steve rb

Trying to optimise my OBs and room with the XTZ mic...

Recommended Posts

Made the fatal mistake of getting out the XTZ room analyser yesterday. Four hours later :wall:

I'm also on a detox, which means no coffee, so I was just about ready to rip my own face off every time I walked past the mirror....

Anyway, started out with the room measurement :

6634346225_6a47c4dd04_z.jpg

Clear main mode at 72Hz and another at 37....

Then, the response from my OBs, just plonked where I left them :

6634345813_1d35579934_z.jpg

Total crap right? How the hell can a frequency response be this wild??? It's also pretty repeatable...

So anyway, started moving things about and was staggered at the major changes that small differences made :shock:

shifted them right into freespace and toed them halfway between straight and direct to listening position. Still a really spiky treble and the room mode at 72Hz is creating a big lump. Midrange still down.

6634345911_41be6028aa_z.jpg

I then left the speakers where they were and moved my 4 monster GIK bass traps around. With them both spanning the corners behind the OBs I got a big improvement in deep bass, presumably as the traps are now absorbing a lot more of that bass cancellation effect from the 18 inchers... blue trace before, green trace afterwards - perfect example of how a bass trap only affects the bass too! Everything else unaffected.

6634346013_23758c1365_z.jpg

I'd also had about 50 moves between this lot, toe in making a huge difference, small movements of a few inches too. Really brought it home just how incredibly sensitive these are to placement....

here's one extreme example where they were moved out 4" and toed in an extra 5 degrees....

6634346119_28fb2dc756_z.jpg

Finally, I'd got it pretty well as flat as I could, then moved the bloody things without taping their positions and I couldn't bloody get it back again :grrr: Very frustrating, but I'm going to have to put half a day into this once I've got the TP60 amps in to be able to tune it.

I ended up with them pretty close to the rear wall, toed in about 10deg towards the listening position, that gave the flattest mid and treble I could manage but with a bass boost. I've currently got acoustic panels on the back of the bass boxes which if I remove will drop some of that lift, still need to address that 72Hz mode though...

I'm also left with two huge suck-outs I just can't fathom at 4KHz and 7 KHz - any ideas why this is? some phase cancellation? if so why so heavily affected by positioning...?

6634345633_0e1a31e8fa_z.jpg

Share this post


Link to post
Share on other sites
Guest sq225917

Steve, maybe just maybe, if you want your music to sound anything like someone intended you might consider some other speakers. +/- a couple of db either way might not be too bad but you've got +/-10db peaks and troughs all over the place. My room is flat to within 1bd above 60hz with the right speakers it's easy.

Share this post


Link to post
Share on other sites

Or some form of active room correction.

Keith

Share this post


Link to post
Share on other sites

Hi Steve.

Did you do the measurements using only one mic position(seating position) or did you use the three mic positions option?

Share this post


Link to post
Share on other sites

I'm also interested in the answer to Juha's question about mic position. Given the extreme sensitivity to where the sources are I wonder if you've optimised the performance within a very small volume of listening space and if you move your head (mic) 20cm left or right you'll find the curve looks quite different. Interference effects change over distances which are comparable with the wavelength of the sound. Wavelength is sound speed (in metres per sec) divided by frequency (in Hz, which is cycles per second). The speed of sound is 340 m/s so at 4 kHz the wavelength of the sound is 8.5 cm and you would expect interference effects to vary over distances of a few cm. At 7.5 kHz the wavelength is nearly halved and interference will change over even shorter distances. If the interference was between the primary source (the driver) and some reflection off a nearby wall then you would expect it to be very sensitive to changes in the driver-wall separation. But those suckouts do look very sharp for a phenomenon (sound-scattering) which I'd always assumed (without thinking much about it, I admit) was more diffuse than specular. Also if it was interference between the driver and something external then I'd expect the frequency of the suckout to shift every time you moved the speaker. This would seem to suggest your suckouts are due to interference within the speaker itself. When I saw them I was reminded of a very simlilar effect in the horn-loaded Audax tweeters use in my WAD KLS8's. This was caused by the four screw holes used to mount the horn and it could be eliminated simply by filling them with silicone rubber. But if your suckouts are caused by the design of the speaker then they shouldn't be affected by moving it, which they are. Confusing. Could it be that there are two effects going on ? You've got a feature within the speaker, which is locking the suckout frequencies, plus an external interference which is causing the suckouts to change amplitude as you shift the speaker around ? Hmmm. Can of worms. I keep thinking I'll have to try measuring my far-from-optimum living room. But I'm afraid that when I do I'll discover exactly the complicated picture that you have ... :nerves:

