Mundorf Accuton Kit - In Theory

[eriksquires]

Hi Everyone,

I went a little nuts and put together some simulations, and a virtual review of the Mundorf MA30 speaker kit here.

I hope someone finds it useful.

I'm not posting the crossover schematics because I think it's a Jerk-ish thing to do, but I did simulate it in XSim and posted the results.

Best,


Erik

 

[Lojzek]

Hi Erik!
Wouldn't you findings be better preserved for future
use on diyAudio rather than on any outside media server?

 

[Cousin Billy]

Erik,
I went to Audiyo to pick up parts. Simon Au had these little gems playing. Wow!. The smoothest clearest most non-fatiguing sound I've ever heard (outside of my own speakers of course).

It was at that point I made a decision: start to use the Mundorf AMT's.

 

[Moovnutz]

What makes you think these are third order filters?
Taken from the catalogue.
"The crossover layout is purely serial, thus there are no parallel components in the entire crossover. Also, the layout is done for bi-wiring resp. bi-amping. Crossover frequency is 3450Hz. High-pass and low-pass are 1st. order/6dB filters, designed as a Notch-Filter in both high-pass and low-pass filters."

 

[speakerdoctor]

There is another thread here that hashes out the details of the MA30 design. Still waiting for Madisound to add the kit to their line.

Erik, did you buy and build a kit that led to your OP?

 

[mbrennwa]

Hi Everyone,

I went a little nuts and put together some simulations, and a virtual review of the Mundorf MA30 speaker kit here.

Thanks for this! It would be interesting to see how it performs in reality (i.e., measurements instead of simulations).

 

[eriksquires]

What makes you think these are third order filters?
Taken from the catalogue.
"The crossover layout is purely serial, thus there are no parallel components in the entire crossover. Also, the layout is done for bi-wiring resp. bi-amping. Crossover frequency is 3450Hz. High-pass and low-pass are 1st. order/6dB filters, designed as a Notch-Filter in both high-pass and low-pass filters."

Hi Moovnutz!

I went from a diagram of the board layout which included part values and copper lands. I then reverse engineered the board using that board picture (also from Mundorf) and FRD/ZMA falues from SPLCopy into XSim. I had to guess at the ground locations but pretty much everything else was laid out for me.

If I had made that big a blunder I would have ended up terrible FR and impedance plots. Also, as others (Bill Poster I think) have pointed out, a crossover point of 3450 makes no sense at all with these drivers. I'm curious about who brags about a serial crossover? Most people brag about having a parallel crossover with no serial components, so at the very least there's some language issues in the description.

The Korean site Fullrange has an FR plot which very closely aligns with the simulation, I've updated my blog accordingly.

So, for all those reasons, I think the crossover description was a leftover from either a previous iteration, or cut and pasted from something else, but I could stand to be corrected.

Best,


Erik

 

[eriksquires]

Thanks for this! It would be interesting to see how it performs in reality (i.e., measurements instead of simulations).

I found a Korean site which posted an FRD plot. I have updated my blog.

Best,


Erik

 

[eriksquires]

There is another thread here that hashes out the details of the MA30 design. Still waiting for Madisound to add the kit to their line.

Erik, did you buy and build a kit that led to your OP?

SD,

After speaking with you re: the Mundorf AMT line I designed and built my own speakers around a 6.5" revelator and the Mundorf AMT. Cabinets courtesy of Taylor Speakers. There's a thread on DIY with Raal about the distortion and linearity measurements.

 

[speakerdoctor]

Hi Moovnutz!

I went from a diagram of the board layout which included part values and copper lands. I then reverse engineered the board using that board picture (also from Mundorf) and FRD/ZMA falues from SPLCopy into XSim. I had to guess at the ground locations but pretty much everything else was laid out for me.

If I had made that big a blunder that I would have ended up with such a perfect FR and impedance plots. Also, as others (Bill Poster I think) have pointed out, a crossover point of 3450 makes no sense at all with these drivers. Also, who brags about a serial crossover? Most people brag about having a parallel crossover with no serial components.

I've updated my blog, but the Korean site Fullrange has an FR plot which very closely aligns with the simulation.

So, for all those reasons, I think the crossover description was a leftover from either a previous iteration, or cut and pasted from something else, but I could stand to be corrected.

