|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
72.89.19.193
I took the plunge and added a set of Class D mono-blocks to my existing tube mono-blocks so I am now running 4 amps for my two speakers. My Artemis Eos 2.1 speakers look something like Watt/Puppies so you can better picture this.
The tube amps are Jolida Fusion which produce 200 Watts each from their 211 tubes. They are powering the mids and tweeters, and the Class D's are Wyred for Sound mAmps putting out 250 Watts each and powering the bass drivers. I am using a DBX 223xs crossover. It is only connected to the mAmps so it is serving as a low-pass filter. The Jolidas are powering the tops full frequency spectrum. So in effect it is like running a pair of two ways and a powered woofer. Except that the speakers retain their internal passive crossovers that blended them together and still do.
The crossover cuts off at 24db/octave - rather steep. Perhaps that is why I find the adjustment of frequency (now at 70 HZ) and the gain adjustments to be very finicky for getting a good mesh of top and bottom. Interesting fact: I am running my pre-amp at around the same gain that I did when I only had the tubes.
I am happy with the results. I have the power and control of bass frequencies that I always felt missing with the tube amps alone. I am listening to Steve Earle's "I Feel Alright" album (ripped CD) at this moment and it is rich and tight as it should be. Listening to Ramsay Lewis's Appasionata album (ripped CD) the bass is powerful and tuneful, delightful. Both albums are recommended.
Hopefully this helps people who are interested in going this route, happy to answer any questions.
Follow Ups:
Better to have 1 good stereo amp than 2 mediocre ones. Biwire for best results
Adding the Base Module - which does 25-55Hz doesn't help - with that overload problem.
Any amp will be working hard in the bass anyway, and the assembly is inefficient. see below.
It's a FR system which won't play LOUD, but needs power. }:-|
Kinda like a FullRange LS35/A but with better impulse response!
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Not sure what the vertical bi-amping component is though. That would suggest you are using stereo amps for each side as Wilson did at RMAF one year using VTL amps.
Thanks for that correction - my bad.
Dear Lifer,This is a quite long response, but I do hope you will read it. It took me most of an afternoon. Please feel free to get back to me and ask questions. You will also need to read the review referenced below.
At the outset I note that the EOS design needs a good deal of power.
From my reading you may have cascaded a 4th order active 70Hz HP/LP filter with the original design's own HP/LP filters designed for the same purpose, and which likely operate between 55 up to above 70Hz, and not symmetrically around 70Hz as your insertion of the DBX active xover assumes.
If I am correct you have added a 4th order HP active filter to a 4th order acoustic filter (the upper module's reflex port) and have a steep 8th order acoustic high-pass slope which does not meet or match the acoustic low-pass slope, which might be 5th order or even 8th. That is, there's a notch in the sound you hear but not a room-cancellation dip. This fits with 'finickiness.'
If that is the case I would not be at all surprised that you have found the adjustments with an added 4LR xover difficult or finicky. Especially if the original HP/LP crossover was not set to 70Hz, as I suspect. Yes, real-world passive HP/LP spkr filter pairings often have a wide pass-band.
If the slopes / pass-band are anything like those reported by Stereophile in a review of the EOS Signature (click below) you have likely damaged the intended sound.
In other words? I think erik-sq is likely to be absolutely right! And, you may well be better off doing it all passively with the original crossovers. But wait, there's more further down.
Even if you have bypassed the Base Module's crossover/s, IE including any High-Pass elements - that does not of itself guarantee that you will get the full benefits of high SS power in the bass - without damaging the overall coherence of the sound as intended by EOS.
So, why do I think that?
i) In the review linked below it is stated that the Base Module's HP slope rolls off slowly above 70Hz (IE not at 4LR) and that the Base Module covers the band-pass 25Hz to 55hz. It seems likely to me that EOS may have done the same thing as they did with the two-way module's crossover's and arranged slopes that will help fill in for the distances between the bass driver below in the Base Module, and the smaller one in the main module.
ii) IIRC this model was an attempt at time and phase coherency by EOS, at least for the upper module. Noting that any high-pass filter that EOS applied to the upper module - to integrate it with the Base Module's sound - would need to be first-order _electrical_ only, so as not to affect the performance of the whole design. ? Ask if you like? You have put in a 4LR high pass which, if I am right, is unnecessary and may damage the time/phase coherence EOS may have been aiming for.
iii) As the 2-way upper module is itself a reflex box, it thus already has its own low-pass but acoustic filter (the port) cascading the acoustic output to 4th order, so EOS probably did NOT add any HP filtering. Addendum? "But as KR stated, there is no high-pass filter in the Eos Signature's feed" referring to the main / 2-way module.
So EOS have 'designed by listening' just a single and likley shallow 'low-pass-only' filter for the Base Module's feed.
