Step Response Simulation Behavior

I have not been able to create a clean step response simulation. No matter what parameter I change in the CPD, I continue to see high frequency content superimposed on the general shape of the step response. Any suggestions on how to smooth this out or is it just an artifact of my room acoustics? The response shown is for a linear phase target. Similar behavior occurs with a minimum phase target.

I would guess this has to do with room reflections for the lower midrange ... around 200 Hz.

Higher numbers will smooth it out a little, but you'll have to knock out those room reflections if you really want to clean it up. Have you analyzed it in REW yet? If you can export your measurement and post the two pcm files here I can take a look at it in REW for you. I'm reasonably good at reading the data to find reflections.

moedra:

I have attached the .mdat file from my initial setup sweeps. My setup is a stereo, 4-way setup (2 subs, woofers, mtm's). I was working with dtune and you were able to fix his response in very short order order, well done! Would you mind taking a look at my AL results in regards to this potential room reflection issue?

moedra said:

Higher numbers will smooth it out a little, but you'll have to knock out those room reflections if you really want to clean it up. Have you analyzed it in REW yet? If you can export your measurement and post the two pcm files here I can take a look at it in REW for you. I'm reasonably good at reading the data to find reflections.

Click to expand...

Sorry, here are the left/right pcm files

There are several reflections in your room that you should look into treating. The biggest one is sitting at 7.55 milliseconds in your ETC graph in both responses. Since it is reading on both sides, the issue is symmetrical. Check the ceiling, the floor, or front wall. It's not the exactly the same on the right, but it's very close. Could also be your side walls if the speakers are centered and the walls and treatment are mirrored/symmetrical. It's too close to be your back wall. The reflection around 3 milliseconds is also very similar on both sides. A desk reflection or something else close, like a coffee table maybe? The front wall? Symmetrical side walls?

At 5 milliseconds there is a reflection in the right channel that doesn't exist in the left. Check for something that is sitting on the left side of the listening area, or the right, that could be causing it. Could also be from asymmetrical walls or lopsided treatment. At 25 milliseconds there is another one, only in the right response. Since it is only present in the right speaker response, and it is later than 25 milliseconds, look first to the back left corner for this one.

See if you can track those reflection points down and treat them. Check your measurements to see if you can get them to disappear. They may be coming from places other than where I suggested, so do your best to hunt them down. Be sure to leave the mic in place while you do this.

Thanks for that great assessment. Confirms what I had wondered about my partial room treatments as related to odd geometry of my room. I should have remembered that REW is more versatile than just looking at FR plots.

sledwards said:

Thanks for that great assessment. Confirms what I had wondered about my partial room treatments as related to odd geometry of my room. I should have remembered that REW is more versatile than just looking at FR plots.

Click to expand...

No problem. I actually do this a lot. Let me know if you want assistance with your room or your target/filter. I don't mind helping. I learn just as much myself by helping others, and I love gaining knowledge about this subject.