REW active filter question

**finalrinse** · **09-16-07, 07:22 AM**

I was wondering, After I measure a sweep and find peaks, I then assign filters and upload them to the FBD. At that point I recheck levels, reset target, and measure a sweep again with the filters active in the FBD. But, are the filters that are active in REW also being output? So in effect am I double filtering? I guess I don't know if you can measure the loaded REW filters without the FBD? If this is the case, I've been doing it wrong.
Thanks,
Tom

**brucek** · **09-16-07, 08:03 AM**

When you take a measurement with REW, it exists without any REW filters. You can easily check that by looking at the filters screen with the EQ filters ICON. If you take a measure and create some filters, then it is associated with that specific measurement. Take a second, third, or fourth measure and there are no filters associated with those plots until you create them (or load a SAVED set into that specific measure). You can have multiple measurement plots present in REW at a time. Each has its own set of filters (or no filters). Keep the EQ filters screen up at all times to view them as you cycle through viewing the different plots.

The sweep of the measurement has no association with any filters. You are measuring your system, and if the FBQ has filters loaded, then the response is altered by those filters residing in the FBQ. REW plots the taken measurement and displays that raw data. There are no REW filters associated with it. Now, if you were silly enough to load a set of SAVED filters into that measurement plot, then yeah, there'd be double filtering of that plot - but you wouldn't do that. The only time you would load a set of SAVED filters over a measurement, would be if you shut the filters in the FBQ off and took a measure...........

brucek

**finalrinse** · **09-16-07, 08:17 AM**

Quote:

brucek wrote:

When you take a measurement with REW, it exists without any REW filters. You can easily check that by looking at the filters screen with the EQ filters ICON. If you take a measure and create some filters, then it is associated with that specific measurement. Take a second, third, or fourth measure and there are no filters associated with those plots until you create them (or load a SAVED set into that specific measure). You can have multiple measurement plots present in REW at a time. Each has its own set of filters (or no filters). Keep the EQ filters screen up at all times to view them as you cycle through viewing the different plots.

The sweep of the measurement has no association with any filters. You are measuring your system, and if the FBQ has filters loaded, then the response is altered by those filters residing in the FBQ. REW plots the taken measurement and displays that raw data. There are no REW filters associated with it. Now, if you were silly enough to load a set of SAVED filters into that measurement plot, then yeah, there'd be double filtering of that plot - but you wouldn't do that. The only time you would load a set of SAVED filters over a measurement, would be if you shut the filters in the FBQ off and took a measure...........

brucek

Thanks Brucek, I'm ok then. One other question if you have a moment, I'm still trying to understand the peaks between 80 and 200Hz. The filters do nothing to tame them and I understand that is because of the 80Hz crossover. Are they there because of room reflections and harmonics? The Find Filters always recommend many filters in that range, should I ignore that area and just concentrate on 20 - 80Hz for filtering?
Thanks
Tom

**Wayne A. Pflughaupt** · **09-16-07, 09:07 AM**

Quote:

Are they there because of room reflections and harmonics? The Find Filters always recommend many filters in that range, should I ignore that area and just concentrate on 20 - 80Hz for filtering?

Yes and Yes.

Actually, I don’t fool with any filtering close to the crossover point until I do a second sweep with both the mains and sub on. For instance, my crossover frequency is 90 Hz, so I don’t do anything above 75-80 Hz until I see what influence the mains are contributing above that point.

Regards,
Wayne

**finalrinse** · **09-16-07, 10:15 AM**

Quote:

Wayne A. Pflughaupt wrote:

Yes and Yes.

Actually, I don’t fool with any filtering close to the crossover point until I do a second sweep with both the mains and sub on. For instance, my crossover frequency is 90 Hz, so I don’t do anything above 75-80 Hz until I see what influence the mains are contributing above that point.

Regards,
Wayne

Thanks Wayne - you and Brucek are very! helpful.
Ok, here are 3 graphs - The first one is a baseline sweep of the subs with no filtering, the second one is with filters, and the third one is a full sweep with mains. I think I have problems???, especially in the 100Hz area. Am I reading this right? What should I do? Like you say, maybe I should back completely off on the filters above 70Hz? Are my mains not doing there job in this region, maybe change the crossover points, I'm currently at 80Hz. I can't change the mains (ceiling speakers, I know, Not good!), I can change the sub.
Thanks!

