Quad 2905

planaria · « **Reply #80 on:** 20 Mar 2009, 06:11 pm »

Hello Hap. Thanks for noting the EHT power supply, and suggesting shorting it with a resistor, using insulated tools. Should have mentioned that explicitly.

Thanks for your other thoughts too. The Mills are non-inductive, and I've had good results with them. Are the Dueland's that much better? My experience is that moving up from Solen metallized is expensive for not much improvement, so I go all the way to film and foil if I want better. Granted 150mikes film and foil is pricey for 4 speakers, so I may have to rethink this. I don't hear much difference between polypropylene film and foil caps, nor between styrene caps, nor between teflon caps, and so tend to buy on price. Are your experiences that much different?

The 70mike F&F array which I used consists of a Russian teflon F&F .056 and an MIT styrene 1mike as the first connected, and the rest in polypropylene.

Surprising how similar our experiences were. My first high end speakers were Mag Tympani 1A's in 1973. I had considered ESL57's but could not afford to stack them.

Since then I have owned many Magnepans, several Quads, Martin-Logan Prodigy's, and now 2x05's. The coherence and ease of modding the 2x05's is going to make these my last speakers, though. (My wife may be listening.)

Using 3B and 4B amps, I find that the bass is plenty good for my taste. That may change with the Plitron step-up transformers, as they are only 75:1. In that case, I may use isobarically mounted Magnepan IIIa's, which may give a satisfyingly fast sub.

Thanks for your thoughts, Hap.

niels · « **Reply #81 on:** 20 Mar 2009, 11:01 pm »

Dueland?? You mean Duelund? Its a danish name.....they give you excellent (but expensive) crossover parts, or you could look for stuff from the 60´s if you want smooth sound...
If you go from a Solen metallized (excellent sound for the money I think) to a Mundorf Mcap Supreme for instance you wont notice much difference, the law of diminished return kicks in. The sound will be "calmer", "smooth", "relaxed", "controlled" and "highend" sounding.

Anyway, I would be careful using the Radio Shack to measure with, I assume you know what you are doing, this is just info for others then.
You cannot use the Radio Shack for higher frequencies, its not reliable, and besides there is really no point in measuring higher frequencies, this is not where you have problems.
From 100 hz to 4000 hz the Radio Shack is pretty accurate though....
On the net there is a set of correction values that must be added to the measurements you do with the Radio Shack.

Hap · « **Reply #82 on:** 21 Mar 2009, 03:17 pm »

Hi planaria,

Life's obsessions are strangely similar in many ways; especially for transparency junkies like us with the tweak bug to boot.

The R/C input circuit parts upgrades are definitely a great place to start; the Mills / Solen parts will be a great improvement over the cheap ceramic resistor and electrolytic cap Quad used; however those bright blue 220uf caps are actually pretty decent...
No question the law of diminishing returns comes into play when taking the cost consideration into account. The high priced Auricaps, Mundorf, Teflons, etc. will yield less positive results over the huge gain you got over the stock parts.
Hey if you like what you are hearing from the Solen/Mills mods you did at the values you currently have, go with it.

The first R/C input cap mod I did on a pair of ESL63's, I used Mills MRA10, 1R5s and two Solen 110uf MP caps which sounded great to me.
Later I did a similar mod for a buddy but used some Duelund 15w, 1R5 ohm resistors and 2 Auricaps (200uf total) and thought his speakers sounded a bit more open than my original pair; but at over twice the cost.

I agree with your assessment of the caps vs. price issue, and also do the same purchase on value when it comes to this type of application.

Believe the rule of thumb on the capacitance values for the ESL63 on up models input circuit is that more capacitance; like up to 330uf will give the speaker more weight in the low end; less bass between 125-220uf.
However there will also be a tradeoff of the placement of the soundstage width and depth when you play with these values.
I have never tried using an array of smaller bypass caps like you have done, so am unable to comment on the results.

Rob Mackinlay of ER Audio in Australia is a guy who I consider an electrostatic speaker guru; suggested trying an array of caps totaling 150uf along with a good WW 1.5ohm resistor for the input circuit; which is way below the stock value of 220uf Quad used.
But going to this lower capacitance opened up the midrange and top end, tightened up the soundstage; but also seemed to move it a bit more forward....just my impressions here.

I'm sure you could detect these changes soon after the R/C parts you replaced burned in a bit, especially with the 3B /4B amps you are currently using with the Quads; which we all know are very amplifier sensitive.

Upgraded signal wiring and binding posts will also help a bit; but being the quads are powered electrostatics, other than the signal input wires, the other wires carry the voltage for the EHT and delay lines, so using high end wire will not give you significant benefits.
Using a good quality Teflon jacket copper stranded wire of the same gauge will work fine in this application if you want to replace the cheap stock wires.

