GoodSoundClub - Romy the Cat's Audio Site - Notch filters remedy

10-26-2008

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	26
Post ID:	8631
Reply to:	7289

To gain some efficiency & LF vs. DEBZ

Like I said, Yoshi's gift of de-whizzered Reps got me thinking of better efficiency and more "weight".

So far, I've got this:

A pair of Eminence Kappa 18" in ~ 6 cu. ft BR box to ~250 Hz; de-whizzered Lowther DX-4 PLUS de-whizzered Reps in OB to ~ 6k Hz; Selenium T-324 tweeter. If it needs more "air"/HF ambience, phase in Fountek/Audaphon JP2 > 10 k Hz.

This should be about 102 dB efficient from ~ 35 Hz. I don't know about small phenolic "bullet" tweeter, but I suspect it would ease electricity pain, and ribbon could be via an L-pad.

Paul S

11-01-2009

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	27
Post ID:	12115
Reply to:	8631

Current "Thinking"

This stuff has been stuck in my mind for some time now, working around in there against my ongoing experience, against what I am now running. The goal is a more dynamic presentation with better spectral balance, with the emphasis on LF and [more] correct harmonic development top to bottom.

At this point I am not at all sure it is possible to do FR with one SET amp, at all. Somehow, everything I model winds up seeming to need at least two amps per side, with one amp just for LF. This notion is based on my cold, objective assessments of the "LF" I now get and mental "simulations" of what it would take to get enough LF energy in my present room to effectively balance off the SPLs presented in the present power band my system provides.

In this case, I mean LF on a par with the best the rest of my system has to offer. While I presently use BR for efficiency, it would be nice to have the LF integration and solidity provided by a sealed system. And I am pretty well convinced by now that this alone makes FR SET all but impossible.

Just now I am thinking for each side: four 15" bass drivers (sealed?); one Supravox GMF 285; 1 de-whizzered Reps; 1 modified Selenium 324 tweeter; 1 Audiphon ribbon.

All of a sudden, we're looking at five channels per side!

But at least we don't have to develop each horn, too...

Paul S

06-17-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	28
Post ID:	13788
Reply to:	12115

Crossing Over

This is sort of a "preliminary" post, to air some of the thoughts that will lead to my next disaster, which I will tell about once I begin to get some actual results. My thinking is still about the same as it was when I made the 11-01-09 post; but some of the ways I have been thinking about the results themselves have warped a little over time.

Basically, I have pretty much floundered when it comes to the subject of crossovers. The hellovitt is that speaker level XOs are the devil's tools. Despite what one thinks when plotting them out on paper, they are actually moving targets, in every sense of the word. Further, especially at lower frequencies, the very nature of the XOs themselves makes it almost impossible for the amp to do what the XO math says it will do.

While I was thrashing around on this subject, I may have caught a small break. I forget exactly what I was looking at a couple of days ago, but a link came up, and that led to another link, to Dick Olsher's site, where I found some ideas for much simpler XOs for speakers much like my own DEBZ; Dick gives all "updates" to anyone who's purchased any of his plans, at any point, and he sent me some more specific information to put in my XO oven. If the examples I got from Dick are not perfect for my own situation, yet they are of the right stripe, and it certainly got me to thinking about ways to try to get some of what I am after with drivers I already have, using mostly parts I also (ahem...) happen to have "on hand".

This post reminds me of those guys on TV, who yell and pump their fists before they ride a small bicycle off a roof...

Paul S

06-18-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	29
Post ID:	13791
Reply to:	13788

Going Through Phases

Crossovers are so annoying, about impossible to nail down, and it's almost as though the more you study the problem, the fuzzier the "answer" gets.

For now, I would like to more or less forget about "phase" considerations, partly in order to get this thing rolling, and partly because I believe there is no perfect solution, in any case. I expect to make final decisions with my ears; I just hope I can get close enough to the XO points and lap curves I aim for to be able to reasonably evaluate the results by listening.

