Seas 5" Aluminum

[MMadrid]

Hi All!

It seems there is plenty of talk about the 5" Seas Excel Magnesium driver W15CY, but virtually nothing written about the (much) more affordable Aluminum L15RLY.

So far as I can tell, the specs seem nearly identical. Both drivers shares the same magnesium basket and surround, and the motor design is very similar as well. In addition, the response and impedance curves are strikingly similar -- both have that hideous metal cone breakup up top, and a nasty peak around 10k or so.

Modeling both in Speaker Workshop, I was also able to get nearly the same theoretical response using the same crossover component values.

Yet nearly all designs that I see that employ Seas midwoofs call for using the Excel.

So my obvious question for anybody who has experience with these drivers: Is the Excel worth the extra $80, or are people simply turned off by the idea that a cheaper driver might be nearly identical to a more expensive model. I'd sure like to save some money on a two-way monitor project I'm designing, but will fork over the extra cash for the Excels once I hear a convincing argument in their favor...

[ShinOBIWAN]

From the accounts of folks who've compared to similar models it seems the differences are small indeed. Infact I can offer an opinion from experience but not one relating the W15/L15, instead I've directly compared the L22 and the W22 late last year.

I can honestly say the difference is small but definitely noticeable, the W22 seemed to have a slightly cleaner sound making things seem more easily heard. I don't have any measured peformance to back this up but my ears preferred the Excel although the L22 was close. I'd go as far to say that some models/sizes of the Excel aren't worth it over the Basic range. Others opinions from the likes of Zaph(has worked extensively with the Seas Basic's) seem to back this up, although I only have one comparison (L22/W22).

My advice is Seas don't seem to know how to make a bad driver these days and at worst you get average and the good ones set new performance levels for the price. In light of this I don't think you'll be disappointed whichever you go for.

[herm]

MMadrid,

I have a pair of the W15s running in a 3 way with Raven tweets
and a Peerless woofer.

Send me an E-mail - you can stop out and listen sometime.
(I'm in Mira Mesa)

herm

[Taco]

You can find designs with the L15, but if you can afford the W15 than I would go for that driver. Especially for the mid/bas units the improved cone and motor are worth the money.

Tony has some designs with the L12/ L15 http://www.humblehomemadehifi.com/

[ShinOBIWAN]

Originally posted by Taco
You can find designs with the L15, but if you can afford the W15 than I would go for that driver. Especially for the mid/bas units the improved cone and motor are worth the money.

Tony has some designs with the L12/ L15 http://www.humblehomemadehifi.com/

Its funny because in Tony's HATT designs he says that out of the HATT SE(Excel) and HATT mkIII(Basic) he prefers the mkIII, citing it as the best of all the HATT designs.

So the Excel looks more of an ego thing rather than anything worth the extra $80. Other respected designers seems to echo the same thoughts and my own experience with the L22/W22 tells me the same thing.

[gl]

Zaph has a really excellent design using this driver on his web site. When he created the design he didn't have any experience with the similar Excel but felt there would be little difference.

In his recent 5" roundup tests he does not include the L15 I was dispappointed because it would have given him an opportunity to re-visit the W15/L15 comparison with hard data. Perhaps he might add that someday.

Check out Zaph's site. It is truly one of the best out there.

Cheers,
Graeme

[Rocky]

I just bargained a pair of the L15RLY for a bookshelf design I am working on. Judging from the published data, that 5'' is probably the driver of the standard series that comes closest to the performance of its excel counterpart.

FYI, I talked to an engineer on the seas factory, and he stressed that it was not enough to attenuate the breakups only so they don't affect overall response. Due to noise artifacts in the breakup area, the cone breakups should be kept 30dB down, he said. At this level he considered them "very hard to notice".

[MMadrid]

Thanks everybody for the advice. The decision is still a tough one.

Yeah, I was a bit bummed that Zaph didn't review the aluminum, only the Excel and the paper cone.

Rocky, you're saying 30db down to keep the cone breakup artifacts out of audible range? If I am crossing at 2000hz with a 12db electrical slope, I should be about 30db down at 10k, right? That's about 2.5 octaves or so between 2000 and 10,000. Am I figuring this right?

Also, sorry to change subjects on this thread, but does anybody have any experience with the offset settings on Speaker Workshop? I am building a 2 way with either the excel or the aluminum, and the baffle will be slanted back with the tweeter flange about 1 inch behind the woofer flange. Does that mean I imput 1 as the offset, -1, or does that mean that I imput 0 since the drivers will effectively be lined up with a slanted baffle?

Thanks again!

[sangram]

Originally posted by MMadrid


Rocky, you're saying 30db down to keep the cone breakup artifacts out of audible range? If I am crossing at 2000hz with a 12db electrical slope, I should be about 30db down at 10k, right? That's about 2.5 octaves or so between 2000 and 10,000. Am I figuring this right?

My understanding is you will still need the notch as the cone will peak up 10dB , bringing the artifacts up to audible level again.

Zaph has a nice notch filter but the coil value is a little tough to get (50 uH, available somewhere on Parts Express AFAIK), which filters out the notch well enough in measurements.

[Rocky]

Originally posted by MMadrid

Rocky, you're saying 30db down to keep the cone breakup artifacts out of audible range?

It is not my sayings, I just quote, but yes.

Originally posted by MMadrid

If I am crossing at 2000hz with a 12db electrical slope, I should be about 30db down at 10k, right? That's about 2.5 octaves or so between 2000 and 10,000. Am I figuring this right?

You will need a notch. Take a look at Zaphs crossover.. it uses a first order filter into the notch. It is a good idea, it does show a dip in the mids due to the filter coil biting in the VC inductance, but this is where your baffle step peak is likely to be, so it looks worse than it is. The crossover I'm working on for mine is of same topology although other numbers, and with the first order filter above 2kHz, simulations shows the main breakup peak at 9kHz between -40dB and -50dB (!!). You could "correct" the midrange dip with a zobel, but the impedance will then be much lower at higher frequencies so you would likely need a higher order filter to kill the breakups effectively.

Hint: You can also use the FR peak from 100hz to 300hz to help you with baffle step correction.