Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
66.36.139.108
To share , from the best of my knowledge:
after living with my modded MMG 1st order /second order for about one year (black gate ,jensen PIO copper , mills MRA and sledgehammer), i had tried the serie crossover 1st order (per SMGA, excepted the capacitor was 28uF), the result was totally unlistenable and this with premiums parts (coil was Solen 0.82 mH 12 GA , and capacitors was black gate AC and Jensen PIO). So i started a small study on crossover for planar. On the net , i had found the schematic of apogee caliper , they were my pinacle loudspeakers,(i got a pair for 6 month in 1988),
so i had adapted the schematic for the MMG ,principally with recalculation of the 4 cross-over frequency points ,from a hypothetical 3 ohms to the 4 ohms hypothetical impedance of the MMG , at crossover points. (i dont had the real impedance plots for each driver of theses 2 loudspeakers).
After listening test ,i supprimed the contour network of the caliper for a simple 2.0 ohms resistance instead ( the MMG tweeter appear to had a flat output relative to frequency). The result : simply amazing , a totally different loudspeaker. The tweeter seem free of
hardness and congestion now ,the dimensionnality ,spaciousness and autority had take a big step. the MMG can play a lot louder now ,
the apogee crossover was and still are state in the art.
The caliper had a 2nd order Riley-Linkwitz ,with a first blend of 6 db per octave per drivers ,at the crossover point ,for approx one octave lenght ,after each driver go to 12 db/octave .This give a smooth blend for the drivers for one octave to integrate well ,and at the same time prevent the drivers from 1st order saturation (for the tweeter) or going in the resonating area(for the woofer). Also the first points appear to be
probably spread apart 10% ,probably to compensate for components tolerance ,to be sure to not have peak at cross-over points.
Also Riley-Linkwitz crossover had differents caracteristics compared to 1st order Butterworth. the woofer section of original MMG was closer to a 2nd order Riley-linkwitz.the recipe :
take the fist order/second order shematic for the MMG ,and :in the tweeter section :
replace the 24 uF capacitor for a 28 uF capacitor , at the resistor jumper ,place a 2 ohm 12W high quality resistor, also run a 3.5 mH inductor in parallel with the tweeter ,at the tweeter posts (not at the amp posts).
tweeter have to be in phase with amp.in the woofer section :
keep the 2.20 or 2.25 mH inductance , replace the 25 uF capacitor for a 16uF. woofer as to be wired phase inverted with amp.place the MMG with tweeters located inside , like the apogee ,
with little toe-in (2 to 3 deg approx).happy listening ,hope this can be helpfull ,i will never return
to 1st order tweeter.
Follow Ups:
On your crossover, a couple of things...Firstly, the woofer panel is a little over 5 ohms impedance and the Tweeter is ~3.8-4 ohms. These speakers are not the same as Apogees. Your crossover calculations are not accurate. That doesn't mean that everything follows exact calculations, but rather that your justification is incorrect.
Now, I think you are kind of on the right track. After testing different values, I ended up with:
30uF on high pass
2.5mH on lowpass inductor
33uF on lowpass shuntThe lowpass approximates a Bessel filter (using a low DCR inductor like Erse). I found the 30uF to sound better than lower crossover points. Your highpass is lowering and your low pass is rising. This may look better in simulations, but most simulators are based on flat frequency response from the drivers- the dip you see in a simulation wouldsimply attunates a rise in output at the crossover frequency in the tweeter section. I think if you do some measurements, you will see a rise in the midrange output(not flat).
Lowering the crossover point of the second generation has a number of benefits due to broader dispersion in the midrange and the ability of the 8 trace tweeter (essentially the same as the MG 12) to handle it. However, by raising the frequency of the lowpass, you will increase the overlap of the drivers, not only increasing output, but also increasing interactions, cancellations, and resonances on the diaphragm caused by two drivers playing the same frequencies on the same diaphragm that are not in a phase aligned crossover topology.
