SmallSyns Sealed Linear Phase/ Linkwitz Transform EQ
Here is the xml file to load into MiniDSP 2x4 HD. There is also a short text file included about how to remove the Linkwitz Transform and/or some room EQ I have applied in the settings.
Here is the xml file to load into MiniDSP 2x4 HD. There is also a short text file included about how to remove the Linkwitz Transform and/or some room EQ I have applied in the settings.
Attachments
Long live the FIR filter and the skilled hands at the wheel! Awesome response Bill.
@ sphykik
Nice work. I think your absorbers look very professional. Did you receive your replacement driver yet?
Yes, I've had the driver replaced for about a week now. Still no reply from FaitalPro regarding warranty replacement, despite several attempts (Sad!). I had to purchase a new one from PE.
I'm exceedingly happy with the Small Syns - never heard a speaker that sounds quite like them, to be honest!
Parts Express - initially I bought 4 FaitalPro woofers but I didn't use or test them until after the 90 day return period. PE doesn't handle Faital warranty returns in-house, they refer you directly to the manufacturer.
I might call and complain to PE and see if they will make an exception.
I might call and complain to PE and see if they will make an exception.
Got the MiniDSP and loaded your config, Bill! Here are my impressions:
1) The Linkwitz Transform is a definite improvement, even with a sub. Just rounds out the low-mids/bass and adds a bit more "umph" to distorted and bass guitars.
2) The linear phase adjustment is tricky. I've A/B'd a number of times (though it is frustrating that the on/off toggle can't be linked to both channels), and ultimately I think I can hear the difference, but you're right, perfect linear phase doesn't necessarily sound better.
I've found that on certain tracks the effect can almost be "too sterile," but this does seem dependent on the mix/master of the song. The Small Syns immediately expose a bad recording.
3) The sub sounds way better now that I'm no longer using the speaker-level inputs. Next up is playing around with crossover slops and taking some measurements. I'm hoping to get one outdoors on Sunday!
1) The Linkwitz Transform is a definite improvement, even with a sub. Just rounds out the low-mids/bass and adds a bit more "umph" to distorted and bass guitars.
2) The linear phase adjustment is tricky. I've A/B'd a number of times (though it is frustrating that the on/off toggle can't be linked to both channels), and ultimately I think I can hear the difference, but you're right, perfect linear phase doesn't necessarily sound better.
I've found that on certain tracks the effect can almost be "too sterile," but this does seem dependent on the mix/master of the song. The Small Syns immediately expose a bad recording.
3) The sub sounds way better now that I'm no longer using the speaker-level inputs. Next up is playing around with crossover slops and taking some measurements. I'm hoping to get one outdoors on Sunday!
If you want to do a quicker with/without comparison, set up two configurations in miniDSP, identical except one has the FIR EQ bypassed. That way a click of the remote can change it (though it takes a few seconds each time).
If you wanted to get more 'blind' about testing it, you could make two without FIR and one with FIR, get someone else to select through them in a predetermined random order and see if you can tell which is the phase EQ'd one. Three instead of two so that the miniDSP goes through a config change each time. (I haven't done this process myself, BTW! It doesn't sound like much fun for the effort. But I like making work for someone else!)
Sunday we had some nice weather with minimal wind, so I was able to get out to my storage unit to do some outdoor measurements. There is a ~50'x50' grassy area, and I set up the speaker to fire away from the metal units into a wooded area.
I wanted to get the speaker up higher (I've read 6-8'), but didn't have anything around to raise it except for some old coolers... not ideal as I figured they would resonate during the sine sweeps. I even tried filling them with acoustic insulation, but I didn't work in the lower frequencies and the measurements went to hell.
Failed cooler setup - amazing how badly this screwed up the measurements.
I settled on having the Small Syn ~4' above the ground - turned the setup here to fire directly into the woods.
I experimented with microphone distances of 1-3m, but found that for the most part there wasn't much difference in results except for a shift in SPL. I opted to point the measurement mic directly towards the center of the speaker, about 2m back.
