Matching drivers...beyond the specs

[Smokinghot]

Ok, here's the deal....

I'm soon to be attempting my first real build. The plan is to utilize old component drivers from my mobile audio days, to save on $$$. I have at my disposal 2 full sets of the Diamond Audio Hex6.6 components. Which gives me a grand total of 4 of these:



and 4 of these tweeters that are still in production by LPG:
http://www.madisound.com/catalog/product_info.php?manufacturers_id=138&products_id= 541

Using these drivers I'm hoping to construct a complete LCR 3way set up for HT use. Yes,.... I said 3way, and yes that means I'm missing something to use as a mid. Which brings me to my question....

The midbass I will be using has construction very similar to the Eton HEXAcone drivers, (http://www.madisound.com/catalog/pro...roducts_id=229). Although I can match senistivity and frequency response to other drivers (front runner being the dayton RS100-4). How concerned should I be with having the mid and woofer cones being of different construction material...? ...and to put it to rest quickly before it's suggested. I will not be considering the Eton4-200/A8. I'll be needing 6 of whatever I end up using, and at $124usd, :E the Eton just throws my "first attempt" waaay beyond what I'd like to spend. :cry:

I have read somewhere, (sorry I can't link...I've done a lot of reading lately), that drivers of like specs will sound different to the ear when constructed with different materials. Is this something that can be compensated with crossover design..? ...and even if it could, I'd imagine the R&D couldn't even begin to happen until I have them placed in a baffle along side each other...?

Can anyone offer a suggestion as to what construction materials I should be focusing on, to match the driver I already have? Or better still...just tell me that it isn't something I have to be worried about....without lying to me that is....lol.

Thanks in advance...

[---k---]

The difference in cone material will likely be the least of your problem.

Proper design of the crossover is more important than the driver selection. You are picking very expensive drivers. It is very likely because of your crossover skill level that a pair of much less expensive speakers will sound much better than the speakers you build.

If you look at JonW's Spassvogle thread, you can see what kind of effort is required to design a proper crossover - and that was just a 2-way. Sure you can do it with less effort, like just tracing manufacturer plots, etc. But, the cost of the midrange drivers you are picking warrent a measured and designed crossover.

(That and that is not a woofer for a 3-way.)

Edit: I reread it and noticed he doesn't want to use the Eton's linked, but similar. I thought he wanted the $124/ea drivers. My statements still stand though.

[---k---]

See: http://undefinition.googlepages.com/diy-mfaq

[Smokinghot]

Edit: I reread it and noticed he doesn't want to use the Eton's linked, but similar. I thought he wanted the $124/ea drivers. My statements still stand though.

No, not necessarily similar. I hoping someone could offer suggestions as to a cone material construction that sonically would play well with the HEXAcone type drivers I will be using.

Just so I'm clear... The drivers in my possession are not an Eton product. I only linked to Eton to illustrate the cone type of the drivers I do have.

My first thought was the Dayton RS100. However being of an Aluminum construction, I did not know if they would match well with the Hex driver I'm using.

I appreciate your reply. Thanks

[Smokinghot]

See: http://undefinition.googlepages.com/diy-mfaq

I appreciate the link, and I followed it to the second section, but it did not address what I asked in my OP.

Again... I'm trying to mate up a midrange driver to the woofer/midbass that I currently own. The driver in my possession is of construction similar to Eton HEXAcone products. Is there any sonic concern about mating my midbass to a midrange driver of a different cone construction...?

Thanks for reading.

[ThomasW]

Is there any sonic concern about mating my midbass to a midrange driver of a different cone construction...?

No..

[HareBrained]

Regarding your OP ...

You don't need to go to a 3-way design. You could develop your xover skills with these drivers in a 2-way TM or MTM, or a 2.5-way TMM design. And then add a midrange once your comfortable with attacking the complexity.

If you "have to" go 3-way, check out the MCM 55-3853 4" cast frame woofer. It's inexpensive but sounds very nice.

