Avoiding the 4k peak in an Accuton 2-way without a notch filter?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi

This is almost a silly question, but I haven't managed to find an answer, and I am starting to work with LspCAD but I don't have proper .frd files for my speakers yet, since I'm considering what to get.

My planned speakers are for a specific situation: I want small, WAF-friendly 2-ways with accuton tweeters and either paper (for simple filters) or accuton woofers, handing over low frequencies from around 100hz to my Velodyne DD10 sub. I have been listening to speakers without filters for quite a while, and I want to try to have as simple filtering as possible, I rather have them dynamic than a perfect frequency response. I genereally like speakers with quite a forward attitude. Lots of ideas and thoughts, but this is my question:

I am considering the Accuton C158-6-851 woofer, because it seems to be the smallest and cheapest Accuton woofer that seems well suited to small enclosures. It does however have a ringing at 4khz, and I am thinking about how to address that, from your experience, would it be sufficient to simply use say a second order filter centered around 1800hz (with C30 tweeters) or will I need a notch filter? With a second order filter, I would be down 18db at 3600hz (if I am correct, 6db down at center frequency and 12db on top of that?), the peak is centered at 4.5k and looks about 3db higher than 2khz response. That should do it right? I would love to go with first order filtering, but I guess the 851 won't really permit that, with first order from 2500hz (C25 tweeters) we are 9db down at 5khz, so something like 6db under the general response... Too little?

I'm planning to get a measuring mik and do this right, but I wanted to try my plan on you before I jump the gun and buy the speakers. If anybody has experience with the 851s, please chime in!
 
Just a point of reference for ya....

When nuking resonances/breakups in drivers, -25dB has been accepted as the minimum reduction from reference level to not become fatiguing. To reduce it to the point of no longer being a concern, -40dB of more from reference is generally pushing it into the noise floor, and -50dB is pretty much dead completely.

And yes- you will have to include a notch of some sort to kill it, or use a higher-order slope; neither of which you seem to want to do.

I did have a friend working with a former model C25, and he was getting what he wanted with 3 parts blended with a SS 7" Kevlar. Coil on woofer, resistor and cap on tweeter. I've not yet heard this beast though, and don't know the values he was using.

Later,
Wolf
 
Pick a different driver, quite simply. Simple filters are fine on well-damped drivers, they're not for accutons or seas excel magnesiums. Notches have to be used and they have to be both deep, and somewhat wider than the breakup peak. Resonances can shift around somewhat, as can impedance profiles and so you need to ensure that that variation is covered as well. I've heard some speakers that instantly go from pretty good to ear bleeding, and the reason for that is a too-tightly-tuned notch filter that didn't cover dynamic variation.
 
Thank you, that is exactly the kind of answers I was looking for!

So, it's back to the friendly 5.5" paper Seas, or significantly change slopes, or wide and deep notch. Minimum 40db in one-and-a-half octave sounds like a fourth order just might be enough, as alternative to a notch filter. I have some reading up to do now. :)

Again, thanks!

*edit*

After some initial thought, it would be interesting to try a symmetrical fourth order filter with even lower crossover point, the C30 tweeter has a resonant frequency of 540hz, and the C25 750hz or so if I remember correctly, and crossing lower might leave more mid range in the tweeter, and that might sound interesting. (Accuton recommends first order @ 1800 and 2500hz respectively, so going a bit lower should be acceptable. In that case, it'll be op amp active filters I think.
 
Last edited:
Pushing dome tweeters really low is a surefire way to have a bad sounding system- they sound much better crossed up in the 3k-4k range, and even then, you want 2nd order slopes at least. For much lower listening levels you might get away with it, but IME, pushing tweeters too low is always a disaster.
 
6" bass is a flippin' disaster! :D

Somewhere between 8" bass and 5" bass lies the best of both Worlds that is the 6" bass. Except maybe it is the WORST of both Worlds. :mad:

TBH, if I had a good sub that takes care of the low frequencies, I might go for a 5" two way.

SEAS 5INCH

Why throw too much money at it? The SEAS CA15RLY and a metal 22TAF/G seems to do the job. As it goes I'd shunt 22R across the tweeter, just to stop it sounding harsh. Probably won't go too loud, but that's the trade-offs.
 

Attachments

  • 5 inch bass Troels Gravesen CA15RLY 22TAF-G.JPG
    5 inch bass Troels Gravesen CA15RLY 22TAF-G.JPG
    18.2 KB · Views: 198
Last edited:
@badman is that also true for the ceramic tweeters in your opinion? My tests with a single C25 (which is now sadly broken) and a CA15RLY sounded quite good crossed first order at 2k... The idea for maybe going with the C30 tweeter is that it could go even lower.

Well, the idea is based on using Accuton tweeters, I am very fond of that sound. Both the C25 and C30 look absolutely terrific from the data sheets and are well reviewed, it's only the woofers that don't have resonance damping cut outs that have ringing nodes, and the cheapest one that does, the 191E, I find too expensive for this project (although I guess I had two of them per side in my old Marten Miles 3. That was an expensive speaker... :p ) Before the Miles 3, I had the Marten Duke, which did not have cutouts, and I believe they had simple filters and sounded fantastic anyway.

Data sheets C25:
http://accuton.com/media/datasheet/Datasheet C25-6-012.pdf

C30:
http://accuton.com/media/datasheet/Datasheet C30-6-023.pdf

This is the 851 I am considering - I am/ was tempted of one of the closed boxes Accuton suggest: "Vb: 2.3L, -16dB @ 40Hz, F-3dB: 98Hz, Q: 0.71 (typical)"
http://accuton.com/media/datasheet/Datasheet C158-6-851.pdf

And this is the CA15RLY which is the contender:
http://www.seas.no/images/stories/prestige/pdfdatasheet/H1216_CA15RLY_SRA_Datasheet.pdf

The idea was born from looking at the Hatt SE and Paper HATT:

http://www.humblehomemadehifi.com/download/Humble Homemade Hifi_HATT Mk-III-C_copy.pdf

http://www.humblehomemadehifi.com/download/Humble Homemade Hifi_Paper HATT_copy.pdf

I even asked Tony Gee about using the C25 in the Paper HATT with the Seas CA12RCY and he calculated a filter with three components per side... An obvious option is to just build a pair of Paper Hatts with C25 tweeters, but then I wouldn't need to buy a measuring mic and walk the walk would I? :cool:
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.