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

---

[ Follow Ups ] [ Post Followup ] Thread: [ Display  All  Email ] [ Planar Speaker Asylum ]

---

[ Alert Moderator ]

---

Are my ears really that sensitive?

Posted by Davy (A) on September 10, 2014 at 10:11:40

I have a set of MGIIIa with stock crossovers. They are the version with 2 ohm wide tweeter and 1 ohm resistor fitted, and the slightly earlier version with a 10uF midrange parallel cap (not 12uF as in latter examples).

I've had these MGIIIas a few years but have not used them much or touched the crossover, as I used my 3.3Rs instead (and did a lot of mods on those). I recently broke a ribbon on the 3.3Rs and have dragged out the MGIIIas to play with in the meantime whilst I wait for some new foils for the 3.3R from Magnepan.

Bear in mind here I am a bit of a caveman and still use (and like) single amp passive drive. I personally find it a bit more coherent than biamp setups I’ve tried. Never tried triamping.

The MGIIIa started life with the older 3 ohm wide tweeter. Magnepan discontinued this tweeter ribbon at serial no 063744 and introduced a thicker 2 ohm tweeter. I’m not exactly sure why. Perhaps because the 3 ohm tweeter was thinner and broke more easily?. They put in a 1 ohm wirewound resistor into the speaker to account for the lower impedance of the new tweeter.
As we know this puts the tweeter output about 3.5 dB down in output compared to the original configuration with the 3 ohm ribbon. I can really hear this decreased output on my MGIIIas with 2 ohm ribbon. The treble with stock config on my speaker is basically muted in the compared to the 3.3Rs. So I embarked on decreasing the resistor value to increase the tweeter output. First tried 0.47 ohms for a while (ceramic 10W wirewound coffin resistor). Much, much better, but actually a bit too bright. The bass seemed a lot less substantial too, no doubt that increased tweeter output changed the whole balance of the speaker a lot. Then I tried 0.68 ohms, and that was not bad, but was too dull and lacking treble detail. Tried 0.56 ohm that was better, but also too little in treble output. So I’d narrowed it down to between 0.47 and 0.56 ohms. I can hear quite a bit different in between these two values. I would not have thought that possible?!

Anyway to get a value in between these two I’ve started using 2W metal film or 2W carbon film resistors in parallel (four or five of them). Tried 0.55 ohms, , - better but still slightly too dull. Then tried 0.51 ohms. Slightly too bright!! I’m now going to try 0.53 and 0.54 ohms, and I suspect (and hope) one of these will do the trick.

I am stunned that such small differences in the tweeter output are audible and affect the balance a lot (in my opinion)

The paralled film resistors do also sound a lot better than the cheap wirewounds, so this is a good (and easy) upgrade to do anyway. The carbon films do lose a bit on the transients (and perhaps have a smoother sound) so I prefer the standard metal films, but the carbon films are still better than cheap wirewound coffins. I also bypassed the tweeter attenuation socket to get better sound.

I changed the midrange parallel cap value in the speaker from 10uF to 12uF and this seemed to decrease the midrange output quite a bit (and changed the balance quite a bit somehow, making the midrange more distant). 12uF was too much so I tried 11uF which was quite a good compromise but I also fine tuned with this parameter and have settled on 11.2 uF. Again small differences are audible.

Will report back when I’ve tried 0.53 and 0.54 ohms on the tweeter.

I think Magnepan was a bit careless by just slapping in a 1 ohm resistor when changing to the 2 ohm tweeter as the audible decrease in output of the tweeter from the 3 ohm without resistor is very large in my opinion. You go from an open, detailed, airy sounding speaker to something that sounds like it doesn't have a ribbon tweeter in it at all. All the fantastic ribbon tweeter transients are lost as well IMO. The 2 ohm tweeter actually measures about 2.2 ohms (measured mine a few years ago) and the original 3 ohm measured 2.75 ohms (measured some I had in a set of Tympani IVa few years ago), so using a resistor of about 0.5 ohms gives roughly the same impedance.

On a separate issue. I am having a slight problem with the sound of the MGIIIa. On the electrical frequency response of the stock crossover there is a bit of a peak at around 1-3 KHz and I think I can hear this. Its making the sound a bit too forward and hard to listen to.

Andy’s MGIIIa tweaks page http://www.integracoustics.com/MUG/MUG/tweaks/AndyR/
usefully outlines a few ideas for getting the FR a bit flatter (and shows the stock response with that peak), but changing those values does involve quite a bit of work (especially if I am using the crossover at speaker level with a single amp changing all the speaker level components). Bit of a longshot but was wondering if anyone had any ideas of how I could try to reduce that excess energy around 1-3 KHz?, without having to resort to EQ or changing loads of crossover values? It might not be possible to easily reduce it unless the stock crossover is changed quite a bit I guess. One thing I had thought of doing which might reduce that energy was to disable the midrange parallel capacitor (turning the midrange bandpass LP into 6dB octave, instead of 12dB/octave), essentially going back to MGIII specification (and also using MGII spec of 35uF worth of caps on the external HP crossover instead of 60uF and the inductor as per MGIIIa specs). The bass LP crossover values are the same in these two models.

The later Magnepan models 3.3R, 3.5R and 3.6R introduce an “audiophile dip” in the presence region around 1-3 Khz, which addresses the problem I am hearing with the MGIIIas. Those crossovers are quite different in terms of values though and also designed around the specific (and different) panel tension of those particular models

If anyone (who is more technical than me!) has any ideas of what might help (and preferably be simple) regarding the peak I am hearing from my MGIIIas, I’d love to hear from you.

Edits: 09/10/14

This post is made possible by the generous support of people like you and our sponsors:

Schiit Audio

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---