Film-Tech Forum ARCHIVE: Speaker Phasing

This is topic Speaker Phasing 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=008297

Posted by Richard May (Member # 2627) on 10-08-2006, 01:36 PM:

Looking back on some of the old posts on this subject, I would like some more info on a few things. I noticed Steve said using a PN generator was risking at checking speaker phasing. I think I found that senario today. I use a 2 piece phase checker myself. While checking phasing with the bass drivers, all speakers were in phase. After setting levels and EQing, I checked phasing again with the generator. Left and right were fine together, but center with either left or right seemed to be out of phase. I say "seemed to be" because of this. When I send pink noise in phase to left and out of phase to lets say right channel, there is a clear drop in low frequency. When I compare left or right channel to the center channel however, its not so clear. Either way I sent the phasing to center, I really dont get much of a drop from the bass. Also, I dont get the same increase as you would when they are both in phase. Follow me so far?? I guess this is why Steve made that comment about not really using the generator to check phasing.
My last question is about HF phasing. Is it as important? I havent really read anything about it. It all seems to be about the LF phasing. Most of the horns I checked so far were wired differently. Some in phase, some out. Also, on a lot of the HF diaphragms I have replaced, polarity is not marked.

Posted by Steve Guttag (Member # 268) on 10-08-2006, 04:55 PM:

Phasing is of absolute importance! It must be correct with respect to each driver within a speaker and each speaker with respect to the other speakers.

Don't believe me...connect your L&R speakers out of phase with each other and turn off the surrounds...they'll sound like they are on!

Phasing is important within the speaker so that at the crossover point, the pressure waves add with each other from the two drivers...otherwise you'll have a more notable dip at crossover.

Since you experimented with phasing...try flattening your EQs and then try the PN phase test...I'll bet it is easier to tell.

Steve

Posted by Richard May (Member # 2627) on 10-08-2006, 05:53 PM:

I do understand that phasing is extremely important. I figured that the EQing probably had something to do with the PN results I saw with the center speaker. The HF phasing is something that has bothered me. You're saying that it's as important as the LF phasing, which I do believe you in that. My question is why I have never heard people talk about it as much as LF phasing. I have found so many more HF drivers out of phase than LF drivers. Also, as I had mentioned, the labeling on a lot of HF drivers that I have seen doesnt tell you + and -.

Posted by Gordon McLeod (Member # 33) on 10-08-2006, 06:18 PM:

Well I always drag out the phase checker and test the Hf and LF units as matter of course
Like the X curve it is part of the drill and when time alignment is completed the system is more transparent

I have a breakout box that goes from the phase checker generator to the D sub for the aux input so I can switch t between cchannels and it passes throught the entire chain ensureing there are no hidden phase reversals

Posted by Richard Fowler (Member # 893) on 10-09-2006, 10:16 AM:

I have experienced so many sound calls where phasing was the issue, especially on HF. Situations such as new speakers from the manufacturer miswired, different brands of amplifiers running in a rack, surround speaker terminations miswired, replacement components miswired.....phasing is the starting point [thumbsup]

Posted by Louis Bornwasser (Member # 3063) on 10-09-2006, 10:55 AM:

Then remember that MOST JBL speakers are made out-of-phase compared to all others. (See THX instructions) Louis

Posted by Kenneth Wuepper (Member # 1174) on 10-09-2006, 01:18 PM:

Louis,

I have a theatre with 2 JBLs and 1 Altec and the phasing was indeed different between them.

Also have a theatre with RCAs and an Altec, must be checked very carefully for proper phase or the results are very poor.

KEN

Posted by Charles Greenlee (Member # 3856) on 10-10-2006, 12:18 AM:

Is this valid for home systems too? I have one that has different branded speakers. Actually my surrounds came from another system, and weren't intended to be used with this one, they were hard wired. I cut the wires, and stripped them and they work, they are right impedance, but I still have to have them all the way up to balance them, and they're still a tad weak. Of course they simply could be less efficent, too. But not by that much.

Posted by Cameron Glendinning (Member # 3516) on 10-10-2006, 05:51 AM:

quote: Richard May
My question is why I have never heard people talk about it as much as LF phasing. I have found so many more HF drivers out of phase than LF drivers.

This has alot to do with the crossover and placement of the horn, 6db (90 degres out of phase,) 18db (270 degres out of phase) traditionally pos to pos termanal. 12db passive crossovers is 180 degress out of phase and traditionally pos to neg of horn, meanwhile the bass pos to pos as a rule.

For the best result the magnets of the speakers should be lined up depth wise, with the horn slightly forward to compensate for the fact that the speaker is up at 2/3 hight of screen. this is time alignment non digital style and has a big effect on phasing.

I have made speakers for cinemas that are wired identially but in one screen, the best result was by reversing the phase of the centre channel only (off the eq equipment) so trust the test gear

[Smile]

Posted by Richard May (Member # 2627) on 10-10-2006, 12:18 PM:

Why are some speakers made out of phase like Louis says. Also, are there any reasons that you would leave a certain driver wired out of phase compared to the other drivers?

Posted by Richard Fowler (Member # 893) on 10-10-2006, 12:39 PM:

Phasing standards vary with some sound manufacturers...with experience you generally know what is required for matching purposes. Sometimes the information is mentioned on the specification sheet, such as + voltage on red = forward excursion.

Posted by Louis Bornwasser (Member # 3063) on 10-10-2006, 01:22 PM:

The story I got was that James B. Lansing was a lefty so he wound the cores anti-clockwise. Some newer JBL product is "normal." This is the equipment from off shore or from a "normal" sub-contractor. Louis

Posted by Edward Jurich (Member # 1787) on 10-10-2006, 02:30 PM:

If you run silent movies you can phase the speakers out. [Roll Eyes]

Sorry for the bad joke. Haven't been here in some time and felt like posting something [Big Grin]

Posted by Bruce Hansen (Member # 281) on 10-10-2006, 05:22 PM:

There is an old trick of putting a 1.5V battery across the terminals. If the speaker cone moves outward, then plus on the battery is plus on the speaker.

Posted by Steve Guttag (Member # 268) on 10-10-2006, 06:19 PM:

Since James B also was Altec...and they are always + on red equals forward driver motion...the JBL being a lefty is out.

However, when JBL was formed, I guess they just had to be different than their previous company and with the + on Red equals reverse cone motion (away from the listener).

This continued on JBL well into the 90s. Since virtually EVERY other speaker manufacturer went with the Altec convention, JBL has made all NEW drivers with the Altec convention, unless they are a direct replacment of a "legacy" type driver.

So...if you have a JBL 2225 driver (pre current manufacture) and replace it with a JBL 2226...all will be well...the 2226 is a direct decendent of the 2225. Likewise for the 2246 vs the 2245 and so forth.

However, new drivers like the 2242 which show up in the current JBL subwoofers but are NOT replacements for say the 2245, carry the current convention (Altec). The 2241, on the other hand is a replacement to the 2240 so it uses the older JBL convention...confused yet?

Presumming that the day comes when they depart from all of their legacy drivers JBL will have the same convention as all other manufacturers but until that day...one has to keep track of which driver they are dealing with.

Posted by Richard May (Member # 2627) on 10-10-2006, 06:50 PM:

So I guess my last question on this matter would be as follows.... Is there any ocassion where you would want or need to have lets say the horn, out of phase with the bass driver? Forget about the markings on the speaker. Lets say you're checking phasing with a 2 piece phase checker.

Posted by Robert Minichino (Member # 3495) on 10-10-2006, 07:02 PM:

Improper phasing between LF and HF units will give you a notch due to cancellation in the crossover region where both units are playing. The proper polarity relationship between the two depends on the crossover.

Posted by Tim Reed (Member # 137) on 10-10-2006, 08:12 PM:

quote: Edward Jurich
Haven't been here in some time and felt like posting something

Hi, Edward! Long time no hear from. Good to see you back, pal!

[Smile]

Posted by Cameron Glendinning (Member # 3516) on 10-10-2006, 10:01 PM:

quote: Bruce Hansen
There is an old trick of putting a 1.5V battery across the terminals. If the speaker cone moves outward, then plus on the battery is plus on the speaker.

yes but please dont do it to the high frequency unit (compression driver)

Posted by Sam D. Chavez (Member # 1841) on 10-11-2006, 10:34 AM:

The only valid way to inject test pulses to test phase is way upstream of the speaker to eliminate the possibility of polarity flips in the wiring to the amps and god forbid, splices in the speaker wire where you can't see them.

And of course, most times you can't see the speaker cone behind the screen.

Posted by Edward Jurich (Member # 1787) on 10-11-2006, 10:50 AM:

quote:
Hi, Edward! Long time no hear from. Good to see you back, pal!

I sorta lost interest for about a year after a bad experience working for a theater owner. But I have been helping out a small town theater with some sound problems and sparking an interest again.
As far as phasing speakers, I always use my ears as a final test. I shut the surround and center channel off. I mono out the sound by clipping the left and right amp inputs together. Then stand center screen, walk around a bit. All the sound should come from the center of the screen if everything is phased right. Then connect the center channel (which should already be left plus right) and check between left / center and right / center.
I was amazed after installing an optical stereo / surround system in that small town theater. Knowing that all the surrounds are a single channel, while sitting through "Open Range", action on the screen would go off screen left or right and the sound would follow off screen. Besides being impressed that phasing audio works that well, it's a good indication everything including the surrounds are phased correct.

Posted by Richard May (Member # 2627) on 10-11-2006, 12:06 PM:

I use a 2 piece phase checker from goldline. The generator sends out the signal. You hold the other unit in front of each speaker cone or horn. I have a patch panel going into the digital inputs of all our processors. This way, like Sam says, you can get an accurate reading at the speaker. The only question I have that hasnt been answered yet is this....Is there any instance where you would need a certain driver of any kind, to be wired out of phase with another driver? And why, on one install, were all the HF's of all the JBL 8330's wired out of phase with the LF's? I read in a post here that JBL did this on purpose......any truth?

Posted by Robert Minichino (Member # 3495) on 10-11-2006, 01:35 PM:

It depends on the crossover type and polarity markings as well as the distance between the effective acoustic origin of the driver and the crossover frequency. The goal is to avoid the drivers cancelling each other out in the region where they're both playing.

A 2nd order crossover (12dB/octave) gives you a net 180 degree phase shift between the LF and HF outputs, so swapping the polarity of one of the drivers reduces cancellation. Some crossover networks reverse their labelling for that and others rely on a reversed connection to the drivers.

Because physical movement of the two drivers relative to each other is the same as an electrical phase shift, a displacement of a half wavelength at crossover frequency will also cause cancellation, so you'd swap the polarity of one of the drivers to reduce that.

Posted by Louis Bornwasser (Member # 3063) on 10-11-2006, 01:57 PM:

We once used a "warble" tone test loop to check phasing of horn vs. woof. You would physically tune for a null in warble level and then reverse the wires to the horn. This assured that the horn/woof added. Louis

Posted by Richard May (Member # 2627) on 10-11-2006, 03:02 PM:

Robert, does that mean that the HF would then be out of phase with the LF to get the proper results? Or are you just saying that the leads would be reversed in order to get both in phase with each other?

Posted by Robert Minichino (Member # 3495) on 10-11-2006, 03:24 PM:

With the crossover in the system they'd be in phase, but (assuming they use the same +/- terminal convention) they'd be out of phase at the terminals.

Posted by Steve Guttag (Member # 268) on 10-11-2006, 03:51 PM:

On the 8330 (wood cabinet) the woofer is indeed wired out of phase with respect to the mid and high.

Oddly enough, it is the mid and high that are wired out of phase on the 8330A (plastic cabinet).

The answer...comes back to the JBL vs everyone else's standards. The 8330 was Postive on BLACK terminal for outward cone motion. The 8330A is Positive on RED terminal for outward cone motion.

Posted by Richard May (Member # 2627) on 10-11-2006, 04:34 PM:

Sorry to keep harping on this. Bottom line, I want to know is, do you want all the drivers in the theater to be the same phase? Are there any exceptions? This includes the surrounds too. If so, do I need to reverse the HF on all of those JBL surrounds? One more thing....why the hell would JBL do that...have the LF out of phase compared to the HF....

Posted by Cameron Glendinning (Member # 3516) on 10-11-2006, 07:35 PM:

Richard phasing is very important, unfortunately there are many variables, often correct phasing is often only found through the test gear showing the cancelation effect at the overlap at the x over point. 6db passive crossovers and a good electronic type are the most phase correct all else is a bit unpredictable. So often the visual inspection of the wiring is not reliable [beer]

Posted by Gordon McLeod (Member # 33) on 10-11-2006, 08:27 PM:

To ensure none of the electronics are causing a phase shift I always connect the phase checker generator to the processors external 6 track input and then go speaker by speaker through the system and yes all drivers must be in phase at the end of the day

Posted by Bob Koch (Member # 1062) on 10-11-2006, 08:47 PM:

I don`t know if I`m getting in over my head here but many, and I emphasise many years ago we were talking about acousticalize phasing here.Altec, with there many systems starting with A2`s A4`s and A4x`s were very susceptable to this. You had to put on your coveralls and go back on the deck and physically move the 288 on its sled and finally find the point where it was acoustally in phase,very difficult to do. If you found the point,and many times it was within an inch or two, you could not believe the difference in the auditorium.I doubt any one in this day and age would want to give any credability to this. Except , Steve I would like to hear any comments you have,

Posted by Robert Minichino (Member # 3495) on 10-11-2006, 09:12 PM:

The A2, for example, has about a 500Hz crossover point between HF and LF units, 1/4 wavelength of which is about 7", so I wouldn't think any changes less than about 7" making a difference in sound. Still, that's not much movement, really.