There are ways to develop the xover in simulations and it can result in very good sounding speakers. For true high fidelity (extracting the most for the configuration), it requires measuring equipment and time to listen and tweak.

When I got into this hobby 6 months ago, I thought I was after hi-fi. When I realized how much I didn't know and how much time I couldn't devote to tweaking, I had to set my sights lower. In the end, there's still much for me to learn but I sim'd and built an MTM design that was less than $200/pair and am incredibly satisfied with how they sound. Could I extract more from the design? Yeah. I think the xover could be designed for a little more "air" and I'd like to try a different enclosure. But, as is, they're way beyond what I was listening to and they addressed all of the wife's complaints without creating new ones.

Do what you can with what you have and spend some time learning what it means and what you're capable of. I too came from a car audio background, and there you can throw things at a problem. But here, it requires a significantly higher proportion of effort.

[Undefinition]

I too came from a car audio background, and there you can throw things at a problem. But here, it requires a significantly higher proportion of effort.

Hmm. That's an interesting way of putting it.

Thomas' response is correct. And --k-- is also right that the sonic differences in driver materials will be the least of your concerns. Only after the frequency response is flat and the drivers phase and/or time aligned is it time to discuss the sonic differences in driver materials.

On the other hand, certain driver materials will definitely make the task of achieving a flat FR much more difficult.

[Smokinghot]

Thanks everybody for taking the time.... ...and I certainly sense the "biting off more than I can chew" overtone of the replies...lol.

I could have written quite a lengthy first post explaining everything from my EE background, board level work experience, to what I hope the end result to be. However, I didn't think that it really had any bearing on the answers I was looking for.

The bottom line is...(so to speak). That I have some equipment already, and would like to get my feet wet in the cheapest manner. However, I saw the next logical step after a 2way to be a 3way design. Will it take time to reach that design goal...?...I sure hope so. But it seemed prudent to me to start with something I can build apon. So I was trying to determine if what I currently have will accept being reconfigured into a 3way design. Hense my question about a third driver.

Thanks again, and I look forward to posting after some serious R&D.

[---k---]

Okay, if you think you're up to it there are people that are willing to help. The folks around here just don't like compromising and settling for less than a perfect design - which requires in box measurements, ect.

Cone material matching won't be an issue. Lots of commercial designs mix and match for a variety of reasons.

The next problem you have, the Hex DA6.6 appear to be very limited for use as a woofer in a 3-way. You should model it in Unibox. But, it doesn't look like it is going to dig very deep, and has small xmax at 2.5mm. Compare it to the RS180 and RS150. I think you'll be dissapointed using that driver as the woofer in a 3-way. It might be better suited to use as the midrange with a low crossover. And, if you want a 3-way then add an RS225 or something like that for bass. This would be much more well rounded.

[Dennis H]

I could have written quite a lengthy first post explaining everything from my EE background, board level work experience, to what I hope the end result to be. However, I didn't think that it really had any bearing on the answers I was looking for.

Okay, so you know what you're getting into. Do you have a calibrated mic and measurement and crossover design software? If not, that's the place to start.

The "sound" of different cone materials is mostly a matter of how they break up at the higher frequencies and that can easily be measured. Stiffer cones have a more abrupt breakup so extra care is needed in the crossover design to suppress that deep into the stopband. At the same time, stiffer cones often have lower distortion in the passband so it's handy to be able to measure distortion when you're choosing drivers and how to cross them. Of course motor design plays into all that as well but again measurements are your friend. Measure the driver and decide the frequency range where its distortion is low. Then start working on a crossover that suppresses all the nasties and gives a good blend between the drivers.

[Smokinghot]

Excellent feedback

---k---:
I can't say I like compromising myself, but I thought it would be purdent to start slow and build momentum. I was aware of what I believed to be the Hex6.6 limitations, but thought if I matched accordingly I could build a low sensitivity 3way LCR. Maybe good for a small room HT..? As it is, I'm going to focus my attention towards a solid 2way, and then test to see if they are worth developing into something else, or just moving on to a totally new design.

