Was the position exactly the same before and after? The treble roll-off in the first (raw) measurement could be from a slight angle change, and not caused by the stuffing.
Is that including the sub? I don't think you should use any positive equalisation to boost the bass of the 8" driver.
That looks well-behaved for a whizzer. The jump up in level at 3kHz is the only bit that looks bad / typical. The raggedness is mostly because it is an in-room measurement. You need to do a close mic or anaechoic test to see if the whizzer itself is ragged.
As GM was saying, this dip could be from reflections. Try (turning off the eq and) lying one speaker on its side for a Ground Plane measurement, as described here:
a mh-audio.nl - Groundplane Measurement
...but (if you cannot do the test outdoors), put the mic much closer - maybe 25cm. This will give a measurement that reduces room reflections; GP test 1.
Then do something to increase baffle size (e.g. stack bricks on top of the speaker) and measure again; GP test 2.
If either of these tests show a flatter midrange response (particularly the 200-300Hz dip), you will know that positioning and / or baffle shape and size are important.
Last edited:
Hi all,
2 picoDumbs: I can't take the back off unfortunately, they are sealed at the back.
Lightbit + hollowboy: All the measurements were done in the same position, but I think at the beginning one of the speakers wasn't aiming at the mic properly (was aiming a few degrees off). I pointed it back at the mic for the last measurement or two and so that's probably why the treble is better.... Good thought and thanks for mentioning it. Feel a bit silly for not thinking of that!
GM: There is a rug in between the speakers and the couch where the listening position is, and reflective floors about a foot in length directly below the speakers to the rug and again another foot length of reflective floors between the rug and the couch.
But measuring the distance is definitely more than 23" from the driver to the floor, as the driver height from the floor is 33" from the floor as it is.
Using the floor reflection calculator, it gives me 1.31ms for reflections, at 382Hz.
Thanks Hollowboy for the ideas for measuring them better. I might get a chance to do more on the weekend - however I live in a very noisy area in the suburbs so I will have to measure inside (thankfully I have double glazed windows which helps a lot with reducing noise that is outside getting in!
Also, the only bass region EQ I am shifting up is some at the 60Hz region, but it's by 7db.
I think I will remove it as it seems it's a room node and is probably bringing the drivers to their max excursion too early.
Otherwise in the bass region I am turning the sub louder than it needs to be dropping the bass frequencies on the EQ down to give the Nirvanas some breathing room.
Seeing as the treble loss was likely due to the speakers facing the wrong way, I think I will remove the stuffing, as it reduced the bass output a bit, and re-eq it to not be pushing up any bass frequencies.
2 picoDumbs: I can't take the back off unfortunately, they are sealed at the back.
Lightbit + hollowboy: All the measurements were done in the same position, but I think at the beginning one of the speakers wasn't aiming at the mic properly (was aiming a few degrees off). I pointed it back at the mic for the last measurement or two and so that's probably why the treble is better.... Good thought and thanks for mentioning it. Feel a bit silly for not thinking of that!
GM: There is a rug in between the speakers and the couch where the listening position is, and reflective floors about a foot in length directly below the speakers to the rug and again another foot length of reflective floors between the rug and the couch.
But measuring the distance is definitely more than 23" from the driver to the floor, as the driver height from the floor is 33" from the floor as it is.
Using the floor reflection calculator, it gives me 1.31ms for reflections, at 382Hz.
Thanks Hollowboy for the ideas for measuring them better. I might get a chance to do more on the weekend - however I live in a very noisy area in the suburbs so I will have to measure inside (thankfully I have double glazed windows which helps a lot with reducing noise that is outside getting in!
Also, the only bass region EQ I am shifting up is some at the 60Hz region, but it's by 7db.
I think I will remove it as it seems it's a room node and is probably bringing the drivers to their max excursion too early.
Otherwise in the bass region I am turning the sub louder than it needs to be dropping the bass frequencies on the EQ down to give the Nirvanas some breathing room.
Seeing as the treble loss was likely due to the speakers facing the wrong way, I think I will remove the stuffing, as it reduced the bass output a bit, and re-eq it to not be pushing up any bass frequencies.
So the thought I have is, since the 8 Alnicos have such a large Qts there aren't many box styles they are suited towards, except very large sealed, and even then they won't make much below 70Hz. And for that those boxers will need to be at least 160L.
At that size I may as well get the 15"s out. They need an even bigger box than that (from WinISD vented will need 230L or so, and sealed 350L or so), but if I'm going big I'll go all the way. And I really want the 15"s in my life.
The 8s might be useful in a open baffles with bass support later on, but there are likely better options than the 8s for that as well!
At that size I may as well get the 15"s out. They need an even bigger box than that (from WinISD vented will need 230L or so, and sealed 350L or so), but if I'm going big I'll go all the way. And I really want the 15"s in my life.
The 8s might be useful in a open baffles with bass support later on, but there are likely better options than the 8s for that as well!
