SPL Meter

[Miamienesex]

I am in the market for an spl meter and have noticed that the prices vary immensely. Is it necessary to spend hundreds of dollars or can it be done with a 70 dollar meter.

Thanks in advance
Armando

 

[bpape]

It really depends on what you want to do with it. If you're looking to level match the system, then a cheaper one will be fine. If you're looking to take some measurements of the bottom end to do some treatments, then as long as you know or can get the correction values for the cheaper one (aka Radio Shack), then the less expensive can be OK.

If you're wanting to do real full frequency range plotting of response curves, then get a better one - or if you have a laptop, get a good calibrated mic and use software that will allow you to enter the calibration file.

Bryan

 

[mike c]

90+% of the people use Radio Shack SPL meter

44.99 USD

radioshack online.

 

[Ethan Winer]

Armando,

> Is it necessary to spend hundreds of dollars or can it be done with a 70 dollar meter. <

As Bryan said, it depends on what you want to do. A $50 Radio Shack meter is adequate for the two most important things people usually use it for:

* Measure low frequency response to assist loudspeaker and listener placement

* Balance speaker levels in a surround system

I have both the RS meter and an expensive calibrated microphone. Below 800 Hz they track each other within 1 dB (no correction curve needed or used), and that's plenty accurate for low frequency testing.

If you use DVD Essentials or some other test DVD for speaker balancing, the absolute response is not relevant. DVD Essentials, which I'm familiar with, plays band-limited pink noise through each speaker. So as long as each speaker measures the same SPL, it doesn't really matter if the meter is actually flat.

--Ethan

 

[acurtas]

Count me in for the Radio Shack analog one as well.

You can also use that as the microphone for some of the software programs like Room EQ Wizard, TrueRTA, and ETF. It's not as nice as a calibrated mic and preamp, but I used this combo to set up my speakers, test different subwoofer locations, and equalize my sub frequency response.

good luck,
a

 

[j_garcia]

For the basics, the analog Radio Shack meter is the way to go, since the correction values are readily available.

 

[GettinDegreez]

Why the analog SPL meter and not the digital one?

 

[mtrycrafts]

Why the analog SPL meter and not the digital one?

For level matching, there is no rounding up or down errors as in the digital. You could be a whole 1 dB off.
The digital has its place on the shelf though for other measurements, how loud is it? Easy to read, each scale covers 20 dB.

 

[Savant]

For level matching, there is no rounding up or down errors as in the digital. You could be a whole 1 dB off.
The digital has its place on the shelf though for other measurements, how loud is it? Easy to read, each scale covers 20 dB.

The digital meter from RS has a simplistic averaging feature that could be used to overcome this. I think you can set the meter to average for up to 100 seconds. This feature is more or less the equivalent of "Leq" in noise monitoring. An explanation of which can be found here.

IMO, this would be somewhat more accurate than eyeballing a needle. But that's just me.

 

[MDS]

IMO, this would be somewhat more accurate than eyeballing a needle. But that's just me.

I agree and that is really the only point of debate as to whether the analog or digital version is 'better'. The digital version has more useful features and it is easier to get a steady reading because you look at a 1" tall LCD number instead of a tiny little scale and needle.

 

[acurtas]

Does the digital meter have the same frequency response of the analog? Or better yet, are there any very accurate correction tables for the digital one?

I know these exist for the analog version and come in handy when using software to help EQ out your bass or locate your speakers. Some software even has pre-made calibration files specifically for this meter.

I would be surprised if nothing of the sort existed for the digital version; but since I've only looked for the analog stuff I don't know.

just something to consider.

 

[MDS]

As far as I know the digital and analog versions of the Radio Shack meter are the same in terms of accuracy and frequency response. If you are using it to balance your system, then the correction factors are not relevant. What is relevant is that the meter is consistent ('deterministic'); ie it gives the same reading each time you use it. In other words it doesn't matter if the readings for the low bass are off by 2 dB because they will be off by 2 dB every single time you use it..

If you want to use it to plot the frequency response of your system then the correction factors may come in handy but then again it is not the right tool for that job anyway.

 

[mtrycrafts]

The digital meter from RS has a simplistic averaging feature that could be used to overcome this. I think you can set the meter to average for up to 100 seconds. This feature is more or less the equivalent of "Leq" in noise monitoring. An explanation of which can be found here.

IMO, this would be somewhat more accurate than eyeballing a needle. But that's just me.

So, you are saying that the whole number displayed are not rounded up or down but averaged and is not off 1 unit if one channel is 74.4, rounded to 74 and the other 74.6 and rounded to 75 on the meter? Hence causing over or under matching?

 

[MDS]

Let's say the meter takes 2 ms to obtain a measurement and it integrates all of the measurements over a larger time window of say 100 ms, thus taking 50 'snapshots' in that window. It then averages all of those values and rounds up or down because it can only show whole numbers on the display.

So what? If the measurement were 75.4 and is rounded down to 75 why is that inaccurate? A. You can't hear a .4 dB difference and B. You cannot read an exact figure like 75.4 with the analog meter either.

Remember it takes time for the needle to move on the analog meter too and any small difference will be subsumed by the time it takes for the mechanical needle to move. Either meter is fine - it's just personal preference whether you want to watch a needle bounce or get an instant numerical reading.

 

[mtrycrafts]

So what? If the measurement were 75.4 and is rounded down to 75 why is that inaccurate? A. You can't hear a .4 dB difference and B. You cannot read an exact figure like 75.4 with the analog meter either.
.

No, you will not hear .4 dB and is not an issue until you have the other channel that really is 74.5 and rounds up to 75. Now you think they are level matched, yet they are .9 dB out of level matching.
With that needle, steady on a mark and the next channel matched to that same mark, that will be extremely close in level, certainly not .9 dB difference.
I think that is what I am driving at. Channels can be out .9 dB with a digital meter and show same number. On the analog, you could make it right on, unless you cannot do fine volume trimming. Some can

 

[Savant]

So, you are saying that the whole number displayed are not rounded up or down but averaged and is not off 1 unit if one channel is 74.4, rounded to 74 and the other 74.6 and rounded to 75 on the meter? Hence causing over or under matching?

I'm sorry, but I don't quite understand the question as it relates to the time-averaging feature. I don't exactly know off-hand the method by which the RS meter makes the calculation. I believe it is akin to equivalent level, as I mentioned above. But I would have to check the instruction for the RS meter to be sure. And even then, I cannot remember what kind of detail it has. I can tell you that the time-averaging for sound level meters tends not to be a straight numerical average. It is typically an energetical average. So, if a sound fluctuates quite a bit over the selected time, the average will be on the high side compared to a numerical average. However, in the contexts of test signals used to calibrate home theater systems, the fluctuations should not be that great and the average will be a very useful quantity.

As for rounding, I believe the time-averaging feature would be more helpful in this regard. E.g., if one channel measures 75 dBA using this sort of feature, then another channel should be tweaked to match the 75 dBA (as appropriate) using the same sample time. IMO, this will be more accurate than simply using a meter (analog or digital) and trying to eyeball a match. Specifically addressing your question (I think), I would be more worried about 74 dBA actually being 73.6 dBA and 75 dBA being 75.4 dBA than I would about your comparison. An almost 2 dB difference is much more significant than 0.4 dB. This again would be a vote in favor of time averaging.

I hope this helps!

 

[mtrycrafts]

I'm sorry, but I don't quite understand the question as it relates to the time-averaging feature. I don't exactly know off-hand the method by which the RS meter makes the calculation. I believe it is akin to equivalent level, as I mentioned above. But I would have to check the instruction for the RS meter to be sure. And even then, I cannot remember what kind of detail it has. I can tell you that the time-averaging for sound level meters tends not to be a straight numerical average. It is typically an energetical average. So, if a sound fluctuates quite a bit over the selected time, the average will be on the high side compared to a numerical average. However, in the contexts of test signals used to calibrate home theater systems, the fluctuations should not be that great and the average will be a very useful quantity.

As for rounding, I believe the time-averaging feature would be more helpful in this regard. E.g., if one channel measures 75 dBA using this sort of feature, then another channel should be tweaked to match the 75 dBA (as appropriate) using the same sample time. IMO, this will be more accurate than simply using a meter (analog or digital) and trying to eyeball a match. Specifically addressing your question (I think), I would be more worried about 74 dBA actually being 73.6 dBA and 75 dBA being 75.4 dBA than I would about your comparison. An almost 2 dB difference is much more significant than 0.4 dB. This again would be a vote in favor of time averaging.

I hope this helps!

The question really comes down to rounding. That meter must round up or down at some predetermined point. If the averages are 75.4, it rounds to 75 on the meter. If it is 74.5, it rounds to 75 again.
So, one channel can be 74.5 and still read 75. The next one will read 75.4 and read 75. Yet, you have a .9 dB channel difference. No?

The .4 db only comes into question when comparing two channels, not how close you are to the desired volume of 75. And, as I suspect, the meter rounds and the channels can be off by .9 dB spl.

 

[Ethan Winer]

Jeff,

> I would have to check the instruction for the RS meter to be sure. <

I just checked my RS meter's manual, and as you'd expect there's very little detail.

However, you can tell the meter to examine the level over a period ranging from 1 to 199 seconds. Then afterward you can see the minimum, maximum, and average values. That still doesn't get you better resolution, but what do you expect from a $49 meter?

--Ethan

 

[Nick250]

Ethan, do you have an opinion on what's the next step up the ladder in meters is beyond RS? For me, I would invest another $50 or $75 more that the RS meter if it would help set up my speakers more effectively.


Nick

 

[Savant]

The .4 db only comes into question when comparing two channels, not how close you are to the desired volume of 75. And, as I suspect, the meter rounds and the channels can be off by .9 dB spl.

I imagine your suspicion is correct. If this is a major concern for you, you should probably consider an SLM with 0.5 dB or 0.1 dB resolution:

Galaxy Audio CM-130 (Same price as the RS meter, with 0.5 dB resolution.)
Galaxy Audio CM-140
Tenma 72-6635

As Ethan pointed out, the Radio Shack meter is a reliable, but relatively inexpensive meter.