Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
64.180.218.155
In Reply to: RE: Marchand XM44 in hibernation ... posted by neolith on January 23, 2017 at 06:14:08
Took me a while; but here is the thread info I read several years ago. Based on Nils advice I went with the symmetrical 280 and have liked it. Perhaps I will order a few more modules and play around with them as my room has changed, along with a few other goodies.
Posted by Neuro (A) on April 9, 2008 at 19:52:15
I have a biamp setup, and I've been playing with crossover slopes and points for quite some time now. I thought I'd share my observations with fellow Magnepan 3.6 owners.
- 250hz 18dB/oct LP, 200hz 6dB/oct HP (stock recommendation): What was Magnepan thinking? My only guess is that a dyslexic wrote the manual and they really meant 250hz HP and 200hz LP. The overlap results in a "wooly" sound. I'm all for warmth, but I was never happy with the resulting sound. Also, phase coherency really suffered.
- 280hz 24dB/oct LP, 200hz 6dB/oct HP: Hmm, even though this looked close to the curves shown in Davey's response model, this didn't sound right at all. For some reason, putting the woofer and midrange panel in phase here helped some.
- 190hz 24dB/oct LP, 190hz 24dB/oct HP (recommended by Andyr): Now we're getting somewhere! Remember, if you're using a symmetrical fourth order slope, the drivers have to be in phase. Bass and midrange in phase, with tweeter out of phase was better for me. I really loved the sound at first--very clean, very coherent, the soundstage seemed very full without any holes. After time, however, I noticed the signature of each driver.
- 190hz 24dB/oct LP, 200hz 6dB/oct HP: After using this, it seems that the first-order high pass is key to the drivers blending well. There's definitely a gap in the low-midrange though. I used this to tide me over until I got...
- 180hz 18dB/oct LP, 200hz 6dB/oct HP (recommended by Andyr): This seems very tonally balanced, maybe a touch warm in the midrange, but NOTHING like the stock recommendation. Detail is excellent.
Conclusions so far:
I might stick with the last configuration above. After hearing the symmetrical slopes, I do find the asymmetrical crossover's soundstage to be not as solid as it could be. It would be nice to see what a triamped, fully in phase setup would sound like. Alas, that's out of my price range at the moment.
-- Nils
Posted by Neuro (A) on August 10, 2008 at 21:05:48
In Reply to: Observations on crossover point experimentation (MG3.6) posted by Neuro on April 9, 2008 at 19:52:15:
Phil Marchand was kind enough to set me up with a 280hz HP to match the 280hz LP. I never tried it until about a week ago. Needless to say, I am VERY happy!
- 280hz 24dB/oct LP, 280hz 24dB/oct HP: The anemic, lean sound of the 190hz crossover point is gone. Coherency through the crossover point is seamless. Also, I'm hearing lots of detail in recordings I've listened to dozens of times. Timbre is so improved that I wonder how I put up with it before. :) :)
It seems that the bass panel is essential in the ~200-400hz range (which is one reason that the 180hz third-order low-pass works, and is possibly the reason that Magnepan recommends a 250hz 3rd order LP).
When I go 3-way active, I'm going to try 240hz LR4 as well, just for fun. I'll get to experiment all over (but at least I'll be going with LR4 for the other crossover as well).
-- Nils
Follow Ups:
- RE: Marchand XM44 in hibernation ... - neolith 09:06:47 01/24/17 (1)
In Reply to: RE: Marchand XM44 in hibernation ... posted by watts on January 23, 2017 at 18:55:36
Obviously you have been doing your homework. The values I gave you are almost identical to the OEM - the 4th order LP is not quite a true L-R filter having a Q of 0.51 rather 0.50. This is really trivial but I can send you the values to get a 0.51 Q if you are that pedantic :).
Magnepan spends some time developing the crossover design and IMO someone would be hard to come up with a better design and that is why I stick as close as I can to the OEM.
From your research you can see there is a considerable variation as to what people think is better and this may have to do with their rooms and ears.
One comment of interest was " 280hz 24dB/oct LP, 200hz 6dB/oct HP: Hmm, even though this looked close to the curves shown in Davey's response model, this didn't sound right at all. For some reason, putting the woofer and midrange panel in phase here helped some." I find this of interest because swapping the polarity of 1st order filter makes absolutely no difference - the absolute phase difference remains the same and I defy anyone, even those who claim to hear absolute phase, to tell the difference.
The point being that a lot of this subjective and what you need to do is find what you like and then just listen to the music and forget about the system.
I married the perfect woman. The downside is everything that goes wrong is my fault.
Edits: 01/24/17
- RE: Marchand XM44 in hibernation ... - watts 16:56:10 01/24/17 (0)
In Reply to: RE: Marchand XM44 in hibernation ... posted by neolith on January 24, 2017 at 09:06:47
Sometime in the near future I will call Phil and see if he has any notes regarding 3.6 owners preferences, and then I will order a few more modules based on you guys recommendations and experiment again.
Maybe I will post and see if anyone has some used ones kicking around they want to sell first...
Thanks for your input guys.
- RE: Marchand XM44 in hibernation ... - Satie 00:20:45 01/24/17 (0)
In Reply to: RE: Marchand XM44 in hibernation ... posted by watts on January 23, 2017 at 18:55:36
The suggest biamp XO in the manual is obviously a typo. The 3.5 model gives 3rd order 200hz LP and 2nd at 200hz HP or 1st order at 250hz HP. I think they just swapped the freq. for high and low pass when the 3.6 manual was written.
Spacing the crossover using a 6db HP for the mids is pretty much the "standard" maggie approach.
From the 20.1 manual (corrected for obvious typo)
"Set the high pass section at 200-300 Hz, 6dB per octave. Due to standing
waves at the crossover frequency, the low pass section may require adjustment
from as low as 90Hz (12dB Bessel) to 150Hz (18 dB Butterworth). Typical
settings are in the middle of this range. Optimization of the low pass can be done
by ear (with music) or spectrum analysis"
For the T IVa they suggest LP Butterworth 3 at 250hz
HP 12db at 400 hz.
They also suggest that symmetrical 12 or 3rd order XOs at 250-400 hz would work well.
I have tried many combos on my T IV/Neo8 for both triamp and biamp. The main deal is to get good time alignment. Aside from the XM 44 that I used with Marchand's suggested LR4 modules I also used an Ashly XR 77/18 3rd order and later a Rane LR4 to experiment further as well as various 1st order and 2nd order symmetrical and asymmetrical crossovers.
The LR4 symmetrical did quite well but you could tell that the bass was crude as you passed from playing the violin in the upper strings to its low string. And Cellos changed character going from one end of the range to the other. B3 was better in that regard, but 1st order was magic. With the drivers time aligned in an arc, imaging was perfect.
I then went on to switch the setup to the short wall and try my hand at wall loading. After much frustration trying the 1st order again just without managing an equidistant setup with the bass panels face forwards to max out wall loading, I swapped to the classic maggie 3rd LP and 1st order HP, just that I lowered the HP to below the midrange driver's acoustic rolloff and let the Neo8 do as much as they can and filled in
with the bass at B3. If you go up with the LP to the -3db acoustic HP point of the mids at about 250-275hz and fix time alignment then you have a full if not slightly ripe tonal balance but lose clarity as the bass panels are audibly losing detail and take over output in the 200-250hz range. Lowering the LP down to below 200hz lightens the tonal balance and makes it a bit lean to the benefit of clarity and imaging. Time alignment is adjusted to compensate for group delay differences between the LP freq) In that circumstance it is a bit better to raise the bass level somewhat to compensate for the leaner subjective balance, particularly if you take the LP down to 150-160hz.
Biamping with LR4 loses you phase info at the mids and damages imaging and tonal texture. It may not be that bad with the 3.x models but with 3 of them I cooperated on the 6db HP was pretty much necessary to make the most of the midrange, anything else squished the soundstage and made the images lose palpability. The main thing is to carefully tune the B3 LP to match with the 6 db HP. It wasn't very practical to do that with the Marchand but once I had things dialed in I redid the boards and it was fabulous, just that it is more so with only a cap rather than the whole Marchand gain structure. So I ended up selling it.
Post a Followup:
FAQ Post a Message!
Forgot Password?
Moniker (Username): Password (Optional): Remember my Moniker & Password (What's this?)  Eat Me E-Mail (Optional): Subject:
Message: (Posts are subject to Content Rules) Optional Link URL: Optional Link Title: Optional Image URL: Upload Image: E-mail Replies: Automagically notify you when someone responds.
To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.
This post is made possible by the generous support of people like you and our sponsors: