Manger

[phase_accurate]

Hi Eric

The Manger project is looking as follows.

The MSW and a Dynaudio 20W75 are currently mounted in an ugly particle-board (I think this is the correct expression - in German it is "Spanplatte") box. It is somehow similar to a Manger Zerobox 109 (not the uglyness for heaven's sake, but the construction).

It is currently in use for casual listening just to break in the drivers. In about three weeks I will make measurements, to be able to design the crossover.
There is currently only a first-order crossover in use, and I am feeling ashamed for using electrolytics for this purpose, but IMO this doesn't play any role when it comes to break in of drivers.
Though the woofer is in this case able to mess up the midrange (1st order lowpass) and the electrolytics are by no means sonically good, the box sounds quite astonishing. Imaging is already quite O.K. and the sound is more detailed as with my other box currently in use (two-way with Dynaudio D260 and 17W-LQ).

As soon as the drivers are measured I will build the prototype electronics. The crossover will be of the subtractive type (+ some equalization), and output power between 50 and 100 watts approx per driver. I will do this on Veroboard only since ist is just for trying any tweaking around.
This will be built into the prototype box and I will listen to it for some time (until the final one is finished). I will keep the electronics for further use (experiments with other crossover/driver combinations also non FR ones).

The final box will not use the Dynaudio driver but one from Audio Technology (currently in the make). This driver is a little larger than the Dynaudio one (it will have about three times it's Vd) and response of the closed box will be equalised flat to around 40 Hz. Maybe I will even make the rolloff selectable, I don't know yet.
So even with this box (that will definitely be larger than the current one) the prototype electronics will again be needed for tweaking before the final electronics is designed and built.
So the whole thing is something like an approximation process.

I don't know how long this hole thing will take, since I don't have much spare time for things like that.

Regards

Charles

 

[cocolino]

Cool

I am currently developing an active system that is using the Manger bending wave transducer (MSW) paired with an LF driver, made by Audio Technology of Denmark, in a closed box. It will use a subtractive crossover (called analog computer crossover by John Watkinson:www.celticaudio.co.uk), like the ones introduced on Nelson Pass' homepage, in order to achieve perfect transient AND amplitude response.

Hello Charles (bzw. "Servus" wie wir hier sagen):

That`s almost exactly what I`m planning and thinking about since a year or two as my next (and likely final for many years) project , unless I`ll use 4 each per channel 10" Scan-Speak drivers (25W) in sealed (of course) enclosures (and also equalized with variable cutoff) and probably a 12db/6dB substractive filter instead of 18dB/6dB. For the bass I`ll build one of Nelson Pass`s designs (thanks to Nelson Pass and the great people and their knowledge here it makes it possible for me) and and an OTL tube amp for the Manger (as the Manger has two 16 Ohms voice coils it should be possible to wire them in series to give an almost perfect load for an OTL tube amp).
Additionally I´ll use an indirect radiating tweeter with variable crossover point/slope/volume for compensating the directivity of the Manger at higher frequency (I intended to put the Manger in an enclosure as well, no open baffle) . I have experimented many years ago with the Manger and found the indirect tweeter method very helpful. Maybe You want give this a try also.
BTW: what do You think about the 3 Manger drivers configuration (per box) to simulate an infinite baffle to avoid the sound pressure drop ?
I believe it`s not necessary (besides that it were incredible expensive buying 6 Manger`s) when I cross somewhat higher than the original Zerobox designs (I thought about 250Hz-350Hz). This also would be beneficial regarding distortion of the Manger and likely makes application of the filter easier too.
I`m pretty sure that four 10" drivers in a vertical aray (WW-Manger-WW) with the higher crossover point will have a comparable effect than the three Manger configuration. And since in this case the baffle width is big enough to avoid the pressure drop at frequencies upwards approximately the crossover range I hope to be able to avoid it to a great extend.

Anyway, You see I`m highly interested in Your project and I`m looking forward hopefully to hear/see something about Your progress once in while (I`m still behind everything and so mine will take much time until I begin at all).

 

[phase_accurate]

Hallo Christoph

I always had the suspicion that you were one of these obscure "Mangerists". (guess why?)

I think your suggestion would be very cool. I wish I could still use such a "monster" at home but sometimes you have to watch out for SAF .

