Beyond the Ariel

I had a similar setup some 25 years ago: Iwata 600 horn and Altec 808-8B with 909-16A diaphragm, crossed over at 1.2 kHz. The plan was to use an Altec 414, but because it was unobtainable, I settled on a Gauss 3181D. Subwoofer was initially a single JBL LE15A, this was quickly replaced by four 10" subwoofers. The in-room balance was never fully satisfying with the 600 horn. This improved a lot when switching to an Iwata 300 horn and a cross-over frequency of 700Hz.

My number one suggestion is to use the Autotech Iwata 300 horn instead of the 600.
I was able to compare all diaphragms for the Altec 802 (the 808 only differs in standard diaphragm) and came to the same conclusion as most of my acquaintances: a preference for the 909-16A diaphragm (Note: impedance is 12.5 ohm in the operating range).
This compression driver is an exception between the 1" drivers in being very good below 1.6 kHz. The drawback is that is a bit less perfect above 9 kHz, so a super-tweeter can improve this range especially with the narrow dispersion of the 300 horn in the upper 2 octaves.

A second suggestion is to do the 2-way with the 416. (The 414 is not the easiest driver to fit dynamically with the other drivers). The larger magnet size of the 604 makes it an even better dynamic fit to the 802, but I have no idea if GPA would make this available as woofer only.

A third suggestion would be to go for the Altec 288 instead of the 802. But this suggestion makes a smaller sound difference than the 2 previous, so it might not be worth the extra budget.

Perhaps there is too much choice, and most of the drivers mentioned since the blog started are now unobtainium and were never cheap or easy to buy.

Thank you for your useful advice on the Fostex tweeters, but I have decided on balance not to use drivers that need a supertweeter support. An amplifier with a super fast rise time will easily put the life into the leading edge transients. A solid state amp can deliver this, a low power DHT amp with a transformer output will struggle. It is here that a Fostex or other tweeter may partly help. I accept however, Pano's comment that they tend to make an unnatural sound, but this may be preferred by some, especially those with loss of treble hearing with the years.
 
Perhaps there is too much choice, and most of the drivers mentioned since the blog started are now unobtainium and were never cheap or easy to buy.
Correct, and this for a thread that started partially because the Ariel drivers were no longer available.

It is here that a Fostex or other tweeter may partly help. I accept however, Pano's comment that they tend to make an unnatural sound, but this may be preferred by some, especially those with loss of treble hearing with the years.
There are indeed too many examples out there where there is an unnatural contribution from the supertweeters. However, this does not need to be. Unfortunately the options in tweeters are so very limited (if good efficiency and dynamic fit is needed) and good integration is not easy: a few cm in height and depth make a difference and every change in position needs a different filter, so it is easy to give up and conclude it can't be done.
Nearly everyone who heard my system asks why I have those tweeters 'since they are not playing at all'. Then when I disconnect the tweeters they are so suprised what these were contributing. Apparently people are so used hearing bad integration in multi-way systems that when they hear good integration coming from a single apparent source they truly believe it has to come from a single driver.
I do agree it is much less effort leaving the supertweeters out, and satisfying results are possible (I did not have them in my system for 2 years) but when correctly chosen and well integrated they are more natural and lifelike than any 1" or 1.4" compression driver can achieve in the upper octave (probably until Materion makes a Be ring diaphragm for the BMS 4552).
 
Member
Joined 2008
Paid Member
Correct, and this for a thread that started partially because the Ariel drivers were no longer available.


There are indeed too many examples out there where there is an unnatural contribution from the supertweeters. However, this does not need to be. Unfortunately the options in tweeters are so very limited (if good efficiency and dynamic fit is needed) and good integration is not easy: a few cm in height and depth make a difference and every change in position needs a different filter, so it is easy to give up and conclude it can't be done.
Nearly everyone who heard my system asks why I have those tweeters 'since they are not playing at all'. Then when I disconnect the tweeters they are so suprised what these were contributing. Apparently people are so used hearing bad integration in multi-way systems that when they hear good integration coming from a single apparent source they truly believe it has to come from a single driver.
I do agree it is much less effort leaving the supertweeters out, and satisfying results are possible (I did not have them in my system for 2 years) but when correctly chosen and well integrated they are more natural and lifelike than any 1" or 1.4" compression driver can achieve in the upper octave (probably until Materion makes a Be ring diaphragm for the BMS 4552).

Hi Peter,
How about showing us a picture ??

I also ask because I have my tweeters added in above a large round horn in such a way that most would say should not work, but they actually do !
 
Correct, and this for a thread that started partially because the Ariel drivers were no longer available.