**brucek** · **09-16-07, 10:37 AM**

Quote:

ceiling speakers, I know

Yeah, and it does look like their natural rolloff is higher than the crossover you have set. It might be wise to raise your crossover to about 100Hz and give that a test.

If you really want to find the rolloff of your mains, you can do a mains only sweep (no sub) with the receiver set to MAINS LARGE (to eliminate the crossover temporarily).

Or you can just bump your crossover and see what it looks like.

brucek

**finalrinse** · **09-16-07, 01:35 PM**

Quote:

brucek wrote:

Yeah, and it does look like their natural rolloff is higher than the crossover you have set. It might be wise to raise your crossover to about 100Hz and give that a test.

If you really want to find the rolloff of your mains, you can do a mains only sweep (no sub) with the receiver set to MAINS LARGE (to eliminate the crossover temporarily).

Or you can just bump your crossover and see what it looks like.

brucek

Thanks Brucek - I bumped my x-over up to 100Hz, re-measured the subs, and adjusted the FBD. My front speakers have a +3dB bass boost that I thought I would also throw in there for good measure, literally! The objective data looks a little better; none of these settings hurt it for sure. Subjectively it sounds pretty good for what it is. I've learned quite a bit in the last few weeks. Because of the large open room I'm in my next investment will be a new SVS or HSU sub. I'm also wondering if an HSU MB12 mid bass sub might help in my weak area, but I can't really use it nearfield which is a strong recommendation. I will certainly start with a new large main sub first. Oh well, as long as we are having fun!
Thanks Brucek and Wayne,
Tom

**brucek** · **09-16-07, 02:02 PM**

Quote:

I'm also wondering if an HSU MB12 mid bass sub might help in my weak area

No.......

Quote:

I will certainly start with a new large main sub first.

Yes - SVS would be my recommendation.

brucek

**finalrinse** · **09-16-07, 02:06 PM**

Quote:

brucek wrote:

Yes - SVS would be my recommendation.

brucek

Sounds good to me!
Thanks

**Wayne A. Pflughaupt** · **09-16-07, 03:48 PM**

It may be that your mains, being ceiling speakers, don’t have enough output down low to blend with the sub. So I second brucek’s motion to do a full-range sweep so we can get an idea what they’re doing. For that graph, you’ll probably want to use some smoothing, maybe 1/3-octave or so.

Quote:

I'm also wondering if an HSU MB12 mid bass sub might help in my weak area, but I can't really use it nearfield which is a strong recommendation.

Nearfield??

Have those people ever heard of phase issues???

Regards,
Wayne

**finalrinse** · **09-16-07, 04:51 PM**

Quote:

Wayne A. Pflughaupt wrote:

It may be that your mains, being ceiling speakers, don’t have enough output down low to blend with the sub. So I second brucek’s motion to do a full-range sweep so we can get an idea what they’re doing. For that graph, you’ll probably want to use some smoothing, maybe 1/3-octave or so.

Nearfield??

Have those people ever heard of phase issues???

Regards,
Wayne

Thanks Wayne - Ok, Here are three graphs. The first one is my latest sub measure with FBD(I set the X-over back to 80Hz, seemed to sound better, maybe not so boomy). The second one is full range sweep with sub, and the third one is full range sweep without sub(mains set to large and sub off). What do think? Any hope at all?

**Wayne A. Pflughaupt** · **09-16-07, 05:30 PM**

Well, comparing your first and second full range graphs…

We can see in the second graph that the sub level has been raised, which probably sounds better. But looking at both charts, the region from 100 to ~250 Hz is about the same – no surprise there, since it’s the sub you’ve been tweaking. But comparing the full-range charts with your main-speakers chart, we can see they are dropping hard below ~250 Hz. Or else the room has some issues there, but giving that these are ceiling speakers, I’m putting my money on them. That’s not something that will be easily rectified, because as you’ve already found, running the sub higher to “fill in” makes it sound bad.

Regards,
Wayne

**finalrinse** · **09-16-07, 05:47 PM**

Thanks Wayne - yes, it is what it is. Well, when I build my home theater in the basement I won't be dealing with ceiling speakers. I'm still going to get a new sub and will take yours and brucek's advise on SVS, sometime before the Holidays. It has to be a box type and would like it to do 20Hz well. I'm thinking of the PB12-Plus, at least that's what my wallet is thinking.
Thanks!
Tom

Advertisement