I have already heard a "complete" three-driver Zerobox but most of the MSW addicts I know are also using only one MSW per box (a big advantage of the tree driver arrangement is that Manger can sell more drivers )
).
The next time I will be able to hear the same box again (Manger's own demo box) at the Swiss high-end fair Iwill ask Mrs Manger if she would kindly demonstarate the difference (their demo Zerobox 103 has a removable bridge exactly for this purpose).
Another possibility would be the use of a spherical shaped "box" but this would be definitely more difficult to manufacture and have even less SAF (during the seventies however such a thingie painted in orange definitely would have been a hit !).

When I have more news I will post it in the subtractive crossover thread or maybe start a new one since it is somehow displaced here a little.

Regards

Charles

P.S. : Have you ever seen the model "Mega" made by Newtronics ?
www.newtronics.de/html/produkte/vorschau.htm#

 

[cocolino]

obscure "Mangerist"

I always had the suspicion that you were one of these obscure "Mangerists". (guess why?)

Yes Charles,
maybe I`m a "Mangerist" (nice word creation BTW) I don`t know (at least it looks good as Avatar methinks).
Since I heard it the first time many years ago (far before Manger was known in Hifi circles but only in recording studios) I felt that this might be THE thing. Short time after this experiance I had the chance to try one at home for a short period. Unfortunately not long enough to try out everything I wanted but enough to do some rough comparisons with own designs that I know well. And I have to say since that time it makes me more and more thinking that all those multiple way (3 or more and even two-way with the common crosspoint upwards 1,5kHz or so) speakers sound somehow wrong independend how "good" they are (but conventional full range drivers with all their compromises also never have me really attracted) . For me the Manger theory definitely corresponds positively with what I can hear (and this does not happen very often to me with theory and regarding Hifi - most of it is marketing hype - isn`t it?).
The only thing what was to critizise IMO was the somehow slightly underrepresented high range (overall, not particular on axis) what I lead back to the directive radiation at high frequencies (very obviously caused by the big 70mm diameter voice coil). But as I said this can be compensated to some extend IMO with an indirect radiating tweeter.
What concerns directivity: I`m sure You have heard from that curved plastic (plexiglass) part Manger mounts in front of his driver since recently. Do You or one of Your friends have any experiance with it?
If it actually helps widen the dispersion without otherwise compromising the performance and as the price for this simple piece of plastic is so incredible, I think about to DIY it (I wouldn`t pay it even if I were a millionaire - and actually I´m a couple of bucks away from it ). All one would need to do it are the dimensions - the rest cannot be such a big problem.

The next time I will be able to hear the same box again (Manger's own demo box) at the Swiss high-end fair Iwill ask Mrs Manger if she would kindly demonstarate the difference (their demo Zerobox 103 has a removable bridge exactly for this purpose).

It would be very interesting for me to hear Your opinion about the difference!!

a big advantage of the tree driver arrangement is that Manger can sell more drivers )

fully agreed (if not the biggest)


I know the Newtronics "Mega" and I guess my project more or less basically will look like it when it`s (ever) ready (though I plan 10" woofers not 12" and my amps and other electronics will be much better ) but I do not like their idea of the additional tweeter radiating from the front baffle as this very likely disturbs the coherent direct waveform and will make the Manger approximation almost useless (since this seems to be one of those small 10mm tweeters, IMO they rather should have mounted it coaxial directly in as short as possible distance in front of the Manger and to equalize the physical disalignment by an electronic time delay).

And also yes, an own thread about the Manger would be great.
(and maybe one of the Moderators can move the according posts from here to where they belong then).

Thus until soon in another thread!

Trigon:
"Welcomen" - that`s another neat word creation

 

[capslock]

Newtronics Manger distortion

I have been wondering about disto measurements for the Manger. The link Charles provided had a very interesting plot taken from the magazine HifiVision.

Unfortunately, there is no information at what absolute sound pressure level this was measured. If it was 100 dB, it is not scary. If it was 85 or 90 dB, it might well be, with 3rd harmonic at 3.2% between 400 and 800 Hz and 2nd at >5% at 1.5 kHz (conventional wisdom has it that midrange disto should be below 1% for a good speaker).

I am also not sure how much of this disto is caused by the additional tweeter. The peak at 1.5 kHz coincides pretty well with the dip in the frequency response around 1.6-1.7 kHz that can be seen in the frequency response plot of the bare Manger driver as well as plots of all finished speakers that I have seen (Zerobox 109, Medea II, Swing).
So the response dip seems to be something fundamental to the driver, and whatever the mechanism responsible for the dip, it may also be responsible for harmonics generation.

I wonder why nobody (to my knowlegde) has tried to compensate that dip electronically. The reason might be that distortion would increase to even higher values.

Eric

 

[capslock]

This is the plot. Note that for the other dips at 300 and 700 Hz, there is also a disto peak at or slightly below this frequency.

The 700 Hz dip can also be found in the bare (open infinite baffle?) Manger data sheet plot, but is less pronounced. The 300 Hz dip does not appear in the data sheet.

Eric

 

[capslock]

I wish I could still use such a "monster" at home but sometimes you have to watch out for SAF

What is SAF?

Actually, I had thought about doing something similar to the Manger sidekick even before I knew they had a product like that. It consists of two very small and flat enclosures for the Manger that can be bolted to a wall and a subwoofer. Rather than using their pretty angular shape, I would try to build something rounder and with a larger diameter. Maybe I'd use a flat wooden box and lots of gipsum to make a small mount. Would be a pretty good approximation of the infinte baffle.

I never understood why the Manger would be more susceptible to the baffle step problem than other drivers. Ideas?

Eric

 

[capslock]

The MSW's Neodymium magnet system guarantees for low IMD and noise.

I can understand that a Neodymium magnet will be less influenced by eddy current induction from the voice coil, hence less distortion (but it would get rid of the Doppler IMD). But how can it result in lower noise?

Please note that I posted something on Manger disto in your introduction. I hope Dave will move those posts into a new thread.

Eric

 

[phase_accurate]

Hi Eric

Neodymium magnet drive systems (and Alnico as well) are less prone to barkhausen noise.

John Watkinson for instance claims that the resolution of a transducer with ceramic magnet can't even achieve that of 16 bit audio !

See the two lowest paragraphs of :
http://www.celticaudio.co.uk/technical2.htm

Regards

Charles

 

[capslock]

Thanks, Dave, for moving those posts.

 

[cocolino]

Yes, thanks to Dave for opening a new thread and moving the posts!

Unfortunately, there is no information at what absolute sound pressure level this was measured. If it was 100 dB, it is not scary. If it was 85 or 90 dB, it might well be, with 3rd harmonic at 3.2% between 400 and 800 Hz and 2nd at >5% at 1.5 kHz (conventional wisdom has it that midrange disto should be below 1% for a good speaker).

Eric: The graph for the distortion You´ve posted is very interesting and me too wonder about the high values
I have the particular Hifi-Vision edition from April 1992 and the test report of the "Mega": I took a look about the distortion measurement conditions. They specified: [email protected] , test signal: sine wave

The frequency response graph of the "Mega" cannot directly compared with common measurements as Hifi-Vision used a somewhat unusual setup. They used to have a rotating mic. and a computer software to integrate and weight on-axis and off-axis data into a single graph.

Also the Manger drivers of the "Mega" from 1992 are not completly identical to the latest version drivers today.

I do not care much about the dip at about 1,6kHz as this is too small and of too narrow bandwith to play a significant (negative) role and equalizing is not really neccessary IMHO but I agree that this might have something to do with the distortion behavior.
The acoustical phase response graph at the Manger site also indicates that there must be some kind of (serious?) resonance.


The dips at 300Hz and 700Hz ("Mega" graph) cannot be judged correctly without knowing how exactly the datas were derived. Also for the lower frequency one would have to know the low-pass xover point and sloop from the woofers. And last but not least the "Mega" has 3 Mangers per box and taken into account the rotating mic. thing, I think these datas from the "Mega" do not tell us much about the Manger itself.


I can`t see a 300Hz dip at the original Manger graph. Also Manger has not specified the baffle and enclosure size for the measurements and the corresponding graphs (I`ll never understand what a graph should be of worth without specifying the measurement setup and test conditions - it`s like to say a thing has a length of 20 .... what?: meters, inches, mm, km...)
Anyway I believe all that (minor IMHO) frequency deviations can be ignored respectively controlled with the crossover but what concerns me much more is the distortion (maybe I`ll email Mr. Manger and ask about if he ever measured distortion versus frequency for his drivers).

I never understood why the Manger would be more susceptible to the baffle step problem than other drivers. Ideas?

I don`t think the Manger is more susceptible but it appears that he just cares more about it than others. And as the goal is to achieve an as perfect as possible step response, this has to be considered (though as mentioned in a previous post, I don´t think it`s neccessary therefore to use 3 Manger drivers per box).

 

[capslock]

Yes, I have write a letter to Manger anyway, and I'll ask about distortion.

The Barkhausen noise thing is interesting. Anybody know more about this?

Eric

 

[planet10]

The Barkhausen noise thing is interesting. Anybody know more about this?

Heinrich Barkhausen, a German physicist, discovered in 1919 that a slow,
smooth increase of a magnetic field applied to a piece of ferromagnetic
material, such as iron, causes it to become magnetized, not continuously but
in minute steps._The sudden, discontinuous jumps in magnetization may be
detected by a coil of wire wound on the ferromagnetic material; the sudden
transitions in the magnetic field of the material produce pulses of current in
the coil that, when amplified, produce a series of clicks in a
loudspeaker._These jumps are interpreted as discrete changes in the size or
rotation of ferromagnetic domains._Some microscopic clusters of similarly
oriented magnetic atoms aligned with the external magnetizing field increase
in size by a sudden aggregation of neighboring atomic magnets; and, especially
as the magnetizing field becomes relatively strong, other whole domains
suddenly turn into the direction of the external field." -_
ENCYCLOPEDIA BRITANNICA

Too late to go digging to-nite, but there is a JoeList thread on this...

dave

 

[capslock]

Hi Dave,

don't you ever sleep? Yes, I would be very much interested in this other thread.

I will open up a new Barkhausen noise thread.

Eric

 

[capslock]

Somehow, the plot I posted has disappeared over the weekend, hence a repost:

 

[capslock]

Two updates

1) Received an answer from Daniela Manger. It is not the drop in SBL due to the finite baffle size they are concerned about but the reflection peak that can be seen in the step response. Looking back, it is pretty obvious in the documentation but I guess I had to be pointed to it.

2) The newest model is the Swing, which consists of a single MSW and a separate, actively controlled subwoofer. Apparently, there is a mike on the dustcap that measures the pressure.
It received very favorable reviews in both Stereoplay earlier this year and Stereo in the current issue. I think Stereo gave it the highest rating of any MSW speaker ever, at 95% which is quite something. The article also reveiled that the MSW, for the first time, now uses a second order lowpass...

Eric

 

[cocolino]

Manger Swing

Hi Eric :

I haven`t read the articles You mentioned (I do not care the magazine reviews anymore - almost ever zero valuable information and meaningless pics ).
Are You sure it is a mic. what is on the dustcup and not an accelerometer?
And regarding the MSW filter: do You really mean second order lowpass (this can be for the woofer only!) or shouldn`t it be correctly, second order HIGHPASS (I mean for the MSW, not the woofer)?

1) Received an answer from Daniela Manger. It is not the drop in SBL due to the finite baffle size they are concerned about but the reflection peak that can be seen in the step response. Looking back, it is pretty obvious in the documentation but I guess I had to be pointed to it.

I`m not sure about to what refer this (the Newtronics Mega?). What has been the initial question to Daniela Manger?

 

[capslock]

Christoph,

you are right, it should read high pass. In addition to the text, there was a plot of MSW and woofer response measured seperately, and the slope was very close to 12 dB/oct.

The reply was with regard to the baffle step/1 vs. 3 MSW question that was discussed in one of these threads here. Regarding disto, she said that disto could be high below 1 kHz but did not provide any figures.

Greetings,

Eric

 

[phase_accurate]

Hi guys

I have read the Stereoplay article as well. I am not sure whether it is really true what they wrote about the 2nd order highpass. To me at least the amplitude rolloff looks more like something below 20 dB/ octave that would speak for a 6 dB highpass added to the MSW rolloff.

I too think the thingie on the subwoofer is an electret microphone.
I remember that there was once an article (including a simplified schematic diagam) in "Funkschau" about a German active Speaker using this MFB technique in the LF range. I am not sure whether it was a speaker made by B&M.
Since I kept most of the electronics magazines I bought so far this might probably still be around but it will take some time to dig it out.
The mic was also mounted at the same place (at the joint of cone and dustcap) and it was also mounted the same way so that it really reacted to the sound pressure and not to the acceleration(i.e the aperture is looking sideways).

Regards

Charles

BTW: The breadboard prototype of my subtractice crossover was finished last Saturday. I will most probably be able to try it out next Thursday.

 

[Richard Moon]

use with Raven..

Hello.

I read your notes which were not easy to understand.

Anyway, I have a plan to build 3 way system with

Raven2 : above 2.5 kHz
Manger : 150 or 200 ~ 2500
Zerobox 103 double woofers : below 150 or 200

I guess if I position the Raven close to Manger the mischief of Xover point would be cured.

And I am considering circle shaped (diameter of 60 cm) front baffle for Manger and Raven. It would help.

Am I doing right with two expensive materials?

I apprecate your advice.

Richard