This kind of imp curve is a evolution over the first gen CF models, even not being exemplar. Also this driver has a small VAS, it will work in a small enclosure, for those like pocket boxes.
Mount them on a flat panel 700mm wide and 1.5 tall and remeasure.
I also got a weird suckout in the bass region when mounted in a box with sound absorbing material.
If that is a pain in the ****, consider installing a reflector type of panel internally with holes (eg pegboard) at 30 degrees to normal to reflect most of the energy downward and not straight back out of the cone.
This was something I was going to try but open baffle sounds better so didn't bother.
I also got a weird suckout in the bass region when mounted in a box with sound absorbing material.
If that is a pain in the ****, consider installing a reflector type of panel internally with holes (eg pegboard) at 30 degrees to normal to reflect most of the energy downward and not straight back out of the cone.
This was something I was going to try but open baffle sounds better so didn't bother.
Last edited:
Start drilling holes in it. Start with a few, add as needed.
A typical rug is thin enuff that it only affects higher frequencies. TO deal with a floor bounce that is typically in the lower midrange it needs to be inches thick.
dave
You should have shown us the measurement before.
I thought your speakers are booming. Actually your problem seems to be room, not speakers.
Last edited:
What will he achieve with that? Reduce bass?
I don't think this is what he actually wants.
Last edited:
From my sims there is no point in doing anything but sealed, aperiodic or a boffle. A maximally flat sealed response requires a 500+ litre box. IIRC correctly Bigun did an ~200 litre boffle for his. This should get him into the 30s for bass extension.
dave
Attachments
Perhaps, but 'very nice' it is not.
Are you sure? Without major help from either the amplifier or series resistance, the driver you refer to a/ wants a box so small it won't actually fit into it, and b/ would have the square root of jack output in that or any other modest sized box with a response that even vaguely approaches acceptable. It's got a -3dB mass-corner frequency of 574.11Hz. Don't get me wrong, it's probably a decent driver of its type, but based on the published spec., that is certainly not what it was meant for, and if it was, the people who developed it didn't know what they were doing as the electromechnical spec. is fundamentally unsuited to such use.
Last edited:
From my sims there is no point in doing anything but sealed, aperiodic or a boffle. A maximally flat sealed response requires a 500+ litre box. IIRC correctly Bigun did an ~200 litre boffle for his. This should get him into the 30s for bass extension.
dave
Hi Dave,
Will the boffle/aperiodic enable having smaller cabinets than the sealed? Do you know of a good box calculator for boffle/aperiodic style cabinets? Which would you choose if you were doing a cabinet for the 15s?
An aperiodic enclosure tends to flatten the peak you see in the Q=1 sim, a smaller box will have a bigger peak, and up some point the aperiodic box can’t deal with it. A Boffle takes the aperiodic idea further, having no back and layers of good damping (fiberglass or denim wall insulation) with a small passage to the outside world.
http://p10hifi.net/tlinespeakers/forum/boffle-RadioElectronics.pdf
http://p10hifi.net/tlinespeakers/forum/Boffle-patent.pdf
dave
http://p10hifi.net/tlinespeakers/forum/boffle-RadioElectronics.pdf
http://p10hifi.net/tlinespeakers/forum/Boffle-patent.pdf
dave
Yes, Dave has simmed (and quite rightly) that going from a small to an enormous box will reduce the bass peak by about 3dB. Post 1 gave the impression this is exactly what you wanted:
In post 33 you showed that you'd found an easy fix - you flattened the bass peak with your stuffing tweaks (pictured). The result looked like a smooth rolloff from 150-50Hz, which looks good to blend with a sub somewhere around 70Hz.
However, instead of saying "hooray, I fixed the massive peak I was complaining about in post 1", you decided (post 44):
...so, if you don't actually want to fix the bass peak, why are you speculating about a huge box? What are you actually trying to achieve?
Attachments
Last edited:
Hi Hollowboy,
When I simmed the boxes in WinISD, that's where it told me I should expect a huge bass peak at ~90Hz, however when I measured it with the mic that bass peak wasn't there. The bass was ok to around 70Hz and then dropped like a stone, but it didn't have a huge peak like the sim said it would, it was rather "balanced".
After adding stuffing the bass output reduced from 100Hz or so down, that is why I am considering removing the stuffing.
I guess the problem was in between my simulations of what the cabinet would do, and the measurements in my room.
In terms of what I am hoping to achieve, I am wanting ideas on what cabinets would maximize the potential for these drivers. There are lots of designs for medium and low Q drivers but I was having trouble finding builds/advice for high Q drivers like mine. I'd also read reports that the base spec cabinets from CommonSense Audio were sub-optimal - which makes sense as they have a "one size fits all" approach to their line of drivers and they are all quite different.
But it sounds like the ideal cabinets these drivers need are huge...
They are. I have rejiggered a few for better performance.
The specs of their drivers are all over the map.
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.