While I would still like to increase overall speaker efficiency, it's tougher than it looks at first. Part of the connundrum is that although additional drivers and channels raise the efficiency on paper, in fact the amp is further compromised by the added burden of all the added networking and feedback. Not to mention the "slice and dice" issues... I don't know if this makes "expanding" a wash, but I was thinking that for the present I'll just try to shift the frequencies around some while reducing XO parts count, and I'll see how that works to accomplish +/- the same things. I'll say more about LF to close, but generally speaking, one has a much easier path to hack if one simply lets go of real FR, going in.

Regarding the mid-top spectrum, a while back Yoshi sent me some Reps drivers that he had "castrated" by cutting off their whizzers. As it happens, I have been targeting the whizzers on my Lowther DX4s for a while and I am pretty sure that I could do better for that part of the frequency range. Simply put, I plan to sub the de-whizzered Reps for the DX4s and run the Audaphon ribbon down a little. I'll keep the lower notch filter I used for the DX4s for the nonce, and I'll just let the Reps run out of HF steam on their own, hoping they will meet the ribbons 1/2 way near 6.5k Hz. Yes, this violates my own ">8k Hz Ribbon Rule", but not by much, and so what? More on the XO changes, anon.

Regarding the low-mid spectrum, I mentioned in another post that my 15" Audax woofers were short on tone over about 125 Hz. Taking a harder look at this (ie, actually looking...), it's no wonder, since they are presently burdened with a very complex 3rd order XO at 100 Hz! I will run them up and I will run the Rep's high pass XO up to a 1st order 450 Hz from the DX4's present 2nd order 150 Hz high pass.

The aim here is to add weight to balance, gain better high-SPL-upper-mid clarity and balance, and make a more direct connection with the ML2s. If I wind up with "real", "spiky" "treble", I will be stoked! This will not address LF, of course. In the context of this thread, I apologize, but I am resigned that true LF will take more amps and more boxes. Present "LF" target is -3 dB @ 40 HZ with no help from the room; ie, business as usual.

Paul S

07-03-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	30
Post ID:	13913
Reply to:	13791

Coddling One's Inner Nerd

The new parts I needed to build the new crosovers arrived about a week ago, but I have been otherwise busy (for a change...), and now I've managed to get sick, so I am still only mentally toying with the project. I will try to work in some bread boarding, at least, this weekend. Anyway, here is what I actually ordered parts for: For the 15" Audax PR380M2, I decided to start with 1st order low pass at 200 Hz (for a 300 Hz XO). This will be an octave higher than it's crossed now, and I'll also be relaxing the slope from the present 3rd order XO. Again, the idea is to add weight in that range. I will also add a notch filter at this driver's 1,500 Hz resonance peak (better safe than sorry...), even though that big motor effectively stifles HF on its own. For the Reps, I will high pass it at 6 dB at 400 Hz (ie., 300 Hz XO),and I will use a 2,200 Hz notch filter for its peak. Present Lowther high pass is 200 Hz, 2nd order. The Lowther-alike Reps will be allowed to run as high as it wants, without its whizzer, and the Audaphon ribbon will come in at about 8k Hz, but 3rd order, rather than it's present 2nd order high pass near 15k Hz. Here, the idea is to sluff "HF" (and quell HF noise). Although all this should allow decent phase relationships (once I flip the ribbon's leads...), who can be sure of these things ahead of time? This ribbon does have a built-in circuit that keeps its own impedance constant, however, which is way cool compared to the wild swings of most ribbons, which make figuring XOs for them extra tough.

I'm presently hatching a plan for a straight-to-the-point "separate" ULF that some purists here are probably going to sniff at.

But, first things first...

Paul S

07-04-2010

Post does not mapped to Knowledge Tree

haralanov

Bulgaria
Posts 130
Joined on 05-20-2008

Post #:	31
Post ID:	13914
Reply to:	13913

Do not use ribbons sooo low

Paul S wrote:
For the Reps I will use a 2,200 Hz notch filter for its peak

Paul, why don’t you try to fix the problem of the Reps driver instead of notching it? There is a reason for the peak and I do not consider the problem could be solved using rejecting filters.

