Weird Phase Cancellation - any ideas on fixes?

[Text_Edifice]

Hi All,

in the process of setting up a room at home to mix / edit on - nothing fancy just want to be able to work from home on projects more regularly and do so with a level of confidence that the decisions I make won't have to be re-made when I'm back in the studio proper.

Anyway, have been setting up speakers / etc. in the (rectangular) room with the listening position set at about 38% of the room's length and the (nearfield) monitors in a tight (800mm) equilateral triangle with the back of the speakers about half a metre (560mm) from the rear wall. (room dimensions are 242mm wide and 410mm deep - haven't measured floor to ceiling accurately but it's approx 200mm). Walls are very thin (post wwIII queeenslander so the bass tends to escape rather than build up - good for me, bad for the neighbours). Floors are polished wood so I've put a carpet under the speakers / listening position. The room is not 'empty' and contains some guitars / amps and a cupboard as well as a spare bed - a REAL bedroom studio...

Anyway have been running through some test discs / reference material to get an idea of what's going on in the room..

First thing I noticed was that the high's were doing something funny so put up some acoustic foam panneling to absorb the most prominent early reflections from the side wall. This helped.

Again, ran through the tests etc. this time I noticed something funny with the levels between the left and right speakers requiring some extra juice from the amp for the left channel to get the image to sit 'right'. Thought this was a little funny so I put up a 1k sine wave generated from within tools and noticed I was hearing some obvious phase cancellation at the listening position concentrated around the mid-range.

2 questions

1) what am I doing wrong? - have done the usual google searches and set up the room according to the advice I found.

2) how do I fix it? - any ideas or pointers in the right direction would be great. Please note this is a DIY job - I want to learn how this stuff works for myself in an environment where I can afford to stuff around a little, rather than hiring in someone else to do the job for me.

EDIT: should add I've been reading everything I can find on the acoustics of control room design. I have been basing my placement etc on Ethan Winer's articles put up on his site. http://www.realtraps.com/articles.htm (worth a read if anyone's interested).

Cheers.

[smash]

Try some acoustic foam behind the speakers/mix posisition on the rear wall.

My guess the wierd midrange you mentioned is phase cancelation between the monitors and the wall directly behind them.

Also try hanging the guitars, and whatever else you can on the opposite wall. This should act as a diffusor, and rather than having a slap echo which depending on the size of the room will be percieved as phase cancellation rather than a distinct slap, will give you a short, but randomised, reverb tail.

Smash.

[Text_Edifice]

Cheers - will try this tomorrow.

Just to clarify, do you mean hang the foam on the wall in front of me when I sit in the listening position (front wall - behind the speakers) or the wall behind me (rear wall).

Have tried moving the listening position and speakers a bit and this helped improve things somewhat (moved everything 2 inches left and 2 inches forward and the problem wasn't as evident.

Wondering if I've inadvertently set up my listening position in a null point or similar?

[Henry]

You'll need a diffuser on the ceiling too just forward of speakers.

[rob]

I'm confused, you said "I put up a 1k sine wave generated from within tools and noticed I was hearing some obvious phase cancellation at the listening position concentrated around the mid-range."

How can you determine some phase cancellation around the "midrange" by putting up a 1K tone?
One thing you never want to do to is put up single higher frequencies in a room to get a tonal picture. Unless your room is an anechoic chamber you'll get a whole spread of peaks and dips as the tones reflect off surfaces and phase add and cancel. If I put up a 1K tone just by wobbling my head around i can hear the peaks and dips as my head moves through these points of addition and cancellation.

As to hanging guitars up as diffusors. I hope these arent' accoustic guitars you'd be using? I've got an old accoustic guitar propped up in the corner of my bedroom and i can lie in bed and get it to "sing" just by humming the right tones. Last thing i'd be putting in a control room would be such a strongly resonating thing as an acoustic instrument.

Rob

[rick]

i dont actualy think this is a room problem

are your speakers ok...?
are you sure ...?
try flipping the phase of both speakers and if you still have wierd phase things
if you still do perhaps its time to have a play inside the speaker
it would it not be the first time i have discovered one of tweeters was out of phase with the other one internally in the speaker ( from the factory)
sometimes you would not notice this untill you start getting your room in order
test each speaker as a mono source do they actually sound the same ( i am betting they dont- most speakers dont when you being hyper critical)
the balance thing on the amp makes me more sus on the speakers or the electronic chain then the room.

btw if you hang your guitars on the wall ( well supported at the bottom )
in a few years time all that singing they are doing will make them MUCH better guitars

a very famous us guitar builder ( ren fergerson- gibson acuostic custom shop) told me once that guitars still think they are trees for many years untill you teach them otherwise
and insisted that i always leave his acoustic guitar OUT of its case !
and that the only time they should be in the case is for transit

so i left my already good but not great gibson j-200 hanging around my studio singing away to itself and obviously out of its case
and it two years it turned into the best acoustic guitar i know of
all by itself !
its 25 years old now and for the first 15 years it lived in its case and nothing "changed" a couple of years singing away really made a world of difference

btw the guitars sit in my air locks off the main room not actually in the mastering room

[Text_Edifice]

Thanks Henry - have been putting off diffusers on the ceiling as I'm not sure how I'm going to hang them yet...

Hi Rick,

Thanks for your comments

Rob - sorry should have clarified a bit further.

Statement should have read I put up a 1k test tone and noticed phase cancellation. Then played a range of other tones and noticed the phasing was most prominent around 1-2kHz at the listening position.

As you describe, when I move my head around the room I can hear peaks and dips but it's especially bad at the listening position so I'm thinking maybe I need to move the listening position or tweak the speaker placement a little more.

Also to clarify when I say the left side of the amp needs more juice it's kind of 2 clicks more than the right side - I am thinking this could be a hardware problem but I'm thinking it could be the amp as I bought it ex-demo.

It's possible that the two observations - uneven l/r balance and phasing - are unrelated (amp / room) problems. My train of thought went:

1) sounds like the left speaker needs more juice to centre the image

2) I'll put up a click in tools as an easy way to centre the image (misguided perhaps)

3) ok, the click's sitting allright, what happens if I put up a test tone

4) I can hear phasing - wonder if that's why the amp has to work harder?

5) There are obvious holes in my thinking :-) I'll ask for help.

[Jeremy H]

I had a similar thing in my room - (after changing things round.)

I (and as Rick suggested above) eventually chased the problem to the speakers themselves. One tweeter is a little louder than the other.

On further investigation, I established that the magnet in the not so loud tweeter was much weaker than the magnet in the other tweeter.

Now saving for a new pair of tweeters!

[Text_Edifice]

Messed around with speaker placement for a couple of hours today and the problem has improved. I think the phasing has something to do with centering the listening position between the two (parallel) side walls - I moved the whole setup a little to the left and the phasing wasn't as bad. Slightly widening the distance between the nearfields also helped - maybe something to do with having my head just inside the equilateral triangle (between speakers and listening position) rather than at the tip??

Anyway, next step is to trouble shoot the speakers - I'm going to try a simple swap of the left and right speakers and see if the amp settings remain similar - if so I'm thinking amp / room. If not I'll send an email to the speakers manufacturer.

Rick - I've got a mono version of Pet Sounds on disc, what should I be listening for in terms of an out of phase / dodgy tweeter? [Edit: sorry stupid question, misunderstood what you were saying] The speakers are bi-wirable so if the tweeter is wired out of phase internally would it be possible to bi-wire the speakers from the one amp and just flip the connections to the tweeter?

Cheers.

[Chris H]

Speaking of speakers when i was auditioning the B&W's the shop assistant put up a pair of 602's and immediatley i could hear they were out of phase. He insisted tthere was no problem. I cheched his wiring and he had wired them ok so i asked him to put up another pair of 602's on the same wires and the 2nd pair were ok. The 1st pair were wired internaly out of phase.
So even B&W can get it wrong on the quality controll. At a studio i worked in after a speaker repair i could hear a phase problem but not like the speakers were out of phase. It turned out to be the tweeter out of phase.

[Text_Edifice]

Ok,

swapped the left and right speakers today, ran through some now familiar listening material / tests and...

same problem, left side of the amp still needs a little more juice to centre the image (blindfolded I arrived at exactly the same settings as before) and still getting some phasing at the listening position, most prominently between 1-2 kHz.

Good news: probably not a faulty speaker (unless I've missed something crucial). 8)

Bad news: could be the amp :? or cabling??? (though not between amp and speakers as I just rewired the speakers at the amp) or room? :x

Will try swapping cabling next - any ideas on the best way to test the amp?

This is a finnicky problem because on the whole I think the speakers / placement are giving me a good accurate representation of what's going on in my mixes (based on old mixes I've pulled up on the new speakers).

Any further ideas?

[chris p]

Light a candle, put it at your listening position, turn off the lights and put Quadrophenia on your turntable.

It won't fix your problem but it might tell you your future.

[JulienG]

Any further ideas?

Get your ears checked.

You may well find (like I recently did) that one eardrum is damaged and the way it manifests is a slight level difference that you can only hear when listening critically.

[Text_Edifice]

Was thinking it could be my ears - but got a couple of friends to repeat the tests I'd been running and they got the same results.

Even had a non-tech friend come over to have a listen and as a double blind test I set the amp levels flat and gave him a rave on how good the speakers where sounding (removing any bias that I thought there was something amiss etc.) - he put on a favourite disc and the first thing he said (about 30 seconds in) was "wow the bass is further to the right than I remember it". I returned the amp levels to where I'd had them previously and he confirmed it was as he remembered.

:evil: :evil: :evil:

[Chris H]

If you are going balanced out of your desk into the amp there might be something amiss with the cable or the electronic balancing circut in the console,,?

[Text_Edifice]

yep balanced out of interface (home setup) into amp - will try swapping the cables around today and see if that makes any difference.

[PeterR]

another idea -

Set a multimeter to AC volts, play a test tone and check the voltage on your audio interface outputs. 1 probe on sleeve the other probe on tip of a TRS cable. Each output should obviously have the same or similair AC voltage. +4dBu is about 1.2v

this will possibly show whether you have a level mismatch with the outputs on your audio interface.

I have the same problem - my amp needs 3 more clicks on the right to balance the stereo image at low volume. As I turn up my audio interface the difference is less noticable to the point where 1 click is enough. I suspect the outputs of my mbox is the culprit but it could also be the amp. I guess I should just do my test that I mentioned and find out.

I dont get phasing problems though.

[rob]

gee this is getting really confusing.
i'm confused, i imagine you are too.

you need to break down your testing to simple steps where only one thing at a time changes when you make your comparisons
use one speaker, one channel of the amp, one output from your soundcard or whatever is your normal source, one speaker cable, one audio cable.
Exchange things one at a time looking for what it is that causes a difference.
What sort of amp are you using, does it have separate level controls per channel. If so, don't trust these. If it does have level controls, can you run them flat out and control listening level from the source?

for example to compare the speakers : you need to everything else to be identical, so you need to prove your wiring, source, amp first.

Rob

[toddd]

Light a candle, put it at your listening position, turn off the lights and put Quadrophenia on your turntable.

It won't fix your problem but it might tell you your future.

I thought it was 'tommy'!

[Text_Edifice]

Hi Rob,

yeah agreed - need a better approach to problem solving this sucker.

Will run some more tests with cabling etc. and report with some (hopefully) clearer data.

:?

[rick]

what about this for a thought
take your speakers and amp out of the room into another room and see what happens, it will only take an hour which is a lot quicker then we are going to take guessing away like we are playing charades

still the same wierdness in another room?, them its your system

btw when listening to just one tone ( like 1k) every room will bounch and bleet phase wise in different positions, listening to white or pink noise will be worse, so try just listening to the kind of music you dig , thats what you were trying to do before you got sidetracked being a sound scientest right..? :)

[Text_Edifice]

Curioser and curioser...

moved gear, discovered three things.

1) phasing is not present in the same way in a different room.
2) imbalance between l / r speakers is present in different room - regardless of arrangement of speakers (swapped l / r speakers while all else remained static).
3) if both amp channels are turned up all the way and I control the level from my interface the image sits fine...

To here's recap what I have tested so far:

Maintaining source material, speaker location and gear settings.

1) swapped left and right speakers around - result: unchanged
2) swapped leads from speakers to amp - result: unchanged
3) swapped leads from interface to amp - result unchanged

Maintaining wiring and gear settings.

1) moved speakers and listening position 2.5 inches left of centre- result: slight improvement in phasing around 1-2k, amp level problem remains

2) widened the speakers by 5 inches - result: improvement in phasing around 1-2k, amp level problem remains. White and pink noise sound ok too on cursory listening.

3) Moving whole setup into another room - phasing around 1-2k not present (I'm sure the room would have had other problems but didn't test for them), amp level problem remains

Maintaining speaker placement and wiring.

1) turned amp up full, used level control on interface - result: image sits front and centre without the need adjust the amp.

Question: am I going crazy? Is the amp legending wrong? Am I just experiencing some sort of placebo effect brought on by too many hours trying to track down this problem?

Thoughts?

[Chris H]

Time to try another amp. Sorry if i missed what type of amp you have but if you haven't got a good studio amp then a good option is to get a cheap 70's hi fi amp, like a sansui au555. they were and still are regarded as a good guality unit. The bigest thing to aim for is an amp that has the option of bypassing the preamp stage and conecting directly to the power amp stage. the signal is noticably better and there is no attenuator so you need to send the level from your interface.... just a thought

[JulienG]

Sounds almost like the input pots on the amp are not matched at all well. I'd look at running it full pelt and getting a good dual-gang pot or even just a pair of well-matched resistors to get the level right.

[Text_Edifice]

Luckily the amp (PEAK Audio studio amp) is under warranty. So will contact the store I bought it from and see what happens.

Should add, the problem isn't the 'sound' of the unit. It works nicely with the speakers I have and was a good price. Would be happy to keep using the unit I have at full pelt but something tells me it's better to try for a replacement while the unit is still under warranty.