Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
32.212.76.154
I decided to move the speakers to another wall of the room for their initial trial -- not easy since the feet won't slide on this floor (furniture glides coming Wednesday). Mostly, I wanted to see if everything worked. Which, I'm glad to say, it did, despite the fact that the delam hasn't been fixed.My HTPC still isn't assembled so I had to jury rig my phone, which has to make these the biggest earbuds ever. I found a wonderful $8 program called USB Audio Player Pro that allows you to send high res audio from your phone to a USB DAC with a standard OTG cable and can control the DAC's hardware volume. So I stole the Dragonfly 1.2 from my computer and put it on the phone, then I used ES File Explorer to stream the music directory on my computer over wifi to UAPP. Instant access to all my music files and a pretty good DAC, isochronous USB at 96 kHz, ESS chip with analog volume control. I hooked them up to the A-21. It all worked wonderfully, both DAC and amp came through with flying colors -- if you don't count me reversing the woofer and mid tweeter feeds as a result of the upside-down diagram on the crossover boxes!
While I plan to use the Tympanis in their split configuration, I decided first to try them in their default in-line tweeters-in arrangement go get a feel for their baseline performance. This means that the mid-tweeter panels are too close together and the woofers too close to the wall, so I wasn't expecting a proper stereo spread and in that I wasn't disappointed -- I had to settle for a laterally-compressed image. Today, I'll experiment with split configuration and see what I can do by way of imaging. This shows the speakers as I listened but note that they actually wrap in more and are aligned as per the manual, you can see by looking at the floor how much geometric distortion there is in the panorama shot:
Also,the felt is currently missing from the ribbon tweeter slots, I don't have felt to replace it with yet. So there will be some serious diffraction issues and also presumably an undamped cavity resonance. My main goal was to test the speakers and test the Crowns while I can still send them back if they don't pan out.
I had anticipated having too much bass both because the room is too small for Tympanis and the woofers only about a foot from the walls and in that, I wasn't at all disappointed -- the speakers sounded boomy and muddy. That took about three seconds to fix, using the graphic equalizer in USB Audio Player -- this is the rare kind of bass reinforcement that's really easy to equalize out. Bass was what I've come to expect from Tympanis -- extended, remarkably even in the absence of bass trapping or careful positioning, and with a planar naturalness to it rather than dynamic woofer mush. I didn't push levels to the point of compression but I did listen to an organ recording and it sounded excellent. I sensed that there was a suckout somewhere between 100-200 Hz, presumably as a result of rear wall cancellation, that old dipole bugaboo. But that's all very preliminary, since I haven't played with position and had done the seat-of-my-pants EQ.
My impression overall was one of marvelous tonal balance and at times, the speakers exhibited a remarkable sense of palpability, in that you had the sense the piano was really there sort of way. Also, unlike my MMG's and even my 1-D's, the IVA's sounded great with even substandard commercial recordings, while preserving the detail that dynamics lose. These speakers are beautifully voiced and that's particularly impressive in a bare room, without tweeter resistors.
The ribbon tweeter was its usual remarkable self, transparent without tizziness, just a subtle metallic twang and the very occasional resonance (though fewer than the quasi-ribbon tweeters). And they don't beam like the 1-D's or even the MMG's. The midrange didn't quite match the ribbon as expected, and had some minor fuzz that makes the transition obvious -- however, note that there's some delamination around the segment tape, so I'm guessing that I'm hearing that, rather than just the normal Mylar noise.
Like other 3-way Maggies of earlier vintage, it did sound like three separate drivers, rather than having the more unified presentation of the 1-D's or 3.7's. Again, some of this may be do to fuzz and resonance from the delam. Either way, it will be an interesting problem to tackle.
There was some glare from the ribbon tweeter when I pushed the levels; according to Satie, this can be avoided if you move the XO up an octave, although listening position then becomes very tweaky.
Speaking of tweakiness, I had thought my old 1-D's were tweaky, but the IVA's set a new record for tweakiness! The 1-D's have only two drivers and single first order crossover, whereas these have three drivers and wo higher order XO's. So you have some pretty substantial lateral lobing, which means that they have to be set up for your listening position. The single panels are a lot easier in this regard, since you only have a single angle to adjust.
To my disappointment, the IVA's don't have the dynamics of my old 1-D's. By that I don't mean they won't play as loud -- they will (the fuses set the practical limit) -- but rather that the ribbon driver sounds hard at higher levels and as Satie has pointed out, the midrange is smaller than that of the 1-D's, which runs the bass panels up to 1000 Hz. So while they're far superior to the 1-D's in clarity and have much lower IM on complex passages, they don't have the sense of "feed me all the power you want" ease in very loud passages that gave the 1-D's their dynamic-like slam. But I expect I'll be able to solve these issues with the Neo 8's, and by tweaking the XO as Satie suggests.
Depth of sound stage was surprisingly excellent, better than the MMG's for some reason and much better than my 1-D's, which didn't really do depth at all.
As I said, lateral imaging was compromised by setup and it was also somewhat fuzzy and confused -- the room is too live and zingy, there's no furniture or carpeting or treatment in it yet, so that's to be expected. And the speaker setup was rough, distances, which are critical with these speakers, were only approximate. I'm impressed that they sounded as good under the cirucmstances and expect to do much better today.
Female voices aren't yet where they should be -- Maggies usually excel in that area and these were not quite there. I'd give piano an A, and should be able to get it to A+. Balance on orchestra was superb and even screechy violin recordings were tolerable. Beatles sounded horrible, but those were technically horrible recordings so I'm not sure what to expect! I did miss the famous bloomy midbass -- but again, only after a rough EQ and with no real attempt at optimal placement. One of the best -- and hardest to reproduce -- recordings of all time, the magnificent Klemperer performance of the Deutsches Requiem on EMI, sounded magnificent, stunning -- and damning, if you consider when it was made and how it blows away most contemporary attempts. Even better than on my 1-D's, where it had once so impressed the chief engineer at Sony Classical.
Definition was superb on the ribbon and good on the mid with some planar noise (that's absent from the ribbon, which however has its usual subtle twang), superb in the bass -- much better than the MMG's of course and better than dynamics. Bass was superb overall, with the EQ/suckout caveats. Depth rendition excellent, lateral imaging needs serious work. Dynamics need work, unlike the MMG's they will play loud but the ribbon glare makes it annoying to do so -- but this is not something you would hear at typical audiophile listening levels, it sounds great at those.
Of course, I'm jumping the gun here, these are very preliminary impressions based on brief listening to broken speakers after a coarse initial setup.
Edits: 10/27/15Follow Ups:
Having finished the Crown trial I decided to try tweeters out, and boy, did that improve the imaging! They're actually a bit too wide as they're set up now but I'm getting a great spread. I did lose some depth, I think because the woofer panels are blocking too much of the reflected backwave.Anyway, while the setup is still rough with many things to go, after doing this and fixing a phasing problem in the bass the sound had gotten to the point at which I could just kick back and listen. First, I wanted to see if the could rock out so I listened to everything from acoustical Hot Tuna to metal and boy, they were spectacular! Anyone who thinks planars don't do chest slam hasn't heard Tympanis pushing 110 dB in a small room (the midbass is really up in this room, as one would expect). And yet they retain that planar clarity. Precisely what I love about big planar dynamics.
Studio pop is always a curious Frankenstein combination, one instrument will sound preternaturally real and another will sound seriously strange, while the recordings vary all over the place, from surprisingly clean to hard panned stuff that sounds like it was recorded that through the janitor's telephone mic. But still, somehow, it all worked beautifully and I just forgot about technical setup crap for the moment and spent several hours listening to old favorites, many of which I'd heard many times in the mixing room (loud, flat, and clean, but not nearly as satisfying) and on my 1-D's (a much more interesting comparison).
Then I ended by listening to some Bach organ -- spectacular, although I'm going to have to do some work on the bass -- then some Byrd -- spectacularly realistic on the chorus -- then the Ninth Sympthony -- tried the Karajan first because it's what I grew up with and I heard him do it live from the first row of Carnegie Hall, but the recording was such overcompressed crap that I couldn't listen so I switched to a different version -- and then listened to Kubelick doing the New World Symphony, and to my delight the recording was superb.
So basically I was just hopping through music and wishing I had more time to listen. But technical status now:
- There's some very subtle noise in the background from the delam in the midrange and likely the bass drivers (I know in the midrange because I can hear it when I put my ear up to the area). I'll fix that with some Super 77 when I get back. I had a bad buzz on piano yesterday, but I fixed that by pressing the midrange diaphragm back down on the double stick tape. I assume I'm hearing some Mylar/metal noise too but with the loose wires it's hard to tell which is which. The midrange and woofer don't have the marvelous clarity of the ribbon. However, these are superbly balanced speakers and without losing detail they're a pleasure to listen to even on bad recordings with pushed-up treble or screechy violins. Finger shows the extent of the delamination, so you can see it isn't bad (a few bigger strips on the bottom):
- Imagining isn't as deep as it was tweeters in, spread ss I said is good but a bit too wide. No wraparound on the walls, which I had with my MMG's. However, depth and imaging were superb on that Kubelick recording. I am guessing that that's because the recording itself contained the right ambient cues, so I didn't have to rely on the backwave for artificial reverb.
-- Bass in this small, almost-square room is as expected very modal and rough, and the large panels overload the room meaning midbass will have to be equalized down. The balance as is works well with rock but with classical or jazz the irregularities can be some annoying. I did some quickie equalization that took care of a gigantic peak at about 65 Hz for now. Since it's elevated, this stuff can be equalized out and I'll likely get some bass trapping in there too, broadband behind the chair (since I have to sit too close to the wall) and maybe some tuned tube traps in the corners behind the speakers which are waste space anyway.
Unfortunately, bottom octave performance in this room sucks. While there's output in the 20's and 30's the -3 dB point is closer to 40 Hz, which happens to be where the lowest mode is. Satie's IV's do a lot better, they benefit from modal reinforcement in his larger room. I may be able to get some more output with positioning and by bracing the panels but I may just have to use the sub.
Edits: 11/05/15 11/05/15
There are a few things you might want to try to get better wrap around for the imaging and increase depth in this placement.
First is to get a bit of distance between the tweeter and the sidewalls. perhaps by moving the mid tweet panels just ahead of the bass panels for some overlap. that bigger gap to the wall should allow for better depth and get you better realism in the lateral spread which is probably a bit caricature like in this placement.and does not image well beyond the speaker boundary.
If there is no effect on the lateral presentation, I found that when placing the tweeters too close to the sidewalls in the outie setup plenty of absorption on the sidewall is needed at the tweeter first reflection area. I used any of a number of things to get that from stacked sofa cushions to hanging fiberfill comforters to fiberglass insulation covered in cloth.
The ideal distance for the ribbon from the wall appears to be in the 32" range and to 3' or so without wall treatment...
I didn't like the bass panels in convex setup and prefer them to be concave. If you have not tried it please do. IIRC it helped with image clarity specificity and saturation. Not that dramatic a difference and it did change the bass balance some so may not necessarily be a net benefit in your room.
I tried overlapping the panels a few days ago. They can't be overlapped much because the feet interfere. Also, it didn't sound right, not sure why.
With funrniture glides all in the speakers are now substantially easier to manipulate than they were before, it really wasn't possible to position them accurately before. I tried the concave arrangement, and it did pretty much what you said -- improved the imaging. But the freq response wasn't as good. It was also kind of tweaky.
Then I decided to play a bit, I pulled the woofers way forward to get them out of the way and just put the tweeters where I wanted them. There was a huge improvement in imaging since the tweeters were at the right angle and out from the wall and the backwave was completely unencumbered. Indeed, it was pretty apectacular.
For the most part, the woofers were just attached mentally to the M-T image, it sounded as if the woofer panels weren't playing at all. But but there were a few times when I could hear that the bass image was a different width -- the panels really should cross over at 80 Hz!
I wish I'd taken a picture from the chair because it gives a better idea of how they woofers no longer block the backwave:
Since I was just experimenting with the front panel position and don't have my DSP yet so the woofers aren't in time, and I didn't focus much on bass performance. There was a suckout around the XO where they're apparently out of phase. But overall it was surprisingly good given the awkward position of the bass panels.
So this is getting closer to where I want them to be -- not only is the imaging spectacular but they won't block the projection screen in my narrow room.
After that, I tried leaving the treble panels where they are and moving the bass panels towards them so they were touching, and then apart some to leave more of a gap for the second reflection to get through. But they didn't sound as good as they had. There's still lots of experimenting to do!
When you have the woofers more than 2' closer than the MT then you should swap polarity on the woofers.. You won't be in time but you would be in phase.
And back again if they are more than 5' out.
Have you tried the woofers with the deep bass at a right angle to the sidewall and adjusting the midbass angle by ear?
The deep bass panel was roughly 3' in front of the mid panel's plane and phase on the bass was reversed. I don't remember the exact figure but I did it by placing the bass panels in front of the mid tweets where the XO suckout was worst and then reversed polarity on the bass panels. Then I started playing with their particular arrangement.
Well I know they were a bit out of phase because I could hear a bit of a suckout and just guesstimating the distance to the acoustic centers told me they were off -- we're talking what, a 4 foot wavelength at the XO point. I probably could have improved things some just by flipping tweeter phase but my objective was to see what it did to the imaging when I got the woofers out of the way. But I remembered your post about the woofers in front arrangement and had it in mind, just didn't have time to experiment with woofer position -- actually first I had them parallel to the side walls which actually worked and then tried folding the panel closer to me so it touched the wall, but that was it.
Once I have a feel for what works both by way of imaging and sight lines I can fine tune the woofers but my feeling was that it isn't going to be right until I get the DSP/biamping up and can keep things in time. Even then, it's going to be tweaky around the 250 Hz XO point if I have that much axial separation. A more modern crossover would likely improve that some, e.g., L-R, since the lobing will be symmetrical and won't have the rise. (I still haven't had time to run the numbers so I'm not sure if I can do FIR at 250 Hz. without too much latency to use my keyboard.)
All of this reminds me that I want is my M-T panels to go down to 80 Hz!
Well, you will have some flexibility with the Neo8 line - Below 200hz is dfinitely possible, but I don't see below 100 hz as realistic without baffle reinforcement for the mids a la IRS mid tweet section. I would guestimate a 1' baffle on both sides (with tweeter and mid array) might get you close to 100 hz..
I did do the woofers in parallel to the sidewalls on the long wall but got a discontinuity in the bass with a huge lump centered on 33 hz - that was when we were discussing the 1/4 wave room mode cancellation - which did not occur when you are sitting well away from the plane of the bass panel output.I EQ'ed it out but still lacked energy in the octave centered around 80-100 and it didn't change much with any reasonable degree of EQ.
I did not try it on the short wall arrangement since that would lose me the deep bass modes I like so much. That would obviously not be an issue with your squarish geometry. So that is probably something worth optimizing.
I don't think I'd want to cause diffraction and resonances by adding a baffle. Or lose too much max SPL. Even the Neo 10 would be a stretch. I think it would be more realistic to use the mid woofer panel. I'll have to experiment but that's more work because I'll have to disconnect the panels and kludge some extra feet.
When I had them parallel to the wall they were actually in front of me. So I must have been getting some dipole attenuation but in my brief listen it sounded surprisingly good. If I have some time I'll experiment with the RTA but chances are I won't be able to get to it for at least three weeks since I'll be out of town.
Wondering if I might have the Kubelik/Dvorak #9 you mention in my collection, I'd like to learn the orchestra and/or label, its catalog number? (I want to try hearing what both you and the Amazon reviewer report.)