Doing modifications to damp the speaker frame structure and stiffen up the panel supports will also improve the sound; but that means hacking up your speakers and coming up with a design that will be visibly appealing; if you are into that sort of thing.

Let us know how your Philtron step-up transformer mod goes; I would like to hear how you plan to implement that into the system.

Cheers.

planaria · « **Reply #83 on:** 21 Mar 2009, 03:47 pm »

Thanks Niels, Hap. Appreciated.

Hap · « **Reply #84 on:** 22 Mar 2009, 02:53 pm »

Hey planaria,
You are most welcome.
Please keep us updated on your progress.
Despite the wife bending an ear to your hifi discussions, doubt the 2905's will be your final pair of speakers...
Cheers.

planaria · « **Reply #85 on:** 11 Apr 2009, 03:47 am »

The Plitron transformers by Vanderveen are superb in the Quad 2905.

So far I have only jury-rigged one transformer, desoldering connections to the two step-up transformers and soldering positive and negative terminals to the leads, and ground to ground. That leaves part of the protection circuitry intact (zener overvoltage), but excises the triac (anti-arcing?) circuit.

The sound was obviously superior to stock and well worth the cost. Then I played with the signal shaping resistor (1R65) and capacitor (220uF) in parallel. Best results were obtained by getting rid of them altogether. The sound is now superb, esses sounding perfectly natural (for $300 !!!). I played the other 2905 with improved resistors and capacitors for comparison, and it now sounds relatively poor: muddy, distorted, shrill. Of course, volume suffered because the Plitron has 75:1 turns ratio instead of 100:1; but so what?

A reminder: this is best done by a technician who is familiar with ESL's, as modding ESL's is dangerous. See the previous posts.

That is all I have to write so far, with less than 10 hours listening. Will keep posting.

planaria · « **Reply #86 on:** 3 May 2009, 05:43 am »

Progress report. Please see my earlier posts for safety warnings. This is a report of my experiences and not a suggestion - I will not be responsible for any harm or loss incurred while conducting modifications or experiments on equipment. Any use of this information is strictly at your own risk.

Permanent installation of Plitron PAT-4134-ES step-up transformers in Quad 2805 and 2905 pairs complete. Used WBT Nextgen binding posts and Mills MRA-5 resistors to replace factory components. Bypassed all protection circuitry (not recommended unless your amplifiers are exceedingly stable. More on that later).

No signal shaping components required, except for Mills resistors. The 2905's sounded best with the recommended .82 Ohms (from 4 x 3R3 in parallel). The 2805's sounded best with 1.1 Ohms (3 x 3R3 in parallel). Now for the first time the 2805's and 2905's have similar sonic signatures. I hypothesize that this is the source of the difference between the factory editions of the speakers.

Quad 2905's are angled inwards at about 10 degrees, with 2805's angled inwards at about 70 degrees. 2805's are raised 15 cm. Both pairs are used for stereo.

Sound is glorious. Voices are human, cello sounds edgy, pan-pipes are raspy, the make of piano is obvious (not possible to confuse Steinway and Yamaha).

Although I was quite satisfied with the Bryston SST's, I decided to try for better with a JL Hood inspired design. I settled on a Class A version modified for Quads, using two pairs of output transistors. Since I was making my own, I limited the power supply rails to less than +/- 30 volts, which is where the Quad protection circuitry intervenes. Thus, by eliminating that protection circuitry, I was eliminating nothing of any use or importance WHEN USING AMPLIFIERS OF THIS DESIGN.

The obvious weakness in the JLH design is the 220 mike cap in the feedback circuit. This can be reduced to 70 mikes without much consequence, and 70 mikes is just possible with a film-and-foil array. Moving the power supply out of the music room removed any sources of mechanical hum. I also changed the power supply to a Plitron isolation transformer and step-down, and a HexFred bridge followed by LCR filters.

The sound is magnificent, magical. It is by far the best I have ever heard. I wonder about the new Brystons, though ...

planaria · « **Reply #87 on:** 24 May 2009, 05:24 pm »

RE: Quad 2805, 2905, Plitron step-up, JLH amps; see above

Extended listening is proving very rewarding. Maurice Andre vinyl (trumpet) shows an astonishing range of virtuosity, and an extraordinary brass sound. The trumpet is at times rude and nasty, just as it should be. A little more polite than live performance, it is true, but nevertheless unique in my experience. Polyphony is exquisite. Brass bells are striking (Solar Winds), and the Renaissance Spagna is weirdly alien (La Spagna).

