67 Hz notch with KH 120 II

[lsanbourne]

Hello, I used Neumann MA 1 Alignment Software to configure KH 120 II pair in a poorly treated studio. After analysis it showed a massive dip at 67 Hz and very poor bass response. REW shows the same. Is the answer just to get a sub or is there some room treatment pitfall that could cause this?

The room has a desk (which will soon go away), some GIK panels (2 on right side, 1 on left), curtains on asymmetrical walls, mattress in back, and my laptop screen...not ideal.

As an aside, my living room setup improved when I used Wiim's iPhone based room correction. What's the easiest way to do this using my MA 1 measurement mic for best results?

 

[kemmler3D]

You can help fill in the notch / null by moving things around in the room, if that's an option. If that's not an option, you *can* try boosting with EQ to fill the null, but those speakers will get stressed pretty quickly at that frequency so it's only a realistic option if you never plan to listen loud. A more reliable fix would involve adding a sub.

Additional panels etc will not tend to work low enough to affect that frequency. If you are dead-set on treatment over sub, you might need to look into helmholtz or membrane traps tuned around 65hz. That said, you can probably get a solid secondhand sub for less money than effective room treatment will cost here.

 

[lsanbourne]

Thanks - I'm happy to get a sub, and first, move the room around. I'm just not sure how to look through the room for typical culprits for a 67Hz null?

 

[lsanbourne]

REW output attached with 1/12 smoothing.

 

[lsanbourne]

and KH 120 II spec. Seems like an issue at 67 Hz should be my room and not the speaker?

 

[staticV3]

REW output attached with 1/12 smoothing.View attachment 507161

That's w/o smoothing.

For an honest look at in-room response, I'd highly recommend using the Moving Microphone Method. Details here: https://www.audiosciencereview.com/...ltra-to-rme-adi-2-pro-fs-r.68904/post-2496560

and KH 120 II spec. Seems like an issue at 67 Hz should be my room and not the speaker?

Most definitely.

 

[lsanbourne]

Will look at that, thanks. How do I search for possible room culprits for a 67Hz null?

 

[Flaesh]

how to look through the room for typical culprits for a 67Hz null?

Measure in several different places.
What are the dimensions of the room and its layout?

 

[lsanbourne]

Thanks!!

The miserable little room:
3.3m wide
3.8m long
2.4m tall

I was seeing comb filtering before, so I moved speakers to front of desk and tried moving my whole setup back. I'm not sure if this is much better or not. the huge dip at 67Hz remains and I'm not sure if I will be able to fill that in with just a sub? I also see wild swings higher in the spectrum even post alignment. Definitely not good enough for high-end mastering work..

 

[janbth]

How do I search for possible room culprits for a 67Hz null?

I can think of two possible reasons for that null:

1. You have a room mode and the speaker is located in a null of the mode. You could try to use a room mode calculator to see if you get a mode at 67Hz. You could also try to temporarily put the speaker in a corner to excite all the modes, and see if you then measure a peak at 67Hz.
2. A boundary effect like SBIR. Check this by moving the speaker closer or farther from the wall and see if the frequency of the null moves up or down.

By the way, how does the L speaker look like?

 

[Flaesh]

Don't attach too much importance to this picture:

 

[kemmler3D]

I'm just not sure how to look through the room for typical culprits for a 67Hz null?

It's probably not an object in the room, dips like this are typically due to room modes, which are inherent to the room based on its dimensions. Moving the speakers or listening position can help but moving a shelf or something generally won't.

 

[lsanbourne]

Don't attach too much importance to this picture:
View attachment 507232

Awesome!!!! Thank you so much! How did you make this, and does this software calculate a "best" position in the room? The room is so full of desks and mattress and suitcases of recording equipment and such - it's a nightmare.

At the moment I'm able to do my audio mixing and just ignore the < 70Hz content but that won't be the case in the future. I also am finding it very uncomforrtable to work with laptop on lap instead of on desk - how do people avoid comb filtering with desks?! The loudspeakers are currently standing on piles of nearly identical-sized books, ear at center point between tweeter and main driver. There is a 1cm discrepancy in height of the two. I was about to buy the desk stands that Neumann makes, but then I figured I might someday want floor stands if I want bigger than 0.9m equilateral triangle...

Thanks a lot for the education from everyone!!!

 

[Defi]

That screenshot is from REW and it can be a good tool to find starting points for speaker positions. It assumes rigid boundaries (I... assume) and it is also hard to know what exactly to enter for surface absorptions, unless the walls are completely bare of course. But it would be informative if that null showed up on the simulator with your current positions entered.

 

[lsanbourne]

Thanks, this is super helpful!!! I used it and it predicted the null. The null reduces a lot if I increase floor absorption. I added a fairly thin shag carpet (2m x 3m perhaps) but saw almost no change.

I also noticed there is basically no place in this room where I can listen and have good frequency response. I'd have to have my head like way to the side of the room, or dead in the center, or something annoying like that. So do you think this room size is just doomed? I have a larger bedroom but it's a bit sad mixing and sleeping in the same room.

 

[Defi]

Do you have any panels on the ceiling? A couple of 4" panels above you with a 4" gap between them and the ceiling does get some action down to that frequency, and thicker is better if you can. I don't know if your chair/desk can be adjusted for height that might help. If it's a vertical mode problem then a sub on the floor might catch it as well.

The carpet won't touch anywhere near that frequency.

 

[lsanbourne]

I own only 3 GIK panels right now (model 242, 24.1875″ x 48.25″). I put 2 on the wall to the right of me, up almost against the corner, leaned at a slight angle against the wall, and one on the side of the wall at left. The left wall has 2 big windows and a big wooden desk. The mattress is at the back left wall.

Unfortunately just to make this room worse - there's a ceiling fan above the center where I am sitting. I am also renting so I'm afraid of ruining the ceiling. I could the desk more against the front wall (eg closer than the 0.8m Neumann recommends) and put some panels above me there. I also think I'm up for buying the 750 - I am mixing audio that includes very low frequencies and I guess that's my only hope of hearing them anyway?

 

[Flaesh]

Try playing around with RoomSim yourself. Compare the simulation with actual measurements. Often, such deeps are related to the placement along the length of the room. As shown in my pic. The similarity between the simulation and reality is greater in an empty, cuboidal room, and less in the opposite case.

 

[lsanbourne]

Thanks!! I was unclear - my observations above were based on using the sim:

> I used it and it predicted the null. The null reduces a lot if I increase floor absorption. ...

> I also noticed there is basically no place in this [simulated] room where I can listen and have good frequency response. I'd have to have my head like way to the side of the room, or dead in the center, or something annoying like that. So do you think this room size is just doomed? I have a larger bedroom but it's a bit sad mixing and sleeping in the same room.

However, I'm also limited in where I can be in this room and still have an ergonomic work space. For example, I eliminate comb filtering by bringing the speakers to the front of the desk, but I increase my shoulder and back pain and then can't work.

So, I'd love to find some solutions.

1. Before I buy thick panels and hang them to the ceiling, is there a way to "prototype" what impact this would have on the sound?
2. Would a subwoofer fill in the 67Hz dip? I could put the sub anywhere more easily.

 

[Defi]

you can add subwoofers to the REW sim, I would try that. You could try changing the height of it as well.

As far as modeling the treatment, you could try increasing the ceiling absorption to whatever coefficient the panel you want gets at 67Hz. I know you mention it goes down with floor absoprtion, did you try ceiling? The biggest square panel GIK sells at 9.5" thick (unless I'm missing something) is about 0.8 at that frequency.