spl charts questions!

[nick mineau]

i have been trying to figure something out.
i have smaart software, in smarrt there are peak spl measurements and rms spl measurements.

if i play a 60hz sine wave on my diy subs @2.83v , and measure it @1 meter on the (regular db ) setting. i get 100db, but if i use the peak setting i get 103db. i understand that this is crest factor. but my question is this.

danley (for example shows) a tef frequency response curve @ 2.83v. which way should i measure my subs to compare to a TEF chart like danley uses? flat db peak, or flat db regular/rms?

do bills charts use 2.83v 1 meter db peak or rms/none peak?

[Bill Fitzmaurice]

Unless specified otherwise my charts are 2.83v/1m half-space, measured with a pink noise signal.

[Seth]

...which way should i measure my subs to compare to a TEF chart like danley uses? flat db peak, or flat db regular/rms?

do bills charts use 2.83v 1 meter db peak or rms/none peak?

I'm gonna stick my neck out and say RMS on both accounts...

I did a quick search through the SMAART User Guide and found this...

Full Scale (dB FS) versus Full Scale
There exist two competing references for decibels in digital audio signals. One convention references dB
FS to the largest positive and negative amplitude values obtainable from a given integer sample word
size – e.g., ± 32768 for 16 bits – normalized to a range of ± 1.0, such that 0 dB FS denotes the maximum
possible digital amplitude value. We will refer to this as “normalized Full Scale.”

The second convention, preferred by the Audio Engineering Society (AES), references 0 dB FS to the RMS
value of a full-scale peak-to-peak sinewave (i.e., 0.7071 normalized Full Scale, rather than 1.0). We will
call this “AES Full Scale”. In Smaart, Full Scale decibel values are always referenced to normalized Full
Scale, meaning that the RMS magnitude of a full-scale digital sinewave is -3.01 dB FS.

I've got to imagine, being the AES "preferred" value, that anything from a reputable manufacturer would (should) be the RMS value. And any professional software or hardware measuring equipment (such as TEF) would more than likely also follow those guidelines and default to RMS. Although, from what's written above, I'm not clear whether SMAART defaults to RMS (AES Full Scale) or not. It reads as though they default to Peak. So, a little wishy washy.

Not 100% definitive, but my best educated guess is RMS would be the correct value to use for comparing to other standard response charts.

Side note FYI, when you measure an AC voltage with your DMM, the number you see on the screen is RMS voltage which uses the same 0.7071 factor... 70.7% of peak to peak voltage = RMS Voltage. So, a wall socket that shows 120v is actually 169.7v peak to peak... which, non-coincidentally is a 3.01dB difference.

[Seth]

Unless specified otherwise my charts are 2.83v/1m half-space, measured with a pink noise signal.

I remember having the conversation, but forget what you said you use for your graphing... is it HOLMImpulse?

[Bill Fitzmaurice]

I've been using HolmImpulse for a while, though most of the older designs were measured before I had it.

[Seth]

Searched the HOLMImpulse user guide and can't find any indication of the value (peak or RMS) it uses.


Also, here's a little excerpt I found in the SMAART SPL Quick Start Guide that may be insightful

Peak Measurement
Peak C
Peak C shows the absolute maximum value reached within a measurement period. It is useful for
measuring compliance with regulations that seek to protect employees against sudden large peaks, like
those created by industrial machines, explosives, and firearms.

Peak C is unique because it is the only metric that is not measured over some period of time. By
definition, Sound Pressure Level is an RMS quantity. RMS, or Root-Mean-Square, is a mathematical
averaging useful when dealing with waves having both positive and negative values. Peak C is not an
RMS value, which is why it’s called simply Peak C, rather than Peak C SPL.

[Bill Fitzmaurice]

All I know is that when the word 'peak' is used in audio it's usually as a method of inflating the value.

[Bruce Weldy]

All I know is that when the word 'peak' is used in audio it's usually as a method of inflating the value.

Dang.....is that why even when I'm at my peak performance, I'm still not very good?

That explains a lot....

[AntonZ]

Thank you Bruce for confessing. I'm not alone.

[DaveMacKay]

Having benefited from the advice given generously by forum members, I'd like to contribute to the forum.

I thought that it might be helpful to add SPL charts for my 12" THTLPs since I don't believe any exist on the forum. However, to be useful, I'd like to ensure that my charts were consistent with how the other charts were made so that they can be compared apples-to-apples.

Is there a way that I could take measurements (in my listening space) with REW and a UMIK-1 calibrated microphone that would provide the forum with useful charts? If so, please let me know:

• what parameters I should use for the measurements (dB level, frequency range, # sweeps, microphone orientation and distance from the THTLP, and where on the THTLP the microphone should be aimed, etc.).

• how to structure the chart (units to use (e.g., SPL, DBFS, etc.), scale (e.g., 40 to 110), what smoothing to apply (if any), etc.)

I'd rather not contribute a chart than contribute a misleading one, which is why I'd like to be certain that my measurement and charting methods would be suitable.

[Bill Fitzmaurice]

Measurements taken indoors will be dominated by boundary reflections and cabin gain. What they're good for is determining the optimal combination of placement and EQ for the best possible results at the listening position in your room, and bragging rights if they look really good.

[DaveMacKay]

Measurements taken indoors will be dominated by boundary reflections and cabin gain. What they're good for is determining the optimal combination of placement and EQ for the best possible results at the listening position in your room, and bragging rights if they look really good.

Thanks!

I'll leave the charts to others then

Happy new year to all!