Digital is also improved, and even 44KHz is tolerable (with a filter). Gorecki's 3rd Symphony is a brooding masterpiece, as it should be: the bass viols are heavy and dolorous, the soprano is alternately sad beyond humanity and threatening (I don't understand a word, mind). This is all reproduced through a godless CD, filtered at 13KHz.

Last night a friend heard Dylan from his first albums, and remarked that it sounded like his first live concerts (Bob Dylan, Times they are a'changin', Freewheeling). Even through those 1960's CBS records!

The speaker/amplifier combination is terrific, but I also want to put in a word for the ESL57 - Bryston SST combination, which is beautifully musical. Nothing close to the resolution, mind, but it makes a wonderful surround's impression of the concert hall, and is very satisfying to listen to on its own (for a lot less money and a LOT less trouble).

planaria · « **Reply #88 on:** 30 Jul 2009, 05:44 pm »

MG IIIa as an isobaric sub

Put two Magnepan Magneplanar MGIIIa together in an isobaric configuration (actually had a pro do it - the damn things are heavy!). Rebuilt, rewired, and reseated the panels with hybrid elastic glue, then reinforced one frame with 1" aluminum channel, and sealed the two frames together with the same glue. Power is from a Bryston 4B SST. Low pass filter is 12 dB/octave at 70Hz (6db point), using Solen metallized caps (600 uF) and Alpha-core 12AWG inductors (8.6 MH).

Sound is improved on some music, providing a lower floor than the Quads, and providing an extra 6 square feet of fast radiating area. At these frequencies, the isobaric IIIa's are almost a match for the Quad 2905s' speed, and the sub does not notably detract from the electrostatic snap. Bass to treble coherence is mostly unaffected. A worthwhile tweak - but not for all music.

studley · « **Reply #89 on:** 30 Jul 2009, 06:20 pm »

Quote from: James Tanner on 15 Dec 2008, 09:29 pm

Hi All,

I have been playing around with my 2905 Quads recently and one thing I have noticed is that no matter what room I have them in there is a bit of a bump in the frequency response at about 50Hz. So I tried a 10B electronic crossover over the weekend and crossed over the Quads at 50Hz into a set of 'subs' as an experiment. Thought reducing the bump (resonance?) at 50 Hz and taking the load off the electrostatic panel may provide a more open sound.

Well so far the results are just excellent. The soundstage improves and the overall volume level of the system is much greater with less distortion. I am wondering if the 50Hz bump (resonance?) is part of the reason why some find the 2905's just a bit 'thick' and not quite as 'fast' or 'detailed' as the older Quads?

I will investigate further.

James

James
any updates on this as I'm thinking of doing something similar using a modded DCX2496 before my 989s
Ian

James Tanner · « **Reply #90 on:** 31 Jul 2009, 05:07 pm »

HI IAN,

YES I SETTLED ON MY SMALL ROOM - 14X12 - AND THE 10B ELECTRONIC CROSSOVER AND 2 SUBS. ALL I CAN SAY IS MAKE SURE THE CROSSOVER HAS PLENTY OF FLEXIBILITY FOR SLOPES AND CROSSOVER POINTS AS WELL AS BEING AS TRANSPARENT AS POSSIBLE.

JAMES

ruggerobc · « **Reply #91 on:** 8 Jan 2010, 12:28 pm »

Hi All,

this is my my very first post on AudioCircle, so let me start by saying hallo to everybody.

I have a pair of Quad 2905 since February 2007 and I must say that I am very happy with them. I share all of James's findings and impressions on this great speaker.

Curiously enough, my listening room and the way I have positioned the 2905's is similar to James's set-up. Actually, my room is even smaller than James's: it is just 12 feet and 10 inches wide per 9 feet and 8 inches deep, with an height of 12 feet and 1 inch (old Florentine building). The speakers are placed along the longer wall, which is entirely occupied by a bookcase (with a chimney in the middle). Behind the "listening sofa" there is a relatively large archway. The speakers are 3 feet away from the back wall (i.e., the bookcase) and more or less the same from the side walls. They are 5 feet and 9 inches from the inner edges of each other. The distance from the listening position is also 5 feet and 9 inches. So, this is nearfield listening, isn't it! (The room is not treated for the time being. In the future, I will place at least a column of three Acustica Applicata DaaD3 in each of the corners behind the speakers.)

While I do not rule out and, indeed, assume that the 2905's would work even better if I were sitting farther away from them, I must say that even now I am extremely satisfied. I cannot think of any better speaker, at least in that price range, for listening to the music I love (chamber music - and classical music in general -, small-combo jazz, female voices) in my little room. I should add that I drive them with a pair of 845-based tube mono amps manufactured by T-Audio, a small Italian firm (almost unknown even in Italy) that designs and produces some truly exceptional stuff. Despite their moderate power (33 W RMS per channel) these amps drive the 2905's easily and to great sonic results. Needless to say, volume levels can only be relatively moderate (although I like to "pump up" the volume, when possible).