Dennis:
That's a big no on the calibrated mic, and I'm playing with a trial version of LspCad. Although, from what I've read, when it comes to R&D SoundEasy could be a better way to go. ....any suggestions, or personal preference on SW...?

You've repeated what I did read on cone material and break up, and it's that info that prompted my initial post. Figured...why shoot myself in the foot with differences in cone material before I can get into stride with building crossovers.

[Undefinition]

Dennis' suggestion is sound, however, you can get pretty good results without a measurement rig.

Roman Bednarek has a good primer on his website for using the free tools to design a XO without measurement gear.

http://www.rjbaudio.com/Audiofiles/FRDtools.html

In my opinion (and experience), this is the best way to get a good handle on XO design before you have to grapple with learning how to measure (and spending several hundred dollars on a measurement rig). The investment is basically free, assuming you already have a Windows computer and MS Excel.

While it is true that the only way to extract the very most from drivers is to actually measure them, simulation can get you 80-90% of the way there. I did the original Aethers using only traced SPL and simulation (and my ears).

When you start getting the hang of XO design, you'll probably want to invest in some measurement gear. And if you really enjoy speaker design, you owe to yourself to do so.

[Winter]

Smokinghot, just some observations on your drivers from the datasheets. The degree that datasheets reflect the actual characteristics of the driver in hand vary from manufacturer to manfacturer.

DA6.6 Hex Mid-bass
The Xmax (linear excursion of cone) of 2.5mm is fairly short for a woofer, so using two woofers in series (due to 4 ohm drivers) would be useful to increase power handling, through not sensitivity.
The frequency response peak at 4.6kHz is a cone resonance, so you want the signal to the woofer to filtered down quite a bit by then, i.e. crossing over at a lower frequency, say no higher than 2,500 Hz tops, lower is recommened. The Qts (driver damping coefficient) of 0.495 lends the woofer to a seal enclosure application.

The LPG 25NFA tweeter has a fairly high resonance of 1850 Hz. General rule is crossing over a minimum of one octave above the resonance frequency. This puts you in the 3,800 to 3,900 hz range, which is a little high for a two way, especially one using a woofer with a strong 4,500 Hz cone resonance. It can be crossed lower, just not usually recommended. A three way is technically the best configuration if you must use this tweeter with this woofer, though must more complicated to design, next to impossible without measuring and design gear. A two way can be done with compromises in fidelity.

Attempting to design speakers without measuring equipment, drivers with flat frequency responses with controlled resonance breakup modes are recommended. That generally means polypropylene or paper woofer cones and textile dome tweeters with a low resonance frequency (below 1000Hz).

[Dennis H]

That's a big no on the calibrated mic, and I'm playing with a trial version of LspCad. Although, from what I've read, when it comes to R&D SoundEasy could be a better way to go. ....any suggestions, or personal preference on SW...?

I like the user interface of LspCAD. ARTA is a current favorite for measurements if you want more than LspCAD provides. People seem to have a love/hate relationship with SE -- love the capabilities, hate the user interface. I don't have any experience with it. And don't forget Jeff Bagby's free Passive Crossover Designer for Excel. It's pretty slick but it doesn't have an optimizer feature like LspCAD and SE.

[Smokinghot]

Undef:
Great link, thanks. It only makes sense to educate myself on XO design via software before attempting to R&D in the real world.

Winter:
I appreciate the assessment. I'd like be able to say "that's what I thought", but, of course it would be a flat out lie...lol. Taking your suggestions it does look as though I may progress into a 3way which I thought (without reason) would be a good idea to get the most out of what I have.

Dennis:
So far, at best, I'm treading water with the LspCad demo. I did try to find a detailed tutorial online without any luck. Several walk throughs of examples are available, but I personally perfer a "start from scratch" approach. However I was warned about the learning curve, and I'm certainly not discouraged.

