Re: Three Pi and Three Pi Sub Plans [message #91441 is a reply to message #91269] |
Sat, 15 February 2020 12:36 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
timothyeyster
Messages: 11 Registered: October 2019
|
Chancellor |
|
|
Hi, Wayne.
After trying a few positions, I just couldn't get my subs far enough behind or below my 3Pis and ended up with a huge hump in the 80-120hz region using a 12db/octave 100 hz crossover on the sub. Ultimately I decided to just stack the mains on top of the subs, cross them over at 60hz 24db/octave, and push them into the corners. They seem to be integrating pretty well now, with only a minor slump due to the lack of baffle step compensation. I expect this will be mitigated once I build out the screen wall, but, until then, I'm really happy with the results; By far the best low end I've had in my room!
I'm still wringing my hands a bit over the top end response, though; They still sound like the highest octave is rolled off.
I decided to take some nearfield measurements to try to confirm what I was hearing. After reading this article on the miniDSP website on speaker measurements using my Umik1 and REW
https://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements
I made this measurement:
![/forum/index.php?t=getfile&id=2507&private=0](/forum/index.php?t=getfile&id=2507&private=0)
Per the miniDSP article, I placed the 3Pi in the corner, turned it in 45 degrees and raised it up until it was roughly centered between the floor and ceiling. I placed the mic at a 1m distance from the speaker baffle and pointed it directly between the bottom of the horn and top of the woofer. I then gated the measurement to eliminate the floor bounce.
The measurement seems to indicate that the response starts to rolloff at 10khz, and that comports with what I'm hearing. Does this still seem likely to be a malfunctioning crossover to you? I triple and quadruple checked the schematic against my wiring and verified each connection with a multimeter. Although it's not impossible, I would be really surprised if I still had a wiring error at this point.
If this response looks correct to you, could you suggest a way I might be able to raise the top octave response a bit to better suit my tastes? For example, would it help to install a cap at the C1 position?
Meanwhile, I'm really enjoying my setup! So much so that I've built boxes for a 3Pi center and third 3Pi sub and am in the process of finishing them with black laminate. I also have a CNC router file ready to roll for a pair of 1Pi rears as soon as I get a little more free time.
Thanks again for all of your great help!
|
|
|
|
|
|
Re: Three Pi and Three Pi Sub Plans [message #91450 is a reply to message #91446] |
Mon, 17 February 2020 11:19 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18753 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I've always loved ribbon tweeters too. Back in the 1970s, the EMIT tweeters in Infinity speakers were oh-so sweet to me. They offered much greater extension than compression drivers of that era. But they lacked dynamic range. They were relatively fragile, and couldn't handle a lot of power. So that was the trade-off.
Most compression drivers don't have the above-20kHz output that ribbon tweeters do. Most don't even hit 20kHz. There was a time when I wouldn't have used compression drivers for hifi because they didn't even hit 10kHz. But since the late 1990s, you could expect to reach nearly 20kHz with a relatively inexpensive titanium or polyimide 1" exit compression driver. That combined with their superior dynamic range and their ability to be used on a waveguide to set directivity are the advantages of a compression driver tweeter.
I wanted to mention a few thoughts about your subs too. I neglected to say this on my last post, so I thought I would add it here.
Subs should always be set at an amplitude that almost makes them invisible. When a subwoofer is set right, you shouldn't even know it is on. About the only thing that should be noticeable is the deepest bass, and that shouldn't be glaring. There should only be a very subtle difference in the sound when you switch the subs on and off when they're setup right.
If you can hear noticeably louder bass or midbass with the subs on, then they're set too loud. Thus is true no matter if they're set as multisubs, flanking subs or traditional subs.
Since flanking subs are run to relatively high frequency, the system will sound throaty if they're set too loud. That's an obvious sign that the subwoofer SPL is set too high. If you can measure a hump in the 80Hz to 120Hz region, then the flanking sub volume is set too high. That's the region where we expect the mains and the subs to blend transparently. There should be virtually no SPL increase in this region. It should be the same SPL level as the 200-300Hz region. If it's too loud around 100Hz, then the SPL of the flanking subs needs to be reduced.
So if you ever feel like experimenting, please keep these thoughts in mind. If you can dial it in with a traditional sub setup, that's awesome. I would expect a little bit of a dip below baffle step and I would expect some SBIR. But if those aren't troublesome - excellent - enjoy the system. On the other hand, if you notice something lacking in the upper midbass and lower midrange, you might try a flanking sub setup again, but reduce the SPL of the subs to the point where you almost can't even tell they're tuned on.
|
|
|
|
|
Re: Three Pi and Three Pi Sub Plans [message #91453 is a reply to message #91451] |
Mon, 17 February 2020 15:13 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18753 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
OutOfSpace wrote on Mon, 17 February 2020 14:25"If you can hear noticeably louder bass or midbass with the subs on, then they're set too loud. Thus is true no matter if they're set as multisubs, flanking subs or traditional subs"
I guess most car 'audio' guys that didn't get the memo. Some of them are seriously rude about it, too.
timothyeyster wrote on Mon, 17 February 2020 14:22With all of that in mind, I may try implementing the flanking subs again when I install the center channel. If I do, could I check for proper integration using a measurement with REW from the main listening position? Sounds like I'd be looking for a relatively flat response at 100hz, and fewer peaks and nulls below that compared to the 3Pi on its own- correct?
That's right, exactly. Set the SPL so that 100Hz is the same level as 200Hz and 300Hz. What you'll also get is a reduction in the amplitude of the dip that results from self-interference from the wall behind the speakers and also reduction of the dip from the interference from the sidewall nearest to the speaker. The subs will provide extension as well.
You won't get much modal smoothing below 80Hz though. To smooth the lowest frequencies, you need one or two multisubs placed far from the mains. They might be placed at the opposite end of the room.
timothyeyster wrote on Mon, 17 February 2020 14:22I also had a thought: I'm using a miniDSP for the sub crossover and volume, and could easily implement a delay with respect to the main speakers- could that be used to approximate moving the subs back and beside the mains, or does that just open another can of worms?
That's actually an excellent suggestion for your room, since you're kind of limited in fore-aft placement. You could delay the subwoofer signal a smidge to make it act similarly to the fore-aft displacement required by flanking subs. That's what we're looking for - We want the phase difference to not be 180° in the 80Hz to 120Hz region where we normally find it.
The reflection from the wall behind the speakers is usually 180° from the main speaker's direct sound somewhere between 80Hz to 120Hz, depending on the speaker's distance from the wall. Having the flanking sub a different distance makes it not have this same null. So the null is partially filled in.
|
|
|
|
|