SPL Meter Correction Tables - Page 2

Today I got my professionally calibrated microphone. Even without any correction files, it's only 0.5 dB off at 10 Hz. Pretty impressive. It was calibrated against B&K 4133 microphone.

I have now measured it against my analog RS meter. As I have said, the RCA output jack is weighted (C or A). But as Ethan showed, it appears that the output of the RCA jack of the digital RS meter is unweighted when set to C-weighting, when set to A, it's A-weighted. When set to C, it has a flat FR atleast down to 20 Hz.

Here is a TrueRTA screen of my results.



As you can see, even with those general CFs, the RS meter is still below the true FR. It tracks the mic down to 40 Hz, but below that it's off. You need to compensate even more than for example 7.5 dB @ 20 Hz. Around 11.5 dB seems to be more closer. Of course every RS is different, so this might not apply to your RS. Although me and my friends have measured total of three RS meters (analog) and they all give pretty similar FR.

http://personal.inet.fi/private/ilkka/RS_calibrated.txt

Here is a calibration file which turns my RS into calibrated measurement microphone. When using this file its FR is within 1 dB (10 Hz - 200 Hz) from my calibrated microphone. It may not be as perfect with your RS, but much better than the general CFs. Notice that general CFs are way off below 15 Hz.

Analog version:
Use atleast general CFs or this file when measuring with the TrueRTA or smilar. Same thing if you are measuring manually (looking at the needle). You can use the rough corrections below, if you don't want to check the file (if you are measuring manually, add those corrections, do not subtract them).

Frequency / Correction
10 Hz +5 dB
12 Hz +6 dB
16 Hz +14.8 dB
20 Hz +11.5 dB
25 Hz +7.8 dB
30 Hz +5 dB
40 Hz +3.2 dB
50 Hz +2 dB
80 Hz +1 dB
100 Hz +1.2 dB

Digital:
Do not use any correction files when measuring with TrueRTA or similar. The output of the RCA jack is unweighted when set to C-weighting. When measuring manually, add corrections shown above to the readings.

Awright, I'm confused. Why would you not use any CFs when feeding a C-weighted, digital RS meter to TrueRTA or ETF? The deviations are due to the meter, not the weighting contour.

Yes I admit, it is confusing. But as Ethan's (and my friends) measurements show, the output of the RCA jack is unweighted. I don't know how and why they have done it, but the proofs are pretty solid. I'll try to get a digital measure to myself and measure it too.

And the deviations are due to the weighting contour and the meter (analog version). As you can see from my TrueRTA screen, if you compensate for the C-weighting, it is not enough. As we all know, the general CFs compensate a little bit more than for the C-weighting, but still not enough.

Which correction table can be loaded for use with RoomEQ Wizard and the analogue RS meter? Or do you need to load one?

dD

http://personal.inet.fi/private/ilkka/RS_calibrated.txt

You can use that one. Just rename it to .cal.

Thanks!

I'll let you know how I make out.

dD

When I used this in RoomEQ Wizard a response plot of the radio shack meter showed up.

I also got a plot for the subwoofer measurement.

What I don't understand: Do I want to then make the final EQ corrections to match the corrected Radio Shack meter plot? Or is the goal to make the plot close to a straight line across the graph?

Does anyone have any correction factors that go below 10hz? Even just a couple of hz would be helpful.

I could do some measurements using my analog Radio Shack, but unfortunately I can't promise that they are accurate for your meter. I definitely recommend some more expensive SPL meter or microphone, if you need to measure frequencies that low. Although most microphones do come with a calibration info only down to 10 Hz (mine too).

Thanks for the response. It's not really THAT important. My in-room response starts to get a bit jagged towards 10hz, so I was really just curious what it does below that. Not worth any work though.

Thanks again!

I have measured my analog RS meter mic against my calibrated Behringer ECM8000 and came up with quite the different corrections from above. John Mulcahy (REW) and Ken Bruce (brucek) calculated new correction values based on this measurement and created a newrs.cal file graphed as follows:



-------


When these correction values were used with a different RS meter mic and compared to the calibrated ECM there were very slight variances, thus indicating there are slight variances in response from one analog RS meter mic to the next, which we pretty much already knew this. The above correction values can be downloaded as a .cal file...

newrs.cal available here.

-------

So now we have "corrected" correction values - serveral times over? The values posted at the beginning of this thread were all in agreement - (e.g. at 20Hz the meter is down 7.5dB, etc.) Are we now saying all those are wrong? Realizing that the TRUTH of a matter is sometimes difficult to come by, I'm just hoping we can determine who's correction values are to be trusted.

Ok Maybe it is just me, but what correction should I use? For exemple I saw for 20hz: -6.5, -7.5, -11.5 and -12.4

What are the globaly accepted corrections for the R.S. Analogue SPL Meter 33-2050 ?

Good point... and I don't know what to tell you other than to show you what we've done above. I sent my ECM8000 off and had it calibrated. It was not off much to begin with. From the several calibration files I've seen from various members who have had their ECM calibrated, they only varied slightly from mine.

Here's what my ECM8000 calibration file looks like (-1.82db at 20hz):



I basically followed the instructions of John Mulcahy (RoomEQ Wizard) and Ken Bruce. Here's a quote from John:

Therefore we took the drastic peaks and dips out of the response (flattened it to as best we could) and measured the three mics.

Whether there is that much variance from RS mic to RS mic, I just don't know. If you compare Ilkka's corrections vs. our corrections:

10 Hz +5 dB vs. 27.00
12 Hz +6 dB vs. 21.91
16 Hz +14.8 dB vs. 16.15
20 Hz +11.5 dB vs. 12.38
25 Hz +7.8 dB vs. 8.80
30 Hz +5 dB vs. 6.50
40 Hz +3.2 dB vs. 3.46
50 Hz +2 dB vs. 2.21
80 Hz +1 dB vs. 0.98
100 Hz +1.2 dB vs. 0.27

We are within 1.5db of each other from 16hz to 100hz which would be a very reasonable difference from one RS meter mic to the next. What I would question is possibly his calibrated mic's accuracy at 10hz and 12hz. I have a problem believing that any RS meter mic is only -5db at 10hz or -6db at 12hz. Our correction values would be more rational when considering what the old ones were... according to the old corrections, 10hz was down 20db... we adjusted it to 27db. At 12.5hz it was down 16.5db and we adjusted it to ~21db.

Sonnie,

I will do more measurements soon. I have also noticed that the way you measure (manually, with program) affects on needed corrections. My corrections are good for TrueRTA, but not possibly for REW or manual measurements. I have both programs and I will do more measurements soon.

Of course it is also a fact that you can't find two identical RS meters.

And what comes to accuracy of my mic, I trust it, it has been professionally calibrated. It's better than ECM8000 even with no corrections at all. It only needs ~0.6 dB correction at 10 Hz.

I think the bottom line is (similar to what Ilkka stated a few posts above) if you want to make absolute certain you have accurate measurements, you should acquire a good mic or have your mic calibrated. Unfortunately I don't know of anyone who will professional calibrate and SPL meter mic for a reasonable price. Picking up a mic like the Behringer ECM8000, mic amp, mic cable, and having it calibrated, will add to your sub eq'ing about $150 or so. You still have to have an SPL meter.

Thanks Ilkka... I definitely agree on finding two identical RS meters. It would be a miracle.

I wonder how accurate the two mics are that we had our personal mics calibrated against. My ECM was calibrated against a ACO Pacific and your mic was calibrated against a B&K 4133. As I stated above... I see a lot of the ECM's being used that have been calibrated. It appears that all of the calibration files appear fairly close to each other with very slight variances. This makes me feel fairly confident with its' calibration as well.

I believe they were so accurate, that we don't have to worry about them. If there are any variations, they are probably well less than 1 dB. So calibrated ECM is a very good and accurate mic. Other variables cause much bigger differences.

I just stopped by Radio Shack to pick up an SPL meter. As others have pointed out they no longer carry the 33-2050. They have a 33-2055 which is the digital one as we all know, but they also had a 33-4050 which seems to be a new analog model.

My question is as follows - does anyone know if the 4050 uses the same innards as the 2050 and can therefore use the same correction tables?

Thanks,
McP

See post #51 in this thread ... when I mention Rodny's meter, his is the newer style like you are referencing. There was very little difference.

Beautiful.

Thanks much,
McP

How is this Scosche Boom Stick SPL1000 Competition Style Sound Level Meter at Walmart (search for "sound meter") compare to the one from Radio Shack?

sorry to beat a dead horse but there is a lot of differing information, just wanted to triple check:

I use a Digital RS meter, use the RCA output into my laptop and use RoomEQ. Since im using the output jack of the Digital meter, i do NOT apply correction values?

Wondering before I purchace a spl meter. The Radio Shack one is go enough instead of getting an expensive one for homeaudio ? Because even if the CV don't match as long as the speakers say the same db reguardless of correction thats what we are trying to do? If I'm not worried about graphs ? (for now )

Yes... you would need to apply correction values.


If you want to simply level match your speakers then no, you would need correction values. You really only need the correction values if you are tryig to level the frequency response of your speakers.

But Sonnie, I think we are all confused because of the above quote by Ilkka which is located at the top of this page. With your tests recently, have you actually tested a Radio Shack digital SPL meter vs an analog one?

I have not personally tested between the digital and analog, however, another forum member did and there was a difference in his testing.

I'm not sure on Ilkka's quote there. I thought the RCA output was in-line after C-weighting. I also didn't think C-weighting effected frequencies as low as what we are correcting. IOW's, corrections are needed for all the meters. Maybe Ilkka could elaborate because I may not fully understand myself.

How about the analog meter from RS? There's an older version.

I believe the model number for the "current" version is 33-4050. But what if you have the one from the 90s? It's model number 33-2050. Do you know if the corrections are the same for that model as well?

I believe that if a meter is calibrated to the "weighted curve" in should be similar in performance; i.e. RS, NADY, the blue and yellow fleabay one, etc. They all look to be built by the same OEM and then outsourced to others for sale. I've attached a scan of the curves with an explanation. HTH

Sonnie?, Ilkka?,

Any further comment about the recent posts by me and Irfan?

Also, Sonnie, you mentioned that you didn't think that the C-Weighting compensation affected the frequencies that we are trying to correct with subwoofer equalization. What frequency range does C-Weighting affect then?