Valvebloke

  • Upvote 1

Share this post


Link to post
Share on other sites

Would be better to isolate a speaker from the room to take measurements. This is how I do it: http://www.lonesaguaro.com/speakers/testing/Testing.htm (I do mine indoors and with the mic attached to a long pole to avoid reflections from the mic stand)

If you want any advice i'm more than happy to help...

Share this post


Link to post
Share on other sites
Steve, maybe just maybe, if you want your music to sound anything like someone intended you might consider some other speakers. +/- a couple of db either way might not be too bad but you've got +/-10db peaks and troughs all over the place. My room is flat to within 1bd above 60hz with the right speakers it's easy.

Hi Simon, couple of points - having had a shit load of speakers,

1. a flat FR is not a receipe for a perfect reproduction, it is merely a guidance on placement and treatment of major nodes. These OBs have some of the cleanest most detailed sound I have had in my room, and as you say, they are all over the place.

2. I can tell almost no difference between the positions above when playing music except for a generic 'bass is up' 'mid is down' or 'treble is harsh' - unless your're running a slow frequency sweep I suspect many people will not be able to hear anything other than the odd resonance or reediness to it.

3. I know many sound engineers who'd love their test chambers to be flat to 1db. A normal room with speakers in it? In fact any speakers themselves? flat to 1db? Really? what did you measure this with?

Share this post


Link to post
Share on other sites
Hi Steve.

Did you do the measurements using only one mic position(seating position) or did you use the three mic positions option?

Only one position. The room analsysis (ie the 'up to 250Hz' stuff) was done with 3 but I don't think there is an option for the full spectrum tests. OBs do beam like hell though..

Share this post


Link to post
Share on other sites
I'm also interested in the answer to Juha's question about mic position. Given the extreme sensitivity to where the sources are I wonder if you've optimised the performance within a very small volume of listening space and if you move your head (mic) 20cm left or right you'll find the curve looks quite different. Interference effects change over distances which are comparable with the wavelength of the sound. Wavelength is sound speed (in metres per sec) divided by frequency (in Hz, which is cycles per second). The speed of sound is 340 m/s so at 4 kHz the wavelength of the sound is 8.5 cm and you would expect interference effects to vary over distances of a few cm. At 7.5 kHz the wavelength is nearly halved and interference will change over even shorter distances. If the interference was between the primary source (the driver) and some reflection off a nearby wall then you would expect it to be very sensitive to changes in the driver-wall separation. But those suckouts do look very sharp for a phenomenon (sound-scattering) which I'd always assumed (without thinking much about it, I admit) was more diffuse than specular. Also if it was interference between the driver and something external then I'd expect the frequency of the suckout to shift every time you moved the speaker. This would seem to suggest your suckouts are due to interference within the speaker itself. When I saw them I was reminded of a very simlilar effect in the horn-loaded Audax tweeters use in my WAD KLS8's. This was caused by the four screw holes used to mount the horn and it could be eliminated simply by filling them with silicone rubber. But if your suckouts are caused by the design of the speaker then they shouldn't be affected by moving it, which they are. Confusing. Could it be that there are two effects going on ? You've got a feature within the speaker, which is locking the suckout frequencies, plus an external interference which is causing the suckouts to change amplitude as you shift the speaker around ? Hmmm. Can of worms. I keep thinking I'll have to try measuring my far-from-optimum living room. But I'm afraid that when I do I'll discover exactly the complicated picture that you have ... :nerves:

Valvebloke

Very valid points... Obviously these are home made, with vintage drivers and NOS c/o components, I have a nasty room with low modes and I am measuring far field (6m) with the pair running.

Perhaps I should measure freespace, near field, one speaker to get an idea of the speaker's FR before I jump straight into trying to assess listening positions. Also as you say, the dips and troughs are repeatable in one position but NOT when speakers are moved, so this is not necessarily a speaker issue but a room interaction with what are potentially a very fussy design.

Thanks for the input

Share this post


Link to post
Share on other sites
Made the fatal mistake of getting out the XTZ room analyser yesterday. Four hours later :wall:

I'm also on a detox, which means no coffee, so I was just about ready to rip my own face off every time I walked past the mirror....

Anyway, started out with the room measurement :

Clear main mode at 72Hz and another at 37....

Then, the response from my OBs, just plonked where I left them :

Total crap right? How the hell can a frequency response be this wild??? It's also pretty repeatable...

So anyway, started moving things about and was staggered at the major changes that small differences made :shock:

shifted them right into freespace and toed them halfway between straight and direct to listening position. Still a really spiky treble and the room mode at 72Hz is creating a big lump. Midrange still down.

I then left the speakers where they were and moved my 4 monster GIK bass traps around. With them both spanning the corners behind the OBs I got a big improvement in deep bass, presumably as the traps are now absorbing a lot more of that bass cancellation effect from the 18 inchers... blue trace before, green trace afterwards - perfect example of how a bass trap only affects the bass too! Everything else unaffected.

I'd also had about 50 moves between this lot, toe in making a huge difference, small movements of a few inches too. Really brought it home just how incredibly sensitive these are to placement....

here's one extreme example where they were moved out 4" and toed in an extra 5 degrees....

Finally, I'd got it pretty well as flat as I could, then moved the bloody things without taping their positions and I couldn't bloody get it back again :grrr: Very frustrating, but I'm going to have to put half a day into this once I've got the TP60 amps in to be able to tune it.

I ended up with them pretty close to the rear wall, toed in about 10deg towards the listening position, that gave the flattest mid and treble I could manage but with a bass boost. I've currently got acoustic panels on the back of the bass boxes which if I remove will drop some of that lift, still need to address that 72Hz mode though...

I'm also left with two huge suck-outs I just can't fathom at 4KHz and 7 KHz - any ideas why this is? some phase cancellation? if so why so heavily affected by positioning...?

6634345633_0e1a31e8fa_z.jpg

Hi Steve,

I have a few questions for you:

Why is the Anechoic button activated on image #2?

Do you have any change of measuring the speaker's response outside?

Are you measuring a single or both speakers at the same time?

Do you have a photo of your speakers and the cross-over frequencies?

Could you describe your room (dimensions)?

Cheers,

Ric

Share this post


Link to post
Share on other sites
Hi Steve,

I have a few questions for you:

Why is the Anechoic button activated on image #2?

Do you have any change of measuring the speaker's response outside?

Are you measuring a single or both speakers at the same time?

Do you have a photo of your speakers and the cross-over frequencies?

Could you describe your room (dimensions)?

Cheers,

Ric

Hi Ric

Why is the Anechoic button activated on image #2?

- because if I understand it right there are only the 3 options, anechoic, ambient and raw - raw is unusable, ambient is room related .... I think....

Do you have any change of measuring the speaker's response outside?

- I've only measured them in the listening room, no change of listening position and only moving the speakers about their general positions so far

Are you measuring a single or both speakers at the same time?

- both...

Do you have a photo of your speakers and the cross-over frequencies?

- All on the Lampizator OB thread here : http://www.hifiwigwam.com/showthread.php?49307-Lampizator-open-baffle-speakers

Piccy here :

6587783759_98c1de3747_z.jpg6587784513_ca2dedbf88_z.jpg

Crossover freq for the bass is 2nd order around 300Hz, mid bass filters off around 6Khz, again 2nd order.

Could you describe your room (dimensions)?

Big room... 31ft x 16ft with an appex glass lantern 2.8m to the edge, about 4m to the centre.

Share this post


Link to post
Share on other sites

For rough measuring of the loudspeaker's FR select Anechoic and if possible take the speaker outside (no boundary reflexes there) and place it on a stool (preferably at the lawn) to avoid floor bounce cancelation.

The microphone should be place at tweeter heigh and distanced 1m from it and try to play the test signal at least some 40/50dB above background noise.

Make a plot on axis, maybe a couple more 10º and 20º to one side and one between the tweeter and the mid-range drivers to roughly evaluate polar response (and estimate the way your speakers interact with the side walls).

For positioning I would choose Ambient (longer time window) instead and place the mic at ear height; it's also best to measure a single speaker at a time:

from the manual:

Ambient – Room information (300/8 ms, Ambi-Win)

Displays room effects – phase errors and reflexes are clearly visible.

Cheers,

Ric

  • Upvote 1

Share this post


Link to post
Share on other sites

P.S.: with such a big room you will be tempted to listen from an overly long distance and position your speakers too wide apart but this will have a tremendous(ly negative) impact on sound quality. I would try, even if it's just for experimental purposes, to listen and measure room response of a 2/2,5m sided equilateral triangle.

Share this post


Link to post
Share on other sites
For rough measuring of the loudspeaker's FR select Anechoic and if possible take the speaker outside (no boundary reflexes there) and place it on a stool (preferably at the lawn) to avoid floor bounce cancelation.

The microphone should be place at tweeter heigh and distanced 1m from it and try to play the test signal at least some 40/50dB above background noise.

Make a plot on axis, maybe a couple more 10º and 20º to one side and one between the tweeter and the mid-range drivers to roughly evaluate polar response (and estimate the way your speakers interact with the side walls).

For positioning I would choose Ambient (longer time window) instead and place the mic at ear height; it's also best to measure a single speaker at a time:

from the manual:

Ambient – Room information (300/8 ms, Ambi-Win)

Displays room effects – phase errors and reflexes are clearly visible.

Cheers,

Ric

Really appreciate this Ric, thanks. I might go short of moving the bloody things outside for now, although with a new TP60 amp here it'll be a lot easier than lugging all the pre-power gear around...

I could shift one into the middle of the room and try there, will give that a bash over the weekend (or today depending on how bored I get...). I was measuring at the listening position, which is roughly 18ft back from the front wall, just past half way to avoid sitting in a node.

Just a thought, isn't my objective (and I know there is a process here) to get the speaker/room FR as flat as poss from my listening position, that's where I'll be listening from. presumably doing it this way tells me how screwed up my actual speaker FR is and the rest is then down to positioning and treatment...?

P.S.: with such a big room you will be tempted to listen from an overly long distance and position your speakers too wide apart but this will have a tremendous(ly negative) impact on sound quality. I would try, even if it's just for experimental purposes, to listen and measure room response of a 2/2,5m sided equilateral triangle.

Yes... think you could be right there too, but for sure I don't listen the length of the room, I'd found out with several speakers (the Zingali and Avantgardes) that really wide placement was way better, the Avantgardes were almost touching the side wall - this was way better than closer placement, but there could be a danger of adopting that for all. Problem is, to move any further forward and I'm sitting in the middle of the room length, bad for bass cancellation, any further than that and I'm like the guy from the memorex advert stuck out in free space, that just won't wash. I'll try them closer together and perhaps with less toe, although these buggers really beam...

Share this post


Link to post
Share on other sites

My room is very 'difficult', I have tied all sorts of passive treatment but I couldnt get rid of massive boost at 40Hz (20dB on the right speaker and 15 db on the left one) and more at 110 Hz. Also there were suck outs at 63Hz and some more at 170Hz. The midrange was also relatively sucked out by few db. No reasonable passive room treatment could correct this. I now use a digital room correction (Trinnov) and the improvement is massive, I can get the bass so much more controlled, the 40Hz doesnt dominate the sound anymore and I can hear instruments that I never knew where in the recording! The Trinnov seems to be one of the best DRCs and used correctly it doesnt introduce any significant unwanted digital artifacts. It does take time to learn how to adjust it properly but once it is done, for me there is no way back.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.