Well, I apologize, it seems Google led me to reviews of the wrong recording! So that was the wrong review. The version I was listening to is from the DG boxed set, CD 6.
Thanks for that information. Not having the box set, I see this recorded performance (with the Berlin Philharmonic) might be available on a single CD in DG's 'The Originals' series. However having way too many satisfactory performed/recorded 'New Worlds', I won't be giving it a shot.
They look great in your room, they do command attention, but then they should. I sometimes wish I would have waited for a set to show up before I bought my new 3.6's way back when. Maybe one day I'll make that switch... enjoy!
Thanks, Grant. :-)
IME there is nearly no end in trying alternate positions. I did that with my IV-As for twenty years in my current listening room. (One complication is that what might seem good for one given recording doesn't work too well for another.) Finally I reached a point where further changes are detrimental, so I arrived at a point where I'm done with it.
I'm in admiration of the work you reported, having enjoyed reading it as well as viewing the pictures you posted. Thank you very much.
I think I had my 1-D's in every possible position when I had them! Even at angles, etc. At one point, I put them against a row of windows and opened them to hear what they sounded like without a reflected backwave. :-) And as you say, eventually you settle on a position.
This time around, I did the same thing with my MMG's when I still had them because I figured that the best position with the MMG's would likely also be the best position for the IVa mid-tweeter panels, and MMG's are a lot easier to move! And that turned out to be pretty conventional -- best was as far out into the room as practical, standard 60 degree stereo spread (I know a lot of people depart from that but it never sounds right to me) -- and I discovered to my disappointment that the configuration I'd wanted to/planned to use, fireplace behind, didn't work.
But of course, once you have the actual speakers, everything changes -- I thought the split configuration would work better than it did, but maybe I just didn't put enough time into it.
BTW, your pictures served as inspiration, too. These really are beautiful speakers, aren't they? I think the best looking that Magnepan ever made.
I confess it's been many a year (or decade) since I've heard Tympanis, but I have fond memories of the T-III. With them, the bass panel was centered with the tweeter/midrange on the outside. It seems a bit odd to have the tweeter/midrange panels situated so closely together.
I can't imagine positioning current versions that way for optimum apparent stage width.
The mid-tweeter panels are definitely too close, but then, my room is too small for Tympanis. I oould put them on the outside, but then I'd have the opposite problem -- the mid-tweeter panels would be too far apart.
I got them though with the intention of using them in the split configuration, with bass panels against the wall and the mid-tweeter panels in front of the low bass panel. That should put the speakers in the right position for a 60 degree stereo spread while leaving enough room to see the 114" projection screen in between. At 18", after all, the single panels aren't much wider than my MMG's, which were 14". So --
....................................................PROJECTION SCREEN
HIGH BASS - LOW BASS
......................MID/TWEETER
Another possibility would be to join the high bass and mid tweeter panels, and then join the two low bass panels together. Active crossover would limit the low bass panels to 100 Hz and they would live in the wide arch at the entrance to my living room. Basically a planar sub.
Meanwhile, I did some more experimenting today, first pushing the bass panels over so they were touching the wall -- a big improvement in image spread and the additional bass level was easy to equalize out. The spread is still a bit narrower than I'd like but this was the best they've sounded so far, and that was wonderful indeed:
Then I tried the split configuration, but it sounded sick at the crossover point so I gave up for now. I may be able to do better when I have DSP and can delay the mid/tweeter to bring it into time with the woofer:
is the way the TIIIs were designed to be placed: bass panels centered next together (who needs bass separation?) with mid/tweeter panels in optimum positions.
I was thinking of trying that today, it's definitely on the list. There are a couple of drawbacks I can think of -- one is that it would tend to trap the backwave, reducing image depth -- and the other is that I'm trying to find a setup that accommodates a projection screen, which this won't do since I don't have enough throw length to use an acoustically transparent screen in front of the speakers.
Another three driver possibilty would be low bass - mid-tweeter - mid bass, with an XO between the low and mid bass panels. That would put the tweeters where I want them and the low bass panels would be against the wall so would be at the center of a virtual six-baffle speaker (12, if you include the floor reflection), partly compensating for the reduction in bass surface area. Power response would be better too. But it would it would interfere with the projection screen.
A modification of that is just to remove the low bass panels, cross them over at 80, and put them together against a wall. So:
WALL -- MIDBASS -- MID-TWEETER................MID-TWEETER - MIDBAS -- WALL
ANOTHER WALL OR SIDE OF ARCH -- LOW BASS -- LOW BASS
I'd lose some bass and midbass SPL capability, but not I think serious in that small a room. The midbass panels could be angled so that dipole are at the first sidewall reflection, reducing lobing and time smear but perhaps exciting more room modes.
...one is that it would tend to trap the backwave
While the T-IIIs I heard years ago were in a fairly large room and away from the front wall, they used the typical "W" pattern for the bass panels to minimize that effect as seen in these diagrams for both the TIII and TIV. I can see how a flat arrangement would do as you suggested.
Clearly, experimentation will yield the optimum arrangement. Good luck!
Thanks, I'll definitely be experimenting!
It's kind of humbling to hear so many major changes with even minor changes in configuration. But I worked it out with my 1-D's and will work it out with these! Doing that is half the fun. :-)
BTW, my understanding of the zig-zag arrangement of the woofers is that it improves power response in the bass panels. And then you swing the tweeter in to keep it on axis and equidistant so the crossovers work correctly.
Interestingly, I didn't have a manual for my 1-D's, which were a year old when I bought them, but after much experimentation ended up with exactly that zig zag arrangement.
The problem with Tympani's is there are too many configurations. The later models take this experimentation option away from you....which is probably a good thing. :)
I farted around with a friend's set of IV's many years ago and ended up with all the panels arranged in a straight line. Sounded excellent.
Good luck.
Dave.
Do you recall whether you had tweeters in or out and whether the entire linear arrangement was angled to any degree of toe in?
The tweeters were inside. No toe-in.
IIRC, an equilateral triangle of about 13' between the listener and the two T/M panels.
It was a good friend of mines system. He ultimately sold them and purchased a pair of Scintillas.
Dave.
: )
Heh, yes. "With great power comes great responsibility." :-)
OTOH, it also makes the Tympanis very versatile. I couldn't fit a 3.6 in my small room with a projection screen in the middle.
Reverse split config
Pull the bass panels up from the front to the plane of your seat and play the mid tweet panels tweeters in and figure out the best placement. Do not connect the bass panels
Then place the bass panels 3' in front of the mid tweet with the deep bass perpendicular to the sidewall just a few inches away from it and the midbass panel inwards and play with the angle of the micbass to match with the mid tweet. The soundstage should "click" together at the right angle. No promise that it will happen but it did work for me with the stock T IV early on when trying my short wall.
Equidistant arc
Take the bass panels to the walls mid room and arrange them into an arc with the mid tweet too - tweeters in, not far from the spread out arrangement you had before just that the mid tweet panels will end up closer to the front wall. The XO is not designed to align the drivers this way so you will have to move the chair around to find the right position (it is not so much a right position as a less wrong one). Tell us what you think.
Reverse split is an interesting possibility, I had the midbass panel on the outside but with this speaker against the wall that shouldn't be necessary or even the best way to do it. I'll have to try it.
Equidistant arc isn't far from what I've tried, I've moved the seat forward and that means pulling the woofers closer in and changing the angle of the tweeter to keep the acoustical centers of the 3 drivers equidistant. The difference would be the direction of the fold between the bass panels -- I followed Magnepan's folding recommendation because it improves power response -- but it's worth trying both ways.
Having the DSP will make this stuff a lot more practical because I'll be able to adjust the relative timing of the drivers.
I think that was done of necessity due to the narrow room and wanting to keep the "official" placement.scheme .
There is probably more room to distance the mid tweets in this orientation by placing the woofers closer to the sidewalls - you don't need that much width of the slot between the bass drivers and the sidewalls to get the sense of depth.
Precisely what I did (partly as a result of your recommendation).
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: