Another SPL Meter/Rives Question

[Dumar]

I’ve armed myself with a Radio Shack analogue SPL meter and Rives Audio Test CD and am almost ready to do battle with my room. I just have one question regarding the plan of attack.

When setting up the meter Rives instructs us to set the dB level to 80, play the 1kHz tone, and adjust output to obtain 0dB. Then check the reading again at a higher frequency around 4kHz. This seems straight froward enough, but how do I know if the sub’s output level is correct for the lower frequencies (my mains are set small and the cross over is 80Hz)?

Thanks.

 

[mtrycrafts]

I’ve armed myself with a Radio Shack analogue SPL meter and Rives Audio Test CD and am almost ready to do battle with my room. I just have one question regarding the plan of attack.

When setting up the meter Rives instructs us to set the dB level to 80, play the 1kHz tone, and adjust output to obtain 0dB. Then check the reading again at a higher frequency around 4kHz. This seems straight froward enough, but how do I know if the sub’s output level is correct for the lower frequencies (my mains are set small and the cross over is 80Hz)?

Thanks.

It is best if a band of pink noise is used to level match. But, in this case, if you don't have a pink noise for the low band, use something around 50-63Hz and set the level to 0 or up to 3dB higher for taste. The problem with single tone low frequency is that you will activate room modes and it may not be accurate overall.

The other problem with the test disc is unless you know what the signal level on the disc is, that 80dB spl is not meaningful. You may want to read the instruction to see if there is a reference to something like a -20 dBFS . If not, call and ask. If the test signal is -20, then when you advance the master volume control so the spl meter reads 0, that will be your reference level you read about. Start with one speaker to set the master volume control, then the channel pots to balance the others.

 

[Dumar]

Thanks, mtrycrafts.

It is best if a band of pink noise is used to level match. But, in this case, if you don't have a pink noise for the low band, use something around 50-63Hz and set the level to 0 or up to 3dB higher for taste. The problem with single tone low frequency is that you will activate room modes and it may not be accurate overall.

This is what I thought I should do but did not think about the modes. However, I should be able to calculate the modes and choose a "safe" frequency, right?

The other problem with the test disc is unless you know what the signal level on the disc is, that 80dB spl is not meaningful. You may want to read the instruction to see if there is a reference to something like a -20 dBFS . If not, call and ask. If the test signal is -20, then when you advance the master volume control so the spl meter reads 0, that will be your reference level you read about. Start with one speaker to set the master volume control, then the channel pots to balance the others.

My intention is to use the meter and test disk to graph the acoustic properties of my room to see what frequencies require attention. I understand the purpose of setting the tone level to 80dB is to establish a reference level for all the other subsequent tones.

Thanks again for your help on this.

Dumar

 

[Buckle-meister]

My intention is to use the meter and test disk to graph the acoustic properties of my room to see what frequencies require attention.

Dumar, have you read this? If not, you might find it interesting.

Regards

 

[Dumar]

Hi Buc,

Yes, I have read most of that thread and did find it interesting. So basically you were not concerned about frequencies below the cross over, right?

I started asking myself about the sub's level adjustment after reading this. The author makes no mention of it, so I'm thinking maybe it doesn't matter all that much if you just leave the level set where it sounds best during normal listening.

Dumar

 

[Buckle-meister]

...basically you were not concerned about frequencies below the cross over, right?

Not quite. I was interested in frequencies below the 63Hz frequency that I used as a cut-off; just not for my front towers since they are currently crossed over at 80Hz.

I started asking myself about the sub's level adjustment after reading. The author makes no mention of it, so I'm thinking maybe it doesn't matter all that much if you just leave the level set where it sounds best during normal listening.

I would imagine there are two schools of thought:

1. The level does matter if you are to correctly integrate your speakers with the sub for a smooth transition of bass from the former to the latter.

2. The level doesn't matter, and you should let your ears decide what sounds best.

Regards

 

[MDS]

I would imagine there are two schools of thought:

1. The level does matter if you are to correctly integrate your speakers with the sub for a smooth transition of bass from the former to the latter.

2. The level doesn't matter, and you should let your ears decide what sounds best.

How about the third school of thought (mine). The level doesn't matter all that much because our ears aren't all that sensitive to minor amplitude variations at low frequencies. I say get it as close as you can and don't worry too much about it.

 

[mtrycrafts]

Thanks, mtrycrafts.


This is what I thought I should do but did not think about the modes. However, I should be able to calculate the modes and choose a "safe" frequency, right?

Yes, you can calculate it but not sure you can pick a safe freq just to level match.


My intention is to use the meter and test disk to graph the acoustic properties of my room to see what frequencies require attention. I understand the purpose of setting the tone level to 80dB is to establish a reference level for all the other subsequent tones.

Thanks again for your help on this.

Dumar

Yes, that is the purpose of a test signal, regardless of its intensity, 80dB spl 85dB spl or 75, is to level all channels the same as that is what is done in a studio and how they then master it.

Yes, checking the frequency response with single tones is helpful but only if you intend to EQ at some point. Otherwise it may be frustrating to see the variations and may distract from listeing. Or, it could be a good excuse why it doesn't sound so good

 

[Dumar]

Yes, checking the frequency response with single tones is helpful but only if you intend to EQ at some point. Otherwise it may be frustrating to see the variations and may distract from listeing. Or, it could be a good excuse why it doesn't sound so good

I’m really just checking the frequency response of the room more out of curiosity than anything else. I think it sounds not too bad the way it is … I guess I was just born to tweak. If need be I may treat the room in a passive way.

The thing that is frustrating me right now though, is how variable my results are. I thought I read somewhere that the Radio Shack meter was a good one to use because it was repeatable: unfortunately that’s not what I’m finding. For example, if I establish a reference level at 1kHz and then run through a bunch of frequencies and then re-check the reading at 1kHz, the value is way off. What’s up with that?

I’ve also found that I need to keep changing the dB scale on the meter just to be able to read a value. For example, at some of the lower frequency tones I need to use the 70dB setting (instead of the 80dB reference range) to get some deflection on the meter. If these readings are to be believed, this means I have several frequencies that are > 10dB below my reference level. This can’t be good news!

One other observation is that the output from the Rives CD does not seem to be constant for a given tone. While “listening” to a tone with the meter you can see the needle settle out at some point for a few seconds, and then move away from that point by several dB. I should point out that I’m not running around the room while this is going on. Actually, I am able to take my readings from behind the listening position and don’t believe I am influencing the readings in any way.

Does anyone have thoughts on these observations?

 

[MDS]

If you could get every single frequency you test to play back at the exact reference level you set, you would have an acoustically PERFECT room. Not gonna happen...

When you calibrate to any reference level you are using pink noise (equal amplitude at every octave) and are in effect measuring an average level. When you run frequency sweeps to plot the response of the room you are using specific frequencies and the amplitude will vary from frequency to frequency so it's normal that you would have to adjust the range of the meter to get a reading.

 

[Dumar]

When you run frequency sweeps to plot the response of the room you are using specific frequencies and the amplitude will vary from frequency to frequency so it's normal that you would have to adjust the range of the meter to get a reading.

Yes ... I understand, thanks.

But why is it when I re-measure at my reference frequency (0.0dB @ 1000Hz = 80dB SPL when the range selector is at 80dB), the meter no longer reads 0.0dB? And why do I see variations while the tone is playing? Shouldn't the meter be "steady"?

Maybe I'm expecting too much from this device, or perhaps it's NG.

 

[Tom Andry]

Yes ... I understand, thanks.

But why is it when I re-measure at my reference frequency (0.0dB @ 1000Hz = 80dB SPL when the range selector is at 80dB), the meter no longer reads 0.0dB? And why do I see variations while the tone is playing? Shouldn't the meter be "steady"?

Maybe I'm expecting too much from this device, or perhaps it's NG.

One question: Are you holding the meter or is it on a stand? I had the same problem until I mounted mine on my camera stand. It's funny how you can see the reading changes as your hand gets closer.

 

[Dumar]

One question: Are you holding the meter or is it on a stand? I had the same problem until I mounted mine on my camera stand. It's funny how you can see the reading changes as your hand gets closer.

Sorry I didn't mention this before.

I did attach the meter to my camera tripod and place it on the couch. The meter is very close to where my head would normally be (when I’m sitting upright that is).

I also played with the position of the microphone: first pointing it forward and then at an angle towards the ceiling, but that didn’t seem to make a difference. I should also mention there is virtually no ambient noise at my place: it’s really nice and quiet.

Room dimensions are 8’ H, 14’ W, and about 17 L. Wall-to-wall on the floor, walls and ceiling are drywall. One couch and two chairs (all upholstered). One 3’ x 5’ window, which was covered at the time the measurements were taken.

I have a friend who has the same meter. I'm going to see if I can borrow it and make some comparisons.

 

[mtrycrafts]

Sorry I didn't mention this before.

I did attach the meter to my camera tripod and place it on the couch. The meter is very close to where my head would normally be (when I’m sitting upright that is).

I also played with the position of the microphone: first pointing it forward and then at an angle towards the ceiling, but that didn’t seem to make a difference. I should also mention there is virtually no ambient noise at my place: it’s really nice and quiet.

Room dimensions are 8’ H, 14’ W, and about 17 L. Wall-to-wall on the floor, walls and ceiling are drywall. One couch and two chairs (all upholstered). One 3’ x 5’ window, which was covered at the time the measurements were taken.

I have a friend who has the same meter. I'm going to see if I can borrow it and make some comparisons.

How much variation do you get when you reset the tone to 1kHz? You have not moved any of the settings, right? I am not sure how good the CD is, how constant the signal is, etc. Without a reference quality CD, a known level of accuracy and level consistency on that CD, it is hard to know what may be going on.

How close do you get to make your readings, even though it is on a stand of some kind. That may have an effect. Just remember, this isn't a micrometer but a dynamic instrument

 

[Pyrrho]

Yes ... I understand, thanks.

But why is it when I re-measure at my reference frequency (0.0dB @ 1000Hz = 80dB SPL when the range selector is at 80dB), the meter no longer reads 0.0dB? And why do I see variations while the tone is playing? Shouldn't the meter be "steady"?

Maybe I'm expecting too much from this device, or perhaps it's NG.

I have found that if I move very, very slightly (just turning my head a little, for example), I affect the meter reading. This is likely due to the fact that sounds are bouncing off of me, and are absorbed by me, so when I move ever so slightly, I am changing the acoustics of the room, and therefore I change the reading. This is with me off to the side, not in a direct path between the meter and the speaker (and I also use a tripod for the meter, so it isn't moving). This effect is also frequency dependent; a slight move of my head makes no difference in my readings for subwoofer frequencies.

As for variations while a tone is playing, assuming that you are perfectly still and not having the problem mentioned above, it may be that you have an acoustically live room in which the sounds are bouncing off the walls, and you get an increase or decrease as the reflected sound "mixes" with the direct sound. The increase or decrease would result from the reflected sound being either in phase or out of phase with the direct sound. (This problem, by the way, is extremely common with very low frequencies.)

 

[Dumar]

Thanks a lot you guys, I think you hit the nail on the head.

After reading your posts and paying particular attention to my position and movement during the measurement procedure, I have seen a major improvement in the repeatability and stability of the meter. I am quite amazed how little movement it takes to change the readings!

I have found that after establishing my reference point at 1kHz and 4kHz, if I change the position of my body only slightly the effect on the meter can be several dB.

After this revelation sank in I ran through the tones again while remaining as still as possible, and am happy to report that with only a few exceptions most of the readings were between 0dB and -3dB.

Thanks for helping me clear this hurdle. Now on to the next …

Dumar

 

[Pyrrho]

Thanks a lot you guys, I think you hit the nail on the head.

After reading your posts and paying particular attention to my position and movement during the measurement procedure, I have seen a major improvement in the repeatability and stability of the meter. I am quite amazed how little movement it takes to change the readings!

I have found that after establishing my reference point at 1kHz and 4kHz, if I change the position of my body only slightly the effect on the meter can be several dB.

After this revelation sank in I ran through the tones again while remaining as still as possible, and am happy to report that with only a few exceptions most of the readings were between 0dB and -3dB.

Thanks for helping me clear this hurdle. Now on to the next …

Dumar

Now, all you have to do is sit EXACTLY in the same position you were, and then the sound will be great (at least where the meter is, if not where you are)! This is something that you will never get perfect, and even if you could, you could not know it, because you will never be able to occupy the place where the meter is while also having your body in the location where it was when you took the meter reading. Happy listening!

 

[Dumar]

Now, all you have to do is sit EXACTLY in the same position you were, and then the sound will be great (at least where the meter is, if not where you are)! This is something that you will never get perfect, and even if you could, you could not know it, because you will never be able to occupy the place where the meter is while also having your body in the location where it was when you took the meter reading. Happy listening!

Hey Pyrrho, I'm already working on this one:

First, I was thinking I would remove the transducer from the SPL meter, attach a couple extension wires, and then glue it to my forehead.

Next, I would sit down in the sweet spot and play the test disk while holding the meter comfortably in my lap so I can read it.

Then, while moving only my mouth, I’d call out the readings to my trusty assistant, Igor who is in the next room with the spreadsheet and pencil (I don’t want him in the room with me because he drools on everything).

Might work, what do you think?

 

[avnetguy]

First, I was thinking I would remove the transducer from the SPL meter, attach a couple extension wires, and then glue it to my forehead.

Better yet, have two SPL meters and put one transducer on each ear!
Of course you'd have to run two tests and swap the transducers to remove reading errors.
After that have your ears tested and adjust the outputs accordingly.

Steve