Jump to content

Comments on building something close to the DQ-10


Pete B

Recommended Posts

I posted this to another list when someone asked about building something similar or as close as possible to the DQ-10, for what it's worth:

It's interesting that there are sites for several of the classic speakers,

Dyna A-25, Spica T-50, LS3/5a, and so on but I've not found anything for the

DQ-10s. Perhaps I'll add a brief page to my site.

Here's the crossover schematic:

_http://www.arsenal.net/dc/user_files/205.jpg_

(http://www.arsenal.net/dc/user_files/205.jpg)

I think you could get close copying from pictures and advertised physical

specs, and with a bit more information on the original drivers.

The woofer is from the large Advent, I would think that you could buy a pair

used, there are many on ebay, and prototype something by just laying the

Advent on it's side. However, the Advent woofer is very good but not as low in

distortion as some of the better acoustic suspension drivers. This

distortion is heard as the bass thickening up with increased level.

The AD5060 mid bass is no longer made, I've used it before and I have 1 of

the SQ8 version and 2 of the W8 version. This was a very popular midbass back

in the 70-80s but I often saw them blown, in Infinity towers, and one in my

own first order 4 way. Many comment that the performance of this driver is

not that good and I'd think that it would be easy to find a better substitute.

I'd rather see a smooth 6.5" here to get a bit more VD, something like a

SEAS CB17RCY or Vifa C17WG-69-08 if one wants to stay with a paper cone.

Don't have specs for the .75" or 1.5" dome and they're no longer available. Kevin's (planerguy) comments are very

helpful, Audax has one, as does Scan Speak, and there are several 2" domes

available. I actually prefer a 2" for thermal capacity and more VD since the

crossover slopes are shallow.

The .75" dome seems to be somewhat unusual as Kevin describes it, but it

seems to me that no super tweeter should be needed if a good modern .75" dome is

used.

Still might be interesting to experiment with a ribbon, and on the other

hand don't see why a good ribbon can't replace the .75 and the old motorola

super tweeter.

Driver substitution is not that hard in many cases if you pay attention to

detail and have the original design as a reference. This is especially true

when the response is dominated more by the crossover than the acoustical

response of the driver, depends on the design. I've done it.

There's a lot of negative talk about DQ-10s on the net and so on, but I

agree that it was a significant accomplishment for the designer to come up with

something so different, with such a complex crossover, that sounded very good.

It was advertised as a time aligned linear phase system, and I question

this, but they did show a decent looking square wave in the sales literature IIRC

so I'll wait to do more analysis before I comment further.

I do see many things that I'd probably do a bit differently.

It would be interesting to try to do something with closeout drivers from

Madisound, PartsExpress etc.

Link to comment
Share on other sites

Some DQ-10 pictures for what they're worth:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewI...me=STRK:MEWA:IT

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewI...me=STRK:MEWA:IT

While I've tried to be positive about this design there are many glaring

design oddities. The design is supposed to be time aligned, yet the drivers are

spaced fairly far appart (lower to upper mid and upper tweeter to super

tweeter in particular) and are laid out horizontally. The early units were NOT

symmetrical mirror image for the right and left speakers so if one was time aligned in the sweet spot the other could not

possibly be. They corrected this at some point in production by making them mirror image. The lower cone mid which one would expect should be closer to

the listener for "time alignment" is actually further away as seen in the later mirror imgage right and left pairs.

Don't know how the baffle sizes were chosen, or if they're optimal in any

way?

There's an 8 ohm resistor across the woofer, not sure if this was required

to pass more mid and upper bass to the series connected mid woofer or to act as

a power level indicator, and to flatten the impedance curve to imply some

unique form of damped woofer which was implied in the marketing literature.

I've heard that these resistors smoke if driven too hard. I think the designer

thought there was some advantage to series connected drivers yet had to use

compromise measures to get the right response. A parallel XO without

compromises would be better.

It seems that replacing the .75 and horn drivers with a good ribbon perhaps

with frequency response shaping would be a much better wide band solution to

the original combo.

That ten inch woofer is really a budget driver, good for what it is but

there are much better choices.

Link to comment
Share on other sites

Thanks for the insight on this landmark product. I only heard it on a few occasions briefly but was impressed by its clarity and the surprisingly large sound it gave for what seemed like a small and rather shallow speaker. I've always been a little puzzled by the Motorola piezo tweeter and the rationale for using it. You can see from its 4 way acoustic suspension design with an dome upper midrange and a cone lower midrange the first glimmering of the concept for AR9. Here is a clear recognition that there is an advantage to not pushing the dome midrange too low in frequency or the woofer too high. The crossover is most unusual with the drivers in a series configuration. This is certainly not in keeping with most other crossover design philosophies. In fact, I don't recall ever having seen another anything like it.

I have never been convinced that the so called "time aligned" speaker is inherently better that an identical non time aligned system. Are you aware of any papers or research proporting to seriously investigate what is otherwise just an advertising claim for superiority? Is there more than just intuitive rationale for this?

The problem with modifying the original design I suppose is that you can never be sure that you have duplicated the attributes that made the original so attractive in the first place. Not a problem for me since I listen to each speaker on its own but I guess purists would shudder. BTW, do you own a pair?

