Help needed to finish studio build designed by John

[limitedheadroom]

Not sure if this should be here or in the acoustics sub forum, feel free to move it if that's a better fit.

A couple of years ago John to designed a studio for me, and I've been busily building it since, occasionally contacting John with questions when they arose. I recently tried to get in touch and of course was shocked and saddened to learn that he had passed away. I've now reached a critical point in the build and need some help. I've just built the speakers into the front wall, and taken the first measurements of the control room with REW. There are definitely some issues but I have no idea what to do about it to make things better.

So I really need some help to finish this project now and hoping people here with more knowledge than me will be able to advise, for many reasons I'm running out of time to get the place finished and open.

The room is about 4m from the speaker positions to the rear wall, by roughly 4.5m wide at the front, 5.5m wide at the rear. Ceiling height at the rear is approximately 3.3m. I'm not sure what other info to give, so please let me know what is relevant, I've attached an image of john's sketchup of the control room for reference.

The control room here is my immediate concern but I'd also like advice on what the best method is to take measurements of the other rooms, and what, if any, adjustments to their acoustics might be advisable. I have a live room, a drum room and a vocal booth, but measuring these is of course not as obvious as a control room.

Thank you in advance, let me know what further information would be helpful.

I can't attach the REW file, apparently too big, so added a screen grab of one of the traces from that also (1/24 smoothing applied), I hope I've done it correctly, it's the first time I've used it and tried to follow the guide in the forum here.

[limitedheadroom]

Thinking about adding some details. The studio is built in and old cotton mill. It has extremely thick brick outer walls, it's on the 4th floor, again, the construction of the floor below is extremely solid. The studio rooms are built within this as per John's design. Inside out walls with double layer fire board. All joints sealed with acoustic sealant etc. Due to space restrictions there isn't a lot of depth behind the rear wall treatment as you can probably see on the sketchup. I have a reclaimed hard wood floor down that I managed to get very cheep.

The construction is basically all done as per the sketchup, I've just not covered the walls or ceiling in cloth yet, just the plastic covering John suggested keeping the mineral fibres at bay and keeping the HF alive in the room.

I've clearly got a big issue in the room at around 97Hz, and I can hear it clear as day, but generally the low end is very lumpy when listening, and the entire frequency response is not nearly as even as I would like so I hope some improvements can be made.

[DanDan]

REW has become very complex and the squiggling graphs are usually very disappointing to newbies. You could copy the .mdat and remove measurements from the copy until it is small enough to post. We only need Mixing position, Left Speaker, Right Speaker, Both.
Equipped with that, I am pretty sure we can see the responses in their best light, and manage expectations. e.g. Floor Bounce is kind of unavoidable in this type of design. What are the speakers?

[limitedheadroom]

Thanks DanDan,

I've attached the three measurements suggested, had to do each individually to get in under the 3mb attachment size limit, a bit inconvenient for people to view but kind of out of my hands there.

I have Neumann KH310 speakers, with KH80 subs, a pair of subs and I've mounted the subs with the speakers in the wall, one sub per side working only with the speaker on it's own side. I realise this may be an unconventional approach, but through listening tests with the subs I realised that when they are sitting together with the speakers like this they sounded great. So I decided to try this, they effectively operate as a larger 4 way, full range speaker in this configuration.

However, when taking these measurements the subs are not connected, this is just the KH310s on their own so as to be sure there weren't any other factors confusing the measurement. I've attached a photo of one side showing it as they are at the moment, the cloth frame still need to be put over the lower part, but that's just aesthetic. There are bass hangars in the space under the speakers, and the rock wool you can see on the lower half is in front of OSB backing behind the frame uprights.

There is a space above the window that still needs slots building to it. I was wondering if I try and tune them to 102Hz which seems to be the worst problem at the listening position. I'm guessing big null at that frequency is a standing wave. The space above the window is approximately 450mm deep, so it should be deep enough to tune to 102Hz going by John's Helmholtz calculator excel sheet.

Thoughts on this plan? And any other suggestions for improvements? Looks like there are also issues at 43Hz, 350Hz and 630Hz.

It looks from the waterfall plot like I've actually got to much energy in the room above 3K, so maybe actually removing some of the plastic covering the Rockwool, cover it in 2 layers of cloth instead?

Thanks

[DanDan]

Very unusual measurements. Let's deal with my biggest concern first. You have 200mS Decay 80-400Hz But 400mS above that. Also the 310s seem inclined to roll off at 40Hz. Mine are still happy down at 35 and even 30. This and the HF lingering would translate to a very bright harsh room tone. Does it? Or is this a measurement mistake? Is the room fully treated now or close to it? Some pics might help. In the ETC I see some reflections sticking up around 4mS and 12mS. They need to be found and suppressed if possible. If it is Floor Bounce, there is a case to be made for placing the subs on the floor.

