Home Theater Forum and Systems banner

1 - 20 of 21 Posts

·
Registered
Joined
·
39 Posts
Discussion Starter · #1 · (Edited)
Just getting to work with REW and picked up the Dayton UMM-6 microphone with calibration files for various axis positions.

Though it would be neat to see if a laser sight would fit onto the microphone itself. For room measurements I should be using the 90° cal file and pointing the microphone straight up (Further review calls this into question) per REW instructions. So the laser sight is perhaps over kill. But looks like a solid idea if you need to point the micro at the same position with high repeatability.

I don't think it would impact the measurements unless it happens to vibrate in response to an audio frequency during sweeps?

This one was intended for paint_ball_guns.

DSC00696x.jpg
 

·
Premium Member
Joined
·
5,817 Posts
How expensive are they? Might be good for setting up your speaker locations too.
 

·
Registered
Joined
·
39 Posts
Discussion Starter · #5 ·
Just to be clear - I have fired up and used REW only once so still wanting to understand proper mic placement for various measurements. This kind of surprise me too.

Maybe the use of the 90 degree cal file does not mean that the mic itself should be orthogonal to the sound source(s)?

Below is the note that I was referring too. It's from the "REW 101 C3.1" PDF file that is around. Was referencing this since it was specific to USB mics.

Note 3: The two USB mic’s can be purchased directly from the vendor (MiniDSP for the UMIK-1, or Dayton Audio for the UMM-6), in which case you can download a single calibration file from the vendor’s web site. However, this calibration file is for a zero-degree orientation. Both mic’s are also available from Cross-Spectrum Labs, which runs a custom calibration for each mic they sell. The mic’s are provided with several calibration files, including a 90-degree calibration file, which is the angle used for REW measurements. Does it make a difference which type of calibration files you have? Check out this post (post 4926), which provides measurement data comparing the various calibration files available for the mic’s.

It looks like I will need to read the help file REW V5.0 that is part of the REW download.

Thank you,
 

·
Super Moderator
Joined
·
9,283 Posts

Below is the note that I was referring too. It's from the "REW 101 C3.1" PDF file that is around. Was referencing this since it was specific to USB mics.
Hmmm – can’t say I’ve ever heard of it, and an internet search turned up nothing. Nevertheless, here’s a good discussion on the topic that starts at this post and continues on down. Other related reading can be found here.

Regards,
Wayne
 

·
Banned
Joined
·
22,577 Posts
What are you measuring?

When we measure for our speaker evaluation we aim the mic straight forward at the front of the room and use the 0° calibration file.

It may be that pointing the mic straight up and using 90° for measuring Audyssey results might be better, since the Audyssey mic is pointed up too. And... for checking Audyssey you are probably only measuring subs with the center speaker... and you might also check subs with the left and/or right speaker just to compare what Audyssey did.

With my subs being in the front and back of the room and having the Audyssey mic pointed up during setup, I prefer to point my mic straight up when using REW. I had been pointing this way for ages, up until the speaker evaluation and the last time we ran Audyssey we pointed it forward and used the 0° file before we setup the Seaton speakers. I did not think much about it at the time... but realize now that it may have been better to point it up and use the 90° file. Probably not a drastic difference either way.
 

·
Registered
Joined
·
290 Posts
Pointing the mic up is sometimes used in a reflective environment to mitigate HF response issues; essentially the goal is to simulate a larger room to target (I can only assume) a rolled off response on the higher end a la sound power. Sometimes. I've tried this method in the car and the difference is nothing more than altering the high frequency roll off.

Not that it matters... What does matter is what you're measuring and what you're trying to achieve with the measrpurement.

Warning: the following is a gross simplification and uses some pretty broad generalizations, so those of you who may try to chew me to pieces, keep this in mind. ;)


If you're using this method for RTA purposes then what matters is your target curve. The target curve is the response you're trying to achieve in your sound system. Most curves involve a high frequency (HF) roll off starting in the neighborhood of 5khz and decrease in output at about 3dB/octave. (This is the part about broad assumptions).

If your target curve has this roll off then pointing the mic up would match that curve more easily because the response of the mic itself is rolling off as well. Why? Because the mic is not on-axis with the speakers. (As you go further off axis, sound intensity decreases; this is known as beaming and is a function of they driver size). So, you have the measured response roll off and that matches your target curve. Great! Nope. It wouldn't be entirely accurate because the curve falls off artificially due to the mic orientation. This is where the 90 deg cal file comes in to play. It would natively boost high frequency response so the measurement wouldn't have that roll off to it. You'd then find that your measurement wouldn't have a HF roll off so you'd EQ it down.

If you point your mic forward, you don't get a HF roll off. It'd be flat. And if your target curve is rolled off, then you would EQ the HF down.

So, if your cal file is set up properly and your mic orientation matches the cal file, it's essentially a wash. And pointing the mic up in a reflective environment doesn't yield any real benefit. Your target curve combined with this will dictate how pleasurable your results are.