There are indeed too many examples out there where there is an unnatural contribution from the supertweeters. However, this does not need to be. Unfortunately the options in tweeters are so very limited (if good efficiency and dynamic fit is needed) and good integration is not easy: a few cm in height and depth make a difference and every change in position needs a different filter, so it is easy to give up and conclude it can't be done.
Nearly everyone who heard my system asks why I have those tweeters 'since they are not playing at all'. Then when I disconnect the tweeters they are so suprised what these were contributing. Apparently people are so used hearing bad integration in multi-way systems that when they hear good integration coming from a single apparent source they truly believe it has to come from a single driver.
I do agree it is much less effort leaving the supertweeters out, and satisfying results are possible (I did not have them in my system for 2 years) but when correctly chosen and well integrated they are more natural and lifelike than any 1" or 1.4" compression driver can achieve in the upper octave (probably until Materion makes a Be ring diaphragm for the BMS 4552).

Clearly you have had success with your own projects. But if I was going to build another speaker set, I would use drivers that do not require a super tweeter. I would put the money into the mid and tweeter to do that, and then add my woofer choice perhaps the 414 or 416

CD could easily be designed to preclude any requirement of a supertweeter for certainly an 18mm or 25.4mm throat. We know too well that Fostex design their drivers to integrate with their own designed packages, so if used with other drivers the matching the unusual FR's can demand too much in the cross over to get a decent sound. Marc Gea has a fully Fostex based system and that works well for him and he is pleased with it. I like both what he has done and your good self, but I would again avoid the supertweeter by using other products
 
Last edited:
As a Mac fan going back to the Mac Plus (back when I was tech writer at Tektronix), I've enjoyed the platform's strengths in TCP/IP connectivity, desktop publishing, digital photography, digital audio workstations (DAWs), and straightforward interfaces to the Unix community.

But ... I have to be honest. The Mac has never been strong for desktop engineering applications, going as far back as MS-DOS 3.1, the ISA bus running at 8MHz, and the joys of managing low-memory I/O ports and IRQ interrupts. That has been a PC specialty from the beginning, with broadest varieties of measurement gizmos under the sun. Windows 7 has the merit of continued support from Microsoft for several more years, and can interface with just about anything.

ASIO drivers are the recognized driver-of-choice for professional studio applications in DAW workstations supporting sample rates right through DXD (a 352/32 PCM <-> DSD editing format). 96/24 PCM is now considered the basic-quality studio format and nothing very special. Fortunately for us, 96/24 covers just about any measurement need for loudspeakers or amplifiers.

The main thing to look for in a measurement ADC/DAC is a lowpass filter at 40 kHz (look for S/N specs that extend to 40 kHz), stable, up-to-date drivers, and a good interface hardware interface to the PC. The FireWire interface has gotten dodgy over the years, leaving USB 2.0 as the most reliable choice. If a PC can't support USB, you have a hardware problem, or a buggy driver. Even the best hardware is useless if the drivers are no good.

In other words, use the right tool for the job. When it comes to instrumentation, measurement, or signal bending, Windows 7 PC's are the strongest platform. Driver hassles, for better or worse, is just the price of admission.
 
Last edited:
A second suggestion is to do the 2-way with the 416. (The 414 is not the easiest driver to fit dynamically with the other drivers). The larger magnet size of the 604 makes it an even better dynamic fit to the 802, but I have no idea if GPA would make this available as woofer only.

The GPA 515 is essentially that. The magnet is the same as in the 604, and I believe the cone and surround are as well. Not sure about the spider, but you could ask Bill easily enough.

I found the 416 easier to mate with a compression driver. The 515's larger magnet lowers Q enough to result in a rising response, not the ideal for my situation, but a good choice for a bass horn, where it can work against the mass roll off.

Gary Dahl
 
If I was working as a graphic designer I would probably have a Mac and for some other creative endeavors. For CAD software until recently it has all been either Windows of some sort or at a much higher level it has been some form of Linux or Unix. Apple is really a BSD derivative so really a unix based system from the beginning. My Clio system and the earlier SysId system from Bell Labs were only windows based so you had not choice. What errks me about Mac has been some of the prices charged for repairs if it hasn't been something under warranty. I've worked on my daughters 27" Imac, one of the nicest screens around, but when a hard drive dies and they want almost $400.00 to replace a $60.00 Seagate drive that would have a 3 to 5 year warranty in windows but has only a one year warranty because it was sold to Apple that is BS. Of course I fixed the computer myself, I wasn't that crazy to pay them that kind of money. The crazy part is that over a year later they offered to replace those same drives for free as they realized so many were defective but I had already taken care of that. I've fixed someone else earlier Imac when all the caps on the motherboard swelled and failed and Apple denied the problem for a long time before admitting there was a problem. Unremoveable batteries in Iphones, and things like that just make me look and keep my money in my pocket. I'll give Apple the two thumbs up on industrial design but not the cost value side of things. I would buy one of the 27" screens sans computer though, you can't beat Apple's screen quality. I happen to have a high end HP Envy laptop with 17" screen and haven't had any problems besides the fact the usb ports are getting funky, not sure how much of a pain it will be to have those connectors changed but I don't know if I will attempt to do that myself. It would be nice with the Apple screen quality and as lite as an Apple but it has been a rock solid computer. Of course I can't really stand the Synaptics track pad but I always used an external track ball, that is why my usb ports have taken so much abuse. To each there own, but I'll stick with Windows based machines and 3 out of 4 people seem to agree with that idea as far as sales go.
 
I only use XP for legacy devices so wonder why anyone is still using that OS for any other reason? For the fact there is no real security on that OS why would you still run that today, it seems more than a bit strange to stick with a legacy OS when security today is so important and there are no updates for critical application if you are using that on a daily basis. I'm no fan of Windows 10 and will stay with Win7 as long as it is still supported but after that I will just have to move over to Linux of one flavor or another when Win7 is no longer supported. Forget about Win10, not going to a cloud connected OS if I can help it. Put your stuff on the cloud and you have lost all control to hackers, can't understand how IT people can be so blind to those problems today.
 
Win7 is just nice as it loads drivers so much better than XP ever did. I don't use Win8 and never saw any reason to move to that at all. I run security software on my XP install but know it is a dying OS and doesn't support so many newer hardware pieces anymore. As long as it works and you have some sort of security software, not MS, that is updated, then I guess you are still okay.
 
The GPA 515 is essentially that. The magnet is the same as in the 604, and I believe the cone and surround are as well. Not sure about the spider, but you could ask Bill easily enough.

I found the 416 easier to mate with a compression driver. The 515's larger magnet lowers Q enough to result in a rising response, not the ideal for my situation, but a good choice for a bass horn, where it can work against the mass roll off.

Gary Dahl
Thanks Gary,
I could only compare the 604 Alnico woofer section and 416 Alnico a long time ago.
The rising response of the 515 v 416 can be flattened with a series inductor, the Q can be increased with a series resistor. If both are needed, an inductor with small diameter wire can be used to keep the cost down. If passive filtering is used, this can be integrated in the filter design.
I preferred the 604 woofer section with the extra resistor and inductor over the 416. But this was in a direct comparison and I can't say if it would be worth the experiment.
 
I preferred the 604 woofer section with the extra resistor and inductor over the 416.

I find this quite surprising. I didn't attempt to alter the response of the 515 to match the 416; it was an obvious possibility but didn't seem worthwhile, since doing so would throw away the 515's additional efficiency.

If your comparison was between Altec units, there may have been more differences between the drivers than there were between the GPA's that I was working with. Altec woofers used many different cones over the years, not to mention the variations present in all the other details. My GPA 416 and 515 differed only in magnet size (and vc impedance); they were otherwise the same, according to Bill.

Interesting stuff!

Gary Dahl
 
Member
Joined 2004
Paid Member
Hi all,

one thing that I noticed in this wonderfully informative and thought-provoking thread is a rather puzzling dearth of actual complete build examples, supported by actual measurements.

In an attempt to buck that trend, I hereby offer my own design for scrutiny, which, even if not fully "beyond the Ariel", still by and large shares the same fundamental principles.

The same compliments as above apply here.

Interestingly I started with a similar 6th order low pass filter on the woofer, but this created too much delay in relation to the shorter front-to-back physical offset of the AH425 horn. My LPF is 4th order now.
 
Last edited:
I've got a question for Gary, Lynn and everyone else out there who have experience with the 416-8B.

I know at least Gary and Lynn are big fans of classical but I'm not so how do these driver fair on other types of music? I listen to all kinds and want a midwoofer that can deliver the goods on everything from heavy metal, pop, blues and jazz. I'm sick and tired of, and having lived too long with, speakers that sound great on some Chesky recordings but that fall apart when Maiden, Dio or GnR gets cued up (previous Maggies best case in point). I want drivers that allow me to rock out and dance most of the time while still sounding amazing on Thessink, Barber, Cassidy or Mo.

Spare me the "best drivers give you what's on the CD mate and if that sounds crap well then it's crap" rant, please. I'm getting too old for the objective camp me thinks... :)
 
Last edited: