Maximize a Scanspeak 18M/4631 T00 Revelator midrange for a 3 way

220-165 uF on woofer, still 4,7 mH must try some smaller.
 

Attachments

  • 220 to 165 uF on woofer.png
    220 to 165 uF on woofer.png
    84.8 KB · Views: 36
I think you are trying to show us that as you let the woofer play higher, and reduce the capacitance the 2Khz resonance rises. Its 25 dB down so maybe not immediately discernible, or you can in fact hear it? What do you hear? can you see it add to the response at 2kHz?

My earlier suggestion was to see what movement in the VituixCad programme did to the response as you moved the mid woofer cabinet backward and forward, in the hope that it would produce a better fix than playing with the electrical components.

There used to be a good tutorial for the LoudSoft FineX croosver programme on youtube. Peter Larsen talks through how to use the software and in the final stages, introduces a design process where he uses the programme to adjust the positions of the drivers in the Z plane fore and aft to finalise a design.

Definitely worth a look while you sip a well deserved Beer. Cheers.
 
  • Thank You
Reactions: jawen
Do you see that because the behavior att crossing LF to MF ?
My red ring in the picture?
Before you adjusted the Z axis of the drivers correctly, yes the red circle in the picture.
The total summed output of the 2 drivers was lower in amplitude, then the individual drivers output.

For tweaking crossovers the Vituixcad Optimizer is a very helpful tool but takes a little getting used to.
https://kimmosaunisto.net/Software/VituixCAD/VituixCAD_help_20.html#Optimizer
 
  • Like
Reactions: jawen
For tweaking crossovers the Vituixcad Optimizer is a very helpful tool but takes a little getting used to.
Yes Arez it does hahaha.
And probably now with the medications I take for the nerve pain, my short-term memory and reasoning ability don't work (just as age & the deterioration of my learning ability weren't enough)

So superhard/impossible for me now to try learn new complicated things. (but im stubburn)
And notice on my way "learning speaker-diy" journey, that the more you learn the more complex it gets.

Think for most people here, maby better to only understand the basics.
With that i build my first speaker and passive xover, and i lowed the result/sound.
 
Hi @jawen,
I can recommend adding bucket magnet to the 18M. Found some magnets in the basement that had a perfect fit. The magnet stick fine secured only with painters tape.

Soundimports even have original Scanspeak magnets with big enough inner diameter.

I really like 18M and I am confident I will never part with it. My search for ultimate midrange in 3-way is over.

IMG_1768.jpeg
 
  • Like
Reactions: profiguy and jawen
@jawen: You are welcome.

I got inspired from @Kwesi to try adding bucket magnet. So far I have tried it with SB17MFC = nice result. with dayton RS225 4 ohm (Alu) - Bit more grunt in midrange.

With 18M difference was not night and day. I tried back and forth with and without many times. I cant measure TS, but I could clearly feel more force from the driver with added magnet. I could also measure a bit more dB above 200 hz.

As with subjective difference in sound with magnet:
Driver sounds a bit more grown up. More grunt. More agile. Better transients. Sound of voice is kinda moved a bit in front of the driver.


Magnet i tried was something similar to this:
We are talking 4 EUR a piece!
https://www.supermagnete.dk/ringmag...aA_H4GA2Fyphtq_avragwaxGrx2TjiPYaAmcpEALw_wcB

This one should also fit nicely:
https://www.soundimports.eu/en/scan-speak-107003.html
 
  • Like
Reactions: jawen
I just mount it on this surface:
1746559129142.png



As with direction of magnet it you mount it like this:

At some distance away from the driver, like 20 cm, the magnet and driver wants to go away from each other.
Moving the magnet closer to the driver, like 10 cm, the same thing happen.
Moving magnet very close to the driver, the magnet and driver pull to each other and basically magnet snap to the driver. Be careful at this stage with both fingers a magnet. It may chip a bit.
 
  • Like
Reactions: jawen
I always use a guide tube to mount bucking magnets onto drivers. Its basically a plastic sleeve made from a disposable kitchen cutting board wrapped around the speaker magnet. Its then secured with either 2 worm gear (jubilee style) clamps or clear packaging tape.

I then use medium viscosity cyanoacrylic glue, coating the magnet with a squiggly bead, then two thinner beads on the outside and inside edges of the same magnet face. Its best to double check your magnet polarity to the woofer so you're 100% sure its correct. The two magnets will be repelling each other initially, then attract each other when the poles are very close.

CAUTION - watch your fingers and wear safety glasses!!! Tiny magnetic shards and chips embedded n your eyes is a bad thing. You have to make 100% sure the added magnet is going to do what you want it to.

On some applications, an extra magnet mounted with same poles against each other, it can reduce magnetic flux in the VC gap, reducing sensitivity and raising Qts. It may also affect flux linearity and geometry in the gap, possibly increasing distortion.

Mounting an extra magnet with opposing N+S against each other will reduce flux in the gap. This is due to redirecting the flux away from the gap. Sensitivity will also decrease from this.

Sometimes nothing will happen and you've wasted a perfectly good magnet on an already great driver needing no help.

I'd verify any of these magnet additions are first tested via TSP analysis (ie. Dayton WT3) and measuring driver acoustical output behavior.
 
  • Thank You
Reactions: jawen
I should add that in the 5th paragraph, I meant to say opposite (N+S).poles, not opposing poles.

Most drivers will respond positively.to a bucking magnet installed with the same repelling pole.facing the driver's magnet. If the VC gap is already magnetically saturated, it won't do any good installing an extra opposing magnet. It may actually change the flux geometry in, above and below the VC gap. That can change BL symmetry with increased VC excursion.

The bucking magnet shape and diameter can also change things unpredictably. A smaller diameter magnet won't do much for overall driver performance.
 
  • Like
Reactions: jawen
@Rokytheman The SS 18M8631T/4631T is truly one of the all time best larger cone mids. They just sound great and are easy to implement. It sounds beautiful with vocals, piano, brass, woodwind, percussion, etc. Very clean and analytical sounding, yet relaxed enough to not be fatiguing. The only other one of similar size that comes close is the Peerless NE180W, but that one needs at least one LCR at 4.5k to be as flat.

That said in most cases, with paper cones, the art and magic of getting those to sound good when crossed higher is having the rolloff region be as smooth as possible. The slits in the cone are a genius idea to tame the breakup region.

The other cone material I really like is woven carbon fiber, specifically the kind used on the Audax HM170CO. The 210, 130 and 100 are also excellent. I think the combination of carbon fiber and resin coating are what make it so great. They're sort of reminiscent of the Focal Kevlar cone mids, specifically the 5K013L. That was IMO the best sounding composite cone mid of its time and beyond. I still have a matched pair from 1989 and they still sound like a modern high end cone mid should.

Most people who like paper cones usually don't like composite synthetic cones. That is a hard comparison these days with paper cones having come so far and diverging from the typical paper/pulp recipe. Now that materials like fiberglass, nylon, Kevlar, etc are being mixed in along with various binding agents. Some of these cones sound nothing like a traditional paper cone.
 
  • Like
Reactions: jawen
Most people who like paper cones usually don't like composite synthetic cones. That is a hard comparison these days with paper cones having come so far and diverging from the typical paper/pulp recipe. Now that materials like fiberglass, nylon, Kevlar, etc are being mixed in.
Have you looked at wavecor's new woven kevlar/carbon fiber cones?
They have a few models.
Smooth and easy response, breakup and time domain behavior looks excellent etc.