IF you're trying to measure the speaker response (free from room interaction) then you really need to consider the impact of the mic and laser proximity, and the stand itself. Anything near the mic will introduce a reflection that will show up in the impulse response. The degree of damage to the measurement varies, but it's a legitimate concern. But, I have a feeling this isn't your goal. I'm assuming you're essentially using it for RTA purposes to see what the in room response curve is. And if that's the case, you need to be doing multiple averages to get a more representative measurement. A single mic location measurement is only a snapshot of what you hear. Unless you're in a HUGE room. :D


I typed this from my phone, so please show mercy. Lol.
 

·
Banned
Joined
·
22,577 Posts
I think what has been tested (the post Wayne P. linked to above), is that pointing the mic up at 90° actually increases the upper end response. I would think it would decrease it, but obviously not.

I might try to do a little bit of testing of that in my room too. I have several mics that measure exactly the same, and several boom mic stands, so I can set up a couple of them where the mic tip is exactly the same... one at 0° and another at 90°. Moving one of the other out of the way for the measurement.
 

·
Registered
Joined
·
290 Posts
I think what has been tested (the post Wayne P. linked to above), is that pointing the mic up at 90° actually increases the upper end response. I would think it would decrease it, but obviously not. I might try to do a little bit of testing of that in my room too. I have several mics that measure exactly the same, and several boom mic stands, so I can set up a couple of them where the mic tip is exactly the same... one at 0° and another at 90°. Moving one of the other out of the way for the measurement.
Right. That's what I was saying in my post. 90deg cal file will increase HF response to account for the roll off that occurs when the mic is essentially placed off axis.

Ultimately, the orientation has little effect as long as the correct cal file is used. You shouldn't use a 90deg cal file for an on axis measurement and vice versa (or anywhere in between).
 

·
Banned
Joined
·
22,577 Posts
Okay... yeah... see what you are saying now. Did not quite grasp that at first, thought you were saying the opposite.

If I am understanding the linked test right... he used the proper cal files and there was still a difference. He did admit to possibly being an inch off on his mic placement.
 

·
Registered
Joined
·
290 Posts
Okay... yeah... see what you are saying now. Did not quite grasp that at first, thought you were saying the opposite. If I am understanding the linked test right... he used the proper cal files and there was still a difference. He did admit to possibly being an inch off on his mic placement.
Agreed. I have no idea why a 90 cal file paired with a mic pointed up at 90 would result in higher output at higher frequencies vs a 0 deg orientation and corresponding cal file. Unless the cal file was wrong or the speaker was in the ceiling. Lol. Maybe there's just something in the puzzle I'm missing.

No matter... That's the only time I've seen that and I've got mounds of data showing the relationship of on vs off axis measurements to help keep me sane. Lol.
 

·
Banned
Joined
·
22,577 Posts
I wonder if a cal file at 90° can account for room effects? Perhaps that is part of it.

Like I say, I would like to test it myself with my mics. I have a Behringer ECM8000, IBF Akustik EMM-8 and the miniDSP UMIK-1. All have both cal files for the ECM and UMIK... but only one for the EMM-8.
 

·
Banned
Joined
·
4,838 Posts
On the laser sight, I love the idea. Probably total overkill, but with all the crazy things we are doing for speaker setup, what is one more laser? Its main value is probably more for repeatable placement of the mic. How about 3 lasers in a X, Y, Z setup so you can repeat mic placement with precision? Having the mic pointed a few degrees off target will not matter. Having the tip of the mic a few inches from its previous measurement location can make a significant difference.

You can clamp a string into the mic clip and a plain key ring as a weight on the other end to form a plumb line hanging to marks on the floor as a way to get sub-inch repeatable mic positioning when moving it back and forth between speakers, for instance.

For Audyssey, the reason for pointing the mic straight up is clear - no speaker in the room will be at an angle disadvantage. The build-in correction file is undoubtedly made for that 90° orientation.

For speaker measurements it seems best to point at the speaker. Some people say to use some other angle for room EQ measurements to get more "room flavor," but that would only apply if you use the 0° cal file and change the angle without changing the cal file, and only becomes significant above 5 kHz or so. I don't understand doing that, though, it just means you are rolling off high frequencies in the measurement, which will lead to HF emphasis when you EQ. But others recommend it, so there might be something I am missing about the reasons.

If you use the 0° file when pointed at the speaker and the 90° file when pointed at the ceiling (for example) you should get the same result. Getting the tip of the mic in exactly the same spot is difficult and is usually a bigger source of error than the angle.

Our speaker evaluation measurements have been done with the mic pointing straight forward toward the front wall, angled slightly left when in the left ear position, and right in the right ear position, trying to keep the measurements quick, because re-aiming the mic would be slow. That might be giving us a dB or so of error at 10 kHz with a 45° angle (the MM-1 is down 3 dB at 10 kHz when turned to 90°). We are already looking at better ways to do our measurements to reduce that error.
 

·
Registered
Joined
·
290 Posts
What kind of "speaker measurements", though? If you're measuring response of a speaker the loading the mic up with lasers is a BAD idea. Here's a good read: http://www.troelsgravesen.dk/measurements.htm

If you're simply doing this for response in the room then using lasers is a wash because you should be using a spatial average anyway which would essentially negate the purpose of exact repeatability.

It just depends on what you're trying to achieve.
 

·
Banned
Joined
·
4,838 Posts
So you are saying a laser mounted on the mic can cause reflections that mess up its HF performance. Good tip.

Even when averaging spacially for room EQ purposes, repeatability of those individual measurements can be useful. As you say, it depends on what you are after.
 

·
Registered
Joined
·
39 Posts
Discussion Starter · #17 ·
Thanks for the comments all. Would like to response with more details but pressed for time at the moment.

Multiple objectives. Measure overall room response from multiple seat positions. Overall review of room and speaker locations. Room changes if needed etc...

Also have a DIY sub that I would like to benchmark in a half space position. Also some experiments. Build thread is here.

http://www.avsforum.com/t/1438061/transmission-line-isobaric-subwoofer-for-home

My mains are bipolar so lots of reflections. Hence I may need to be double sure of my measurements and what they may be telling me.
 

·
Registered
Joined
·
109 Posts
On the laser sight, I love the idea. Probably total overkill, but with all the crazy things we are doing for speaker setup, what is one more laser? Its main value is probably more for repeatable placement of the mic. How about 3 lasers in a X, Y, Z setup so you can repeat mic placement with precision? Having the mic pointed a few degrees off target will not matter. Having the tip of the mic a few inches from its previous measurement location can make a significant difference.
Agreed on the repeatability. The trick is the offset of the laser vs. the mic at various distances. The TEF 25 measurement system I use has the ability to take 4 or 6 measurements using a cardioid microphone in each normal axis on a fixture which allows the tip of the mic to remain in the same location for each measurement. For large rooms where you need to determine where a reflection is coming from, a laser pointer can then be placed in the same fixture and aimed at specific elevation and horizontal angle to point to the reflection point.

Having seen how this works, I would suggest it might be easier to place a laser pointer in a body the same size and length as the microphone. Load the pointer in the stand, and aim it, then replace the laser with the mic. Of course the quality of the stand and clip matter, but there are good ones for reasonable money.

For speaker measurements it seems best to point at the speaker. Some people say to use some other angle for room EQ measurements to get more "room flavor," but that would only apply if you use the 0° cal file and change the angle without changing the cal file, and only becomes significant above 5 kHz or so. I don't understand doing that, though, it just means you are rolling off high frequencies in the measurement, which will lead to HF emphasis when you EQ. But others recommend it, so there might be something I am missing about the reasons.

If you use the 0° file when pointed at the speaker and the 90° file when pointed at the ceiling (for example) you should get the same result. Getting the tip of the mic in exactly the same spot is difficult and is usually a bigger source of error than the angle.
You would get the same result only if measuring outdoors or with a very narrowly gated measurement. As soon as you have reflective boundaries and a time window that allows them in the measurement (required if you want to see low frequencies), sound hits the microphone from most all angles. In effect you have an average of the on and off axis response, with the angle getting hit with the direct sound more likely to dominate so long as the direct sound is stronger than the reflections (not always the case).
 

·
Banned
Joined
·
4,838 Posts
Agreed on the repeatability. The trick is the offset of the laser vs. the mic at various distances. The TEF 25 measurement system I use has the ability to take 4 or 6 measurements using a cardioid microphone in each normal axis on a fixture which allows the tip of the mic to remain in the same location for each measurement. For large rooms where you need to determine where a reflection is coming from, a laser pointer can then be placed in the same fixture and aimed at specific elevation and horizontal angle to point to the reflection point.

Having seen how this works, I would suggest it might be easier to place a laser pointer in a body the same size and length as the microphone. Load the pointer in the stand, and aim it, then replace the laser with the mic. Of course the quality of the stand and clip matter, but there are good ones for reasonable money
Good ways to maximize accuracy and minimize interference.

You would get the same result only if measuring outdoors or with a very narrowly gated measurement. As soon as you have reflective boundaries and a time window that allows them in the measurement (required if you want to see low frequencies), sound hits the microphone from most all angles. In effect you have an average of the on and off axis response, with the angle getting hit with the direct sound more likely to dominate so long as the direct sound is stronger than the reflections (not always the case).
Thanks, Mark, you caught me over-simplifying a bit. Let's just say a 90° angle measurement will usually be most accurate with a 90° calibration file, although exceptions are possible. For that matter, most indoor measurements have limits to their accuracy and therefore limited usefulness.
 

·
Registered
Joined
·
109 Posts
Thanks, Mark, you caught me over-simplifying a bit. Let's just say a 90° angle measurement will usually be most accurate with a 90° calibration file, although exceptions are possible. For that matter, most indoor measurements have limits to their accuracy and therefore limited usefulness.
Well understood and I recognized it would be a bit nit-picky. I guess my point was to insure users realize that unless you only have sound arrival from that particular angle (ie 90 deg off axis), you should consider more of an average of the on/off axis response. Since this is mostly just a tilt of the HF, this is easy to consider when making any final voicing choices of the top end shape.

While certainly nothing new, those following along should remember that some of the most important info you can know about your measurement setup are its limitations. Just because you can create/take a measurement, doesn't mean it's telling you what you would like it to! :scratch:
 
1 - 20 of 21 Posts
Top