Paul S wrote:
Audaphon ribbon will come in at about 8k Hz, but 3rd order

Is that the same tweeter like this:

http://www.madisound.com/store/product_info.php?manufacturers_id=132&products_id=1557

If it is, I have to tell you I had an opportunity to make experiments with this driver. The first thing I noticed is that the driver had an absolutely horrifying tone! It doesn’t worth even $1…
The ribbon is 100% tonally blind and if the crossover point is lower than 15kHz the driver sound honky as if you speak with your nose closed, despite the big effective size of the ribbon. If you use this tweeter lower than 10kHz your system will be totally unlistenable because you cannot reproduce the proper tone of harmonics.
Think of the tweeter as a natural extension of your midrange driver – they must operate on the same principle and they must have identical materials for their moving systems. The Reps demands paper HF driver with very small voice coil and alnico magnet system - something like this:

my alnico tweeter v2.jpg

"If you can't explain it simply, you don't understand it well enough." -A.E.

07-04-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	32
Post ID:	13915
Reply to:	13914

Death by Ribbon?

Haralanov, as far as I know, Madisound does not offer my ribbon. Mine is not a stock Fountek, but it is a re-worked version of the old, discontinued Fountek JP2, a very narrow "naked" ribbon (ie,. no coating), and 97+ dB efficient. I got my Audaphon JP2 from the Lautsprecher Shoppe, in Deutschland. I think it is "rated" to XO at 2K with a 3rd order curve, although I would not dream of doing this. As it happens, however, I agree that ribbons lack tone, and they also lack texture, so this will be an experiment for me to run it this low. I am counting on the Reps to parallel , of course. I had thought to try the 101 dB (phenolic) Selenium ST324 for that range, and I may yet give it a go, depending on how this re-arranging of the spectrum works out in terms of balance and dynamics. This is just the first round, after all...

Regarding the notch filters, I consider them the lesser of evils rather than a "cure". I have no serious idea how to completely "solve" the Lowther-alike problems in order to get what I like from them without penalties. Although I have, over many years, found a few drivers I've liked better in terms of tone through a fairly narrow range, I have not found other direct radiator drivers this efficient and dynamic over anywhere near this wide a range (4 1/2 octaves, in this case).

I also agree intellectually about the paper tweeter, or maybe even a silk dome. Now, find me one that is ~99 dB efficient. I haven't seen anything like this in over a decade, and I don't have the patience to wait until I find and acquire something like this, IF it exists.

From your own posts and photos, it appears that you are presently interested mostly in tone. So, you must be ever on the hunt for the great old Peerless, etc. Also, judging from the photos you have posted, you are presently settling for a more sedate presention and a narrower frequency range than I would want at this time.

This thread is titled "Speakers for a Powerful SET", and the opening idea was to try for FR with one SET. At this point, I am still struggling with the one amp limitation, even though I've let go of anything significant under 40 Hz.

If/when I add real subs, I will also try to shrink and seal the Audax boxes and run them only to 50-70 Hz, something like that. While I do not totally disregard Thiel-Small info, neither do I regard it as Gospel. One does have to listen at some point.

Best regards,
Paul S

07-04-2010

Post does not mapped to Knowledge Tree

Romy the Cat

Boston, MA
Posts 10,156
Joined on 05-28-2004

Post #:	34
Post ID:	13920
Reply to:	13914

High frequency and sharp filters

haralanov wrote:
The ribbon is 100% tonally … If you use this tweeter lower than 10kHz your system will be totally unlistenable because you cannot reproduce the proper tone of harmonics.

It is pretty much what I found as well during my experiments with Water Drop tweeter and other ribbons
I played with. Ribbons are toneless by nature and I wish people understand it. Well, they do have tone but it is extremely primitive tone and very predetermined overly hygienic tone – making all music to sound the same. I think a ribbon must not be used for anything by HF artifacts, which in my view start after 10K. High frequency and sharp filters (ribbon and first order does not exist in nature) is the only place what I think true ribbon need to be used. It all was in the older corresponding posts….

