advertisement


E-III Revision

Rusty,

I don't think there is anything inherently superior about the ES14 mid-woofer. I believe the key to its beguiling midband performance is down to three factors: (1) complete absence of filters, (2) break-up free polypropylene cone, and (3) wide bandwidth that covers the full range of the human voice.

The most obvious artefact of the M15 is the complete absence of glare, hardness yet there is no loss of presence or detail. I think it's quite a special driver. Yet, because it does not draw any attention to itself whatsoever, I do wonder if the characteristic edge of paper cones might not be more, um, entertaining. It's hard to decide.

When are you down in Wellywood next?

James
 
Addressing your points:

1) Undoubtedly a factor. Would be interesting to hear the unit with a conventional, but well-spec'd crossover in front of it.

2) Not entirely, and I quote: "...resonant modes can be seen just under 2 kHz and at 3 kHz, these presumably due to the woofer cone misbehaving".

3) Agreed. Moving the X away from the usual 2/3 kHz is a good thing.

As to the worth of the ES14 woofer, it has some unique build features, and is certainly a cut above the average driver quality found in like speakers, but you would have to reason that the latest high-tech offerings from Seas and Scan should be superior.

No current plans.
 
Can you pin-point exactly how you think the E-III midrange comes short of the ES14? Is it a matter of tonality (frequency anomalies), clarity (transparency), definition (articulation, pitch, dynamics) or coherence (integration with woofer/tweeter)? If it is not the latter, I'll see if I can effect a simpler change.
Click to expand...

They ace on clarity - the ES14 midrange is so open you could fall into it.

Really though, that is not the whole story. What they are best at is making me forget all about the fact that I'm sat in front of a stereo.

Mr Tibbs
 
Mr Tibbs said:
They ace on clarity - the ES14 midrange is so open you could fall into it.
Click to expand...
Hmmm ... that could be hard to emulate with a bandpass filter that has MORE components than before.

Really though, that is not the whole story. What they are best at is making me forget all about the fact that I'm sat in front of a stereo.
Click to expand...
The M15CH001s do that. I'll have to search the world to find you a pair.

James
 
Mr Tibbs,

Working on the premise that one day, you will find a pair of M15CH001s, I have worked out a new filter network for your E-IIIs that will enable you to improve their midband performance with the existing drivers. This new network comprises an identical low pass and high pass sections to my revised E-IIIs, and features a simpler band pass filter. The midband level is easily adjusted with changes to R2011, which can be varied from around 2R5 to 4R0 without screwing up the transfer function or phase alignment. I've modelled this on 3R0.

Here is the filter schematic and system impedance plot.

6647030-md.jpg


The filter is essentially a first order electrical filter with a resonance trap, and a zobel. But it yields a proper 2nd order acoustic transfer and the phase alignment between the midrange driver and its companion driver is not too bad, as you can see in the following picture.

6647029-md.jpg


Here is a picture of the E-III filter network before and after surgery.

6631996-md.jpg


6631997-md.jpg


Yes, you do spot three bi-polar electrolytics, but these are used in shunt and does not affect the quality of the sound. Besides, I'd struggle to fit a pair of 50uF and a 280uF poly caps in there.

James
 
What are you bods argueing about?

The E IIIs that I heard when Mr Tibbs plugged them in for the first time were nothing short of magnificent.

Accept them for what they are ... a pair of beauts.

quincy
 
James, I have to hand it to you, you NEVER say die :)

That's a very clever piece of lateral thinking!

quincy, We are not arguing, we are just trying to build the perfect beast ;)

Mr Tibbs
 
Mr Tibbs said:
... you NEVER say die.
Click to expand...
I'm almost tempted to put the band pass filter together just to prove it works. I'll rummage through my inventory of parts and see what I can come up with. It should be a simple matter of disconnecting my new BP filter and hooking up a bypass circuit for the CA15RLY if I have the parts.

James
 
The value of C2101 appears to read 500 uF!! Is that right? And if so, where are you hiding it? :)
 
RustyB said:
The value of C2101 appears to read 500 uF!! Is that right? And if so, where are you hiding it? :)
Click to expand...
Rusty, you are mistaking my revision with the one I've come up with for Mr Tibbs. Mine does not have a 500uF, but a 280uF on the bandpass filter.

James
 
If those el-caps were in the signal path, perhaps, but even then the gains from better quality by-passing caps are directly proportional to their contribution. I remember reading somewhere that it's a fallacy to think putting a 0.1uF poly in parallel with a 100uF elcap makes the latter sound like a poly. Such a combination would sound exactly like 99.9% el-cap and 0.1% poly.

But since they are in a shunt circuit, they play bugger-all role in the signal getting through to the drivers. The revised E-IIIs sound pretty damn good, even if I say so myself. Makes the E-X even harder to get onto the drawing board at this rate.

James
 
James, would it be at all possible to extend the XO point of the mid-tweeter further up the range, to say 5kHz? I'm wondering if this may actually be more important than what's going on at the bass-mid cross. Just a thought.

Mr Tibbs
 
Anything is possible, Mr Tibbs. Whether it is worthwhile or not is another matter. I'll need to model the 5kHz cross. The M15CH001 won't reach that high up, and so my lateral plan for the CA15RLY would not be tenable. I'll see what I can come up with.

James
 
James, let me know if some CA15RLY's become available; I quite fancy a CAOW1.

Mr Tibbs: have you ever heard an ES30?
 
Rusty,

I'm NOT going back to the CA15RLY version again, so if you fancy building the CAOW1s, the pair is yours - after I've had a chance to prove the Tibbs Revision.

It'll be interesting to compare the CAOW1 against the E-IXs.

James
 
Anything is possible, Mr Tibbs. Whether it is worthwhile or not is another matter. I'll need to model the 5kHz cross. The M15CH001 won't reach that high up, and so my lateral plan for the CA15RLY would not be tenable. I'll see what I can come up with.
Click to expand...

Gotcha. Don't let this mess with your plans to try what you've already decided on - probably 5kHz is asking a bit much of the CA15 anyway.

This speaker designing lark has so many variables that the possibilities are endless. My ideal 3-way speaker might now centre around a really high-spec mid that covers all of the 'vital' (vocal) area, with the tweeter and bass units having a secondary supporting role. In fact, if a little of the bass extension could be sacrificed I could get away with a 2-way design, using a minimal crossover. A few scribbled fag packets later and I'll end up with an ES14 copy ;)

Mr Tibbs: have you ever heard an ES30?
Click to expand...

A long time ago. I didn't pay much attention to them as they were outside my price range and needed a better amp than my NAP110 to sound really good (the ES14 was fine with the 110).

Mr Tibbs
 
Mr Tibbs,

Mr Tibbs said:
My ideal 3-way speaker might now centre around a really high-spec mid that covers all of the 'vital' (vocal) area, with the tweeter and bass units having a secondary supporting role. In fact, if a little of the bass extension could be sacrificed I could get away with a 2-way design, using a minimal crossover. A few scribbled fag packets later and I'll end up with an ES14 copy ;)
Click to expand...
What you are looking for is the E-IV Revision. Of course, that has yet to be built, but that is essentially the current E-IV with Scan-speak 18W/8535-00 mid-woofers instead of the Seas CA18RLYs.

I have proven in the E-VII that the 8535 is a stunningly good driver that delivers an utterly beguiling midband and sacrifices very little by way of bass extension. Putting a pair of this per side in a 2.5-way will give you full range coverage of the human voice, and more than sufficient bass coverage in the lowest octaves.

The only trouble is the Scan-speak driver is slightly bigger than the Seas and hence a whole new pair of cabinets will need to be built. And I'm tossing up between an identical pair (same volumetric requirements) or a wide-baffle variant.

This speaker designing lark has so many variables that the possibilities are endless.
Click to expand...
Too right. It's better to be kept awake at night thinking about DIY loudspeakers than fretting about mundane stuff like work, life and responsibilities.

James
 
don't you find the 8535 gets a bit ropey at around 100hz?

from -

freq_18w_8535-00.gif


to my uneducated sight, it looks like a mid-bass unit, that could do with a bass driver underneath..?
 
You must log in or register to reply here.


advertisement


Back
Top