Speakerplans.com Homepage
Forum Home Forum Home > General > Newbie Discussion
  New Posts New Posts RSS Feed - Reverse Polarity?
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Reverse Polarity?

 Post Reply Post Reply Page  <12345>
Author
Message
mitchiemasha View Drop Down
Registered User
Registered User


Joined: 27 August 2009
Status: Offline
Points: 466
Post Options Post Options   Thanks (0) Thanks(0)   Quote mitchiemasha Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 1:43am
Originally posted by toastyghost toastyghost wrote:

By the way, a POLARITY flip does cause a shift in the phase response of a speaker, it’s more complex than that diagram makes out. 


Care to expand. All one is doing is flipping the +/-, how can it change?

With an actual sound waveform, the reverse of the wave isn't an exact copy of the other side like a sine wave but i'm curious to where the phase shift is. it's not shifted, it's inverted.

Or is it because the actual speaker won't work the same in reverse. You don't have to explain. Links will do. Thanks in advance.

Edit: Wait a min. toastyghost... slight confusion there. It is exactly the diagram. What the diagram shows is flipping the polarity will not introduces any time delay. Obviously the phase has changed, in regards to its original self, it is inverted like 180 but not 180 out of phase in the form of a time delay phase shift, which is the point of the diagram.  

Edited by mitchiemasha - 18 October 2017 at 5:56am
Back to Top
mitchiemasha View Drop Down
Registered User
Registered User


Joined: 27 August 2009
Status: Offline
Points: 466
Post Options Post Options   Thanks (0) Thanks(0)   Quote mitchiemasha Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 2:51am
Warbles... Yes, I just googled that and watched the video. Obviously, when using a sine wave, if moving to the next cycle one will be inphase but out of time, quite like the 360 on my diagram. The same for 180 out of phase, polarity reversed, will be in phase but out of time (I might add that to the diagram to address this issue).

https://www.youtube.com/watch?time_continue=1&v=HXLAzVDCf2Q

It took a while for the ball to drop. I felt he could of explained the issue a bit better at the start, more like i did above. I was confused at first but had no reason to be. As it's obvious! Knowing which cycle you are in phase with, to know if you are in time or not is a problem with the sine wave test (which can be seen on my diagram)

My current set up Has a Drive rack PA, only 10ms so you only get 1 cycle to match up anyways. If the subs/tops are any further apart than what can be corrected within 1 cycle, it's tough. 

For reflex loaded stacked ontop of each other, moving to the first cancellation, with a sine wave test, should be intime but you are right it would be guess work. How would one know they weren't actually 180 out of phase and polarity inverted, same result. My only guesswork there would be, if the divers are almost in line physically, visually and one flips the polarity, if you get some summation, this would be the correct polarity, if you got some cancellation this would be the incorrect polarity, now you could move to the first most cancellation point. Which way you'd go would depend on which one was further forward or back (Edit:The ports are out of phase anyways so there's always that issue to throw in the mix, lol!)

Disclaimer... The system i currently work on is out of time and badly tuned. The bar speakers are prosounds (or were! they died along time ago, now they have fane tweeters, paudio drivers, original cross overs (urgh)) Even if I wanted to align these I couldn't. The dance area has 1x Mach dual 15 1xRcf art 300's, with the same doing the other area. I can't align 1 with out disaligning the other. Not that i can go in all day testing sound, it serves food too. To the owners it's just not important.



Edited by mitchiemasha - 18 October 2017 at 3:25am
Back to Top
Earplug View Drop Down
Old Croc
Old Croc


Joined: 03 January 2012
Location: Europe
Status: Online
Points: 7199
Post Options Post Options   Thanks (0) Thanks(0)   Quote Earplug Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 10:21am
I´ve used one of these for basic polarity checking:

https://www.thomann.de/es/phonic_paa3_personal_audioassistant.htm?ref=search_rslt_phonic+paa3_180832_0


Quick and easy.




Edited by Earplug - 18 October 2017 at 10:25am
Earplugs Are For Wimps!
Back to Top
toastyghost View Drop Down
The 10,000 Points Club
The 10,000 Points Club
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 10920
Post Options Post Options   Thanks (0) Thanks(0)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 12:54pm
Originally posted by Earplug Earplug wrote:

I´ve used one of these for basic polarity checking:

https://www.thomann.de/es/phonic_paa3_personal_audioassistant.htm?ref=search_rslt_phonic+paa3_180832_0


Quick and easy.






And as mentioned, polarity is not phase, so at best, you're sort of in the right city but far, far from the right building let alone street.
Back to Top
Earplug View Drop Down
Old Croc
Old Croc


Joined: 03 January 2012
Location: Europe
Status: Online
Points: 7199
Post Options Post Options   Thanks (0) Thanks(0)   Quote Earplug Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 1:46pm
Originally posted by toastyghost toastyghost wrote:

Originally posted by Earplug Earplug wrote:

I´ve used one of these for basic polarity checking:

https://www.thomann.de/es/phonic_paa3_personal_audioassistant.htm?ref=search_rslt_phonic+paa3_180832_0


Quick and easy.






And as mentioned, polarity is not phase, so at best, you're sort of in the right city but far, far from the right building let alone street.



No, the PAA3 DOES check phase and time alignment by sending a pulse through the system and making sure that everything is running in time. I know I said 'polarity', but meant phase, ok, silly error. RTM and it is plain.

And mitchiemasha, flipping the polarity doesn´t help when you really need a phase shift. Time alignment will be out. Some of the frequencies (bass end) may sum, but will still be out of phase.



Earplugs Are For Wimps!
Back to Top
toastyghost View Drop Down
The 10,000 Points Club
The 10,000 Points Club
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 10920
Post Options Post Options   Thanks (0) Thanks(0)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 1:48pm
Yeah, I owned one, and no, it doesn't display phase nor handle alignment.
Back to Top
Earplug View Drop Down
Old Croc
Old Croc


Joined: 03 January 2012
Location: Europe
Status: Online
Points: 7199
Post Options Post Options   Thanks (0) Thanks(0)   Quote Earplug Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 2:02pm
Then you must have owned a different one to what I´ve used. Page 17 of the manual. Maybe you didn´t get that far. Too much too read, hey...



Earplugs Are For Wimps!
Back to Top
mitchiemasha View Drop Down
Registered User
Registered User


Joined: 27 August 2009
Status: Offline
Points: 466
Post Options Post Options   Thanks (0) Thanks(0)   Quote mitchiemasha Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 3:15pm
Originally posted by Earplug Earplug wrote:

And mitchiemasha, flipping the polarity doesn´t help when you really need a phase shift. Time alignment will be out. Some of the frequencies (bass end) may sum, but will still be out of phase.


WOW. I never said anywhere that it did. Did you see my Diagram? "Time alignment will be out." this is exactly what i was contributing.
Back to Top
mitchiemasha View Drop Down
Registered User
Registered User


Joined: 27 August 2009
Status: Offline
Points: 466
Post Options Post Options   Thanks (0) Thanks(0)   Quote mitchiemasha Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 3:33pm
Toastyghost... back to those Warbles. I'm not going to be able to tell the difference when listening to a inphase but 2nd/3rd cycle misalign, especially once factoring in the room (not that I class an inphase out of cycle wave as true phase anyways).

What has been overlooked here is a tape measure and some basic understandings of sound. The low frequencies are so large one would have to make a colossal mistake to be aligning the sine wave to the 2nd/3rd cycle. Especially so for stacked systems. I can see it being an issue on larger system, more spread out. 



Back to Top
Earplug View Drop Down
Old Croc
Old Croc


Joined: 03 January 2012
Location: Europe
Status: Online
Points: 7199
Post Options Post Options   Thanks (0) Thanks(0)   Quote Earplug Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 5:39pm
Originally posted by mitchiemasha mitchiemasha wrote:

Originally posted by Earplug Earplug wrote:

And mitchiemasha, flipping the polarity doesn´t help when you really need a phase shift. Time alignment will be out. Some of the frequencies (bass end) may sum, but will still be out of phase.


WOW. I never said anywhere that it did. Did you see my Diagram? "Time alignment will be out." this is exactly what i was contributing.



Ok, just clarifying for the OP and original question with the Tannoys. This has all gone off at a tangent (as usual).  LOL


Earplugs Are For Wimps!
Back to Top
toastyghost View Drop Down
The 10,000 Points Club
The 10,000 Points Club
Avatar

Joined: 09 January 2007
Location: Manchester
Status: Offline
Points: 10920
Post Options Post Options   Thanks (1) Thanks(1)   Quote toastyghost Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 6:21pm
Originally posted by mitchiemasha mitchiemasha wrote:

Toastyghost... back to those Warbles. I'm not going to be able to tell the difference when listening to a inphase but 2nd/3rd cycle misalign, especially once factoring in the room (not that I class an inphase out of cycle wave as true phase anyways).

What has been overlooked here is a tape measure and some basic understandings of sound. The low frequencies are so large one would have to make a colossal mistake to be aligning the sine wave to the 2nd/3rd cycle. Especially so for stacked systems. I can see it being an issue on larger system, more spread out. 





A tape measure doesn't take into account the combined response of the driver, the cabinet / horn, any venting, bandpass elements, or filtering (EQ or highpass / lowpass) and their affect on the phase response of the box.

It gets even harder to do without FFT when you try to maintain being within +/- 30 degrees for 1 octave either side of the acoustic (not electronic) crossover point, because subwoofers in particular are devices that are very stretched in time, even the very very best ones, and some much worse than others.

Here's a simple video showing why this is important:
https://www.youtube.com/watch?v=HXLAzVDCf2Q

And from the same guy, an explanation on why warbles are better than the sine wave method, including links to download them:
https://www.merlijnvanveen.nl/en/study-hall/110-bink-audio-test-warbles

Honestly, the barrier to entry for REW is so stupidly low that there's really not much excuse not to at least try it. A recent-ish old laptop or computer will do, software is free, and an adequate quality mic that plugs straight into USB is about £60 if you order directly (MiniDSP UMIK-1). If you have access to even a crappy old M Audio 2ch interface, with phantom, then the Behringer mic is what, £30?

The REW manual in PDF, and the website, are exceptionally helpful and simple to follow too:
https://www.roomeqwizard.com/help/help_en-GB/html/gettingstarted.html

How about a real world example? This is SMAART but the data would be pretty much the same in REW, just look different. This is a reflex top, dual driver midrange and HF comp on a horn, aligning to a BPH sub/kick section. The mid and high section of the top was already aligned with FFT previously, without subs.

First up, the result that was achieved by somebody doing it by ear:


Doesn't look that bad eh? Until you run both cabinets at the same time and make a trace – looks fine at the place they likely ran the sine wave, say 80-100Hz, but then you see the holes punched in the lower midrange and the lower response of the sub:


Actually measuring the boxes from no relatively delay, with mic on the floor to counter reflections (hence the laps in the HF response which aren't relevant to this part of the process), and adjusting not only the delay but the electronic crossover point and filter type got us these nicely overlapping traces:


Which results in this combined result in purple:


Now, I know what combination I'd prefer to listen to…

Edited by toastyghost - 18 October 2017 at 6:38pm
Back to Top
mitchiemasha View Drop Down
Registered User
Registered User


Joined: 27 August 2009
Status: Offline
Points: 466
Post Options Post Options   Thanks (0) Thanks(0)   Quote mitchiemasha Quote  Post ReplyReply Direct Link To This Post Posted: 18 October 2017 at 9:48pm
I watched that video. I even linked it already. I think we're having a misalignment ourselves.

I don't dispute anything in the video. What i think you've missed is the "1st cycle, 2nd cycle". The only issue with the sine wave test is you'd have no way of knowing which cycle you are on. That is all he is showing it that video. However, if you aligned the sub using a sine wave test and it was the correct cycle, the sub and the mids would be intime just as much as the warble test or any other test, hence a tape measure. The 2nd cycle, in phase but out of time is around 7 meters of incorrect delay. No one in their right mind would do this. The maths allone would set off alarm bells... especially how a simple Driverack wouldn't even give you enough ms to achieve it.

What he is saying is right but so much is being over looked here.

Any extra issues due to port alignment and the delay of the frequencies at 30/40hz is an issue with the cab itself. No adjustments we can make will fix that. all we are doing is matching the main part of the sub to the mains.

The room itself is doing so much damage to the audio, which is pretty much impossible to get perfect, regardless to the pseudoscience people believe. I'm not sure any impulse response, fft or smart will give you a true image of the speakers. i don't know this though as I've only use REW once, in my studio. I have a mic specifically for it and i read the endless links supplied by the DBX forum.

Just to be clear. If using the sine wave test 100hz, xover at 100hz and you get the cycle match correct. You will not get the issues that the video is pointing out. Your sub will be in the best possible alignment with the mids, that can be achieved. However, the room itself may destroy your response due to room modes, nulls etc at and around the xover frequency. 

But please continue. 

Edited by mitchiemasha - 18 October 2017 at 10:01pm
Back to Top
 Post Reply Post Reply Page  <12345>

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.06
Copyright ©2001-2023 Web Wiz Ltd.

This page was generated in 0.109 seconds.