The Cat

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche

07-05-2010

Post does not mapped to Knowledge Tree

haralanov

Bulgaria
Posts 130
Joined on 05-20-2008

Post #:	35
Post ID:	13921
Reply to:	13915

Tone and tonal integration

It’s a bit offtopic, but:

Paul S wrote:
I haven't seen anything like this in over a decade, and I don't have the patience to wait until I find and acquire something like this, IF it exists.

The problem is that it does not exist… Simply there are no customer demands on the market for such a product, not to mention it does look like shit. If it is not shiny and good looking there will be no sales, so… And the most ridicules thing is that it is unbelieveably cheap to produce!

Paul S wrote:
Also, judging from the photos you have posted, you are presently settling for a more sedate presention and a narrower frequency range than I would want at this time.

I think the picture misleads you. The cone of my HF/UHF driver is only 55mm in diameter (it is deliberately ragged around its apex) and the voice coil has almost no inductance so it has almost no limitation of frequency response and because of the very low moving mass and strong motor it has 98dB efficiency. I have listened almost all of the tweeters that industry offers with all imaginary extended frequency responses but all of them sounded “closed in”. An exception of this list is Fostex T500a Mk2. A friend of mine bought a pair, so I had an opportunity to do some experiments:

The subjective feeling it gives is the harmonics goes to forever – very open sound, but also very very sharp. It is like someone shoot you in your face with needles, no matter how high you cross it and no matter if you have good SET and good electricity. In addition to that, it does not integrates tonally if you use paper midrange - no matter what you do, so you need very specific tweeter for your Reps driver. You may not believe me, but my small paper tweeter sound even more open – it allows you to hear very subtle nuances of the original acoustic space of recordings. T500a also present this effect, but to much lesser degree – as you look through very small audio window and actually it does not transports you at the original location where the music was played.

One year ago, I also used a ribbon tweeter, the I trashed it and now I don't even want to remember that time :-))

Best Regards,
Haralanov

"If you can't explain it simply, you don't understand it well enough." -A.E.

07-10-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	36
Post ID:	13967
Reply to:	13921

You Can't Always Get What You Want

Haralanov, let me know if you actually have something, otherwise I will continue with the Audaphon ribbon for the first round of the downward-pitching exercise. On the one hand, I wish I'd kept the Peerless 2" paper tweeters I had; on the other hand, they were only about 89 dB, or something like that. In fact, all those great old Peerless drivers were no more than 90 dB, as I recall. The Audaphon is actually very good for "non-present" HF; but... it is a ribbon.

Not the most useful place for this warning, I suppose, but folks should be careful what they stick parts to boards with. I made the mistake of commissioning some crossovers from Madisound a few years back, and whatever glue they used has fused with the cap wrappings, and it sticks everything together like SuperGlue! Who needs anything so serious for sticking parts down?!?

I hope to get most of the hack work done tomorrow.

Today, I hate DIY more than usual!!!

Best regards,
Paul S

07-30-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	37
Post ID:	14146
Reply to:	13967

Wow! It Actually Works!

Too soon to begin in-depth descriptions, but after a good deal more work (and expense...) than I'd bargained for, it did not blow up, and there was not the all-too-usual painful silence in lieu of the first expected sounds. I admit I was as surprised as I was pleased when it started making music, right away; no problems of any kind!

Ironically, after all the changes, it is not so very different at this point. The main benefit now is better integrated upper bass and a good deal more "ease" through the midrange. The Reps is Alnico (versus the Lowther DX4's Neodynium magnet), and so far the Alnico makes for more upward-striving harmonics; nice and comfy, but upwards, like my previous Alnicos. So far, the Reps seems to be slightly "faster" than the DX4, but this might just be a certain clarity the Reps gained when Yoshi cut its whizzer off. I can't say yet.

Ironically, despite the upward stretching (so far...) harmonics, plucked and bowed basses and cellos have more deeply resonant bodies, and left hand piano is richer. Of course, this may have more to do with running the Audax up to 200 Hz (from 100) and using a 1st order filter (instead of 3rd order). Just maybe...