I have a number of questions, for James, Planaria, and all of you.

@ James: Can you tell me more on your experience with the subs? Of course, which subs are you using? How (I mean, where) have you placed them in your listening room? How have you set the 10B electronic crossover?

@ Planaria: The idea of modding my Quad's is intriguing; do you know anyone one could do the job in a very competent manner here in Italy or in Europe?

Furthermore: What cables do you use to interface the 2905's with your amp(s)? I use a pair of old MIT MH-770 Twin Tube Series 2 - they do a pretty good job. What would you recommend, as a possible alternative? Also: what power cords do you use? Do you use any power filter/conditioner for your 2905's?

Thanks in advance for your feedback.

Ciao,

Ruggero

James Tanner · « **Reply #92 on:** 12 Jan 2010, 01:32 pm »

Hi Ruggero,

WELCOME!

I am using two of the PMC - TLE subs - see below.

Its is very frustrating sometimes because there are cuts where I like the sub in and others where I do not. The 10B is crossed over at 50Hz because with the 'raw' Quads there is always a slight bump in frequency response around 60Hz in any room I have tried them in.

ciao,
james

ruggerobc · « **Reply #93 on:** 12 Jan 2010, 04:59 pm »

Hi James,

thank you!

Extremely interesting. I believe the bump you mention was also reported by Jonathan Valin in his TAS review.

Have you tried other crossover settings?

May I ask you if you an idea why the subs sometimes work, sometimes not? Is it true that no sub is able to go as fast as the Quad's?

(Please accept my apologies if my questions are naive or even nonsensical - I am afrad I must be the most ignorant among audiophiles ...)

Ciao (and grazie!),

Ruggero

James Tanner · « **Reply #94 on:** 12 Jan 2010, 05:05 pm »

There are no 'nonsensical' questions ---just answers.

The subs certainly allow the Quad to play slightly louder and deeper and most people can not hear them in the system - they integrate well.

My sense is though that with most material the Quad goes low enough - at least in my small room - to satisfy all but the biggest dynamic changes. So the Quads tend to sound a bit more coherent when I use them without the subs?

james

JRace · « **Reply #95 on:** 12 Jan 2010, 06:08 pm »

Those subs look tiny beside the Quads!

ruggerobc · « **Reply #96 on:** 12 Jan 2010, 07:14 pm »

Thank you, James!

Ciao,

Ruggero

planaria · « **Reply #97 on:** 15 Jan 2010, 04:07 am »

Hello Ruggero, welcome.

Sorry, I don't know - have you tried the nearest Quad and Martin-Logan dealers? I understand there is a company in Germany which manufactures ESL panels - they might do the mod. There must be lots of Brits, but then shipping might be an issue.

The main concern I would have is liability - I would not want to be responsible for any accidents, so I would not hire anyone directly. I would use a shop which is supposed to understand the hazards and is therefore responsible.

As for line conditioning, I tried a $1000 cord and a $3000 power conditioner. When feeding the Quads, neither made any difference whatsoever. When feeding the Brystons, the cord made a small improvement. The power conditioner made a slightly larger improvement. The best was a Plitron isolation transformer wired into the house wiring (Plitron 8575-X0-02, 2000VA, if I remember correctly). It was also the cheapest. Since these can hum quite loudly, especially if they are working hard, I placed mine outside the listening room in a metal box. Even the Electrical Inspector was impressed with the quality of the power!

In my opinion, changing the step-up transformers gives 10 times the improvements obtainable from line conditioning, so that is where I would begin. I have never heard any speaker which comes close to these modded Quads, except in the bottom octave of course.

I use Goertz 13 AWG ribbon cables for the Quads, as short as possible. I find them transparent and inexpensive. The only problem is the insulation on each ribbon, which is a bit hard to remove. A wire wheel on an electric drill should work for you, though. If you want to terminate them, I found Audio Note banana plugs to work well. Make sure that you don't use anything magnetic.

Hope that helps. Enjoy your Quads!

planaria · « **Reply #98 on:** 15 Jan 2010, 04:11 am »

Hello James.

You mention that you find the Quads more coherent without the subs - that is my experience too.

ruggerobc · « **Reply #99 on:** 15 Jan 2010, 07:51 pm »

Hi Planaria!

Thanks for your kind and highly informative reply. I truly appreciate. Let's now start the quest for an experienced "modder" (via a reliable shop - good point).

Ciao,

Ruggero