Posted by Richard May (Member # 2627) on 10-11-2006, 09:16 PM:

Thanks Gordon...that was the answer I was looking for....ALL drivers regardless, need to be the same phase. BTW, thats how I check phasing also. Going throught he external inputs.

Posted by Gordon McLeod (Member # 33) on 10-11-2006, 10:25 PM:

Robert the industry standard for stage speakers has been for a long time 500Hz
only the smaller compact speakers has there ever been much of a variance from that
500Hz was choosen as it allowed the male dialogue range to be pumped throught the horn

Posted by Cameron Glendinning (Member # 3516) on 10-11-2006, 11:42 PM:

quote: Robert Minichino
The A2, for example, has about a 500Hz crossover point between HF and LF units, 1/4 wavelength of which is about 7", so I wouldn't think any changes less than about 7" making a difference in sound. Still, that's not much movement, really.

I'm not sure but ideally would you not want the horn to be more than 7 inches from the woofer. Personally my experience agrees with the comment about moving back and forward to a fairly precise spot, altec do go to a lot of detail in their installation guides, ie reverse polarity to horn, move back and forward until cancellation of frequencies is most noticeable, then swap hi frequency phasing and your done [Big Grin]

Posted by Steve Guttag (Member # 268) on 10-12-2006, 03:39 PM:

Richard,

Speaker phasing should be done by "system"...that is, don't rewire your surrounds. Their woofers should match every other woofer in your auditorium.

Now having said that, on the stage channels, they generally should all be the same phase throughout the system if there are no passive crossovers in the system.

And that is a key element...the crossover can create time-domain shifts which is why some speakers will have drivers deliberately wired out of phase.

So if you go around with your phase checker, all of the woofers should match...on the stage channels the highs (and mids) should match the woofers.

As to moving the HF sled to achieve proper match...yes that is correct. If one knows the crossover point, all one has to do is play a tone at the crossover frequency and move the sled until maximum level is achieved.

Nowadays, most people use active crossovers and can adjust the delay on the driver that is arriving sooner to electrically achieve similar results. Rane, in particular, makes a nifty crossover, the AC22B (and AC23B) that has a continiously adjustable delay so one may precisely tune it.

However, Altec had it right (as they normally did) to mechanically get things aligned properly .

Steve

Posted by Richard May (Member # 2627) on 10-12-2006, 04:16 PM:

Thanks Steve,
All of our speakers use passive crossovers. Should all the woofers and HF's still be made to be the same phase using the phase checker? You said not to rewire the surrounds. What about the HF's being out of phase with the LF's?

Posted by Steve Guttag (Member # 268) on 10-12-2006, 04:27 PM:

Richard,

On the surrounds...they have a PASSIVE crossover in them...the phase shift in the crossover accounts for the need to wire the other drivers out of phase to achieve the proper response.

As to your stage channels...since they have passive crossovers there are no absolutes...MOST have all drivers in phase but in the case of several speakers...like the JBL 4670 and those that look like it from EV and EAW...I believe they instruct you to wire the HF out of phase with their current crossovers.

If you know the crossover point...put a tone out at that frequency...and measure the level with the phase reversed...the correct phasing should have the highest level. Note...all of your stage channels should match though...that is, if reversing the phase on the HF unit of Center is best....then it better be reversed on Left and Right too.

Steve

Posted by Robert Minichino (Member # 3495) on 10-13-2006, 11:03 AM:

The reason why you hook up the LF and HF units out of phase when trying to physically align them is because it's much easier to hear the null created when they cancel than it is to hear the peak when they're perfectly aligned.

Graph of attenuation in dB versus wavelength

Posted by Richard May (Member # 2627) on 10-22-2006, 07:16 PM:

Steve, thanks for all the info. My final thought on this subject is the following. Lets say, like you said, the HF's on the surrounds are wired out of phase for a certain reason. If the HF's on the stage speakers are wired IN PHASE, wouldnt there be a problem with the HF between the stage and surrounds or no?

Posted by Steve Guttag (Member # 268) on 10-22-2006, 08:23 PM:

Snappy answer...NO.

The reason is left as a homework assignment.