SB acoustics SB12MNRX25-4 "full range" Build thread

[5th element]

It's nice to see that you've managed to make some progress in figuring this out. Of course other parameters will shift besides fs through break in, but often some of the relationships between parameters shift together. I'd definitely break in the driver fully and then see how the measured and simmed results compare.

Reducing the port CSA, but keeping the length the same, will lower the tuning and do pretty much what you measured it doing. As to which is better depends on the loudspeaker in question and its interaction with the room, so listening would be required before any decisions can be made. You might find however that with the higher port tuning, that you can get away with less BSC than with the lower. Of course if you're using a PLLXO for BSC, there's less incentive to go this route than if the solution was entirely passive.

[wintermute]

yeah it really bugs me when things don't turn out like I expected I did about 30 minutes of 30Hz this evening and then did the nearfield measurement again (with the 70Hz tuning). It flattened out the response somewhat. Still not going as far as the sim, but definitely better I might leave it running over night with some lowish level 20Hz.

Note that the upper frequencies are more likely affected by a change in the foam, as I kinda ruined the first bit when I stuffed it in the port to try sealed this measurement is with what is left of that foam sitting in the top of the box with none on the rear. It did only affect after about 1K before so I don't think it will be changing anything down low.



Tony.

[5th element]

That foam certainly smooths something or other around 1kHz that's for sure.

[wintermute]

Yes I have to do some more testing with foam (and maybe some of my good stuff as well, but it is rather thick and may not be suitable).

After running for 12 hours there is a slight improvement in low end response compared to last night. Blue trace is this mornings measurement. Apart from below 100Hz it is matching the sim pretty well now. I might leave it running 20Hz for a bit longer.

note I also change the sweep to start at 20Hz this morning. Last nights measurement had the sweep starting at 40Hz.

Second pic shows THD of the NF woofer and port measurements. I find it interesting that the THD on the driver is very flat above the box tuning frequency, but the port shows big variations... all have 1/16 octave smoothing.

Tony.

[wintermute]

Although I haven't really progressed on this, I thought I would upload a measurement I did of the (still clamped together) box, on Easter Monday.

The measurement was taken outside off my Verandah pointing into the yard, giving quite a bit more usable distance before reflections.

Attached is a 1/8th octave smoothed response curve. All in all it looks pretty good to me. I decided on 1/8th smoothing because gated I can only get down to about 200Hz. It is interesting comparing it to the measurements above.

I think that with a broad notch (assuming I can develop one) between 5 and 10K it should smooth out quite nicely.

All I have to do now is get off my backside, glue in the braces, and glue on the back panel. Oh and make the second one It's looking like this is going to hit the one year mark soon if I don't get moving.....

One thing that looks like it has worked as planned is the baffle geometry, The baffle step sim had a lump where there was a natural dip in the drivers (factory) response between 1 and 2Khz. I was hoping that the Baffle step might fill it in, which it appears to have done nicely.

Tony.