[limitedheadroom]

I wonder if removing the plastic over the insulation will help with the longer HF decay, should I try this and take a measurement? I was concerned that it might be thicker plastic than ideal, it is pretty heavy duty. Any other suggestions to try?

I'll also double check I didn't catch the LF roll off on the speakers when I put them in the wall, in case this is the reason they roll off higher than yours. Not sure what else could cause this.

Attached are some images of the room, wide angle camera so perspective is a bit warped but it shows the state at the time the measurements were taken.

[DanDan]

Unless the long HF Decay is caused by some error or mistake, it seems likely that the plastic is at fault here. My CR is mall and mostly concrete. If your space has light boundaries then this difference might explain the LF roll off.

[limitedheadroom]

OK, I was able to spend some time today to look into the extended HF decay, and it was indeed caused by the plastic. I pulled it off gradually in stages at took tests as I went, but it wasn't until all of it was off that the HF decay was the same as that of frequencies below 1k.

So I now have bare insulation on the ceiling and rear walls, I'll cover it in cloth. I've added some images below showing the change, as you can see the RT60 times are looking a lot more consistent across the frequency range. I also had a listen to some of my references tracks, the imaging feels much tighter as a result and it all feels less aggressive. A great improvement!

Since my last post I've built the desk that will house the mixing desk, racks and touch screens etc. This is a significant piece of furniture in the room, so I think it's good to have it in place now for measurements. It seems to have helped mitigate the 100Hz hole that was there before, definitely not as deep as it was previously. So I suspect you are absolutely right that is caused by floor bounce. The desk will have reduced the strength of the waves bouncing back off the floor slightly.

What are your thoughts on these measurements now? What else can I look at to help improve things? There seems to be something really odd going on around 10k when I use L & R together, either speaker on it's own is much flatter up the top end there. Any ideas what might cause this?

Also I am not sure why the low end rolls off so quickly below 40Hz when you say your speakers roll off more gently. I guess I will try the speakers on stands in case it is something odd in the way I built them in to the wall, lost for other ideas as to what it could be.

Thanks
Dan

[limitedheadroom]

I forgot to upload the last set of measurements from today in that post, but I just tried. A single measurement saved mdat file is 3.2MB but 3MB is the maximum upload size allowed.

Maybe it is because I took the measurements at 96K? I don't know why, but my first attempt (at 44.1) everything rolled off very steeply above 14k and I couldn't work out why. I switched to 96K and took a measurement and it went all the way up to 20K. I changed nothing else other than the sample rate of the measurement.

But because of the file size limit I can't post the mdat files from todays measurements. Any way to make them smaller?

[DanDan]

Very interesting. How thick was that plastic? I just tried an .mdat of over 13M. It failed and I got an error message saying max upload is 12MB. My CR is concrete. The lowest mode is 35Hz so I expect that is giving me some LF extension. I recommend you sort out your 44K problem. If you are using a UMIK-1 I would go 48K. Omit Loopback and go as simple as possible. Be sure that you do not have C Weighting engaged in the Mic prefs. Nor HPF either in prefs or an external preamp.

[limitedheadroom]

How is it that you get a 12 meg upload limit? Does it increase with input to the community? MB in return for posts?

Yeah, the plastic was fairly thick, not at all like shopping bag plastic. That's why I was concerned about it, but it was there and so I used it. I've been building this thing on a shoe string, so any chance no to have to buy more materials has been taken.

I'm using the measurement mic that came with sonarworks, It may not be the best measurement mic in the world, but comes with a calibration file and again, it's what I have. I'll try your suggestions, won't use loopback and go with 48k and take another set of measurements. Will upload monday evening.

[limitedheadroom]

While I'm redoing the measurements for the control room, what's the best method to measure the recording rooms? of course you're not listening to speakers in there and there is no set configuration. Best to put a speaker in one corner and a mic somewhere near the far side of the room or something like that? I'd like to get an idea of the two recording spaces and see how they are so I can move on getting the cloth on the walls if they are sounding OK.

Thanks for any advice on what method to use here

[gullfo]

try to do a few measurements only and/or save the main ones into a separate file (copy -> remove extraneous ones). people can do the rendering of the various bits later. this should keep the file to under 5mb in most cases.

[limitedheadroom]

Thanks for the tip Glen. That is what I tried though, the upload I tried at 3.2mb was just one single measurement (ie. just the left speaker) not a collection of 2 or 3 measurements. But the forum will only allow me to upload 3mb as a max file size. Now I had looked at the measurements, as in opened them in the different visualisations. So I wonder if the rendering of things like the waterfall plot etc increase the file size.

Unfortunately I didn’t get a chance to retake the measurements yesterday due to family matters, but am on my way to the studio now armed with all the tools.

I won’t look through the measurements until after I’ve saved them. Fingers crossed I can keep them under 3mb this time!

[gullfo]

maybe host the file somewhere (like Google Drive, OneDrive or a file hosting service like DropBox etc) and post a link to it?