• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Rebuild of Purifi SPKx speakers

OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Well, listened to frequency sweeps extensively without hearing much difference between the different vent configurations. Given this, my inclination is to do less rather than more to try to mitigate the 240 Hz resonance. Spreading it into more audible midrange seems questionable at best.

Since the port is rear mounted, unless the output is high enough to affect the front firing response, also seems best to avoid adding parts that may eventually shift or fall off and potentially result in uglier problems.
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
2,136
Likes
1,495
Location
Sweden
Great work. Did you ever run sweep and distortion test?
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Great work. Did you ever run sweep and distortion test?
I had as part of the questions over the driver resonance concern brought up elsewhere. Amir did in the SPK5 review before I modified.

Currently neither speaker is in a state to test.
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Have you seen brandons extensive research on cabinet and port reaonance?

Yes, thanks. I have seen before and still have the same reaction…

Apparently has a lot of time on his hands as is a huge data dump. Also is more a new design and am working within the constraints of the original SPK5 box.

Does he ever summarize his findings somewhere?
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
So much of what I had done to the SPK cabinet was improvements and yet was still seeing a resonance around 224 Hz. This was worse than when Amir measured it. One major change had been to use screws rather than bolts to mount the woofer. As other remediations fell short, decided I would use some nutserts and bolt the woofer to the baffle. Am using the longer port so you may see some higher resonances, but the 224 hz one has improved...

1663630293858.png


Still need to tweak the port length and likely will line it with felt but am hoping to button up the speaker before the end of the week.
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Getting closer- here is a comparison of the latest port options...

1663949499396.png


The greeen trace is a 4x2.5 inch port with flares on both ends. The blue trace is a 2x6 inch port with only cabinet-end flared. The orange trace is same port as but except with an inner lining of acoustic felt. Am leaning towards blue or orange. Need to give them a listen next!
 

alex-z

Addicted to Fun and Learning
Joined
Feb 19, 2021
Messages
633
Likes
1,144
Location
Canada
Getting closer- here is a comparison of the latest port options...



The greeen trace is a 4x2.5 inch port with flares on both ends. The blue trace is a 2x6 inch port with only cabinet-end flared. The orange trace is same port as but except with an inner lining of acoustic felt. Am leaning towards blue or orange. Need to give them a listen next!
In my opinion, the blue trace makes the most sense. The felt is giving you substantial loss from 40-200Hz, yet the leakage from 300-800Hz is no better.

In order to substantially improve the overall performance I suspect you would need something 3d printed. A continuous flare seems to be offer the best overall result. KEF has also done some interesting work with introducing a flexible inner portion, which might be possible to replicate with extruded polystyrene.

 
Last edited:
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
In my opinion, the blue trace makes the most sense. The felt is giving you substantial loss from 40-200Hz, yet the leakage from 300-800Hz is no better.

In order to substantially improve the overall performance I suspect you would need something 3d printed. A continuous flare seems to be offer the best overall result. KEF has also done some interesting work with introducing a flexible inner portion, which might be possible to replicate with extruded polystyrene.

Yes, I concur. Am keeping open to the felt as was used in Troels design and need to assess it more from a system response and distortion perspective before I make a final call.

My overall takeaway is more sympathetic to commercial offerings, We often see resonances in commercial BR speakers. Some of these have ports with unique shapes or other more extreme measures to remediate resonances but very few really quell them entirely. The main tradeoff is comparable to cabinet bracing. While better bracing can improve a design, sometimes you are just shifting the resonant frequency (sometimes to a more audible frequency). So, while applying something make you feel better, you really do not know if it is an improvement unless you measure.

A more exotic port shape would be interesting to try. Also possible relocating the port might help. May try the former in the future, the latter is really not an option for this cabinet design. As with other Purifi-based designs, a passive radiator is another approach. However the SB one used for Directiva r1 does not fit well on the back and the angled brace does not support side-mounting one or more as has been done in other designs.
 
Last edited:
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Some further progress on the resonance front today. When I noticed the improvement made by the using nutserts rather than simple screws for the woofer, decided to apply the same mounting to the tweeter. Fortunately, the tweeter holes were big enough that I could use the same nutserts, but the bolt were not going to be as nice. Early last week I ordered the new machine screws but had some delivery delays and just got the last night.

Was very pleased to see the new screws were worth the wait. Here is a comparison...
1664815224101.png


The upper cabinet damping is less than before, but the blue is before and the brown is after. The 240 Hz resonance peak is pretty much eliminated. After I add some upper cabinet damping, expect even better results! :)
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
I settled on the internal damping (at least for now). Here is what I did starting with no damping…

E6AF22B6-A38C-4ECF-93A5-625210350E20.jpeg


Appears I forgot to take the empty front, so here it is with the first application - a layer of 19mm Sonic Barrier on the bottom of the brace and inside of the bottom…
555FA5B8-41FB-4B65-A1A5-6B1EF48EB6F8.jpeg

Next was 50g of Acousta-Stuf behind the woofer…

27397445-04F5-4F97-BAF2-51A8F727FDBB.jpeg


As you can see, the nutserts are already in place so next will flip back over and finish the upper chamber first with a layer of a Sonic Barrier across the top…

AC37FAC8-9901-4F2F-B16F-E3782418EA90.jpeg


then the 130g of cotton damping around the port…

B386C3D0-A97F-4A22-A237-5E5B6BE7374F.jpeg


and put the back on…

image.jpg


and the drivers in…
image.jpg


I had a computer lose power and do not have a before and after shot handy. Will add later. For now, here is how well the pair match...

1665760098580.png


Still may be some adjustments after I listen to them, but the sweep sounds clean. The 380 Hz woofer resonance is 30 dB down from the port's max output and the vent resonance (around 880 Hz) is about 25 dB down. :)

Now will get in some listening and overall measurements!

P.S. Here is my attempt to recreate a before and after view (using some older saved measurement just after adding the nutserts) for the damping adds...

1665761088033.png


Blue race is after and brown one is before. Notably the port output is much cleaner from 300 Hz up 4 kHz.
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
2,136
Likes
1,495
Location
Sweden
Interesting with the nutserts results. Why do you think it affects? Did you apply a defined torque?
 
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
Interesting with the nutserts results. Why do you think it affects? Did you apply a defined torque?

I suspect it is a matter of how the woofer is coupled to the front baffle and how that energy can shake it, the tweeter plate and the rest of the cabinet.

Yes, the woofer screws are torqued at about 6 ft-lbs but the tweeter screws are hand-torqued. When I was playing with different damping behind the woofer, some of the 240 Hz resonance started to re-remerge. The cotton damping helped this, but later I replaced it with the Sonic Barrier.
 
Last edited:
OP
Rick Sykora

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
2,419
Likes
4,461
Location
Stow, Ohio USA
A look at on-axis FR at 1m, 5 msec gate... EDIT: updated with corrected gating
1666102441270.png


Blue trace is left speaker and green is the right one. Looks like we are off to a pretty good start! :)
 
Last edited:
Top Bottom