PDA

View Full Version : Sub response measurements (RS SPL and ECM8000 ETF compared)



BradJudy
04-03-2005, 07:01 PM
This is cross-posted to AV123 here: http://forum.av123.com/showthread.php?s=&threadid=8129

I got around to doing FR measurements of my sub (Hsu VTF-2) this weekend. I posted some thoughts in another thread that I will copy-paste here to keep things together.

First, I charted my sub using my RS analog SPL meter in a few situations (180 phase port A plugged, 180 phase port B plugged, 0 phase port B plugged, 0 phase port B 1' from corner, 0 phase port B 2' from corner) and then I measured it using a Behringer ECM8000 mic, M-Audio MobilePro and ETF 5.x software (0 phase port B in corner and 180 phase port B in corner).

I'll copy and paste my prior post about the impact of phase and position, but for now, here's a pic of the different measurements:

http://www.bradjudy.com/AV123/ETFvsRSSPL_forum.JPG

The ECM8000 line is somewhat arbitrarily placed in relation to the two RS lines - I placed it close to the other two for easiest comparison.

What do I take from this comparison? That the RS values are pretty good for general EQ work and that the corrections are probably not that critical.

For my personal use, I will probably need to do a significant wide cut centered around 30Hz and a narrow boost around 60Hz.


Quoting my earlier post in another thread:

I just did some measurements with jimsfield's 1/12 octave tones (big thanks for those) and my RadioShack SPL meter, and found a couple of interesting items.

First was the phase. I have a Hsu VTF-2 which has a 0/180 phase switch instead of a variable phase knob. I noticed that when set to 0 I had pretty seemless transition to my mains, but when set to 180, the transition turned into a massive hole (as to be expected). I didn't expect the hole to be down 20db at 100Hz (my receiver is set to an 80Hz x-over)! Make sure you all check your x-over range. The phase switch had no impact below ~63Hz in my testing.

Second was placement. I have my sub in the front left corner of the room. I tried moving it forward along the wall and found that the drop in response was more pronounced at higher frequencies (almost no effect below 40Hz). In the end, the corner was much more even than one or two feet into the room along the wall (I only charted these two additional positions).

Lastly, I tried applying the correction factors (interpolating for 1/12 octave) and I don't think I really find them that important. I'll know more once I run tests with my nicer equipment. But, now it's time for weekend chores.

Here are the two raw ETF graphs for 0 phase and 180 phase. Note that the absolute values (e.g. 90db) are not valid since this system is not calibrated, but the relative values (one point is 5db higher than another) are valid. Also note the different scales from the graph in the first post. ETF limits the options on the X axis scale and 20-200Hz is the best option.

0 Phase:

http://www.bradjudy.com/AV123/ETF0phase.JPG

180 Phase:

http://www.bradjudy.com/AV123/ETF180phase.JPG

It's crazy what a little phase does. :D

curtis
04-03-2005, 07:51 PM
Wow Brad...that is a good argument for me just to use my RS Meter and not worry about getting TrueRTA or ETF when I EQ my sub.

bikeman
04-03-2005, 07:55 PM
"Second was placement. I have my sub in the front left corner of the room. I tried moving it forward along the wall and found that the drop in response was more pronounced at higher frequencies (almost no effect below 40Hz). In the end, the corner was much more even than one or two feet into the room along the wall (I only charted these two additional positions). "

Just to show how much influence the room has on a sub, I had done the same thing that Brad did. I started out with the sub in the left front corner and moved it about 4 feet out along the wall. I wound up with a 2db drop below 40hz but with a much flatter FR. I was then able to EQ using only two filters. I'm looking forward to trying this same experiment with a new sub and comparing the results. Now if my wife lets me have some of the tax refund.... http://forum.ascendacoustics.com/images/smilies/wink.gif

David

PS: Brad, my DDE 3.0 will be here Tuesday. Hint, hint...http://forum.ascendacoustics.com/images/smilies/smile.gif

BradJudy
04-04-2005, 06:22 AM
Wow Brad...that is a good argument for me just to use my RS Meter and not worry about getting TrueRTA or ETF when I EQ my sub.

Yeah. While they can differ by about 5db in relative measurements, that ECM8000/ETF combo is rather expensive (although it can do a lot more than this). I just got some links to free measurement software and I will probably test some of it and see if any of them are good. If they are, that cuts the cost a lot since the mic and a pre-amp total about $100 and the software is an extra $75-150+.

Of course, it's a lot easier to tweak and retest with ETF since the test takes five seconds and is automatic instead of manually taking down meter readings and charting them in Excel.

Next up, add measurements with the Aperion SPL meter to the mix. In a quick test before, the Aperion meter was as much as 10db different than the RS one.

David, I'm looking for your manual, but I haven't had any success. I'll be pretty annoyed if I lost my electronic copy.

BradJudy
04-06-2005, 08:36 PM
A copy-and-paste from another post on my measurements:

Aperion SPL meter NOT recommended for low freq
Ugh....I just finished doing another set of low frequency measurements, this time with my Aperion SPL meter (which I was given at the RMAF).

The good news? According to Aperion, my response is +/-3db from ~25Hz to ~63Hz.

The bad news? According to both the RS SPL meter and my ECM8000/ETF setup, this is completely inaccurate.

The real problem appears to be a complete lack of sensitivity below 100Hz on the meter. It was measuring levels 10-30db lower than my RS meter! Most of my RS measurements were in the mid-upper 80dbs and the Aperion ones were in the mid-low 60dbs. Once it got above 100Hz, it started to get closer to my RS meter measurements.

This meter should definitely not be used for EQ measurements and even when used for level matching would lead to a MASSIVELY hot LFE channel.

Ugh....perhaps I got a bad apple, but ugh!