AudioRoundTable.com: Pi Speakers » 4 Pi Design Questions

Home » Sponsored » Pi Speakers » 4 Pi Design Questions

Show: Today's Messages :: Navigator :: Polls
E-mail to friend

Re: 4 Pi Design Questions [message #62304 is a reply to message #62303]

Wed, 07 April 2010 21:55

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

That's it!

Report message to a moderator

Re: 4 Pi Design Questions [message #63208 is a reply to message #62304]

Sun, 20 June 2010 08:18

Nichol1997
Messages: 21
Registered: December 2009
Location: Virginia

Chancellor

Wayne,
It has been brought to my attention that I did not use the latest crossover design for my 4 Pi build. Specifically, C4 is labeled as 10uF in the plans dated January 30, 2008.

I used a 20uF capacitor. I am wondering if I need to change it to correspond with the latest design. Can you tell me the impacts of this change?

Report message to a moderator

Re: 4 Pi Design Questions [message #63211 is a reply to message #63208]

Sun, 20 June 2010 08:54

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

I would swap it, yes. It's an easy change worth doing. The position of the forward lobe is shifted up slightly, more aligned with the baffle normal, i.e. straightforward. With the 20uF cap, it points slightly downward but with the 10uF cap, it's almost perfectly centered.

Report message to a moderator

Re: 4 Pi Design Questions [message #64648 is a reply to message #63211]

Thu, 11 November 2010 21:11

Nichol1997
Messages: 21
Registered: December 2009
Location: Virginia

Chancellor

I thought I would post some pictures now that I finally finished putting the finish on them.

http://img513.imageshack.us/img513/4232/dscn1441g.th.jpg

http://img168.imageshack.us/img168/6747/dscn1442h.th.jpg

http://img535.imageshack.us/img535/5870/dscn1447t.th.jpg

http://img507.imageshack.us/img507/3356/dscn1439u.th.jpg

By the way, I swapped the 20uF caps to the 10uF caps and I haven't noticed much difference. Of course it has been 5 months since I heard them with the old caps so that could be part of the reason.

Report message to a moderator

Re: 4 Pi Design Questions [message #64649 is a reply to message #61737]

Thu, 11 November 2010 22:06

Psychoacoustic
Messages: 75
Registered: May 2009

Viscount

They look great! Well done on finishing them.
I've still got 7 Pis in beautiful MDF!

Report message to a moderator

Re: 4 Pi Design Questions [message #64650 is a reply to message #64648]

Fri, 12 November 2010 07:52

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

Nichol1997 wrote on Thu, 11 November 2010 21:11

By the way, I swapped the 10uF caps to the 20uF caps and I haven't noticed much difference. Of course it has been 5 months since I heard them with the old caps so that could be part of the reason.

It's very subtle, in fact, I'm not sure that even describes it. In some listening positions, it does nothing at all. The only thing that's changed is the position of the forward lobe, shifted slightly upward.

The original cap had the forward lobe pointing down about 5°. The pattern was good from about -30° to +20°. The new cap centers the forward lobe, so the pattern is good from -25° to +25°. If you're not listening way above or way below the speaker, it sounds the same.

Report message to a moderator

Re: 4 Pi Design Questions [message #64651 is a reply to message #61737]

Fri, 12 November 2010 09:41

Doc Jr 8156
Messages: 26
Registered: May 2009

Chancellor

Great looking speakers! Wayne, what is the consequence of not having R3 installed? My crossover (from a kit) has none of this. Is it imperative to use a 100 watter 8 ohm resistor or could I get away with a 15 watt non inductive 8ohm resistor? Thanks and sorry for asking these here.

Report message to a moderator

Re: 4 Pi Design Questions [message #64658 is a reply to message #64651]

Fri, 12 November 2010 14:44

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

If you're using the TD12S woofer (used only in the three π loudspeaker), you don't need R3 but on all other models, you do. If you omit this resistor on any of the models that call for it, you'll have a pretty nasty peak in the upper midrange, near the crossover frequency. This is true for all versions of the four π loudspeaker.

You can derate any component, provided power input is reduced accordingly. That said, I would not suggest derating any of them because it never hurts to keep things cool.

See the Speaker Crossover document for more information. An example of the peak you'll get if you remove R3 is shown on page 43 and the power transfer curve for this component is shown on page 66.

Report message to a moderator

Re: 4 Pi Design Questions [message #72745 is a reply to message #64658]

Thu, 17 May 2012 19:38

Nichol1997
Messages: 21
Registered: December 2009
Location: Virginia

Chancellor

I finally got around to measuring my modified Pi4 speakers. Attached is the nearfield frequency response using Room Eq Wizard version 5.

I am still learning how to use the software and hardware so I am not sure if I am measuring it correctly. I had the mic mounted on a stand that was 16" from the baffle and just a little bit below the throat of the compression driver on the vertical.

My measurement gear consists of:
HP laptop w/ Windows XP
MobilePre USB external soundcard
Behringer ECM8000 mic
Behringer EP2500 (ran the MobilePre USB directly to the amp)

To me, the measurement makes it look like my woofers are out of phase. I only measured the right speaker. I will have to wait until it is quiet in the house again before I can measure the left speaker.

Attachment: Pi4 Measurement.jpg
(Size: 90.03KB, Downloaded 487 times)

Report message to a moderator

Re: 4 Pi Design Questions [message #72746 is a reply to message #72745]

Thu, 17 May 2012 22:46

Wayne Parham
Messages: 18786
Registered: January 2001

Illuminati (33rd Degree)

You may have the woofer or tweeter leads reversed. But redo your measurement first, because measurements made like you are doing always create notches from boundary reflections.

Lay the speaker on it's back and position the microphone one meter (or more) above it, facing the upper half of the midwoofer cone. This orientation will prevent a floor bounce notch, and it puts the microphone in the center of the forward lobe. You can usually get away with indoors measurements when done this way, if all you care about is response in the crossover region, but you'll have no visibility below a few hundred Hertz. It will be peaky if ungated, or meaningless if gated. Of course, it never hurts to take the speakers outdoors, which is truly anechoic. You'll still want to lay the speaker on it's back, or you can put the microphone on the ground and angle the speaker down to face it.

This is what you should expect to see, measurements from two different people using two different measurement systems:

four π loudspeaker measured on Ominmic and LMS
four π loudspeaker performance data, LMS data, including off-axis charts

I realize your speaker is a modded version, but if the only difference is box size, then I would expect most changes to be below 200Hz. So setup your measurement as described above, and if you don't see a chart like this, check your connections and crossover.