Thanks for the feedback guys

[Dennis H]

LspCAD is really easy once you get the hang of a few basic concepts. The Edit tab of the schematic window is where you click and drop components to make your circuit. A minimum circuit is a generator, a driver, and a ground for the driver. You can hook them together with a wire component or just click and drag the leads to stretch them. The Simulate tab is where you change component properties by clicking on the components. For a driver, it's properties will include its SPL and impedance text files as well as its XYZ location on the baffle. For caps and coils, the properties include their values as well as checkboxes for whether or not to include them in an optimization run. Right clicking on a component sometimes gives you more options than left clicking.

The pdf tutorial in the doc folder sort of assumes you know too much but the pdf manual is pretty straightforward if you have the program open and can click around.

[Smokinghot]

The Edit tab of the schematic window is where you click and drop components to make your circuit. A minimum circuit is a generator, a driver, and a ground for the driver. You can hook them together with a wire component or just click and drag the leads to stretch them. The Simulate tab is where you change component properties by clicking on the components. For a driver, it's properties will include its SPL and impedance text files as well as its XYZ location on the baffle.

I really should dig into the info provided, but what's got me confused currently is the difference the driver under the passive tab and the loudspeaker unit under the box cabinet tab. What's the point of the driver (under passive tab) if you need the T/S values (which are only available with loudspeaker) for sake of graphing responses..?

I'll focus some time into it tonight...(it's my day off and I'm working nights...If that makes any sense )

[Dennis H]

There are two kinds of drivers. One uses actual measurements of SPL and impedance (preferably measured in the box) and the other uses T/S parameters to try to predict the response in some box you specify. You'd use the latter to design a box and the former to design a crossover after you've built the box and measured the mounted driver.

[Smokinghot]

hmmm....

Just so I'm clear... The drivers in my possession are not an Eton product. I only linked to Eton to illustrate the cone type of the drivers I do have.

Slightly OT but an update none the less...

I've been doing some digging among the WWW, and it seems as though my Hex6.6's are in fact an Eton product.

Would any here by any chance be able to comfirm that for me...? :lol:

Reason I ask is... I have an opprotunity to purchase another "barely used" pair at a very reasonable price, (considering they may actually be an Eton product).

Although, to be honest, the feedback I've received so far on what these drivers may be good for...(not much from the sounds of it), is discouraging me from pulling the trigger. I'm really on the fence on this one, considering having 3 pairs of these really opens up some possibilities in design.

Any feedback guys would be very appreciated.

-are they junk, and don't bother
-for a couple of $100, how don't you
-$$$ can be better spent
-considering age, a matching pair is too good to pass up

:boohoo:

[Winter]

Smokinghot,

The Hex6.6's appear fine, pending actual testing. Eton is a well respected manufacturer. The crowd on this forum is very informed and demanding, in a positive way. We/they will look for the strengths and weakness in any driver, $20 to $280. You design to the strengths and work around the weaknesses in all drivers.

If you only plan on building one pair of speakers, and you want to proudly say it's the best, then buy other components for an already designed system.

However, I recommend you consider using the Hex6.6's, two woofers per speaker due to the low Xmax, in a two way system with another tweeter that plays lower. Tweeters such as the Seas 27 TDFC, Seas 27 TDFC/TV, or Seas 27TFFC along with others could be considered, in the $40 range. The HEX6.6's are very good for a sealed enclosure, with a -3dB around 63 Hz. I question the recommended enclosure volume of 7 liters on the datasheet, due to the given Vas of 25.4 Liters and the Qts of 0.495. You already have the drivers so use them to build a good system, while you learn. And speaker designing and building is a long process. The best situation is finding someone in your area that uses speaker testing equipment and design software, usually LspCAD or SoundEasy.

Driver tests by John "Zaph" Krutke: http://www.zaphaudio.com/
Good general information: http://ldsg.snippets.org/sect-14.php