Film-Tech Forum ARCHIVE: SG-1 pink noise generator

This is topic SG-1 pink noise generator in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=004347

Posted by Tom Wienholt (Member # 1510) on 12-25-2002, 01:23 AM:

I recently purchased a component engineering SG-1 pink noise generator. The instructions explain that the channel selector switches (L,C,R,S) are off in the center position and are "of opposite phase on each side of center". What does this mean? Should the switch be in the left or right position when putting pink noise through a channel?

Posted by John Walsh (Member # 168) on 12-25-2002, 09:07 AM:

If the instructions saay: "of opposite phase on each side of center" - I think that is a mistake in printing. When the switches are up, the pink noise in in phase. When down, that channel is out-of-phase.

Posted by Steve Guttag (Member # 268) on 12-25-2002, 09:56 AM:

No Mr. Walsh, it is you that is mistaken on this one....

The CE SG-1 is the black framed PN generator for Dolby Analog processors (except the CP-45). The switches are more tightly spaced on the unit to allow for the tone generator knob and THX jack (for RT-60 tests). Thus, the switches operate in a left/right orientation instead of the traditional up/down orientation.

What the instructions are saying is that if you wanted to check phasing of your speakers (at best a risky subject using a PN generator), you can check relative phase by moving one or the other switch to the opposite direction and see if there is a level increase or decrease (as measured on your RTA). The idea being that if there is an increase when the two switches are moved in the opposite direction, you have the two speakers out of phase with one another.

With the SG-1, moving the switches to the right is considered "in-phase" (equivalent to the Cat. 85C from Dolby with the switches up). However, with any one channel selected, it really doesn't matter.

There...confusing enough?

Steve

Posted by Tom Wienholt (Member # 1510) on 12-25-2002, 10:50 AM:

Steve
Thanks for your help.

Posted by John Walsh (Member # 168) on 12-25-2002, 01:28 PM:

Oops,... It must be the eggnog.....

Posted by Rick Long (Member # 213) on 12-26-2002, 11:22 PM:

".....if you wanted to check phasing of your speakers (at best a risky subject using a PN generator)," - Steve Guttag

Steve, if you wanted a quick check of proper phase polarity, what method would you use (other than visually checking the wiring)?

Posted by Steve Guttag (Member # 268) on 12-27-2002, 07:23 AM:

There is no substitute for using an actual phase checker. It will thump the speaker every second or so...the pick up will tell you if the cone/diaphragm motion is inward or outward.

Steve

Posted by Steve Kraus (Member # 476) on 12-27-2002, 07:45 AM:

Let me expand on that and explain that Steve is referring to a 2-piece device consisting of a signal generator that produces the uni-polarity thumps (line level, connect it directly to amp channel) and the receiver which has a mic on it and is held near the speaker (or driver of an array). A typical design has LEDs that flash one color or another depending on the phase of the thumps and being very small can be lashed to a pole and held up to surrounds and other hard to get to speakers to check the phasing. As always it doesn't matter what the phase is so long as they are all the same.

Posted by Gordon McLeod (Member # 33) on 12-27-2002, 07:50 AM:

Yes the phase checker is the best unit and is available from smart devices [Smile]

Also the cp650 has the thump part built in

Posted by Bill Purdy (Member # 202) on 12-27-2002, 09:46 AM:

There is a happy little bit of serendipity for those of you who have both an SG-1 and a
phase checker. If you put the Pink Noise/Tone switch in its middle position and connect
your phase checker to the pink noise output test jacks (hot to Blue and ground to Black)
you will be able to then use the channel select switches on the SG-1 to feed the phase
checker to the amplifiers. The phase can also be reversed by these same switches just
as discussed above.

Posted by Steve Kraus (Member # 476) on 12-27-2002, 01:54 PM:

Hmmm...speed of sound is about 340 m/s. So a 1000 Hz tone would have a wavelength of .34 meters. Which means that when the same sound eminates from more than one speaker if the listener is in a position whose distance to either speaker differs by half that distance will have phase cancellation. Is there a flaw in my reasoning? It sure gives one pause regarding speaker phasing.

Posted by Paul Mayer (Member # 355) on 12-27-2002, 04:30 PM:

Ah, but we don't normally use a steady tone to check phase. The phase checker transmitter puts out just a series of thumps or clicks (hence the nickname). Since the test signal is just a transient (not even a very fast one), the receiver can tell whether the driver in question is going in or out. The phase checkers I've used didn't seem to be sensitive to distance or placement relative to the test source--the closest one is the one that gets read.

Posted by Steve Kraus (Member # 476) on 12-27-2002, 05:25 PM:

Yes but my point is that how important is phasing if moving my head a few feet one way or another will create the same phase error as miswired speakers?

Posted by Mark Hathaway (Member # 1044) on 12-27-2002, 05:35 PM:

I find that normally if something is to be out of phase, it will be the horn in relation to the bass bin.

I just love getting up onto the platform.

Mark Hathaway
Atlab Image and Sound Technology
Melbourne, Australia

Posted by Christopher Seo (Member # 7) on 12-27-2002, 11:23 PM:

Interesting point, Steve. Presumably this is why the multichannel pink noise/RTA test of speaker phasing is not foolproof? Having the stage speakers out of line with each other, as in a curved screen setup, would also seem to throw off the phase.

I'd imagine it's far more important for an array of speakers designed to work in unison to be in phase, i.e. a subwoofer array or the frequency sections of a single stage speaker, rather than a set of stage speakers whose axes may not be parallel, whose sound may not emanate from the same plane, and whose sound information is usually different anyway.

Posted by Steve Guttag (Member # 268) on 12-29-2002, 10:34 AM:

It is very important for any multi-channel system. You'd be amazed at how awful a non-phased system sounds...kinda like permanent surrounds.

As to curved screens...no that actually would help things since the distance to the listener is more similar between speakers. A flaw in Steve's reasoning is that he is listening to one tone from one location...when listenting to wide-band being out of phase will create a bit of a combing effect and NO position will sound right. There is also the psychoacoustics of it all...that is the first arrivial is given more weight by your ears and brain than the second arrival...you also listen with two ears (well most of us) so the direct path to your left ear is biased to the left speaker and such.

But this is an easy experiment...go home and play one of your favorite tunes on your CD player...then switch the phase on one speaker....report back with your findings.

Steve