Please see the frequency response of the 1st gen MMG here:
http://www.enjoythemusic.com/magazine/viewpoint/1199/donibbles.htm
I'd try your crossover if I hadn't already gone with the early style in my MMGs.
I may anyway - just not right now. I'm quite happy with the improvement the 1st gen crossover made (or maybe it was the better components). For now, framing them (in hardwood) is my biggest concern.
Here a quote from Northcreek audio ,from the similar apogee stage crossover :Comments on the original design:
One thing about Apogee was that we were very particular about crossover component selection, and we selected only by ear. Apogee at the time of the Stage's production was quite possibly the only manufacturer in the world using 12 AWG inductors in their woofer circuits, and in fact 12 AWG inductors were not available on the world market until several years later. Likewise, the Sprague and ASC capacitors were chosen by ear simply because they sounded better than any other capacitors available in their day, and still sound better than the great majority of metallized capacitors in use today. Likewise, Apogee was the only high end audio company using Ohmite 1% mil-spec resistors; everyone else was using (and in most cases continue to use) sand cast resistors. Apogee's crossovers were probably the most expensive in production in their time, and would still cost an order of magnitude more than the great majority of crossovers produced today.
The original crossover design was a simple Linkwitz-Riley second order with the woofer in reversed phase. This type of network allows for the gentlest blending of the different drivers throughout the crossover region and also required the fewest crossover components. The second order high pass also protected the tweeter ribbon from overexcursion, and in fact the ribbon will begin to twist at high excursion long before it goes into a failure mode.
Both the planar magnetic woofer panel and ribbon tweeter were the most advanced produced at that time. There have been some advances in planar magnetic technology since then, largely to increase the efficiency of the drive unit. The tweeter ribbon and its longer siblings still represent the state of the art and remain the best ever made.
The Apogee Stage was superbly designed loudspeaker, considered among the best sounding loudspeakers available in its day, and still much better sounding than the vast majority of loudspeakers produced today. What really amazes me is that the loudspeaker was designed almost completely by ear; the time-windowing measurement equipment that is commonplace today had not been developed in 1988, and this loudspeaker is virtually impossible to measure by conventional means.
Hey Jaques,Why don't you write up a step by step process for this with some picts, and have mart add it to the tweaks section?
That way, those wishing to "appogeefy" their maggies will know what to do.
Ho ,im sorry , i was thinking this forum was called Planar Speaker Asylum (in general), not Magnepan speaker asylum. By the way the top
of the line Magnepan speakers MG3.5 MG3.6 MG20 and MG20.1 used some kind of second order crossover for the tweeter section.
Maybe the bottom line MMG dont have the right to sound as good
as the others maggies ,they are only marketing tool ,after all.
(probably why their original crossover was downgraded to 1st order/1st order ,to let the place for the MG12 in the market).
End of line.
.
Please see in this link :
http://www.thomas-schick.com/Apogeed.htm
on the schematic :
the network (0.1 mH inductance ,10R and 8.06R resistors) are remplaced with a single 2.0R 12W resistor, the 4 x10 uF caps are reduced to 28 uF total capacitance.
the 2.7mH inductance is remplaced with a 3.5mH ,
the 2.0 mH inductance is replaced with a 2.20 or 2.25 mH
and the 2 x 10uF is reducted to 16uF total capacitance.The MMG is a remarkable loudspeaker , easy to modify ,
easy to live , incredible potential for upgrade.
.
Yes, everyone can make the maggies sound better even mod or change the XOs even make better frame after all mag's designers haven't get a clue what the h*ll they're doing.
Henry
You're right. I'm sorry. It's just hard to resist the temptation to defend Magnepan against such outrageous statements. Even you shot back one quick reply. Oh well, it's time for a visit from our good moderator! It's just too bad Jacques' original thread will have to be sacrificed (or will it?).
Henry
I know, but I *was* feeding him.
