Crackyflipside's DA Khanspire build

[crackyflipside]

Seems nice and flat is all. Especially for no roundover and whatnot.

Oh ok.... and CJD, just based on the FR I get and that the room has no treatments (just four huge sofas) and has tile floors. I am amazed by your design. :T ;x(

I got two of them running now, third is being wired up, pictures soon.

It was 1/3 octave smoothing just so ya know!

Image not available

[cjd]

Yeah, but that room shouldn't be that flat. No nulls for the sub to deal with? I mean, what gives!

How's it sound in stereo?

[---k---]

Those look big. They are sealed right? Did you increase the volume? Did you end up taking some of the woofer stuffing out and seeing how they sound?

[---k---]

Chris,

While ugly, this pic is good enough to steal. BUT, you're not showing the woofers wired with reverse polarity. If you fix this, I'll include it in the design thread.

Image not available

You should also think about fixing this thread up with some good photos and stuff. I would prefer to see everything here, and HTguide get the traffic they deserve.

[crackyflipside]

At first I was like, wow I totally forgot about making this thread. :lol:

Chris,

While ugly, this pic is good enough to steal. BUT, you're not showing the woofers wired with reverse polarity. If you fix this, I'll include it in the design thread.

You should also think about fixing this thread up with some good photos and stuff. I would prefer to see everything here, and HTguide get the traffic they deserve.

[crackyflipside]

Yeah, but that room shouldn't be that flat. No nulls for the sub to deal with? I mean, what gives!

How's it sound in stereo?

Before the room was built I had modeled it to death for room nodes and I also have a pretty deep double stepped soffit so that also must help break up room modes.

Centered the drivers because the speaker will be flush with the wall and the top of the speaker will be flush with the ceiling so the tweeters will be at about 5-12" above ear level depending where you sit. Here was the first baffle we began to cut as well as the reason you should not follow K's CAD picture :lol:

Image not available

Then we got the other panels and slopped on some wood glue with some paintbrushes and held it together with some clamps while we gos some drywall screws to secure it shut. Of course glue would have been enough but why not put in some screws as well. Remember to use screws shorter than the total baffle thickness!

Image not available

Cutting out driver holes....

Image not available

Putting in the braces as well as wiring the crossovers with the drivers and through the speaker. Pray that the crossovers work because there is zero accessibility after.


Images not available

[crackyflipside]

- Listened to it for the first time and thought the bass was very lacking so I deviated from the original design and wired the woofers in polarity (wrong way) and while the bass seemed to improve, the numbers told a different story once integrated with the sub, there was a pretty even null across the whole woofer's frequency range:

Image not available

Reversed the woofer polarity to how it should be and look how the 80-400hz area cleaned up:

Image not available

[crackyflipside]

Image not available

So then sanded them and filled gaps with 20 minute compound, sanded again to get compound off, then painted flat black Behr paint. I'll finish the last two speakers tomorrow.

Image not available

Image not available

[Jed]

Seems nice and flat is all. Especially for no roundover and whatnot.

Nice job Chris. :T

[cjd]

Geeze Jed. I meant relative to the normal room problems people have! But I suppose that plays a role too.

[Exocer]

Very nice build, Kudos to you!. I Still haven't painted my Rythmik 12" sub

[crackyflipside]

Image not available




And L/R/sub sweep EQ'd with the Audyssey XT:

Image not available

[---k---]

I think we need a full review now that you've had them up and running for a bit. Are you satisfied with the bass now, with and without the sub? I'm wondering how they do with the smaller inductors.

[crackyflipside]

My cousin and dad putting on the brackets for mounting:

Image not available

Image not available

It is an angle iron screwed into the studs on the ceiling and wall. We also ran angle iron across the entire length on the bottom (not pictured) and also screwed into the studs and speaker.

K, these are definitely speakers to run with a sub <60hz or so. But having it mounted on the wall where it is supposed to be makes it sound many times better.

[---k---]

Those look very cool.

Yeah, I go back and forth on the bass of these. Proper position makes a lot of difference.

[Johnloudb]

I'm just curious why you guys mounted them so high on the wall. Wouldn't they sound better when listening on axis?

[crackyflipside]

I'm just curious why you guys mounted them so high on the wall. Wouldn't they sound better when listening on axis?

They actually are pretty much ear level for the tweeter, first row is about 12ft back and I would estimate tweeter height is ~4" above ear level.

And --k--, no full review until I get the room treated!

[chasw98]

And --k--, no full review until I get the room treated!

And maybe we can get together and I will bring my calibrated mic and other gear for a full analysis? This is looking very, very good Chris! I can't wait to see it.

Chuck

[---k---]

Chuck,
I've been hoping you would make your way down there and give me your opinion.

[crackyflipside]

OK, I am now positive the REW graphs are completely offcause before it would say clipping was detected and I would just click OK thinking nothing of it. I forgot to save the graphs pictures!

-Mark's speakers (the surrounds) measure very good.

-The Khanspires bass is kinda hot, mid is kind of low, but the tweeter is really boosted. But it doesn't look like room modes since it is relatively flat over the x-over ranges for the individual drivers. It's just like the tweeter and woofer are getting too much power.

-The absolute worst one is the IB... I'm pretty sure the triangular shape of the baffles is causing an insane peak around 50-60hz. Like +25dB insane 8O




I do have a BFD so what I am going to try to do sometime this week is see if it's possible to make filters over a very wide frequency range.

[Johnloudb]

I think the boost of +25dB at 50-60Hz is caused by a room mode at that frequency. In our room if we place the subs along the short wall it caused -20dB dip at 50-60Hz. We fixed it by putting the speakers and subs along the longer wall. You don't want to boost a -20dB mode, but you can cut a +25dB room mode with a subwoofer equalizer.

[crackyflipside]

I think the boost of +25dB at 50-60Hz is caused by a room mode at that frequency. In our room if we place the subs along the short wall it caused -20dB dip at 50-60Hz. We fixed it by putting the speakers and subs along the longer wall. You don't want to boost a -20dB mode, but you can cut a +25dB room mode with a subwoofer equalizer.

Yeah, I wasn't to worried with the sub since the peak is like perfectly triangular and symmetrical.

[---k---]

-The Khanspires bass is kinda hot, mid is kind of low, but the tweeter is really boosted. But it doesn't look like room modes since it is relatively flat over the x-over ranges for the individual drivers. It's just like the tweeter and woofer are getting too much power.

I don't know what to make of that. I was still thinking you thought the bass was light. Could be artifacts of putting them on the wall? I think we probably need to see gated measurement to accurately judge.

[crackyflipside]

I don't know what to make of that. I was still thinking you thought the bass was light. Could be artifacts of putting them on the wall? I think we probably need to see gated measurement to accurately judge.

Gated?

Well the bass was pretty flat compared to the mids. It's the tweeter that was very high in level which probably would make the bass and mid bass sound weaker. Sorry about not having graphs, when I get some free time I'll put one up!

[cjd]

Bass being elevated with these mounted right to the wall isn't entirely surprising. Couple dB only?

How were your tweeters rated from PE? They've had wandering around on sensitivity with the ferro issues and not being sure which version of the tweeter was used where, so some additional series impedance on the tweeters may be called for.

C

[---k---]

Somewhere in one of these threads, he says that his tweeters are rated at 90.0db, or exactly the same as what mine are. None of my tweeters have ferro-fluid. I read somewhere, I believe it was Curt, who said that change was made after the first batch and few people have the ff version. No one can tell me why the sensitivity varies as much as it seems to. I just keep getting told that it is that way with all tweeters. :shrug:

I'm still curious what the volume of those boxes are - they look bigger than mine. And have you experimented with the stuffing?

[crackyflipside]

Somewhere in one of these threads, he says that his tweeters are rated at 90.0db, or exactly the same as what mine are. None of my tweeters have ferro-fluid. I read somewhere, I believe it was Curt, who said that change was made after the first batch and few people have the ff version. No one can tell me why the sensitivity varies as much as it seems to. I just keep getting told that it is that way with all tweeters. :shrug:

I'm still curious what the volume of those boxes are - they look bigger than mine. And have you experimented with the stuffing?

The volume of the box is the skinny version you got listed. I haven't really gotten the hang of BFD and it seems the damn thing won't let me adjust more than 3 filters/channel, and this is just doing cuts. The bass issues were just when I was playing the speakers far away from wall.

Image not available

Playing with BFD:

Image not available

[cjd]

There's something very odd going on here. No way I can imagine to achieve that result, though it does sort-of look like the woofers may be wired out of phase. Any chance you have the mids wired out of phase from each-other? This is the most direct road to this result that comes to mind off the top of my head.

Are you able to do an impedance sweep by any chance?

[crackyflipside]

There's something very odd going on here. No way I can imagine to achieve that result, though it does sort-of look like the woofers may be wired out of phase. Any chance you have the mids wired out of phase from each-other? This is the most direct road to this result that comes to mind off the top of my head.

Are you able to do an impedance sweep by any chance?

I wish I knew what that meant!

I have measured the L/C/R and they all basically look the same with a dip in the mid-bass / midrange frequencies but didn't save the graph pictures. All of this is with REW and a radioshack SPL meter as mic. I've been meaning to get chuck (chasw98 ) over here with some proper equipment and experience but have just been swamped with construction jobs and school.

[cjd]

I'm wondering if the mids are wired like this:

+ ----- +
        Driver
        -
        |
        -
        Driver
- ----- +

When it should be:

+ ----- +
        Driver
        -
        |
        +
        Driver
- ----- -

[crackyflipside]

I'm wondering if the mids are wired like this:

+ ----- +
        Driver
        -
        |
        -
        Driver
- ----- +

When it should be:

+ ----- +
        Driver
        -
        |
        +
        Driver
- ----- -

I'm pretty sure it is but I'll double check it. This Sunday I'll start flipping polarities on drivers and measuring responses for all possible variations.

[---k---]

An impedance sweep would be generating the impedance graph with something like the Woofer Tester. It should look like this:



I'm confused, I thought they were measuring and sounding good before you hung them on the wall. Did they just go to crap when you hung them? Do they sound like crap?

The measurements are bad. I'm looking at the tweeter. The tweeter is crossed at ~1800hz. So above 2000 hz it should basiclly be just the tweeter output. But, the measurement shows a 10db rise from about 2000hz to 10k. The RS28 is flatter than this. There is nothing in the design that could cause the tweeter response to peak 10db in that range. I think your RS meter is that far off above 2000hz.

I'm also not sure that a RS Meter (that is what you are using, right?) and Room EQ is the right gear to be taking measurements for this. And, in room graphs are great, but to many things can be screwing them up. To isolate the speaker from the room, I would think that measurements need to be take at 1 meter, with the measurement gate set so that it eliminates room reflections. It would be great if Chuck could come over and do real measurements, but that is a real pain in the butt.

[crackyflipside]

An impedance sweep would be generating the impedance graph with something like the Woofer Tester. It should look like this:

I'm confused, I thought they were measuring and sounding good before you hung them on the wall. Did they just go to crap when you hung them? Do they sound like crap?

The measurements are bad. I'm looking at the tweeter. The tweeter is crossed at ~1800hz. So above 2000 hz it should basiclly be just the tweeter output. But, the measurement shows a 10db rise from about 2000hz to 10k. The RS28 is flatter than this. There is nothing in the design that could cause the tweeter response to peak 10db in that range. I think your RS meter is that far off above 2000hz.

I'm also not sure that a RS Meter (that is what you are using, right?) and Room EQ is the right gear to be taking measurements for this. And, in room graphs are great, but to many things can be screwing them up. To isolate the speaker from the room, I would think that measurements need to be take at 1 meter, with the measurement gate set so that it eliminates room reflections. It would be great if Chuck could come over and do real measurements, but that is a real pain in the butt.

No, the old graphs were useless!

I realized when I was doing the measuring, all of it was wrong. I would just run the measurement and REW would tell me that it may be clicked and I would just click OK thinking not much of it. But now realized that it was recording a clipped measurement so since the actual dB's were much higher than the microphone would measure, it would just give save it's peak measurement that it was able to record.

They sound GREAT to my ears which is why the graphs puzzle the hell out of me as well. Maybe my ears suck?

[dlneubec]

If it is not a wiring problem is there any chance that dip might be floor bounce related? I'm sure the speaker was designed to sit on the floor. For example, guessing that the lower woofer is about 15" off the floor (to the center) and the lower mid is maybe 24" off the floor. That makes their floor bounce nulls at about 938hz and 591hz. Assuming they are being lifted up another 15" when mounted on the wall, that puts the lower woofer and mid at about 30" and 39" with floor bounce nulls at 476hz and 371hz, which is getting closer to the dip in the REW graph. Also, it looks like the ceiling is scalloped up quite a bit, which might mean the ceiling bounce could be reinforcing similar nulls.

I don't know where you have these crossed over and whether changing the null locations would have any effect.

Just a thought off the top of my head.

[---k---]

They sound GREAT to my ears which is why the graphs puzzle the hell out of me as well. Maybe my ears suck?

Not discounting Dan's theory. I thought about the floor bounce myself. But my guess is that the measurements are just bad.

Trust your ears and be happy.

Though, do try and get Chuck over to your place. I would be very interested in hearing his thoughts. Measurements or no measurements.