That's not what I figured you did at all. But I really liked the original too, and I spent a lot of time on it.
I'm more interested in something like this these days:
The crossover is another matter. I say that in a sharing way. It could be metal drivers.
I'm more interested in something like this these days:
An externally hosted image should be here but it was not working when we last tested it.
The crossover is another matter. I say that in a sharing way. It could be metal drivers.
That is a bad start. The AMT is not particular extended in the lower reaches plus the AMT is very much forward with acoustic center. Actually i newer herad an DÁppolito that satisfied me in terms of imaging. That does not say that the therory does not work but you really need a low crossover of a particular type ( phase quadrature ) and a very compact and time aligned physical design. That cheap paper cones do not cut the mustard against the high resolution AMTs.
This had the correct physical design :
Dunlavy Audio Labs Signature SC-VI loudspeaker | Stereophile.com
Unfortunately Dunlavy used the cheapest drivers he could find so they did not impress me eather. Great measurements though.
Dunlavy Audio Labs Signature SC-VI loudspeaker | Stereophile.com
Unfortunately Dunlavy used the cheapest drivers he could find so they did not impress me eather. Great measurements though.
That is a load of hogwash.
I hate it when people give silly advice, so that is why I feel the need to react so strongly to it. Please, fellow speaker builders, mix and mash as much as you want, if that gives you the drivers best suited to the combination you are working on.
To be more precise:
- there are paper drivers that can be used full range without falling apart.
- soft domes usually have no apparent break down, and certainly not as low as 12 KHz.
- what is this thing about Q's matching? Q is related to damping, that much you seem to know, but you place it completely out of context here. What is there to match?
- if there is something like an 'acoustic signature' in a driver, that means it is either not very well made, or not very well applied. The material of which it is made has little to do with it.
In short, use the drivers that best suit your specific application, and consider the material they are made of as a completely secondary consideration.
Last edited:
I have plenty of practical experience with SEAS metal drivers, here's a few notes:
22TAF/G My tweeter of choice. Despite a fr of 1100hz it will crossover very low. I note the Anima is 1900hz - mine is 1750 2nd order electrical and shows no sign of strain. Sound is very smooth, airy and detailed.
27TBFC/G Very low distortion and easy response to control. But to my ears strings are very 'scratchy' and cymbals flat and lifeless. I dumped these for the 22TAF/G.
L18 The 7k peak is easily controlled and the response is smooth. The distortion peak is high compared to other metal drivers at 2.3Khz but is only -35db - not so good.
L22 Said to be no good in the midrange due to the distortion peak being at 1.4Khz, however it's fairly broad and at a manageable -50db. What I've found in practice is you can't hear it. I'm using these now in preference to the L18 - they have the kind of dynamics and 'life' that makes the L18 sound too polite. The notch needs careful matching as it contributes significantly to driver roll-off.
22TAF/G My tweeter of choice. Despite a fr of 1100hz it will crossover very low. I note the Anima is 1900hz - mine is 1750 2nd order electrical and shows no sign of strain. Sound is very smooth, airy and detailed.
27TBFC/G Very low distortion and easy response to control. But to my ears strings are very 'scratchy' and cymbals flat and lifeless. I dumped these for the 22TAF/G.
L18 The 7k peak is easily controlled and the response is smooth. The distortion peak is high compared to other metal drivers at 2.3Khz but is only -35db - not so good.
L22 Said to be no good in the midrange due to the distortion peak being at 1.4Khz, however it's fairly broad and at a manageable -50db. What I've found in practice is you can't hear it. I'm using these now in preference to the L18 - they have the kind of dynamics and 'life' that makes the L18 sound too polite. The notch needs careful matching as it contributes significantly to driver roll-off.
AL130
I had the chance to listen to a pair of D`Appolito speakers using the AL130 and have listened to several speakers using the Seas magnesium and aluminium cones. Its going to be like comparing peaches to oranges as crossovers have been different, but it seemed to sound more like the Seas aluminium drivers, I think it lacked slightly behind the Excel cones. Seems to use an overhung design and has quite high inductance for a 13cm driver. I would be very interested to see a real world measurement of those aluminium drivers but would go for the Seas or Daytons if I had to choose a hard cone.
I`ve never heard a metal cone speaker ( including some Accutons ) that is able to reproduce the human voice in a natural way, they all seem a bit too bright for my taste
I had the chance to listen to a pair of D`Appolito speakers using the AL130 and have listened to several speakers using the Seas magnesium and aluminium cones. Its going to be like comparing peaches to oranges as crossovers have been different, but it seemed to sound more like the Seas aluminium drivers, I think it lacked slightly behind the Excel cones. Seems to use an overhung design and has quite high inductance for a 13cm driver. I would be very interested to see a real world measurement of those aluminium drivers but would go for the Seas or Daytons if I had to choose a hard cone.
I`ve never heard a metal cone speaker ( including some Accutons ) that is able to reproduce the human voice in a natural way, they all seem a bit too bright for my taste
Last edited:
I'm quite new when it comes to speaker design, but the ER18RNX displays a noticeable drop in SPL by about 60hz (and apparently starting to roll off even higher). Would you guys recommend using a third driver for low frequencies (maybe an 8") in a 3-way setup to compensate? Maybe a subwoofer cross-over at ~80hz would be sufficient?
Then again, is there a 6.5" that would be more capable of reproducing lower frequencies for a lower subwoofer cross? Much music does appear to have material below 60hz.
Thanks!
The Usher 8945a has some of the cleanest measured distortion and smoothest midrange response of any 6.5" driver. Its a great deal compared to SS and Seas.
Agree. I don't believe in D'Appolito approach either. Even worse, I don't believe in paralleled woofers unless it is very low in the frequency. It exactly reminds me of an amp with paralleled output stage
I don't know why the OP ask about 6.5 inches aluminum driver. Such cone is difficult to work with. One reason to use it is because the reproduced details will outperform those from paper cones.
Then the task is to make it to sound as natural as possible (with vocals for example). This most probably will require complex filtering that will kill low level reverberation (I may use the wrong terminology here
). Then the implication is very clear for me. There MUST be a third driver that will handle the lower frequency. So at least a 3-way is mandatory for such driver.Well there are high end speakers that is only 2-way (Magico mini for example). But I believe that such is possible because the driver itself is so special that it doesn't need complex filtering.
Scanspeak 18W/8535, 18W/8531 can both do well below 60Hz in a properly designed bass-reflex enclosure with the second working better in a larger than suggested by simulation software enclosure, depending on tuning - 34-38l. I run the 4 ohm version in 34l and port tunning of 35Hz with excellent results. The Usher drvers seem very close ( will abstain of using the word "copy" ) to the Scanspeak Classic series. SB17NRXC35 is another option that you may want to look into. Those are all paper drivers, will save you a few bucks on the crossover and you could get it with less hassle. There`s plenty of DIY projects available too.
I see you`re in the States - ring Madisound and see if this offer is still available
Last edited:
The biggest problem when using aluminum cones in the midrange is the amplification of the motor induced midrange distortion by the inherent resonance peak. E. g. a peak of 10 dB at 5 kHz will amplify the F2 at 2.5 kHz by 10 dB, F3 at 1.7 kHz, and so on. You can see a high F3 with many aluminum drivers betweent 1.5 kHz and 2 kHz, depending on the frequency of the resonance. And the worst part is: you can't filter it out. You can easily filter the resonance itself, but this filtering will not affect the distortions, because they are created AFTER the filter (in reality, because of the acoustical/mechanical/electrical coupling, there will be an effect).
There are only two ways to avoid these distortions:
- keep motor distortions low, e. g. a shorting ring
- get the resonance peak high enough in frequency, I would say 7 kHz, so F3 peaks at 2.3 kHz, which is usually out-of-band for a 6.5" midrange.
For best results, combine both measures.
Paper cones have similar problems, but at a lower frequency (the resonance peak is lower in frequency), and with less level (resonance peak is lower in level), but with a higher bandwidth. Again, reduce motor distortion.
There are only two ways to avoid these distortions:
- keep motor distortions low, e. g. a shorting ring
- get the resonance peak high enough in frequency, I would say 7 kHz, so F3 peaks at 2.3 kHz, which is usually out-of-band for a 6.5" midrange.
For best results, combine both measures.
Paper cones have similar problems, but at a lower frequency (the resonance peak is lower in frequency), and with less level (resonance peak is lower in level), but with a higher bandwidth. Again, reduce motor distortion.
Very well said above. I think that crossover suppression should be the last resort, some sort of cone treatment may help reduce the peaks. Accuton seem to have succeeded with the dual black rings they attach on the cone sides. I`ve thought of adding some very light bitum stripes on the back of the hard cones, never tested it however, no idea if it would work.
agreed. The points which baseballbat makes are the very reasons I opted for Visaton AL130 rather than the AL170. Both have shorting rings, and from what Ive read, good THD measures. The breakup of the 130 occurs at 7k, giving me just enough margin to avoid the worst of it. The 170 would be much more difficult.
With the caveat of vacuphile's kind observation that I speak complete hogwash at times, and er, thankyou man, I have wondered if a notch on the metal bass is the best approach.
It's not really necessary. Let's do some steep filtering on the bass, just like Celestion in the golden age of hifi. Not hard.
As a theoretician, I really couldn't say if this will sound any good, but it lines up VERY NICELY INDEED in a 10L standmounter with the tweeter acoustic centre set back about 3 cms using Visaton's AL130.
BTW, I wasn't suggesting a Mundorf AMT tweeter was my particular bag. But the D'Appolito MTM array has some very nice features. Though it seems everybody just HATES the parallel wired bass proposition. Well we could fix that. Any use mondo?
It's not really necessary. Let's do some steep filtering on the bass, just like Celestion in the golden age of hifi. Not hard.
As a theoretician, I really couldn't say if this will sound any good, but it lines up VERY NICELY INDEED in a 10L standmounter with the tweeter acoustic centre set back about 3 cms using Visaton's AL130.
BTW, I wasn't suggesting a Mundorf AMT tweeter was my particular bag. But the D'Appolito MTM array has some very nice features. Though it seems everybody just HATES the parallel wired bass proposition. Well we could fix that. Any use mondo?
Attachments
haha interesting Steve. Im happy with my current filters, but it might be fun to try, if I have the coils lying around. 3cm is a large offset though, not sure about that. XO point is about the same as mine, similar polar bloom (g20sc would match better higher Ive learned since my build with those drivers). I got good boxsim results with an elliptical filter, phase is tricky then however.
Im not sure about L1 in the tweeter circuit, that short really needs a potential divider I reckon. Just like the double woofers might warrant?
Im not sure about L1 in the tweeter circuit, that short really needs a potential divider I reckon. Just like the double woofers might warrant?
Last edited:
Oh, all I was trying to convey with the 3rd order 12uF possibility, was that I can never quite decide if low distortion at tweeter Fs or good phase alignment is better. Second order tweeter just works better theoretically.
But time alignment is essential IMO. Joachim didn't do the sloped baffle for nothing.
But time alignment is essential IMO. Joachim didn't do the sloped baffle for nothing.
But time alignment is essential IMO. Joachim didn't do the sloped baffle for nothing.
Sloped baffle is usually a sign that the design has narrow directivity, limited sweet spot, only sound nice from certain position between the boxes
Hi all,
Lots of talk going around.
GOWA, I noticed you never reported back, are you really interested?!
Or it's just a random question.
If you are interested in building, what type of speaker are you working on, hope is not "state secret" like other members that have similar secretive business questions...
Or, is it for a mid driver, or as a mid-bass, you have something on to explain and feed back. You need to explain what enclosure size/int. volume you are working on and lots of more details you are not mentioning (SPL is important or not, LF extension if for bass, the type of system BR, sealed, horn, OB...). Without that how do you want us to guess what you have in mind.
Last edited:
To my knowledge the 6.5" aluminum driver with the highest in frequency peaks is the Seas L18RNX/P H1224-08 L18RNX/P
Very impressive, and I believe that this is close to the best result. Improvements may be still possible with other/thicker cone material, but at a much higher price and/or lower efficiency. And I don't think that it will get much better, because the curvature of the cone seems to be very close to the optimum.
The efficiency of this driver is also not bad for a hifi midwoofer (88 dB/W). Unfortunately, the demodulation rings are missing (or does he have them?), I would love to see a demodulated version of this driver.
Very impressive, and I believe that this is close to the best result. Improvements may be still possible with other/thicker cone material, but at a much higher price and/or lower efficiency. And I don't think that it will get much better, because the curvature of the cone seems to be very close to the optimum.
The efficiency of this driver is also not bad for a hifi midwoofer (88 dB/W). Unfortunately, the demodulation rings are missing (or does he have them?), I would love to see a demodulated version of this driver.
Hi, very similar to Joachim Gerhard Studio One's. Is it the same as in the Animas?
Welcome to Spiral Groove
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.