I'm going to let the Reps' notch filter (at ~ 2.3k Hz) settle in before I comment on dropping the ribbon down to 8k Hz. At this point, there are some minuses, but there are also some pluses with the new HF arrangement.

Today, for the first test, I used the ML2s' 8 Ohm taps. I will of course try the 16 Ohm taps again once I establish an aural reference.

Summing: At least as much Music, from less hi-fi.

Paul S

08-01-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	38
Post ID:	14154
Reply to:	14146

Now the Real Work Begins

OK, maybe I got too zealous with the notch filter for the Reps. I aimed to bring it down 6 dB at ~2.2k Hz, using .47 mH and 10 uF against 10 Ohms for the notch, with the Reps itself at 7 Ohms. Right now, drums are too backed off, there is no "hair" to speak of on bowed strings, and I am still waiting for "spiky" treble. I am debating whether to put some kind of clips on the notch filter so I can play with parallel resistors until I get a better sense of what's going on there. Needless to say, of all the parts I do have, I don't seem to have any small-value, 12 Watt Mills resistors...

Not to divide this up like a "reviewer", but HF sounds a little bit "detached" now. It may well be OK, once the upper-mids/lower treble are balanced out; hard to say against an apparent trough. Just now, I suspect that, if anything, I might have to raise the HF x/o a little; it sounds like a slight peak at the HF x/o, exagerated by the trough below.

Midrange soundfield is very clean, calm, relaxed and "extended". It easily "swallows" the LF, and it seems like just the clear, empty soundfield for the "spiky treble" to emerge from, once the "spikes" are there to be had.

Upper bass is indeed much better; I am pleased with the tone that the harder-working Audax is developing; however, BR is still BR. Of course, I would have to double the woofers to go with sealed...

A possible generic point of interest: The extended-range Audax is still not "source-able", at all. In fact, bass imaging is now by far the best I've ever had. Upright bass is now totally detached from the speakers, and it stands alone in the soundfield.

I'm trying to keep in mind that I've used many new parts for this experiment, including the long-unused Reps, and I've added new duties for the Audax and the Audaphon, so I don't want to get too hung-up on any of this too soon. Not to mention that I have to "break myself in" to the new sound, also.

I don't know why for sure, but what I hate worst about tuning is the emperical work (sweeps, etc.). Still, it must be done at some point -- unless I just luck into "success"(!)...

Paul S

08-02-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	39
Post ID:	14157
Reply to:	14154

In Place of My Bowler, a Dunce Cap, or...

...It's about the Curve, Dummy!

This morning I disconnected the notch filter and listened as long as I could stand it. Not long!

Then, I went over my math and remembered that I went the way I did largely because I "already had the parts".

After an extra cup of coffee, I got a good deal more serious with the numbers and I realized right away that although I had centered the filter well enough, it was rather a broad swath compared to the VERY SHARP resonance I want to damp, and my half-assed filter had made troughs on both sides of a softly rounded rise.

So, it's back to Parts Frenzy for the stuff I don't have (...how can there still be anything I don't already have?).

Hard to believe anyone can actually just sit and listen to these things without a filter! I will either temp up the old notch or do something other than hi-fi until I get a filter on that spike.

On the Plus Side, the Reps are imaging significantly better than the DX4s did (and the DX4s are pretty good...).

Off the cuff, to Stock Lowther Lovers and DIY Guys: I guess I just don't get the Big Attraction!

Paul S

08-03-2010

Post mapped to one branch of Knowledge Tree

haralanov

Bulgaria
Posts 130
Joined on 05-20-2008

Post #:	40
Post ID:	14164
Reply to:	14157

Notch filters remedy - the the ultimate illusion

Paul S wrote:
... VERY SHARP resonance I want to damp

Paul, this sharp resonance peak is indication of time domain problem (it could be seen on a CSD plot), and it cannot be damped just by using electrical circuits with the driver, because they only lower the amplitude of the musical signal at that frequency, but the resonance and the processes that form that resonance are still there. Generally speaking, you try to cure one evil with another evil - do you think it is the proper solution? Why don't you try to find out what cause the driver's anomaly and to fix the problem the way it must be fixed?

