Axiom Home Page Forums General Discussion Home Theater EP500 Response

Hey Folks, given the rainy day today I tried measuring the Freq respond of the EP500 using some WAV files of test tones [sine wave] I generated on my PC and burnt to CDRW. I'm using a Rat Shack SPL meter. Values have been adjusted based on the corrections for the Radio Shack meter given on the SVS site. Initially I played a 50 Hz tone and adjusted the volume to 90 db then took the measurements. As of yet I haven't seriously played with placement. This is what I got:


As I don't have a long enough cable yet to move the sub to the listening position I haven't tried crawling around to find the position that sounds best.

After having done this little experiment I'm wondering if I would be better off to do these types of measurements to try and find the position that offers the most linear response instead of just listening to some bass on my hands and knees and decide what sounds better.

Would taking this approach offer better or worse results do you think?

Given the EP500 is front firing in which direction should I face the sub when it is positioned at the listening position while determining optimal placement? Given LFE is non directional does it matter?

Any thoughts on how I can flatten this pic out...?

Cheers, Jag

Room dimensions are 129"W X 199"L X 83"H should this matter.

It looks like you've found a pretty good spot for it right where it is. There are no big nulls to deal with. They are the most difficult thing to fix.

You have a big drop off after 100Hz. Did you have your main speakers connected when you did these measurements? If so, then your sub level appears to be a bit too high.

Hi Michael_A thanks for the help!

No the mains where disconnected and I pushed the recievers crossover up to 150Hz. Is this OK?

I really have no idea what a good or poor in room freq response would be... Is it reasonable to hope for something flatter?

Is this measurement approach better than the listening and crawling approach? I was thinking of placing the sub in the listening position and then taking measurements like these at different places in the room to find the placement that would give the flatest response.

An thoughts on which direction the sub should face if I were to do these tests with the sub at the listening? Face forward or backward?

Or, should I just be happy with the way it is now?

Thanks a bunch, Jag

I think this is probably pretty good. They have huge anechoic chambers for a reason, you know

First guess would be that you have a pretty large room and the sub is near one or more walls. You could probably flatten it out by pulling it out into the room a bit if you want to experiment with something

Michael, you didn't describe the position of the 500 when you measured, but after doing a room mode calculation from the dimensions it looks like the usual suggestion(except for square rooms)to put it in the tighter corner(farther from a doorway or other large opening)behind the speakers should work well.

A consideration which is separate from the room modes is the boundary reinforcement which a sub driver gets when placed within a certain distance from the three nearest room surfaces(i.e. the floor and two nearest walls). There isn't too much available online discussing this, but here's a pretty good description in a slightly awkward translation from the Swedish. The basic point is that contrary to the advisability of keeping a full range speaker out of the corner and at different distances from the three room surfaces, for a sub the exact opposite applies. The orientation of the driver isn't of any real significance from the directionality standpoint, since those very low frequencies are omni-directional, but it should be positioned if possible so that the driver cone is less than about 22" from the floor and both walls. This results in the strongest and smoothest reinforcement from the three boundaries.

Thanks JohnK, this is of great help.

The position was at the front right hand side about two feet from the corner.

Last night I went ahead and took a bunch of measurements and got some very interesting results! Confusing but Interesting. Its going to take some time to sort through the numbers and to confirm things but I'll be posting them soon and would love your input.

At first assesment it would appear the sub freq response becomes incredibly linear ( even down to 12 Hz ! )when set to play at about 10db louder. ie Rather than initially setting the 50 Hz test tone to 90db as my begining point I set it to 100 Hz. This doesn't make sense to me yet but I'm going to try some more measurements and then see if I can figure out whats going on.

I did try the corner position and indeed noticed at least a 10 db gain in volume but no change in linearity of the response. In fact I couldn't find any position that changed much in the linearity.


I'll be getting back to you folks with more measurements in the next little while and would love all your input.

Thanks a bunch, Jag

You have the beginning of a great house curve. Tame the peaks at 40 and 100hz and you will have an awesome response curve.

OK Folk's here's some more thorough measurements. I think these are begining to show just what the EP500 can really do!
The SPL meter was mounted on a tripod and placed on the couch in the listening position. The sub was moved to the different positions indicated on the plot for testing. At each position the 50Hz test tone was played first and the volume adjusted to 90dbs for the first graph and either 90db or 100db for the second. Once the 50hz tone was adjusted, measurements were taken. Values again were compensated for the Rat Shack non-linearity given by the SVS site. Note: the Hertz are plotted from 125 to 10 (opposite to the above posted plot)



As can be seen all the positions in the room (original position= position in previous graph above) yield pretty well the same freq response with the exception of the placement in the Middle Front and in the Centre of the Room. The Centre of the room just plane sucks; clearly there's little benefit from boundary reinforcement. Other than an obvious increase in loudness due to boundary reinforcement, placement in either corner yeilded no improvement in Freq Response. The Middle Front seems very impressive to me particularily in the low end. There's one difference with this placement though; I really had to increase the overall volume of the sub (just like for the centre placement) in order to get a 90db reading with the 50Hz test tone prior to testing.

After getting this better curve with placement in the Middle centre I wondered if increasing the actual output from the sub had any influence on the Freq Response (even though the EFFECTIVE volume at the listening position was the same as the other placements)

Consequently, I decided to put the sub back into my "Original Postion" with out adjusting the volume output. This resulted in the initial 50Hz test tone being 100db (instead of 90db). In so doing I managed to maintain the smoother profile even in the different position!??!



Now I'm actually getting higher output down at 12 and 10 Hz than at 100! And the response has smoothed out quite a bit.

Questions:

1) It seems the Freq Response gets flatter when the sub is made to exert more. I figure that this is either due to the sub itself or some phenomena from the room. Does any of this make sense to you Experts? Would this be a characteristic of the sub or the room. Louder gets smoother...? Given the DSP on the EP500 I'd assume the sub should be equally linear at any volume???

2) Now as nice as I think the Freq Response is at this volume the problem is I can't imagine listening to it this loud and placement in the middle front position is not an option for me since the TV and equipment cabinet containing the DVD, and reciever sit just behind there. I actually saw the TV picture shake in this position. So any ideas how I can get around this? (other than ear muffs) ???

3) Any critique of my measurement method that may offer some more insight on how I get this affect. Louder=Smoother???

I very much appreciate all your input.

Cheers, Jag

The louder=smoother thing is not what I would expect. Might be something in the DSP programming, I guess...

Jag: to answer 1 and 3...:

Actually, that is exactly what I would expect given the type of measurement system he is using. What is going on is that the amplifier or the woofer is distorting, and thus raising up the total SPL. The only REAL way to test the true output when the amp is being stressed like that is to get an RTA and have that tell you on an individual test tone what frequency bands produce how much energy. You cannot isolate a frequency with a simple SPL meter.

Edit: Essentially you are doing a distortion test. The logic behind this is that the lower frequencies require more driver and more amplifier power to produce. When you up the volume by a few dB on the higher notes, the amp will not be clipping and the woofer will not do any distortion and so the volume will go up linearly. When you either reach amp clipping or woofer distortion, the frequencies at which it distort appear to be much louder due to the harmonics being played at a much higher level than usual. This keeps on rising exponentially until the amp either is too distorted and sounds like crap, the woofer bottoms out, or the sub's internal limiting circuitry decides that it cannot produce a louder tone.