Are you serious? Their recipe is absolutely brilliant. Their speakers sound excellent as they are designed and what a design they have for manufacturing and shipping efficiency! It would be very difficult, if not downright impossible, to find anything that can equal their sound in a dollar to dollar comparison. Does that sound like the product of a clueless company to you?Do they cut corners here and there? Yes. Doesn't everyone? Okay, with the exception of a few manufacturers like Wilson, the answer has to be affirmative. Cut corners are a blessing as they contain costs and provide those interested in redressing such shortcomings the possibility of making improvements. Can those of us who are brave (and handy) enough to rework our Maggies improve on their products? Yes we can, but that doesn't mean the original product was crap. As the saying goes, "You can't polish a turd."
Your statement is not only blasphemous, but it demonstrates a complete misapprehension of our goals as tweakers. I'm certain that you won't find one supporting voice among those of us who have modified our Maggies. We'll all tell you the same thing, "Magnepan is a brilliant speaker manufacturer."
Thank you Mr. Whiney!
you're a dick wazoo
Why, because he had a reasoned, astute reply to a stupid one, and rightly pointed out the posters lack of understanding?You could kind of say the original poster took it up the wazoo...
I don't really care what anyone thinks of me. This whole thread has gone way off track and I'm sorry for my part in taking it there. It just irritates me when someone makes such offensive statements about Magnepan. You and I know the truth, and that should be enough for us.I wasn't trying to "fan the flames" as it were. I just hoped I could enlighten a misguided (in my humble opinion) individual about why Magnepan isn't clueless. The last comment (about washing his mouth out with soap) was meant to be funny.
.
____________ // ________________
I'll wager they know grammar a bit better than you do."
Henry
mag's designers haven't get a clue what the h*ll they're doing.I don't think there are any here who think that. The ones who have done the mods, etc. will be the first to tell you that he Mag designers do a hell of a lot with run of the mill parts and for their given price points you simply can't beat their speakers.
If they used better parts they would have to hit much higher price points, and well they probably wouldn't be around.
The fact that you can significantly improve them with some mods, is icing on the cake.
They know what they are doing. They are building fine speakers for reasonable prices, while making enough profit to stay in business.A person can determine where Magnepan has (for good reason) had to cut corners, and investing a little time and money make the speakers better for his own application.
Cars, houses, and most other consumer goods can all be tweaked in similar ways by a DIYer.
You can even take grandma's chocolate chip cookie recipe and modify it. That doesn't mean grandma didn't know what she was doing!
> > > You can even take grandma's chocolate chip cookie recipe and modify it < < <One wouldn't dare!
Good points though!
____________ // ________________
J,I dont doubt that this sounds better than magnepans current design.
But if you have already done this, you might as well try Tazs capless tweeter mod. It really ups the resolution and the tweeter is just amazing in this config.
Capless tweeter mod? What's that?
This is a mod that Taz came up with, and it is in the tweaks section. It is his 3rd mod...3rd diagram.It removes the tweeter caps and replaces them with a .375 mh inductor (in parallel). I'll let Taz describe it:
You could also try drawing #3.1 . ( created 2.12.00 while rewriting this ) You still swap capacitors but for the tweeter you add a resistor, parallel coil to form the high pass filter. For you that have never heard a capacitor less tweeter you are in for a big surprise. My motto is why up grade a capacitor when you can design it out of the circuit. As you can see the signal will only flow through a resistor then straight to the tweeter panel. Start with 4 ohms, you can get this by paralleling 2 Radio Shack 8 ohm non inductive resistors and believe it or not these are very neutral sounding resistors.
I did this, and ended up with the resistors at 4 ohms. I used to have a pict, but I have changed isps, and have yet to get a new site up and running.
All I was saying that except for the caps that he used, this is close to what the original poster, and something he should try. Taz was dead on on this, and by my math, it is one of the cheapest ones to do (caps are pricey), and it is probably the best short of active (YMMV).
It does get bashed here, but I can guarantee you that the detractors are looking at graphs and theories, and have not tried it. They wouldn't be detractors if they had.
This post is made possible by the generous support of people like you and our sponsors: