Some help needed understanding MultiEQ measurements

[ABall]

Borrowed a GIK bass trap, i think its this one. 1200x600 series. GIK
definitely a better response... Ignore the file name for the second run, i forgot to change it. I thought I would try my little IPL's yesterday, no trap, the cushions made little to no difference to these speakers so I doubt they would be much difference with traps. I dont have time to move them around tonight, they do start off a lot lower but they are less efficient, a lot cheaper than the Kurts if you buy a kit, sorry bought, he's sadly retired now. Loving my UMIK and REW and I know i havent scratched the surface.

Actually looking a bit closer its only reduced the dips in a couple of places and its either worse or the same overall, or thats how i see it.

 

[-Matt-]

Let me just check that I understand which curve is which...

The red curve stays the same between the two images - is this your Kurt speakers without any panels?

In the first image the blue line is close to the red one in the bass region - is this the Kurt speaker again with a bass trap?

On the second image the blue line is much higher in the low frequencies - is this the IPL speaker with no panels?

 

[-Matt-]

Also, are you measuring the left or right channel (right had the biggest dip so best to start here)?

...and are you measuring in pure direct or with eq?

It looks like the main improvement with the panel is between 70Hz and 120Hz (this was part of the region that we were targeting where there was previously a deep dip). You probably shouldn't expect much difference below this frequency because the absorption of the panel drops right off.

I think I'd take all the differences above 500Hz with a pinch of salt, these are probably very dependant on the smoothing applied and on slightly different position of the mic.

It would have been nice to see a bigger change on the deep dip at 125ish Hz and above. However, I think the diffusion plate on the panel you linked means that it has low absorption for 120Hz up. You might do a bit better with one of the broadband absorber type.


Edit: I'm refering to this absorption plot (from the link you gave above).

 

[ABall]

Sorry I should of renamed one of the results. The red curve in both images is the Kurt's with the bass trap, i pushed the speaker tight as i could against the trap. The blue curve in the first image is the Kurt without bass trap, I should of renamed that curve. The blue curve in the second image is the IPL without bass trap.

 

[-Matt-]

Hmm, in that case then it did actually help with the sharp dip at 125ish Hz but gave a broader one centred at 100Hz. Not sure if I understand that.

Is the speaker position the same or did you have to move it out to fit the panel behind?

 

[ABall]

Also, are you measuring the left or right channel (right had the biggest dip so best to start here)?

...and are you measuring in pure direct or with eq?

It looks like the main improvement with the panel is between 70Hz and 120Hz (this was part of the region that we were targeting where there was previously a deep dip). You probably shouldn't expect much difference below this frequency because the absorption of the panel drops right off.

I think I'd take all the differences above 500Hz with a pinch of salt, these are probably very dependant on the smoothing applied and on slightly different position of the mic.

It would have been nice to see a bigger change on the deep dip at 125ish Hz and above. However, I think the diffusion plate on the panel you linked means that it has low absorption for 120Hz up. You might do a bit better with one of the broadband absorber type.

Yeah I was thinking of sending the results to GIK, I thought that the scatter plate wont be needed at such a close distance, it was just available to test, I didn't check the right speaker, I was very limited on time, I thought it would give me an idea of how useful the traps are, the right speaker has a corner that the left doesnt have.
I was surprised to learn that GIK didnt ask my employer how the speakers where placed in our customers HT room, we mounted them in wall, when you put me onto SBIR I read that in wall speakers dont suffer from those problems. What I have discovered with REW is that closing the lounge door that normally opens onto the speaker, is very bad! That was an eye opener.

 

[ABall]

Hmm, in that case then it did actually help with the sharp dip at 125ish Hz but gave a broader one centred at 100Hz. Not sure if I understand that.

Is the speaker position the same or did you have to move it out to fit the panel behind?

I had to pull it forward 6" to fit the panel behind, I guess the wall to speaker distance is pretty crucial

 

[-Matt-]

Yes, definitely reach out to them for some advice. Let us know what they suggest. I'm trying to help but am certainly not an expert on this.

 

[-Matt-]

I had to pull it forward 6" to fit the panel behind, I guess the wall to speaker distance is pretty crucial

Ah, yes that could definitely shift the centre of the dip to lower frequencies.

 

[ABall]

Yes, definitely reach out to them for some advice. Let us know what they suggest. I'm trying to help but am certainly not an expert on this.

You have been very helpful Matt so thanks, you have a much better handle on this than I do,

 

[-Matt-]

The nice test to do, if you have time and inclination, would be...

Meas 1: Right speaker, pure direct, as close as possible to the wall with the panel behind.

Meas 2: Remove the panel without moving the speaker, measure again right channel, pure direct.

This should hopefully isolate the effect of the bass trap. (However, it looks like that model of panel has quite a narrow band of absorption - so may not be ideal).

Edit: Suggesting the right channel as it had the deepest dip in the first set of measurements. Also ensure that the lounge door is in the same position for both measurements!

 

[ABall]

The nice test to do, if you have time and inclination, would be...

Meas 1: Right speaker, pure direct, as close as possible to the wall with the panel behind.

Meas 2: Remove the panel without moving the speaker, measure again right channel, pure direct.

This should hopefully isolate the effect of the bass trap. (However, it looks like that model of panel has quite a narrow band of absorption - so may not be ideal).

Edit: Suggesting the right channel as it had the deepest dip in the first set of measurements. Also ensure that the lounge door is in the same position for both measurements!

The lounge door is on the left speaker but I will keep it in the same position for sure. As you say I dont think these are the best panels to test. I'm not sure when we have to install them so I dont know if I will get a chance but I need to do something so I am going to have a chat with the wife. She didnt see me bringing the massive box in the house and there was a right mess of polystyrene to clean up. Lol. I've looked at the contact form for GIK and that's going to take a little time to do properly, life is getting in the way for a week or so but I will do some more if I get a chance, I think we have some 2.5" panels coming in for the same job, no plates, I might get a go on them.

 

[-Matt-]

Good luck on broaching the topic with your wife. Unfortunately for me, a rational conversation on the topic no longer seems to be possible!

So is your day job custom HT installs? Sounds fun!

 

[ABall]

Good luck on broaching the topic with your wife. Unfortunately for me, a rational conversation on the topic no longer seems to be possible!

Lol, I may end up in the same boat Matt, my current wife has put up with my shit for 20 years, hifi could very well be half the reason I am on number 3!

So is your day job custom HT installs? Sounds fun!

No not really, started off 16 years ago doing aerial and satellite installs, we moved with demand to CCTV, WiFi/Networking, multiroom audio etc but we have always installed TV's and projectors. Our latest install has 30k of hardware, I think the PJ is about 11k! its not our typical job though, not by any stretch. Hanging big TV's on walls and making the cables disappear without needing to decorate is our speciality, or a big TV fitted to a reverse lift dropping out of the bedroom ceiling, that was fun..... NOT! lol

 

[ABall]

It does mean I get to buy stuff at trade so I can tell anyone now, the current sales of the 3700 at 979 is a bloody good deal, I can tell you there is not a lot of money been made by Richer sounds ETC at that price. Its a bargain, Ive been bidding on second hand 3600's and they are going for £750 on ebay all day.

 

[Jon AA]

I have to totally disagree with your opinion on the tape measure though, tape measure, string, whatever, its more accurate than the software, I already proved this using Amused To Death.

Well, the impulses either line up or they don't. If they don't, whatever you did to calibrate wasn't good enough and you need some fine tuning. And yes, actually measuring in REW is going to be more accurate than your ears for verification. Because in the end, the only thing that matters is the time arrival of the impulse at the listening position. That's the final answer you're looking for.

Beyond the impracticalities of physically measuring across a room with a tape measure with the required precision (I prefer laser tape measure myself, but alas, they can be finicky with black objects) , speakers of different sizes with different depth cones/horns, etc, that method has a huge problem with most high end--and these days, many mid to even budget systems.

Most for a long time now, have been aware their subs use separate amps and likely separate DSP which adds delay--such that the physical measurement is pretty meaningless. Adjusting the delay for subs is somewhat expected for that and other reasons. But many systems today use separate amps (many with DSP) for some of their speakers (LCR only is quite common). And these days lots of people are mixing in active speakers into their systems--even budget systems. Those speakers are using their own amps and DSP.

These will all add unknown amounts of delay which cannot be measured with a tape measure. The only way to time align these speakers precisely with the rest of the system is to skip to the end, the only thing that actually matters and measure the time arrival at the listening position acoustically.

Since that's the only thing that matters in the end, measuring it directly and fine tuning that result is the most accurate way to do it because it eliminates all sources of potential error in measuring something else and trying to extrapolate to that end result.

 

[ABall]

Well, the impulses either line up or they don't. If they don't, whatever you did to calibrate wasn't good enough and you need some fine tuning. And yes, actually measuring in REW is going to be more accurate than your ears for verification. Because in the end, the only thing that matters is the time arrival of the impulse at the listening position. That's the final answer you're looking for.

Hmm, I still disagree, partially, yes about the impulses. I don't know if the impulses lined up or not, What I did to calibrate was to let the Denon do its job, I kept a very strict 20 inches from the first position, (I made sure it was pretty much exactly where my head is) my results using my ears indicated it wasn't quite right. I lost all the Q Sound effects after calibration, Q Sound manipulates timing, amplitude, and frequency response to produce a binaural image, it relies on the speakers been at an equal distance left to right from the listening position, if that's off even by a small amount its just music. To be honest I never measured the distances using REW, I don't know how to do that yet, I just corrected the distance for one channel (according to my tape measure) and it came back, so I say again Multu EQ calculated the distances slightly wrong but enough to screw things up. Lets call that the "fine tuning" you suggest. Its nothing to do with my ears either, I purchased Amused to Death in 1992, so for 30 years I have been using it to setup my 2 channel system and I have been playing it to everyone who visits my house and is interested in HiFi, they have all thought my rear speakers where playing the effects as it comes out at 90 degrees.

Most for a long time now, have been aware their subs use separate amps and likely separate DSP which adds delay--such that the physical measurement is pretty meaningless. Adjusting the delay for subs is somewhat expected for that and other reasons. But many systems today use separate amps (many with DSP) for some of their speakers (LCR only is quite common). And these days lots of people are mixing in active speakers into their systems--even budget systems. Those speakers are using their own amps and DSP.

These will all add unknown amounts of delay which cannot be measured with a tape measure. The only way to time align these speakers precisely with the rest of the system is to skip to the end, the only thing that actually matters and measure the time arrival at the listening position acoustically.

Since that's the only thing that matters in the end, measuring it directly and fine tuning that result is the most accurate way to do it because it eliminates all sources of potential error in measuring something else and trying to extrapolate to that end result.

I don't have a sub right now, I have played with stereo 15" Hawthorn Augies but I would never, have never, used a tape to set them up, its indirect sound after all!

 

[Jon AA]

It's pretty easy to do in REW, here are a couple tutorials: https://www.minidsp.com/applications/rew/speaker-time-alignment https://www.google.com/url?sa=t&rct...pdf.2033257/&usg=AOvVaw30HQKdaeYdJbqYTFRWCeWp . You can find plenty more if you look around.

I just looked back at some of my older calibrations: It looks like having MiniDSP HDs on my LCR added around 4 ms of delay (equivalent to around 4 feet of distance) to those three speakers relative to all the rest. So yeah, with a lot of systems, measuring the physical distance just isn't an option.

 

[ABall]

It's pretty easy to do in REW, here are a couple tutorials: https://www.minidsp.com/applications/rew/speaker-time-alignment https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiQ8JXv6dD5AhUfHTQIHUksBjsQFnoECDoQAQ&url=https://www.avsforum.com/attachments/time-aligning-pdf.2033257/&usg=AOvVaw30HQKdaeYdJbqYTFRWCeWp . You can find plenty more if you look around.

I just looked back at some of my older calibrations: It looks like having MiniDSP HDs on my LCR added around 4 ms of delay (equivalent to around 4 feet of distance) to those three speakers relative to all the rest. So yeah, with a lot of systems, measuring the physical distance just isn't an option.

Thanks Jon, I will definitely get around to watching more tutorials.

 

[ABall]

Yes, definitely reach out to them for some advice. Let us know what they suggest. I'm trying to help but am certainly not an expert on this.

So I've sent GIK 4 emails with responses from them, they asked for the REW files which I sent and I have to be honest Im not impressed, I wish there were other suppliers of bass traps in the UK. I told them how I arrived at all the measurements, I asked if a better trap could be used because the impression series, while fixing the dip at 130hz created a wider one at around a hundred. This was over 4 emails, the second last reply I got was, "I guess your looking to fix the dip at 60Hz" at that point I've concluded he did not read a single email or looked at all the plots. I still replied again and said I may try to fix the 60hz dip with a subwoofer and was more interest in the one at 120hz, his last email said that's not likely to be SBIR! He then explains how I can move the mike and if the dip disappears then it is, didnt I prove that by taking measurements with the speakers pushed back against the wall! Totally frustrated and pee'd off as it took days to get to that point. I think Maybe they were a bit busy or the janitor has hacked the manager of customer service and sales email account so he can wind people up......