"If you can't explain it simply, you don't understand it well enough." -A.E.

08-03-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	41
Post ID:	14167
Reply to:	14164

The One Way Way

Haralanov, you might not get it from my actions, but I actually understand and agree with everything you are saying, in principle. At the same time, I have no idea how to address such a sharp peak mechanically, since adding mass as usual would just make the whole area sag, like my pervious filter. Even the "Purist", Tommy Horning, uses notch filters on his Lowthers, and he is the kind of guy who makes his own tweeters to pair with them.

Since it's you I'm talking to, I will share that I had thought to work some voodoo with those little paper stick-on "dots"; but this sort of hit-and-miss fishing drives me nuts. Also, I keep having the "vision" of painting "dope" or something in a ring around the center of the cone (at the "critical distance" from the center, of course!), even though - again - I have no solid idea and no tools to determine exactly where the actual problem is. If you know of anyone nearby with a laser inferometer, please direct me!

You have probably played with the Lowthers and Lowther-alikes. Regardless, this sort of thing comes up every so often. What would you actually do, in the real world? I am very open to useful suggestions (while I wait for another round of parts)!

Best regards,
Paul S

08-03-2010

Post does not mapped to Knowledge Tree

haralanov

Bulgaria
Posts 130
Joined on 05-20-2008

Post #:	42
Post ID:	14169
Reply to:	14167

Ok, let see what we have

Paul, I need to see some detailed pictures of your de-whizzered REPS driver including the visible part of its the voice coil and how it is attached to the cone in order to give you any recommendations how to tweak the driver. You can start a new thread because the subject of this conversation does not have relation with the actual thread. Only few people have access to laser intereferometer, so you might laugh at my advice, but try to inspect where is the area of the cone which resonate touching different parts of its surface with your finger, while playing very complex music at loud levels. Have in mind the reason for the peak may be very very complex so some pictures are going to help a lot...
Regards.

"If you can't explain it simply, you don't understand it well enough." -A.E.

08-03-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	43
Post ID:	14171
Reply to:	14169

You Won't Catch Me Laughing

Many years ago, I taught myself to evaluate a speaker's tone potential by scratching its cone with my fingernails and lightly tapping its surface, so I am very up for this sort of emperical exploration. I might also make a very light "tool", some sort of "sliver", that I could rest on the surface and watch it move.

I have been very lazy about pictures, so far. But I do have a digital camera, and I want to take advantage of the insight you are offering, so I'll see what I can do.

Best regards,
Paul S

09-19-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	45
Post ID:	14501
Reply to:	14395

3rd Time a Charm?

Sheee-it... I finally got the new coils wound and the "new and improved" X/Os "worked out" and soldered together. It will take the (removable!) silicone goop overnight to stick the parts to the DEBZ's integrated X/O "bustles"; then I will tip the DEBZs back up, hook them up, and wait for wife-y to go on an errand, or something.

Is it ironic, or a sad state of affairs, or both that I insist on effing with the best pair of speakers I have ever heard?

Remember, this round is not about "adding" anything, since I pretty much got the "weight" last time. Rather, it is about scrubbing "HF".

Cross yr. fingers.

Paul S

09-20-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	46
Post ID:	14509
Reply to:	14501

Ha! Score 1 for Modern Medicine!

OK, it looks like the 3rd time was indeed a charm.

No, the new X/Os do not really "correct" problems inherent in the drivers. That's another story now. Still, a better power contour helps my DEBZs to sound more correct in terms of voicing and overall balance, with respect to pitch and timbre. "Space" has "emptied out", just like I wanted and, if only by contrast, mid/upper bass has come up in the mix. This latter is actually a side benefit; but it is a nice one.

Electricity ranged from not good to bad today, yet it was less obnoxious than usual, I hope this was because of the lower HF volume.

I used only CDs today, because I wanted to be fully remote capable, for testing. I look forward to good electricity and LPs.

According to my calculations, I have given up some "efficiency" in order to substantially bring down the Lowthers' relentless HF. But I really can't say I noticed any significant loss in efficiency, since in the past I always put the "meat of the music" where I wanted it (in terms of volume) and just suffered the extra HF.

No surprise, there is less in the way HF "detail" now; but I don't miss that yet.

BTW, I wound up setting the ribbon down 6 dB at 13.5 k Hz, 2nd order, and this after re-sloping the whizzers' natural response down by 6 dB/8ve, as well. And there's still plenty of HF.

Paul S

11-11-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	47
Post ID:	14921
Reply to:	14509

After the Gold Rush... Another Kind of Success

After another couple of weeks in X/O Hell, I think I at least have the notches where I want them. Yes, I added another notch filter, because the Lowther response is so ragged that using a simpler curve results in bringing down frequencies I need to leave alone. I also used Martin King's response curve and impedance plots as a basis this time, so attenuation levels are better, as well. This time I also tried to correct a baffle step attenuation from last time, and that worked perhaps a little too well, with the low end (amazingly...) PERHAPS a trifle too loud.

All in all, the DEBZXs do sound much better with my now rather complex X/Os than they did with the stock X/Os. Again, these X/Os and filters mitigate response level anomalies but they do not correct mechanical and electrical problems of the drivers, including the Lowther. However, like I have said many times before, I can no longer listen at all to the "raw" Lowther, so if only in this sense the experiment is a success.

I am currently weighing the WAF of spending even more time on this, to take out the LCR resistors and sub in some pots that will give me a range of resistance, so I can play with this stuff a little more without having to R&R resistors every time I want to fiddle with it.

The DEBZ speakers now make a HUGE, rich soundfield, and the limiting factors are the drivers, themselves. I am also attempting to address these more fundamental issues, but "progress" as such does not so far warrent further mention of this gambit.

Paul S

11-13-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	48
Post ID:	14931
Reply to:	14921

Of Speakers, Amps and Expectations

Against long odds, I got another chance to do some more careful listening yesterday. On a hunch, I put the Fluke on the ML2s, and, sure enough, the left channel had drifted up to 340mA and the right had moved up to 300mA. Once I dialed the operating points back to where I wanted them, I finally got to hear what I had/had not accomplished with the DEBZs and my "Powerful SETs". I was a little anxious to get in a good listen before the new caps go into shock for two months, and I have to say I have accomplished most of what I set out to do. Sure, I simply abandoned ULF, but I will not feel bad about this until someone else comes up with a way to get true FR at live volumes from one amp. Right now, I'm good to the high 30's with one amp, "approaching" live volumes; loud enough to get the message, from classical music.

Since no one but me uses full-blown, original DEBZs, this is not so much an advertisement for this speaker design as it is my own attempts to get FR from one SET amp per stereo side, just like the thread header says. Even if I keep playing with contours, I have fed my curiousity about FR with one amp, and I have in the process pushed the venerable Lowthers a good deal farther than I thought they would go. In doing all this, I believe I have come up against and gained a better understanding of the stock electrical and mechanical limits of the drivers that comprise my speaker system and, in the process I have also, finally, pushed the effective limits of the ML2s, as far as these particular speakers, at this particular point, are concerned.

Not to sum, but to recap and re-frame: I am pretty sure at this point that the range from, say, 70 Hz to high enough can be done quite well with a single powerful SET and 3 -4 "way" direct radiators, with 2 amps per stereo side. To gain dynamics and clarity over what I have now would require "better drivers" (as opposed to significantly different topology), and I have actually started working on this, very slowly. In the meantime, my DEBZs work better than ever now, by a fair margin, still a single SET amp per side.

Paul S

11-19-2010

Post does not mapped to Knowledge Tree

Paul S
San Diego, California, USA
Posts 2,657
Joined on 10-12-2006

Post #:	49
Post ID:	14990
Reply to:	14931

Mission Accomplished?

Sorry, I couldn't resist the header...

