Measuring speakers in-room. Practical considerations.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Just a quick update on the room placement suggestions people made.

An externally hosted image should be here but it was not working when we last tested it.


I ran some sweeps to check the frequency response. When located in the corners as seen above, I get significant gain from 40Hz to 80Hz but a major dip from 100Hz to 200Hz. This is the bass heavy sound I've been hearing. When watching movies, you get tons of thud and hip-hop bass...but it seems to come out of nowhere.

The rest of the response was very good except a sharp hump at 7K followed by a gradual roll off. Removing the tow-in on the speakers so you listen slightly off-axis removes the hump, but you sacrifice top end extension with the roll off occurring at 10k instead of 15k. There is definitely a sweet spot where you get just the right amount of treble without it jumping out.

Moving the speakers directly in front of my bookshelves and about 2 ft from the wall creates a much smoother curve in the bass dept. The low end gain is reduced but I get back the 100-200Hz region.

I also experimented with placing them on either side of the TV but it ruins the imaging. The center vocals become about as wide as the TV, rather than focused at the center.

So basically, the optimum spot is a no-go. I think I will likely leave them in the corners for daily use and bring them in front of the shelves for serious listening.

I also experimented with a 2.5w MiniWatt tube amp, a 10w chip amp, and my 50w receiver. All three produced relatively equal output on the RTA. I detected a bit of graininess from the receiver at loud volumes that wasn't there with the 10w chip amp, but it did a better job with the bass. The tube amp sounded great at low levels and had the lowest noise floor, but couldn't play loud enough. I wish there was a quick way for me to switch between them. I had to keep getting up and switching cables around so it was tough to really dig into what I was hearing.
 
Interesting observations. Did you try any listening positions far out into the room, maybe even along the room's center-line? Was the 7K hump related to room position or was it an on-axis / off-axis problem? Was that 7K hump an issue listening to either instrumental music or vocals?
 
The hump is consistent with the reported frequency response. I get the appropriate on-axis rise, but it does not extend to 20Hz like the graph below when measured 1-2 meters from the speaker.

An externally hosted image should be here but it was not working when we last tested it.


When listening on-axis it adds a slightly abrasive sound to trumpets and puts them sort of in your face. Switching to off-axis ~10 degrees removes the abrasiveness but sacrifices cymbals which don't cut through like they should and are slightly muted. Of the two options, off axis is definitely more enjoyable and you likely won't miss much...but it makes me curious to see what a super tweeter would do to the sound.
 
I would say this agrees with my listenings of the 10.2. I have some 7.3's on the way, hoping to get, more soundstage depth (width on 10.2 was fine), and more extended top end while maitaining tonal balance (less directivity). Runnning the 10.2's on tube amps seems to improve the dpeth issue, but this makes sense in keeping with tubes ability to make individual instruments bloom/standout somewhat in their place in the soundstage.
 
I will take some more measurements and post my findings...if it helps. I use a Behringer ECM8000 with TrueRTA.

FWIW, I did enjoy the sound when powered off my 2.5w tube amp. I just wanted more volume out of it. I couldn't tell you if the increased clarity was due to it's construction or simply lower volume. I didn't try it until late in my session and my ears were already tuned into a louder volume. On my next session, I will level match using the max volume of the MiniWatt tube amp so i do a better job analyzing the differences. However, I probably won't make the MiniWatt my primary amp simply because it won't be loud enough for watching movies.
 
I will take some more measurements and post my findings...if it helps. I use a Behringer ECM8000 with TrueRTA.

Hi Chuyler, Guys,
Please qualify your posts by letting guys know what measuring equipment you're using when making measurements. Post No. 141 should have included more information on how the measurements were taken and the equipments used. Many members don't know the limitations of low spec measuring equipment and issues on non-anechoic conditions. They become confused between the anechoic measured output of drivers and the effects of in-room measurement limitations, box design and room environmental factors. I get bogged down with emails following up from posts like No.141 (Chuyler) asking me why the driver performance and in-room measurements differ. Please help by making posts more comprehensive and qualified, as this benefits everyone.

Using a US$50 ECM8000 mic has limitations measuring the complete bandwidth of Full-Range drivers. Behringer's info sheet for this mic excludes the "actual" sensitivity of the device. Its useful range will be in in the regions of 50Hz to 15-kHz.

For more reliable measurement, I recommend Earthworks mics. Their M30 is a unit suitable for measuring reliably in the 20Hz to 20-kHz range. This Mic retails around US$650. We use Earthworks M50 as our primary and Linear's M53 as secondary, both these mics will give us reliable data readouts to +30-kHz anechoic. I can't speak for their performance in non-anechoic (room) environments but expect they would be better than cheaper mics.

In-room measurements can be useful to help you adjust your system set-up but they won't give you accurate data on your speakers themselves. For that, you'll need to isolate your speakers by getting them into an anechoic isolated environment.

Cheers
Mark.
 
Last edited:
frugal-phile™
Joined 2001
Paid Member
Using a US$50 ECM8000 mic has limitations measuring the complete bandwidth of Full-Range drivers. Behringer's info sheet for this mic excludes the "actual" sensitivity of the device. Its useful range will be in in the regions of 50Hz to 15-kHz.

That range is good for a calibrated ECM8000. From the graphs of my 4 calibrated ones i would say that the safe zone for an unclaibrated one is more like 100-7k

dave
 
I cannot agree more strongly with Mark regarding recommended testing equipment. A serious hobbyist needs the benefit of good test equipment as much as a professional designer. As we strive for the best sound we can get, regardless of whether we're professional designers or amateurs, we should use test equipment that is at least as good as the results we expect. Using lesser equipment just makes the job of tuning that much longer and can lead to non-repeatable test results (read: "frustration").

But this is probably best addressed in a different thread, eh?
 
That range is good for a calibrated ECM8000. From the graphs of my 4 calibrated ones i would say that the safe zone for an unclaibrated one is more like 100-7k
dave

Correct, we don't know if Chuyler has calibrated his equipment, or the reliability of any calibration. My worry is Chuyler (and many other guys) assume the ECM8000 is capable of measuring reliably beyond 15-kHz.

I'm not trying to criticise Chuyler's efforts. He's making a very nice project build and commend his efforts. I'm letting guys know that its important for those who embark on measuring their systems in non-isolated environments with low spec equipment to understand the data generated will have limitations that could be mis-read.

Evan us pro's can slip up. Evan and me came close to radically changing the design of the Gen. 1 Alp10 because we kept getting weird test data in the upper mid-range from the sweep tests in the isolation chamber. Turned out to interference from the side walls of a thick baffle plate rather than the driver. Knowing what factors affect measurements is a major task in its own right. We spend many hours setting up for sweep test to isolate as many variables as possible.

Thanks
Mark.
 
I cannot agree more strongly with Mark regarding recommended testing equipment. A serious hobbyist needs the benefit of good test equipment as much as a professional designer. As we strive for the best sound we can get, regardless of whether we're professional designers or amateurs, we should use test equipment that is at least as good as the results we expect. Using lesser equipment just makes the job of tuning that much longer and can lead to non-repeatable test results (read: "frustration").

But this is probably best addressed in a different thread, eh?

Hi Silver,
Many thanks for this understanding. Agreed, its a frustration for hobbyist's wanting to experiment more by having to pay US$hundreds/thousands on a decent mic and associated gear. I don't want to put off guys from experimenting and testing, just flagging up the risks of over-reliance of data generated from low cost hobby measuring gear.

Agreed, we gone off topic (apologies for me contributing). I'll gladly set up a new thread and migrate the mentioned posts if guys would like this.

Thanks
Mark.
 
I would say this agrees with my listenings of the 10.2. I have some 7.3's on the way, hoping to get, more soundstage depth (width on 10.2 was fine), and more extended top end while maitaining tonal balance (less directivity). Runnning the 10.2's on tube amps seems to improve the dpeth issue, but this makes sense in keeping with tubes ability to make individual instruments bloom/standout somewhat in their place in the soundstage.

I thought it might be cool to quote myself. I wanted to be clear that my comments about the high frequency response did not pertain to the extension, but rather the balance from top to bottom. I believe that you can get very good balance all the way up to, or close to 20kHz. Unfortunately, I believe that in order to do so, you have some loss in the areas of imaging and sound-stage due to the towing in of the drivers. If you are willing to give up some top end, you gain what is near perfect side to side sound-stage and imaging. I believe this makes sense when considering directivity as it has been explained to me, and for this reason, I am getting my third pair of MA drivers , the 7.3's. In the end, the previous sentence says it all.;) Oh yeah, the opinions expressed here represent only the poster in question, and in no way represent the truth or even fact.
 
Last edited:
I understand my equipment is sub-par, but mainly i use it to confirm what i am hearing...and that is part of the reason i did not post my response graphs in the first place.

Hi Chulyer,
Thanks for your understanding, its appreciated. Apologies as I meant no criticism, just the nature of the measurement debate can end up killing allot of my time that I need for driver design while I try to sort out all the questions.

I think there is a need for a separate practical thread as a sort of practical guide on home measurements. Not sure if I can spare the time but I do some thinking on how to help.

Thanks
Mark.
 
I thought it might be cool to quote myself. I wanted to be clear that my comments about the high frequency response did not pertain to the extension, but rather the balance from top to bottom. I believe that you can get very good balance all the way up to, or close to 20kHz. Unfortunately, I believe that in order to do so, you have some loss in the areas of imaging and sound-stage due to the towing in of the drivers. If you are willing to give up some top end, you gain what is near perfect side to side sound-stage and imaging. I believe this makes sense when considering directivity as it has been explained to me, and for this reason, I am getting my third pair of MA drivers , the 7.3's. In the end, the previous sentence says it all.;) Oh yeah, the opinions expressed here represent only the poster in question, and in no way represent the truth or even fact.

Hi Buzz, Guys,
Interesting comments.

From my perspective, as a driver designer/maker, I'm trying to make Full-Rangers deliver the widest possible set of frequencies with wide dispersion characteristics. That poses allot of technical challenges. Likely impossible to make drivers that 100% suite every application and individual taste. So I try to make drivers that are relativity flexible in their application with various box designs and room settings.

Typically, from most box designs and room environments, there will be some audible bass gain. Once we factor in the driver to box match, amp, source and room, there's so many factors that its near impossible to accurately predict what a driver will do beyond its anechoic sweep test performance.

As a guide, reckon on a Full-Ranger delivering 80% to 90% of its anechoic test chamber performance once its in a box, in a room. Mother nature and the law of thermo-dynamics will always apply; there's bound to be losses once we start converting from one form of energy to another; And when we re-direct a form of energy.

A typical outcome for many newer Diyers once they start using Full-Rangers is their surprise on getting some useful bass, but struggle with the top end. No surprise given the nature of the losses, so that's why I made many of my models with a rising response in the mid-highs. At least guys got some choice and could adjust to tow-in to get a balance that suites their tastes across most of the speaker's output.

Now things have moved on these last few months due to me listening to the feedback from you guys. There were a large number of requests for flatter and wider frequency delivery. Hence the new Gen.3 CHR-70' and Alpair 7's go further to meet these needs. Likely these drivers will need less tow-in/out than the previous models as there's more range at your disposal.

Conversely, there were a number of requests for "roll-off" drivers. Some guys think of high range output as "sharp and peaky", sometimes associated with cone break-up; Hence the production of the Gen. 2 CHP and the Alpair 6M drivers, the 6M design favouring its use in near-field applications.

When it comes to the bigger drivers (Alp's 10, 12 and 14), the complexities of balancing range output from the increased cone acreage becomes greater. Chuyler's comment in post No.3 are interesting:

"When listening on-axis it adds a slightly abrasive sound to trumpets and puts them sort of in your face. Switching to off-axis ~10 degrees removes the abrasiveness but sacrifices cymbals which don't cut through like they should and are slightly muted."

This is a likely outcome from me extending the range and dynamics of the Gen. 2 Alp10 over its predecessor. As I push the performance envelop of the these larger Full-Rangers, they are becoming more resonance sensitive. These drivers will demand good box optimisation, room positioning, good source material and well matched amplification. Once we get into deploying bigger low-mass cone and power-train technology, everything about the driver's performance becomes more sensitive and demanding. But that what makes this hobby so very interesting; By presenting Diyers with drivers that help push the boundaries of what is further achievable in home builds.

Alpair 12 Gen. 2 is next; new ultra-low profile cone, extended low mass coil, new Matsubara san rear suspension, new ultra-thin Lu Yi Ping front suspension, revised frame seals, revised magnet/motor structure, remains a semi-free-to-air, single mechanically controlled power-train unit. If I can get it work, this one's going to sort out the Diyer Men from the Diyer Boys.

Have fun!

Cheers
Mark.
 
Last edited:
Alpair 14?

Hi Mark,

Reading your last post, was there a hint there of what's coming or a miss print? An Alpair 14 in the works?

When it comes to the bigger drivers (Alp's 10, 12 and 14), the complexities of balancing range output from the increased cone acreage becomes greater. Chuyler's comment in post No.3 are interesting:

Bernie
 
Hi Mark,
Reading your last post, was there a hint there of what's coming or a miss print? An Alpair 14 in the works?
When it comes to the bigger drivers (Alp's 10, 12 and 14), the complexities of balancing range output from the increased cone acreage becomes greater. Chuyler's comment in post No.3 are interesting:
Bernie

Hello Bernie,
Yes, we are in a very early design stage for a Type 14 driver. The time-line for the first prototype is some 6 months away. But the Gen. 2 Alpair 12 is much closer. Hope to have prototypes working in the next 2 to 3 weeks. Will start a new thread to seek feedback.
Thx
Mark.
 
Mark,

Is the 14 being designed for use as a woofer or fullrange? If a woofer then it will be very compatible with the Alpair7 as a two way?
Dave is working on a 20 sided Mar-ken7t enclosure for me and were talking about doing a 20 sided woofer. Will this driver be useable for such a design?

Bernie
 
When it comes to measuring the room, it shouldn't really matter even if you're using a Radio Shack SPL meter... should it? I mean, he's just trying to see what changes occure. He's not after accurate results, just confirmation of the changes occuring.

Having said that, I'm not sure his measurements help anyone other than himself, in the bass region. Every room is different. Room modes cannot be shared on a forum.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.