You could of course return to the original design's intent if you were to use only the original passive crossover's for the Base Module and the 2-way module that sits on it and, passively bi-amp horisontally. I think you should try this, soon, anyway. But, see below first about gain matching.
The power losses involved would likely cost you the same number of db per amplifiers, less than <3db for both modules.
IF however you do really want the full benefit of maximising the power available from using an active xover, AND without losing EOS's intended integrated and coherent sound, which I am guessing you have managed to do.
Well, the only reliable way to do that is to get someone to measure the transfer function of the passive filters as used by EOS to integrate the Base Module to the upper module. And then, build those /that as active line-level filters.
It will likley be just a pair of LP filters.
I would retain any inductance-rise-eq that EOS has soldered across the larger driver's terminals in the Base Module and across the smaller bass drivers in the upper or main module. This is so the matching of the bass/base module to the smaller 2-way module is tightly maintained.
I would expect that active crossovers closely matching the transfer function of the OEM passive items would then have audible benefits. It would play louder and likely more detailed and nuanced, even nicer.
But, I am not convinced that this will always be noticeably better than passive bi-amping, with a big SS power amp in the bass. That is, passive horisontal bi-amping.
There is however one slight downside to horisontal passive bi-amping using a big bass power amp. The more powerful bass amp /amps need to be gain-matched to the other stereo pair of amps used for the mids and HF. It is possible that they will match anyway.
But, you can add fixed-cut passive attenuation inside the amp/amps with the most gain, to achieve matching gain. The MFrs of both models should be able to tell you the amps' gains.
The sole benefit of using matching stereo amps to vertically bi-amp, using the OEM passive crossovers is that the amp's gains do match.
I've had a stressful few days, and TMFW&Finance wants help with dinner, so I am done editing and re-reading.
Lifer, I look forward to your response.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Edits: 10/03/16 10/03/16 10/03/16
Tim, wow, thank you very much for your well considered response that integrated the Stereophile tech info. I do appreciate it. Please note that the model reviewed in Stereophile has external crossovers, mine has internal crossovers in both the top and bottom modules.
One quick and perhaps painful point (for all of the work you did), as I noted in my original post I am only using the low pass output of the active crossover. The mid/tweeter speaker is getting full frequency spectrum.
I tried running both top and bottom full frequency by attaching the mAmps directly without an active cross-over, but when the bass module is run without being attached to the upper module the bass trys to play the entire frequency spectrum and sounds like crap. That is why I got the active crossover - it needs the low pass filter. I don't know why/how the bass module's response changes so drastically when it is connected to the upper module.
One thing that has changed (I am tuning still) is that I have the dbx crossover set to 0 db gain. I might have it modified to bypass the gain pots entirely. Since the mAmps are connected balanced and the Jolida single-ended I believe their is 6 db more signal to the bass amps. I am now crossing over at around 80 hz (tuning).
the Base Module. At 70Hz low-pass in to it you are still not High-pass filtering out the mains module's pass-band down to 40Hz nor the large amounts of power down to ELF being fed into it.
As I read it the EOS design has difficulty playing real LOUD, because the smaller bass driver in the main module is being driven into over-excursion (overload distortion) below its port's tuning frequency. IE signals below 40 or so Hz. There is no 35-40Hz HP filter inside the main module.To put it another way, the original design has a flaw. A high-power FR signal is being fed to the ported main module tuned to 40Hz. ELF signals below that are putting the main module into overload.
On reflection stay with 70Hz but add in active 4LR High Pass at 70 Hz. This will eliminate the Main Module's problem with signals below its port tuning. This will increase the system's headroom / peak loudness. Which was likely your initial objective with bi-amping
BUT you also MUST remove the LP crossover that is inside the Base Module. It is probably set to 55Hz or a bit higher, and it may not be steep.
Adding these two LP filters in series - as I believe you are still doing - is likely giving you a deep notch somewhere between 40 and 70 Hz, unless your room has a peak which cancels it, which is unlikely.
Next, If you have the funds and the DBX unit is out of warranty, I would also convert the 40Hz ELF filter option to 20 Hz. The benefit here is less overload from frequencies below 25 Hz. And once again more system headroom.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Edits: 10/04/16 10/04/16
What a perfect candidate for a DEQX crossover/EQ!
Horizontal would be my choice .....
why?
...describes what one does with ONE amp, that being use it 'vertically' on two ends of one speaker or use it 'horizontally' on the same end of two speakers.
----------
Tin-eared audiofool, large-scale-Classical music lover, and damned-amateur fotografer.
William Bruce Cameron: "...not everything that can be counted counts, and not everything that counts can be counted."
Pls review what I have posted above.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Hi Lifer,
Interesting way to bi-amp, I'm glad you like the results so much. I'm curious why you chose to use a low pass filter. As I read your post, the woofer's original low pass filter is still in place, or am I mistaken?
If so, there are a couple of considerations:
- If you use an active high pass filter before the tube amps you greatly improve the dynamic range of that amp, since you are removing the largest part of the voltage swing which is always in the bass. Of course, not having the woofers in the circuit gets you some improvement by reducing the current draw, but the voltage swing remains. This may help stiffen up the amp.
- If you use no active filters at all you would maintain the original phase matching between woofer and mid/tweeter. By adding another crossover layer in addition to the woofer's low-pass you are probably getting a 6th to 8th order crossover instead of the original 2nd to 4th. Or am I misunderstanding that the woofer had no such filter?
If in doubt, I strongly encourage you to look into Room EQ Wizard or OmniMic to get detailed measurements of each section and examine the crossover region in detail.
Best,
Erik
Ok, I am a bit confused (better than my normal state), since I am not using the active crossover for the upper module I can't see how I am effecting the crossover to it. I am only using the active low pass because I need to limit the signal going to the bass module or else it trys to play it all and sounds awful.
The upper modules were sold separately and many people play them without the bass module so I think my approach is reasonable. If I sound defensive, please disregard.
I think we're all a little confused. :)
What's the "normal" hook up of the bass modules? Is there a passive low pass filter involved?
Best,
Erik
You will find that there is a low-pass filter for the Base Module but with the OP's items it's inside the Base Module and not in an external box.
The real flaw in the EOS design is that the main module is tuned to 40hz by a port but there is no HP filter to the main module.
So the design is prone to excursion overload below the port frequency, by high level (full power -3db) low bass (below 40 Hz signals going to the main module. Adding to this problem is the low sensitivity of the design.
You can either
i) fix that problem somewhere in the wide band-pass between the two modules and go with active low-pass alone, in the existing band-pass between the two modules. Plus a line-level passive HP to the amps driving the main module, 1st or 2nd order is doable.
(Because the MM is time & phase coherent, I think a 1st order PLL feeding the Main Module amps would retain more of its designed in impulse response. The response would cascade to 5th and cone bounce would fall enough to help.) OR
ii) get rid of the existing passive LP xover that is inside the Base Module and actively bi-amp 4LR at say 70 Hz. System headroom would increase even more dues to the higher xover.
Adding a 20Hz HP filter would avoid overloading the Base Module. There's a 40Hz subsonic add-on in the DBX xover which 'might' be convertible to 20Hz. Which would give even more headroom.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
You will find that there is a low-pass filter for the Base Module but with the OP's items it's inside the Base Module and not in an external box.
The real flaw in the EOS design is that the main module is tuned to 40hz by a port but there is no HP filter to the main module.
So the design is prone to excursion overload below the port frequency, by high level (full power -3db) low bass (below 40 Hz signals going to the main module. Adding to this problem is the low sensitivity of the design.
Not a simple active bi-amping proposition.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Thanks for your input to this thread. I agree with you.
Please review my response to Lifer.
I think I've got it right, but I have missed a few errors, etc.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
The 50:50 point is about 350hz.
Too much is never enough
Too much??
Forget my tag line. I don't customize it per post.
My point is that the power even split point for 'normal' music is about 350hz. Power needed falls OFf as a percentage of total BELOW and ABOVE that point. I'd say, however, in NO CASE to have less than maybe 20% of power in the bass end, regardless of crossover.
My panels cross at 600hz. And low-cut about 60hz. So, I don't feel TOO BAD having 2x200 per panel. ONE stereo amp per speaker. My Parasound A23 do well. And since I never redline it, I have NO idea which amp would give up first, the HF side or the LF side. The sub is another 250 or so and crosses about 45hz and is in no danger of running out of power.
Gain match of amps is very important with 2 different amps per speaker. I'd be tempted to work out a DLP solution which can make up for such differences.
Too much is never enough
My advice, from TomServo - the professional horn guy just this year, was to use asymmetric slopes at 150Hz. It was nice to have my own decision to do exactly that confirmed.
And I was planning to, for my QUAD 63s with a SWARM of omni/box subs. I did not want to run my subs above that point.
I will be using 3rd for Low Pass, plus 1/3rd octave Eq.This makes it easier for me to get the wring / phasing right.
The just will be complex, given that I will be switching for absolute polarity at spkr level, driving two big, sealed, DIY, clam-shell drivers, sealed subs (half-built) with a stereo power amp each, running OOPhase. A stereo integrated will drive two tall DIY TL subs using gifted KEF B139s.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
I am so happy to hear that you are listening to these. They were one of the first speakers that I reviewed and they made a great impression on me and on my wife. She still mentions how much she loved them.
It has been such a long time since then but I'd really like to hear the again.
:-)
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Over twelve years ago. We all loved them too and an employee ended up buying them
ET
"If at first you don't succeed, keep on sucking till you do suck seed" - Curly Howard 1936
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: