I may be seeing a bit of hyperbole here as as of course having something is better than nothing and for me the interest is not only a form factor choice but to hear for myself what the possibilities were for a metal cone driver that was not one of the big box store numbers that make your ears bleed ,figuratively This satisfying of my curiosity was relatively inexpensive, less than 40 dollars per driver yet still represented I think, a legitimate effort of a respected company (SEAS) to produce a quality driver .albeit with It seems, WITH well known limitations . Where it ends up I don't know! but it seemed like a entertaining thing to do. Are there safer options available ? of course, but this one scratches my itch .
For general consumption I THINK I agree with you especially considering the many talented designers that have a deep understanding and appreciation for a great, result at a reasonable cost
It seems a kindness to warn others of the risks. NICE
The peak is often strongest on axis. This is the axis which makes the smallest contribution to power and so is least representative. You may find breakup can be overestimated this way.
In any case, multi-angle measurements usually suggest listening to a multi-way speaker on a different axis.
From a circuit point of view the reasons are not highly significant.
I would normally choose the first of the two for an out of band notch like this. On the plus side it makes tweaking easier as it's effect may remain more steady with circuit changes. The impedance is the same around the passband, but lower than the other circuit at the higher frequencies (where it is high enough anyway).
I wouldn't typically choose the last circuit because of the capacitor configuration as I've drawn it (I was making a point in that post)... but there are other ways to make a parallel notch in series with the driver which I'd be readily OK with.
* With any circuit that has a tuned part like the bandstop filter here, it is advisable to tune it on test to ensure it is working at the right frequency. Not that it shouldn't, but broadly speaking this kind of circuit is more sensitive to this than the other parts.
Agreed. I'm thinking about getting a pair of these too. Metal cone drivers have breakup... cross lower and/or steeper and notch as (if) needed.
$40 for a $100 Seas driver doesn't seem like a horrible idea to me, although I don't find glowing reviews of the 8 ohm version.
A lot of the time, I will use a small tank cap across the lowpass coil, AND use an LC series across the driver.
I have had a few instances where the tank cap or parallel notch will not attenuate the problem. As soon as the LC was run across, the blemish subsided and splendid sound was restored.
I have had a few instances where the tank cap or parallel notch will not attenuate the problem. As soon as the LC was run across, the blemish subsided and splendid sound was restored.
Interesting The common practice is to Toe in the speakers I am wondering if not doing so would help to cure this issue or is that not enough ?
This is not the 8 ohm version but rather a to spec order for a for a 4OHM version I don't know if that changes much
Yes I know. Obviously there are no reviews of the 4 ohm but if I found good or bad reviews of the 8 ohm that would be relevant. On another forum I mentioned this woofer as soon as it went on sale and Wolf did mention he has generally not liked the L-series that he has heard. That would rule it out for me if it was near $100, but for the sale price it's still tempting to play with.
That is the way I was seeing the opportunity as well. That is in the light of how well received the 8 ohm version was, No it is probably not going to knock your socks off but perhaps the Priced paid $99 dollars had a person saying well its not as good as XYZ option ,in view of the NEW company (money wise ) it finds itself in there may be a different comparison
In any case I am comfortable with the risk ,if it doesn't work out I can always gift them to someone not so picky
Hey related to this is the Tweeter I also picked up WHAT does anyone think GOOD BAD OTHER its the
In any case I am comfortable with the risk ,if it doesn't work out I can always gift them to someone not so picky
Hey related to this is the Tweeter I also picked up WHAT does anyone think GOOD BAD OTHER its the
SEAS Prestige 27TDFNC/GW (H1462) 1" Textile Dome Tweeter
AllenB> I assume you've seen Lars' (Purifi) comments on how those notches differ. Only the parallel notch (B in your post) can damp the non-linear harmonics. Thoughts?
Wolf> I specifically asked Lars about the small cap across the first inductor, and while I didn't quite follow everything he said, I took his answer as "no, that won't work like the parallel notch for non-linear distortion." Too bad, I like the small cap shunt trick.
Wolf> I specifically asked Lars about the small cap across the first inductor, and while I didn't quite follow everything he said, I took his answer as "no, that won't work like the parallel notch for non-linear distortion." Too bad, I like the small cap shunt trick.
I do try to separate small signal analysis from nonlinear analysis, so don't read too much into my not mentioning one or the other.
Yes, I'd like to read these. Could you please link to Lars' comments.
I also note that Scottmoose has seen work in this area. I can't find a link just now. I skimmed it once, but I'm not sure I discovered a conclusive connecting factor at that time.
Ok, perhaps this is it - https://www.diyaudio.com/community/threads/new-aluminum-cone-purifi-drivers.382663/
I don't see why it would not reduce HD as he says other parallel LCR filters will. It adds resistance, infinite at that, between the F3 points of the cap and coil.
That said, I feel the LC series notch can be much more effective at killing breakups, energy storage, or resonances.
That said, I feel the LC series notch can be much more effective at killing breakups, energy storage, or resonances.
AllenB> that's a good thread for his comments, but not the one I specifically asked about the circuit Wolf mentioned. My takeaway from his answer was that the parallel circuit creates a high impedance at the notch, which is what does the distortion damping at those frequencies. But when I model both, the inductor + small bypass cap creates a very large impedance peak at the notch frequency too, so I don't see why it shouldn't function the same.
I suspect (it's just a guess) there may have been / be some talk at cross-purposes re series and parallel notches. Lars / Purifi use 'series' to refer to a circuit in series with the driver and 'parallel' to refer to a circuit in parallel with the driver. Others (like muggins here) use the terms to refer to the topology of the indivdual circuit itself, i.e. a 'series' LC where the inductor and cap are placed in series with each other, and 'parallel' where the same are placed in parallel with each other. The latter is the more common use in electrical circuit design, but it's not exclusive of course so all we can really do is try to be as clear as possible about what we're describing (that's me doomed then 😉 ).
The Purifi writeup: https://purifi-audio.com/wp-content/uploads/2022/03/220211_R05-Notchfilter.pdf They're far from being the first to do this, but to the best of my knowledge they're the first to do a really decent writeup with examples and concise explanation.
The Purifi writeup: https://purifi-audio.com/wp-content/uploads/2022/03/220211_R05-Notchfilter.pdf They're far from being the first to do this, but to the best of my knowledge they're the first to do a really decent writeup with examples and concise explanation.
I can definitely understand and respect why people choose to build systems with high value per dollar drivers which may have some performance limiting characteristics. I guess it really depends on a person's will to work around these challenges and create a solution that works for them. My only complaint is the necessity for more complex (and costly) crossovers, which sort of takes away from the original value of the driver. I had a similar experience with the (at the time very affordable) Peerless NE180W, which has a significant peak at 4.5k needing to be tamed. Even using it as a midrange between 200 and 1k, the 5 dB peak was still audible. I didn't have much hope to get it sounding good, but once that peak was notched with an LCR, it became one of my favorite midbass drivers regardless of price. Too bad its now impossible to get reasonably priced.
The peak on your Seas driver being so high may be outweighed by the better pistonic performance of the stiffer cone. Some of the posted sims look promising however and the driver could lend itself to a decent midrange driver. I'd be interested in its Rms value.
Agreed.
The thread I linked to also refers to subharmonics, when the correct term is fundamental. There is a thing called subharmonics but it's not related to nonlinearity so it shouldn't be used this way.
Last edited:
@augerpro I've had another look at the papers. Since this issue is proportional to nonlinearities at the fundamental frequencies, I'd want to determine whether distortion is going to be a problem in the first place. Why talk about reduction without considering the threshhold of audibility?
gedlee said: