Planar Speaker Asylum

Welcome! Need support, you got it. Or share your ideas and experiences.

Return to Planar Speaker Asylum

Planar Speaker Asylum

RE: Varkdriver's Gunned 3.6R First Measurements

98.247.165.94

In Reply to: RE: Varkdriver's Gunned 3.6R First Measurements posted by gregmacknass2@netzero.com on July 15, 2010 at 21:04:23

Take a look at the curves on this page and you can see the electrical responses of a stock Maggie 3.6 crossover:

http://www.integracoustics.com/MUG/MUG/tweaks/davey/mg3.6xo.htm

There isn't any bass-boosting going on with a stock 3.6 crossover.

The response measured in the Stereophile write-up is flawed because of poor measurement technique and shows a much boosted bass response which is not representative of how the system performs in a real room. Read the response by Siegfried Linkwitz at the bottom of the page for an explanation.

The crossover of this (modified system owned by Varkdriver) appears to be boosting the midrange of the speaker and not the bass response. This seems to be consistent with the approach Magnestand has applied to all crossover networks.

Cheers,

Dave.

Follow Ups:

RE: Varkdriver's Gunned 3.6R First Measurements - MaggieLover 23:24:40 07/15/10 (11)
I have to respectfully disagree with at least part of your assumptions. I think it meaningless to compare a response curve taken with specific tools in his specifc room with a response curve taken with difference measurement tools in a different space.

Just moving the speakers a few inches could cause a dip or peak of a few db. What I'like to see is a measurement with the same tools in the same room with no changes except the speakers. Anything else is apples and oranges. I had a dip of 8-9 db at around 200hz at one point. Moving my 1.6's and seating position several inches reduced the dip to 4-5 db, not a subtle differnce.

To assume that two frequency response measurements taken in different spaces could be somehow compared, completely ignores the sonic signature of his listenng room, his setup, the furnishing, etc. For instance, we have no way of knowing if the high frequency rolloff is a funtion of his equipment, the microphone, the room, or the new xover. The original stock pair could have also rolled off in his room. The seemingly elevated midrange he's getting now may only be highlighting a high end roll off he was already getting. And, for all we know the midreange may also have been slightly elevated in his room before PGs mods.

Measurements, like statistics, can be interpreted in many ways. I'm not suggesting he is incorrect but sonic memory is very fleeting, and for his measurements to mean anything significant, we need a comparitive set taken in his rooom prior ro the mods.

Maggies, because you can never be too thin!

Mark
- RE: Varkdriver's Gunned 3.6R First Measurements - josh358 06:32:36 07/16/10 (4)
  I agree that moving the speakers can significantly affect in-room response, but I also have to agree with Davey that the Stereophile measurements were botched and don't accurately depict the bass response of the 3.6's. That's because they were taken in the near field. You can't do that with a dipole woofer. Dipole woofers are designed to have a frequency response that rises with the inverse of frequency to offset 6 dB/octave dipole cancellation. If you measure up close, you see a rise in the bass response that doesn't correspond with what the listener hears. (In Maggies, that's achieved by underdamping the woofer.)
  Which leaves open the question of whether the 100-200 Hz suckout is related to the room. Rear wall cancellation and floor reflection problem are two likely contributors to the problem. All speakers seem to have difficulties in this region. However, Magnepan designed its crossover to have a response bump in the midbass:
  http://www.integracoustics.com/MUG/MUG/tweaks/davey/mg3.6xo.htm
  Perhaps to compensate for those phenomena.
  Magnepan's crossover also seems to have a significant dip at the crossover between the midrange and tweeter, and the Stereophile measurements suggest that it's apparent in the near field response of the stock units. The downward slope of the stock curve suggests that like many high end speakers the 3.6's are balanced to favor reproduction of symphonic material, rather than small ensembles. Unfortunately, in two channel stereo, a speaker can't accurately reproduce both, because two channel recordings of orchestras rarely capture sufficient hall ambiance to recreate the tonal balance heard by the ear.
  
  Edits: 07/16/10 07/16/10
  - RE: Varkdriver's Gunned 3.6R First Measurements - MaggieLover 10:47:16 07/16/10 (0)
    My point had nothing to do with Stereophile measurements. I meerly opined that comparing a response curve made by Varkdriver in his room to one made by someone else in a different room with different equipment was like comparing apples and oranges. Very little of real value can be drawn from such a comparison. We know little of the testing parameters and testing technique performed to generate each curve, or the skill of the testers. As a result these curves deserve no more than a "Hmmm, thats interesting" from us
    
    Maggies, because you can never be too thin!
    
    Mark
  - RE: Varkdriver's Gunned 3.6R First Measurements - Davey 07:13:32 07/16/10 (2)
    Hold on there. Just because a summation of the electrical responses yields a bump (or dips) doesn't (necessarily) mean the acoustic response follows. The raw acoustic response of the drivers would need to be included to yield the whole picture. Those graphs are most illustrative of the electrical response of the crossover network and nothing more.
    
    The black (total) curve I should have removed from those graphs originally since guys like AndyR have been confused about it for years. Oh well, my mistake.
    
    Cheers,
    
    Dave.
    - RE: Varkdriver's Gunned 3.6R First Measurements - JBen 07:32:49 07/16/10 (0)
      Glad you brought it up, Davey.
    - RE: Varkdriver's Gunned 3.6R First Measurements - josh358 07:30:41 07/16/10 (0)
      Of course, I wasn't implying that it did. My point was that the electrical response of the crossover was likely tailored to optimize the net response, which includes the acoustical response of the drivers, baffle phenomena, and the reflection phenomena that typically bedevil the 100-200 Hz region. Magnepan knew what they were doing. If PG's crossover (with which I'm not familiar) doesn't have a similar boost in that region, it's possible that the in-room response of the speaker will be excessively depressed in the midbass as a result.
      The result of the dip in the electrical response is also clearly apparent in the acoustical output of the speaker in the midrange-tweeter crossover region in the summed response curve in Stereophile. The near field on-axis response follows the electrical curve quite closely. The lateral response curves though show variation due to crossover lobing.
      
      Edits: 07/16/10
- RE: Varkdriver's Gunned 3.6R First Measurements - Davey 06:07:56 07/16/10 (1)
  Well, that's a rather pessimistic outlook on speaker measurements. Sure the listening room has an effect on measured response, but it doesn't mean well taken measurements are all invalid. When you say the original stock pair could have rolled off in his room then you're making an assumption too. In fact, your whole third paragraph is one big assumption that I might respectfully disagree with. :) The important thing is to determine what's valid and not valid in objective measurements.
  
  Anyways, the key point in Varkdrivers posting is his statement that "the response matches my impressions of the speaker in my review." And that "moving the speaker well out into the middle of the room produces a very similar response." That seems fairly definitive.
  
  As I said in my other response...it would be best to re-measure the speakers with the original crossover in place so a valid relative difference could be seen. I missed the fact that he has already sold the original crossover so I guess that's not possible....easily.
  
  Cheers,
  
  Dave.
  - RE: Varkdriver's Gunned 3.6R First Measurements - MaggieLover 07:35:31 07/16/10 (0)
    Well, that's a rather pessimistic outlook on speaker measurements.
    I'm not a pessimistic about speaker measurements, but measuring speakers to reflect what is being heard at the listening position is dificult at best, especially with dipoles. To compare measurements taken by one person with unknown credentials and specific equipment (the OP) in a space unknown to us, and try to compare the results in any significant way to measurements taken by a different person in a different place, is, too me, a leap of faith. This does not suggest his results are wrong in his room, but without a baseline measurement in the same space to compare to, one must be careful drawing conclusions.
    
    Sure the listening room has an effect on measured response, but it doesn't mean well taken measurements are all invalid.
    Agreed. My concern is not with his measurments, but with their comparison to other measurements.
    
    When you say the original stock pair could have rolled off in his room then you're making an assumption too. In fact, your whole third paragraph is one big assumption that I might respectfully disagree with.
    I made no assumptions The OP indicated, if I recall correctly,that with PGs xover the highs sounded rolled off even before he measured. His measurement seems to confoirm that. All I suggested was that they may have been rolling off in his room even with the stock xover, but with the elevated midrange that roll off might now be more prominent.
    
    Anyways, the key point in Varkdrivers posting is his statement that "the response matches my impressions of the speaker in my review." And that "moving the speaker well out into the middle of the room produces a very similar response." That seems fairly definitive.
    Now I think you're making assumptions about the listeners meassuring and listening abilities, and you seem also to assume the response in his room at the seating position would be similar to your experiences in a different room.
    
    As I said in my other response...it would be best to re-measure the speakers with the original crossover in place so a valid relative difference could be seen. I missed the fact that he has already sold the original crossover so I guess that's not possible....easily.
    That would be the way to go. It would likely provide more meaaningfull results. Even if his tools and measurement procedures are not entirely accurate, they would perhaps at least be fairly consistant from test to test. We could than possibly draw better conclusions from the results. Perhaps Enchantment, who purchased the xover, and has indicated a willingness, could return it to him for that purpose.
    
    Maggies, because you can never be too thin!
    
    Mark
- RE: Varkdriver's Gunned 3.6R First Measurements - sailor321 05:11:45 07/16/10 (0)
  I agree the room is everything. That suck out is due to room dimensions and where the speakers are. My room produces a 40hz bump and 100hz suckout.
  Moving the panels just a few inches into or back makes a big difference on the graphs.
  I will be posting (soon)graphs soon on my IIIa rebuild using AndyR passive xover using RealRTA and the Behringer mic. I have read that the brain does not process suckouts very well but bumps are definitely heard. One cannot EQ suck outs, at least I have never been able too but by moving the speakers around I have been able to lessen them. To tell the truth the music sounds just fine and we should spend more time listening to music rather than sound.
  PG sent me a layout of his xover and I really do not quite understand the design other that is is 1st order. Andy's is straight forward 2nd order. I used parts from Madisound. Sent them the circuit and they recommended parts. Great company.
  So it goes.........
- Hear, hear! :-)) Just focussing on the so-called "HF roll-off" ... - andyr 01:20:43 07/16/10 (2)
  IMO that has nothing to do:
  * with the XO (because it does not have a ribbon lowpass slope )
  * with the ribbon itself, because Maggie ribbons do not start rolling off at 10KHz.
  
  Therefore, it can only be due to either:
  * the room (and its furnishings),
  * or the microphone. AIUI, you need an extremely expensive "pro" microphone to be able to deliver flat-to-20KHz ...so maybe it's varkdriver's microphone which is giving this spurious result?
  
  Regards,
  
  Andy
  - RE: Hear, hear! :-)) Just focussing on the so-called "HF roll-off" ... - Dawnrazor 09:35:42 07/16/10 (1)
    Hey Andy,
    
    I dont think it was the mic.
    
    1st off, I heard a similar roll off at pgs house. I never commented on that because it could have been the tubes in his system. But it was there.
    
    2nd that mic is not bad. Sure there are probably better ones, but I certainly would buy that mic for measurements (actually have 2 of them).
    
    They claim ruler flat response from 15k to 20k:
    
    No one here remembers the bending of our minds
    - http://www.behringer.com/EN/Products/ECM8000.aspx (Open in New Window)
    - RE: Hear, hear! :-)) Just focussing on the so-called "HF roll-off" ... - josh358 10:08:35 07/16/10 (0)
      I have one too but I've never trusted Behringer's measurements, I've seen their supposed response curve and not even a B&K would measure that flat, it looks like somebody made it with a ruler. Which is moot since Varkdriver's has been calibrated but I'm kind of wondering about the accuracy of mine and whether I should have mine calibrated too . . .

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:

Atma-Sphere Music Systems, Inc.