, my suggestions would be that you concentrate on refining what is already an excellent idea.
I would advise you to not waste your time building the "front radiator" version, I already did and it is seriously flawed due to the baffle lift.
My suggestions for refining the design:
1) take the perpendicular to the back of the cone as your "optimized frequency response" and integrate the tweeter with that.
2) Then, start playing with absorptive surfaces on the basket and the magnet to reduce reflections.
3) Then "time-align" the tweeter dome by bringing it close to the center line.
4)Play with crossing the tweeter in at a lower frequency, probably between 3.5Khz to 5KHz
5) Check Martin King´s quarter-wave.com site for ideas to build these as T-lines and maybe extend their bass response... looks like 35Hz is feasible.
My bet is that this will yield superlative performance. Please let us all know what you find! ( I will really be looking forward to Christmas this year!)
I have some other ideas for approaches similar to the Pencil, involving cylindrical cabinets and Mass Loaded Quarter Wave T lines, but those will have to wait until Xmas.... sigh.
Great Thread, please keep it going!
Hi Juspur,
just a few more comments about the "design axis" for measurements.
Dr Floyd Toole performed much research which showed that for natural sounding reproduction the on axis response must be very flat, AND of equal importance, to design for an "average omnidirectional" response that is within about 2 dB of the on axis.
This is important for real room performance because for our ears to hear it as "real", the sound waves that are reflecting in the room need to be very coherent with the on axis response.
In other words, it is important to have the "average power response" and the on axis response both as close to flat as possible.
HBTW, Dr Toole´s research work got him appointed as the head of design for the Harman Group. The designs which most conform to his recomendations are the ones from Revel, and they are the subject of constant rave reviews. His conclusions are right on target.
For those of us that do not have anechoic test chambers and the like, it is well to note that in most cases, the frequency response of the driver at about 45degrees off axis compares fairly closely to the "average hemispherical power response", so if I had only one frequency response measurement to work with, i would select the 45 degree as the one which is most likely to predict how neutral the driver will be.
When measured from the front side of the cone at 45deg the Peerless unit you have selected performs very well, very flat up to about 2.8KHz and then about a 6dB/octave rolloff up to 5KHz where it transitions to a faster roll-off.
If the backside perpendicular response is anywhere close to this, we probably have a serious winner here!
Other comments:
My prediction is you will find that the "direct radiator" version of the Pencil will sound very forward, not nearly as good as the Pencil.
Why? Well, if you use these drivers "unfiltered" and mounted in a regular box they sound very good off axis... BUT they can sound very aggressive, shouty and forward on-axis.
This is because the front baffle creates a gradual 6dB on axis lift.
This is not theory, I already built a couple of these using the same Peerless drivers... and unfiltered they sound very aggressive when front baffle loaded in a small profile box.
To compensate for this many designers use baffle lift compensation or add in parallel a 2nd driver which only operates in the bass range, but this usually throws the "Dr Toole" parameters out of balance.
Another option is to use two of these in "bi-pole" configuration ... which I also tried, and it sounds superb off axis and in the room, but it still sounds too forward on axis and on overall balance.
(I ended up applying about a 2 dB shelf filter from about 700Hz using a Behringer DigitalEqualizer, and it sounds very, vey good, but this requires much more complexity... the Behringer sounds OK in digital mode, but the analog sections sounds bad... so i use it in digital mode only, with an external DAC... and yes, my system now sounds really good...)
but what i really want is to use it in minimalist configuration, no filters, no Behringer, no DAC... just a direct triode amplifier... so, how to do this?
Well, that's where your Pencil idea is just the ticket.
In the case of the Pencil, having the driver mounted in "Walsh mode" brilliantly solves the baffle lift effect.
Yes, the baffle lift still exists, but now instead of having it pointed at the listener, it´s pointing up at the ceiling!
And as I mentioned, the Peerless sounds really great unfiltered when listening "off axis".
Having it mounted in Walsh mode probably brings much more freedom from that dreaded baffle lift "shoutiness". It really fits the "DrToole" criteria much better than the direct front radiation mode does.
Given my prior experimentation with this driver, I can predict that it would sound extremely balanced and natural in the Walsh mode, provided the basket/magnet assembly doesn´t create too many nasty aberrations in the response.
So, I can't wait to get home and find out.... AND, unfortunately, that will have to wait until Christmas. I would love to do this testing myself, but I am traveling almost all of the time and am far away from my system and my test equipment.
As you can see, I have more than a passive interest in knowing if the back of the cone response is close to that of the front !!
If it IS, identical, then we have a truly excellent driver for mounting in "Walsh mode".
Also, to placate all you Walsh purists, I harbor no illusions that this is really close to a Walsh driver, it really is not the same thing.
However, for this application, the way Juspur integrated it with a tweeter, no filters on the mid-woofer, it is a brilliant idea.
So, Juspur, having come this far, my suggestions would be that you concentrate on refining what is already an excellent idea.
I would advise you to not waste your time building the "front radiator" version, I already did and it is seriously flawed due to the baffle lift.
My suggestions for refining the design:
1) take the perpendicular to the back of the cone as your "optimized frequency response" and integrate the tweeter with that.
2) Then, start playing with absorptive surfaces on the basket and the magnet to reduce reflections.
3) Then "time-align" the tweeter dome by bringing it close to the center line.
4)Play with crossing the tweeter in at a lower frequency, probably between 3.5Khz to 5KHz
5) Check Martin King´s quarter-wave.com site for ideas to build these as T-lines and maybe extend their bass response... looks like 35Hz is feasible.
My bet is that this will yield superlative performance. Please let us all know what you find! ( I will really be looking forward to Christmas this year!)
I have some other ideas for approaches similar to the Pencil, involving cylindrical cabinets and Mass Loaded Quarter Wave T lines, but those will have to wait until Xmas.... sigh.
Great Thread, please keep it going!
critofur said:
Amen brother, and that goes beyond capacitors too.....
I have some folks I know who claim HUGE improvements in their system based on changes they make. EVERY change results in a HUGE improvement according to them. Any time I humbly suggest a blind test they either pretend not to hear me, or give me "that look".
Well, speaking for my self, not every change is an improvement - I've tried expensive high-end fuses (paid for 
) with no improvement - and gosh - I REALLY wanted an improvement. The same with tube dampers - positive effect on one amp, negative on another.
So, basically I've learned that THE THING is looking at the equipment as a chain. You can change items, but it doesn't necessarily improve. Expensive isn't necessarily better.
... And sometimes the improvement is in one field, while degrading occurs another place...
I fully respect if people can/cannot hear changes. The truth is between your own ears, simply
) with no improvement - and gosh - I REALLY wanted an improvement. The same with tube dampers - positive effect on one amp, negative on another.So, basically I've learned that THE THING is looking at the equipment as a chain. You can change items, but it doesn't necessarily improve. Expensive isn't necessarily better.
... And sometimes the improvement is in one field, while degrading occurs another place...
I fully respect if people can/cannot hear changes. The truth is between your own ears, simply
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.