Alright, here we go:
And for demonstration's sake, here's the measurement when the speaker was on top of the cooler - green line is with the cooler, blue is without:
I'm not 100% sure how to interpret these results. When "zoomed out," the results don't look too bad, however in the 2nd photo with the y-axis at 10db limits there seem to be some obvious problems. Mainly the giant peak at ~720hz. Bill, I know you have mentioned the peak at 100hz, so I expected that one.
I have a couple theories about what could be causing the 720hz - perhaps it's the amount of stuffing I have in the box (may not be enough), or maybe the putty I'm using for the midrange driver isn't working properly and is causing problems.
I welcome any guidance - these were my first outdoor measurements. Thanks!
I wanted to get the speaker up higher (I've read 6-8'), but didn't have anything around to raise it except for some old coolers... not ideal as I figured they would resonate during the sine sweeps. I even tried filling them with acoustic insulation, but I didn't work in the lower frequencies and the measurements went to hell.
Failed cooler setup - amazing how badly this screwed up the measurements.
I settled on having the Small Syn ~4' above the ground - turned the setup here to fire directly into the woods.
I experimented with microphone distances of 1-3m, but found that for the most part there wasn't much difference in results except for a shift in SPL. I opted to point the measurement mic directly towards the center of the speaker, about 2m back.
Alright, here we go:
And for demonstration's sake, here's the measurement when the speaker was on top of the cooler - green line is with the cooler, blue is without:
I'm not 100% sure how to interpret these results. When "zoomed out," the results don't look too bad, however in the 2nd photo with the y-axis at 10db limits there seem to be some obvious problems. Mainly the giant peak at ~720hz. Bill, I know you have mentioned the peak at 100hz, so I expected that one.
I have a couple theories about what could be causing the 720hz - perhaps it's the amount of stuffing I have in the box (may not be enough), or maybe the putty I'm using for the midrange driver isn't working properly and is causing problems.
I welcome any guidance - these were my first outdoor measurements. Thanks!
You could do a ground plane measurement for the lower end if you want. It's done with the speaker on the ground tilted towards the mic at about 5 to 10 degrees roughly and put the mic on the ground at 2 meters. 78 3/4 inches.
That will get you as good as anechoic response.
The other way I do good measurements is up on a really tall ladder. I have a step ladder that is over 14 feet. I clap a platform on the top and use a webclamp with a ratchet to hold the speaker in place. The mic setup has the same level of creativity. 14 feet will get you an anechoic measurement down to the mid 50's.
That will get you as good as anechoic response.
The other way I do good measurements is up on a really tall ladder. I have a step ladder that is over 14 feet. I clap a platform on the top and use a webclamp with a ratchet to hold the speaker in place. The mic setup has the same level of creativity. 14 feet will get you an anechoic measurement down to the mid 50's.
Agreed, that 700Hz-ish peak looks kind of ugly. Is it in both speakers? (you should be able to find the peak with indoor measurements).
I'd try looking at drivers alone one at a time if you can. See which one is zinging like that. The tweeter shouldn't be active there, but worth a check.
Might possibly be not enough absorber behind woofers?
Is that with or without EQ?
I'd try looking at drivers alone one at a time if you can. See which one is zinging like that. The tweeter shouldn't be active there, but worth a check.
Might possibly be not enough absorber behind woofers?
Is that with or without EQ?
I did only take one speaker with me, although in retrospect it would have been wise to bring both. The measurements were done without EQ - just audio directly out of REW through a USB interface into the amp. I am at a conference through Friday, but will do some more testing this weekend. I'll try the other speaker first, and then go from there.
I'm using a decent "chunk" of rockwool material (maybe 8"x8"x3") behind each woofer, but the box is by no means full. I was waiting until final finishing to add more.
Mark - my brother has a 12ft ladder, so I might have to go over to his place and get creative. Thanks for the tips!
Will return with more measurements!
I'm using a decent "chunk" of rockwool material (maybe 8"x8"x3") behind each woofer, but the box is by no means full. I was waiting until final finishing to add more.
Mark - my brother has a 12ft ladder, so I might have to go over to his place and get creative. Thanks for the tips!
Will return with more measurements!
That amount of stuffing should be fine. I have a little 4x4 strip of 1" bluejean insulation behind each woofer in the ported box but don't get a peak like that.
Really, where that peak is you should be able to see it in your living room, not really a need for a ladder. For that matter, you should be able to close-mic it and see much lower even.
Really, where that peak is you should be able to see it in your living room, not really a need for a ladder. For that matter, you should be able to close-mic it and see much lower even.
I've been considering building these again however they'd be for HT use and my concern is whether they'd have sufficient output at the low end to keep up. I'm not too concerned about the height of the speaker or the cost involved (in more drivers or xo changes), the depth limit is the major factor for me.
Options I've thought of seem to be;
- add 2 more woofers above the horn (as per http://www.diyaudio.com/forums/multi-way/292379-small-syns-5.html#post4940776)
- go with the shallow shelf ported version
- go with the existed sealed version but sit it on top of another woofer (like an MBM type setup?)
- simply go with the existed sealed version because crossed to subs, it will do just fine to HT reference levels
the last point might be wishful thinking but I thought I'd throw it in there anyway
Any thoughts on which way to go?
Options I've thought of seem to be;
- add 2 more woofers above the horn (as per http://www.diyaudio.com/forums/multi-way/292379-small-syns-5.html#post4940776)
- go with the shallow shelf ported version
- go with the existed sealed version but sit it on top of another woofer (like an MBM type setup?)
- simply go with the existed sealed version because crossed to subs, it will do just fine to HT reference levels
the last point might be wishful thinking but I thought I'd throw it in there anyway
Any thoughts on which way to go?
A question.
On a direct radiator system your advice is spot on.
Does the 1 cm distance work as well when trying to measure a horn loaded system?
I had to reread my note in the context of the previous postings so I could remember what I was thinking.
So in looking for a 700 hz peak, close mic'ed would mean measuring at the opening of horn mouth (big end). You can see the woofer and tweeter interaction without the woofer. There would still be horn loading at the mouth opening.
Of course the direct radiator woofer close mic'ed (1 cm) would get the woofer and internal box interaction.
At a 80 cm mic distance indoors, you get all three drivers, baffle, and horn. Even with the room gain you could see the differences between the measurements above. For clarity, gating down to 400 hz would clean up most of the room contributions.
Hopefully this clears up my previous statement.
So in looking for a 700 hz peak, close mic'ed would mean measuring at the opening of horn mouth (big end). You can see the woofer and tweeter interaction without the woofer. There would still be horn loading at the mouth opening.
Of course the direct radiator woofer close mic'ed (1 cm) would get the woofer and internal box interaction.
At a 80 cm mic distance indoors, you get all three drivers, baffle, and horn. Even with the room gain you could see the differences between the measurements above. For clarity, gating down to 400 hz would clean up most of the room contributions.
Hopefully this clears up my previous statement.
I made a start on sketching out a 4 woofer version, a few pics and a few Qs to go with them.
To start with I simply mirrored the slots above the wg so that the upper slots are the same distance from the CD as the lower ones like so
This gives a centre to centre spacing of the top/bottom woofers as about 10" (NB: I am working in mm here), xdir suggests this will produce the following lobe which seems a bit narrower than the mid/cd but not terribly so.
Is this a reasonable approach to placing the upper woofers?
Are there any concerns with space for the mid here? I guess not (because the mid is listed as 120x108.5mm and there is ~150mm between the woofer mounting points).
I also worked out the internal volume of the sealed smallsyns to be somewhere around 30L. Is this about right? It seems like an appropriate size (when modelled) so I was planning on simply extending the baffle down til I hit around 60-65L internal volume.
To start with I simply mirrored the slots above the wg so that the upper slots are the same distance from the CD as the lower ones like so
This gives a centre to centre spacing of the top/bottom woofers as about 10" (NB: I am working in mm here), xdir suggests this will produce the following lobe which seems a bit narrower than the mid/cd but not terribly so.
Is this a reasonable approach to placing the upper woofers?
Are there any concerns with space for the mid here? I guess not (because the mid is listed as 120x108.5mm and there is ~150mm between the woofer mounting points).
I also worked out the internal volume of the sealed smallsyns to be somewhere around 30L. Is this about right? It seems like an appropriate size (when modelled) so I was planning on simply extending the baffle down til I hit around 60-65L internal volume.