Electricity was OK today and either the X/O parts are already breaking in or I had a "lull" in the expected parts shut-down. Anyway, I actually wound up seriously listening to some serious music! Even without full parts break-in, this is as good as I have ever heard, which at least seems to vindicate the approach. No serious congestion in the mid-90s, and more and better than ever LF integration. Space continues to empty out the way I want it to, so details no longer depend on extra volume to come across. This is good because, as I have said, I have brought down the Lowther "detail" peaks.uble notch filters. And today there were many, many of those "startling" sounds at all system frequency ranges.

OK, I am arguably insane... but I have reason to believe that an extremely elaborate, expensive DIY system appraoch I have been put onto would surpass my "perfected" DEBZs. The hold-up here is, how to afford the new system, and what will I listen to while I go through the complex process of building and integrating a new system?

In any case, based on present results (and sticking to the thread), I will be using dynamic drivers and "regular" SETs for the foreseeable future.

Paul S

	Target	Threads for related reading	Most recent post in related threads	Forum	Replies	Views	Started
»	New	$350.000 of audio-scripting foolishness..	Re: $350.000 of audio-scripting foolishness...	Audio Discussions Forum	1	25722	06-15-2004
»	New	Speakers: a hi-fi disaster...	Good writing, T......	Audio Discussions Forum	22	207418	01-16-2005
»	New	Lamm Industries: a special interview with a special com..	Buffer?...	Audio News Forum	105	1323778	09-18-2005
»	New	About Wilson Audio Loudspeakers..	Nagra HD?...	Audio Discussions Forum	53	408547	10-05-2005
»	New	GM70 vs. 6C33C..	Give the GM70 the best shot....	Melquiades Amplifier Forum	13	187311	03-09-2006
»	New	A quest for a better monitor...	Dome tweeters and brightness in SL600...	Audio Discussions Forum	97	961074	06-08-2006
»	New	More power from Melquiades? More powerful tube?..	A new Icon Audio's MB81 Mono Blocks...	Melquiades Amplifier Forum	19	249584	09-01-2007
»	New	A DSET is better then an expensive SET..	DIY Stradivarius...	Audio For Dummies ™ Forum	41	392146	09-21-2007
»	New	NAT Audio Magna -160 Watts of Single-Ended Class A..	The KR Audio ways....	Audio Discussions Forum	5	73112	11-12-2007
»	New	Midrange driver suggestions..	Choosing Drivers...	Horn-Loaded Speakers Forum	34	305782	01-02-2008
»	New	All Active! A DSET and multi-way acoustic system...	Hahaha...	Audio Discussions Forum	14	125048	01-31-2008

	Target	Threads for related reading	Most recent post in related threads	Forum	Replies	Views	Started
»	New	$350.000 of audio-scripting foolishness..	Re: $350.000 of audio-scripting foolishness...	Audio Discussions Forum	1	25722	06-15-2004
»	New	Speakers: a hi-fi disaster...	Good writing, T......	Audio Discussions Forum	22	207418	01-16-2005
»	New	Lamm Industries: a special interview with a special com..	Buffer?...	Audio News Forum	105	1323778	09-18-2005
»	New	About Wilson Audio Loudspeakers..	Nagra HD?...	Audio Discussions Forum	53	408547	10-05-2005
»	New	GM70 vs. 6C33C..	Give the GM70 the best shot....	Melquiades Amplifier Forum	13	187311	03-09-2006
»	New	A quest for a better monitor...	Dome tweeters and brightness in SL600...	Audio Discussions Forum	97	961074	06-08-2006
»	New	More power from Melquiades? More powerful tube?..	A new Icon Audio's MB81 Mono Blocks...	Melquiades Amplifier Forum	19	249584	09-01-2007
»	New	A DSET is better then an expensive SET..	DIY Stradivarius...	Audio For Dummies ™ Forum	41	392146	09-21-2007
»	New	NAT Audio Magna -160 Watts of Single-Ended Class A..	The KR Audio ways....	Audio Discussions Forum	5	73112	11-12-2007
»	New	Midrange driver suggestions..	Choosing Drivers...	Horn-Loaded Speakers Forum	34	305782	01-02-2008
»	New	All Active! A DSET and multi-way acoustic system...	Hahaha...	Audio Discussions Forum	14	125048	01-31-2008