The secret of building a good 2-way

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
diyAudio Chief Moderator
Joined 2002
Paid Member
Especially speakers are a full set of compromises. When the room they play in comes in the equation (which is fundamental to come in), the term ''adequate'' becomes much heavier than the illusive ''superior'' term. There is no best speaker among the most competent ones, there is best suited. The speaker plus room interface dwarfs the preceding chain in importance IMHO. After it is done right, then all the definition considerations from source through amplification can be addressed at leisure.
 
That's why I used "least compromised" and not "the best".

I do believe though that we really need a framework for judging speaker performances based on technical specs. While we've learned here at diyAudio that it's better not to utter absolute statements, I believe that this "postmodern" :) approach has it's limit. And I see no reason not to be able to debate on establishing a set of features that are more likely to be compromised than others when discussing about 2 way designs.
 
Fountek FR-88 looks to be a very excellent driver. Sensitivity is a little low for use as a mid-tweet, however not out of the question.

I wouldn't worry about the high frequency response. Top octave is often not heard to begin with. Listen a little off axis and it is "tamed". I would give these speakers a good listen, and if you find that the high end is a bit much, add a filter to bring it down a bit.
 
The Iron Lawbreaker satisfied the goals at hand for a 2 way - I extend the basic approach to 2 1/2 way configurations for larger speakers.

128598d1236576390-second-approach-maximizing-bass-output-iron-lawbreaker-img_1367a.jpg


For this speaker, in a volume of approx. 3.5 cu ft, I wanted the maximum broadband efficiency, good quality bass with good extension, a flat response and low distortion at high levels, and I think I achieved it, with a response down to 50-60hz flat @ ~99db/w/m. The HF driver is an Altec 288G and the bass driver is a JBL 2220A. As a side benefit, I was able to achieve a very flat impedance of between 16 - 20 ohms between 40 and 240 hz due to adding ~ 1 cu ft of activated charcoal to the enclosure (in gas tight bags) to increase acoustic compliance and using a special xover/frequency shaping circuit in the bass featuring a custom air core transformer that allows minimum component count bass response shaping with the loss of almost no efficiency or damping and also serves as a series inductance to the woofer - the only series woofer xover component (although I also use 600uf of bass polypropylene AC coupling to reduce cone movement from out of band LF). With an overdamped tuning of the box to ~32 hz, the overall bass quality is surprisingly tight and very detailed - more like that of a closed box than a conventional ported system, only in some ways better yet since the impedance is so constant throughout the bass passband (no Zobel networks used). There is some serendipity with this all coming together as well as it did, IMO. With this design, using Elliptical xover filters allowed me to reasonaby approximate a first order xover response with the minimum possible driver overlap & allowed an xover ~550hz which minimizes Doppler distortion.
 
Last edited:
Decades ago I copied Ken Kantor's wonderful NHT architecture for a wide-dispersion surround speaker. Ken's was a 2-way, quasi-dipole, one midbass firing toward the wall spaced about 1-2" away, 2 matched tweeters firing anti-phase of each other fore-aft & flanking the midbass. Ideal siting was on a line parallel to the main seating area as a side surround (this was before simultaneous rear/side surrounds became popular).

Ken's speaker was quite tiny, only 4" midbass or thereabouts. My version was much larger w/ 130mm Audax midbass. For side-firing I used so-called "fullrange" drivers, approximately 3 to 4", IIRC Pioneer brand paper cone, something like what may be found in a TV or other appliance, very low cost (under $5 ea).

They provided exceedingly smooth, musical, satisfying & room filling performance. Still missed every time I think about them.
 
Especially speakers are a full set of compromises. When the room they play in comes in the equation (which is fundamental to come in), the term ''adequate'' becomes much heavier than the illusive ''superior'' term. There is no best speaker among the most competent ones, there is best suited. The speaker plus room interface dwarfs the preceding chain in importance IMHO. After it is done right, then all the definition considerations from source through amplification can be addressed at leisure.
ditto
ditto
ditto
Lynn's Ivor Tiefenbrun was very very smart (see his "audio hierarchy") but he left out the speaker/room interface.

Last year I replaced $11k worth of superb highly regarded separates by two of the world's best designers w/ a $150 used receiver. With other changes, the results are more pleasing to my ear. (I admit a PhD hand picked the reciever, which passed his proprietary amp tests w/ flying colors).

TAS bombastic pronouncements about preamp/amp differences are so much hooey. Don't get me started about cables.
 
My bad: Pioneer VSX-D912 circa '03-'-04 as rec. by Dr. Earl Geddes.

6 x 105Wrms BUT BEWARE 8 OHM RATED ONLY MEANING THE LOWER THE LOAD BELOW 8OHMS THE WORSE IS PERFORMANCE. Current speaker is 5.3 Ohm minimum, 8 Ohm average, 20 Ohm peak @ 2 kHz. Performance audibly improves w/ impedance EQ circuit to flatten the 20 Ohm peak, but same effect even when high current pure class A Plinius SA-50 Mk3 was employed (100Wrms @ 4 Ohms). The 912 beats the SA50 in the treble & overall lower glare/grain/haze.

Preamp replaced was Bongiorno's superb $7500 Ambrosia voiced by Bascom King. Over time no audible difference in the Ambrosia's SE in/out compared to the 912s pure analog input (sum total one) & pre outs (of course the 912 has seven pre channels vs. the Ambrosia's two)
 
Last edited:
TAS bombastic pronouncements about preamp/amp differences are so much hooey. Don't get me started about cables.

6 x 105Wrms BUT BEWARE 8 OHM RATED ONLY MEANING THE LOWER THE LOAD BELOW 8OHMS THE WORSE IS PERFORMANCE. Current speaker is 5.3 Ohm minimum, 8 Ohm average, 20 Ohm peak @ 2 kHz. Performance audibly improves w/ impedance EQ circuit to flatten the 20 Ohm peak,...

So which is it? The first paragraph above appears to contradict the second one regarding the audibility of amplifiers.
 
Pioneer's decision to rate the receiver for only 8 Ohms apparently signifies its output stage has moderate current capacity at best, consistent w/ all the chip architecture with which I'm familiar (which isn't saying much but it's apparently the nature of the beast...notice all the "apparentlies" though). Please correct as needed. The moderate current capacity of chips is the only Achilles heel w/ which I'm familiar.

So what this means is the receiver appears to be a giant killer but it can only slay moderately sized dragons (meaning speakers w/ moderate 8 Ohm loads). So...the lower the speaker load falls below 8 Ohms the worse will be this receiver's performance.

Into my speaker (86 dB, 5.3 to 8 Ohms above 100 Hz, active 80Hz HPXO) it performs very well up to moderate levels but strains when pushed beyond a certain threashold.

A new speaker is due shortly, a clone of the current speaker but w/ bipolar radiation rather than monopole, sensitivity increased about 1.5-2dB & doubled impedance (10.6-16Ohms). I'm hoping the changes will produce a larger maximum SPL capability.
 
Just another Moderator
Joined 2003
Paid Member
I think you will find it is for C1 a 4.7uF cap with a 0.68uF high quality polypropylene cap in parallel with it.

This I would guess is more to do with getting a non-standard capacitance value.
C2 is probably similar to C1

In the case of C3 the 0.22uF high quality is in parallel with a 33uF non-polar electro, this will be likely done to be able to use a cheaper electro for the bigger cap and get some benefit from bypassing it with a high quality polypropylene (note I've not tried this so cannot vouch for it's effectiveness).

u68 is just a way of writing 0.68uF just like 0R5 is the same as 0.5 ohms.

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