Link to comment
Share on other sites

Many have commented on the Motorola tweeter, it is crossed over very high, and probably just adds some air to the sound. I have no idea why that driver was chosen, oh probably so they could offer a spec going to 40 kHz - marketing. The reigion above 10 or 12 kHz is very important to the voicing of a system, however it's easy to shape the response of the main tweeter rather than adding another and another crossover point. This ties in with your discomfort with driver swapping, it was considered a significant upgrade back in the day to replace the Motorola horn with a ribbon, so there you go one swap that was considered valuable. The .75" is commonly replaced by a .75" Scan Speak, and people say this too is an improvement, I've not seen the instructions for the swap so I can't comment about if it was done correctly from a technical perspective. I find that people who fear driver substitution never try it and/or do not have the skill, when I say I've done it you can be sure I know what I'm doing. Believe whatever you want, it's only Hi Fi, no reason to shudder.

I've posted at length on other lists concerning series crossovers when many including experts didn't know the background or completely understand them. There's a classic paper from the AES on them, I probably read it 20 years ago. There was some controversy over this paper since there were a few errors by the author. I don't like series crossovers but I may use one someday if I find a logical application for them. Fried, Medowlark, and a few others use them, IIRC.

Linear phase and "time aligned" systems, and the audibility of phase nonlinearity have been discussed at length on the net, as always consider the source, and in the literature (AES, etc.). I've read much of it and have my opinions. You can Google the subject and there's probably a long list of references in the CALSOD manual to get you started.

You are imagining a problem, if you duplicate a system and A/B it with the original that would be verification. And certainly sometimes it can't be done due to, for example, a very unusual driver but that is rare. While some systems are held to reasonably tight tolerances to the prototype, many are not and the home builder obviously cannot expect this without the tools. The home builder would have to have decent test equipment to match a manufacturer's design/specification and might have to hand select drivers. I don't believe this precise matching to the proto is important for the home builder if you consider room response variations and if a few precautions are taken. It should be obvious that I'm not claiming to be able to match something calibrated and designed for laboratory use again this should be obvious. Improvements can also be made, there are much better drivers available today, and the stock DQ-10 does not have enough head room for my taste.

What does it matter if I own a pair or not, my points are technically valid either way, it's just a technical discussion at this point.

Link to comment
Share on other sites

One potential problem I see substituting drivers is that the LCR characteristics of one driver will affect the response of others because they are components in the series filter networks. This is a consequence of the series approach. So a Scanspeak, Vifa, or Morel 3/4 tweeter may be a much better performer than the original Peerless but it may not be even close in its electrical parameters. Tough problem unless you have all of the data and can make compensations for the differences. The Motorola tweeter as far as I can tell is not used in any other loudspeaker system with any pretense of being a high fidelity reproducer. For it's purpose it is an excellent piece of hardware, very efficient, very reliable, very cheap, but also highly directional. I think most acousticians and designers would agree that response above 20khz is of no benefit to the purpose of accurate music reproduction and if anything can cause problems if there is any interaction within the audible range either electrically or acoustically. I won't go through the Japanese demonstration but their error was to use a tweeter which spanned the upper octave of audible sound and the ultra sonic region causing the ultrasonic signal to create noticable distortion in the audible range. When the error was corrected and they were covered by different tweeters, the expected result was obtained and that was that there was no longer an audible difference. This all extends from the myths used for advertising hype in the 1960s by Harman Kardon (one amplifier actually had a power bandwidth to 1 mhz), University whose "Sphericon" super tweeter extended to 40khz, and Audio fidelity records whose recordings had a response to 25 khz. Phonograph cartridges were not designed deliberately to extend response beyond 20khz until the quadiphonic CD4 system in the mid 1970s which required bandwidth to 40khz for other reasons. FM broadcasts, most magnetic tapes available to the consumer, and compact discs of course have no content above 20khz.

Link to comment
Share on other sites

  • 2 weeks later...

"One potential problem I see substituting drivers is that the LCR characteristics of one driver will affect the response of others because they are components in the series filter networks. This is a consequence of the series approach. So a Scanspeak, Vifa, or Morel 3/4 tweeter may be a much better performer than the original Peerless but it may not be even close in its electrical parameters. Tough problem unless you have all of the data and can make compensations for the differences."

Like I said it's not so difficult if you know what your doing, I didn't do the Scan substitution but many say it sounds good.

"The Motorola tweeter as far as I can tell is not used in any other loudspeaker system with any pretense of being a high fidelity reproducer. For it's purpose it is an excellent piece of hardware, very efficient, very reliable, very cheap, but also highly directional."

Most have a very negative opinion of the MOT horn driver, but I prefer to draw my own conclusions and have not heard it a length, crossed over that high, to offer a reliable opinion. It is still available and is inexpensive for those who want to experiment. I would assume that the designer wanted a directional super tweeter perhaps in an attempt to duplicate the directional characteristics of the flat panel designs he was trying to immitate.

"I think most acousticians and designers would agree that response above 20khz is of no benefit to the purpose of accurate music reproduction and if anything can cause problems if there is any interaction within the audible range either electrically or acoustically."

Actually, some acousticians say that some can hear or "perceive" all the way out to 50 or 80 kHz, this subject does not interest me very much so I've not followed it in any detail. I've never heard of problems with speakers extending to 40 kHz or higher.

Yes it's important to keep inaudible HF noise and interference out of systems with very high bandwidth.

Link to comment
Share on other sites

  • 3 years later...

The Dahlquist DQ-LP1 Crossover is discussed here on page 13:

http://www.avahifi.com/root/audio_basics/ab_pdf/ab1989.pdf

Schematic here toward the bottom:

http://mpbarneyamps.googlepages.com/mpb%27...troniccrossover

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...