Vented pole pieces typically have an oil can resonance from the vent that is not helpful for the sonics. I will often block them up in anything that isn't a subwoofer (or a dome tweeter, but i rarely use those)
MA voice coils are vented out the sides of the voice coil. This also has the benefit of reducing VC mass.
dave
Is there any way you can increase the resolution of the graph by reducing the db range?
And perhaps keep the smoothing to 1/24 oct?
Thanks!
-vertical scale reduced by 30db
-frequency dependent window size increased to 1/24 octave
-mic cal applied
This was just a quick measurement I made out of curiosity. I have not checked for repeatability. I plan to do a more thorough job when the speakers are completed.
edit: I originally showed the response to 30khz, but decided to remove it since my mic is not calibrated in that range.
Attachments
Last edited:
Member
Joined 2009
Paid Member
It's worth watching how this develops over time. I found that cone treatments did not help in the end for my ears. In the beginning my enthusiasm carried me on, but over the few years I've been in this hobby my ears have become less forgiving - I guess they became a bit Victor Meldrew.
Member
Joined 2009
Paid Member
good, hopefully that'll sort out the FR. But cone-break-up modes can be hard to eliminate with EQ, too much going on with the phase etc. - I tried passive EQ and it didn't work out but I never tried a digital approach - it wouldn't have been in keeping with the spirit of what I was doing at the time.
Bigun, think remember you did a A10m gen3 build for a friend that you liked some, and if we compare manufacture data on same graph think Pluvia Seven is high value driver and it actual also have a very beautiful look seen with my eyes. First visual below is usual 100dB range or more as MA normal publish where second one is a 50dB range, green is A10m gen3 and red is Pluvia Seven.
Third visual is gmad's beautiful enclosures http://www.diyaudio.com/forums/full-range/265053-hypercube-loudspeakers-52.html#post4811515 smoother performance verse manufacture's IEC baffle, and fourth one is same but intended info for gmad where FR135EX manufacture data is mixed into. Amplitude dip and impedance blib just below 2kHz is probably because of under hung surround and un-repairable, it falls at a unfortunate frq point seen with eyes what they discuss and try to improve at this thread http://www.diyaudio.com/forums/multi-way/277519-fixing-stereo-phantom-center-61.html#post4745331
Third visual is gmad's beautiful enclosures http://www.diyaudio.com/forums/full-range/265053-hypercube-loudspeakers-52.html#post4811515 smoother performance verse manufacture's IEC baffle, and fourth one is same but intended info for gmad where FR135EX manufacture data is mixed into. Amplitude dip and impedance blib just below 2kHz is probably because of under hung surround and un-repairable, it falls at a unfortunate frq point seen with eyes what they discuss and try to improve at this thread http://www.diyaudio.com/forums/multi-way/277519-fixing-stereo-phantom-center-61.html#post4745331
Attachments
Thanks, BYRTT. One thing I should have mentioned is that the second graph I showed is derived from the minimum phase extracted version of the sweep; I had lost the full measurement which I think is what I used for the first graph. They were, of course, very similar, but I remember thinking that the broad peak around 1khz may have been slightly more pronounced in the min phase version. Sorry for the sloppiness 
. I meant to do a more accurate and thorough job by now but I wanted to wait until the speakers were done. It's taking a lot longer than I anticipated since I decided to redo the finish (I'm still not sure whether I want to spray or brush...).
edit: About the dip that shows up in the impedance sweep: I had assumed this was due to the driver behaving a bit like a BMR. Could this possibly be the case?
. I meant to do a more accurate and thorough job by now but I wanted to wait until the speakers were done. It's taking a lot longer than I anticipated since I decided to redo the finish (I'm still not sure whether I want to spray or brush...).edit: About the dip that shows up in the impedance sweep: I had assumed this was due to the driver behaving a bit like a BMR. Could this possibly be the case?
Last edited:
Member
Joined 2009
Paid Member
Yes, I didn't detect anything irritating about the A10.3 drivers, they were really good.
I was using them (during the brief period I had them - they were built for a friend) toed in slightly and just 10 to 15deg off-axis is enough to level the FR nicely. Their stereo imaging was memorably impressive.
The key is likely not the exact shape of the FR but the cause of the bumps in the treble. If it's a gently rising response unrelated to cone-break-up that might produce a very different subjective experience than a cone break-up resonance.
If the Pluvia sounds as good as the 10.3 I'd be enthused to hear about that impression from anybody who has both drivers to compare. But the test sound files I listened to from XRK were not good at all to my ears. Still, that wasn't exactly first-hand experience.
Last edited:
Broadly similar, but they are not capable of reprodiucing the same level of detail (ie have less DDR) than the A10.3. The Mms is near 2x the A10.3m so they will be physically tougher (i accidentily bounced an allen key off the cone, a 10.3 would have been damaged, the P11 was not). For systems where the lowest level stuff is already lost before it gets to the speaker, that will not be noticed.
John -- who won the CGR P11 cabinets and drivers -- had me EnABL the drivers. Performance is superb. Any isses with top end ringing now seems MIA (his comment).
To my mind, so much is lost, and so much convolved into those files they are not worth much more than entertainment.
dave
Just a quick follow-up to my previous post to clear up any confusion (since I now remembered what I actually did). Both frequency response graphs I posted were derived from the minimum phase portion of the measurement. I intended it to be this way for the first graph since I was showing the response with eq. As I remember it, the second graph would have showed slightly flatter response through the midrange had I used the the full measurement (not just the min phase portion), but I had lost the original measurement. We're not talking about a big difference here, but I wanted to be as clear as possible since BYRTT took the time to make those graphs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.