Best,


Erik

An XO of 3450 hz makes more sense to me because that tweeter is really intended for a 3-way design (read Mundorf's and Madisound's literature) with a high xo point. If you look at the woofer's FR's, the mfgr has posted they aren't very pretty above 2 kHz, hence all of the EQ'ing in the xo to tame those peaks down at that high xo point. My Intimates use the same tweeter with a similar xo and 3rd order xo. The SS 15W/8630K 01 has a much better FR above 2 kHz.

 

[eriksquires]

An XO of 3450 hz makes more sense to me because that tweeter is really intended for a 3-way design (read Mundorf's and Madisound's literature) with a high xo point. If you look at the woofer's FR's, the mfgr has posted they aren't very pretty above 2 kHz, hence all of the EQ'ing in the xo to tame those peaks down at that high xo point. My Intimates use the same tweeter with a similar xo and 3rd order xo. The SS 15W/8630K 01 has a much better FR above 2 kHz.

SD, I see what you mean about the tweeter, but your point about the differences between the SS and Accuton in the top of their range is the limiting factor I think. Still, according to my simulation, the lower XO point is quite a good compromise. The FR plot I posted has the tweeter output, let me post the voltage transfer for you. Mundorf recommends this tweeter for 2,800 at 12 dB/octave, so I think pushing it a little down with a third order is still OK, especially with these tweeters. Unlike ribbons, these are pretty robust.

 

[mbrennwa]

I found a Korean site which posted an FRD plot. I have updated my blog.

Best,


Erik

Do you have access to the speakers? Can you measure them? Polar response, time domain/step response, distortion, waterfall, single driver FRD, and impedance response would be most interesting to better understand things.

 

[eriksquires]

Do you have access to the speakers? Can you measure them? Polar response, time domain/step response, distortion, waterfall, single driver FRD, and impedance response would be most interesting to better understand things.

Nope.

 

[Bill poster]

Check out Vapor audios Nimbus, it's a 3 way which uses this Accuton x2 with a Raal tweeter. I think the X/O is around 2.5khz

 

[Moovnutz]

I talked to Madisound about six months ago the kit was going for $1750.00
Why they haven't put it out there is beyond me. Maybe because they redid there website.
I went for This one instead AMT 2510C, Scanspeak 18WU8741T-00 and 26W/8861T00.
Active crossover. Sounds awful

 

[eriksquires]

Adam at Madi just emailed me that they have the kit and are working on the web page. I think you may be able to contact them directly.

 

[eriksquires]

I talked to Madisound about six months ago the kit was going for $1750.00
Why they haven't put it out there is beyond me. Maybe because they redid there website.
I went for This one instead AMT 2510C, Scanspeak 18WU8741T-00 and 26W/8861T00.
Active crossover. Sounds awful

Moon,

I used practically the same, only no sub-woofer, passive. Sounds stunning.

I suspect you have too much bass. Start a new thread, lots here who would love to help.

Also, Scanspeak T/S specs are non-standard. Use the resonant F from the impedance plot, not the spec. Madisound usually does a good job of posting usable cabinets per driver.

Best,


Erik

 

[system7]

AMT tweeters seem a nice idea. We had a very nice, more affordable build recently using the cheaper Airborne RT-4001:
Airborne, RT-4001

Those Scanspeak low-inductance, high Qms 6" drivers are nice. Filters behave as advertised.

Since an AMT has a more line-array sort of dispersion, it seems to make more sense to use them in an MTM.

Not a million miles away from a Gryphon Audio design:

http://www.diyaudio.com/forums/multi-way/289344-seas-ca15rly-x2-cabinet-volume-suggestions.html

TBH, I thought the Mundorf 2 way crossover was a fairly straightforward low bafflestep second order parallel design. It don't mean a thing without a schematic IMO. Below.

 

[eriksquires]

Steve,

The board design you posted is missing half the parts. i went from one contained inthe set up pdf.

Best,

Erik

 

[eriksquires]

Gadzooks, I have had parallel and serial reversed. I got thrown off by Mundorf bragging about having no parallel parts.

Again, I think it's a jerkish thing to post the schematics in full so the values used are the XSim defaults, except 1 cap, just to illustrate it's a small one used to add up capacitance. Like Steve said though, pretty standard stuff. C3/C4/R2 seem to just do a little phase compensation.