My New SonoSub Graphs ...

[salvasol]

I just finished building a SonoSub, I took some measurements and here is the graphs...

First graph is the final setup (after some tweeking to what YPAO did) ...orange sonosub with mains and gold sonosub alone



Second graph is comparing new sonosub to old subs (Velodyne VRP1000 (front) and TSS750 (rear)) ...orange is what YPAO set up, gold final setup (after tweeking YPAO), purple Velodyne alone and green Velo and TSS playing together ...



All this measurements were taken on the back row (which is not the best seat for audio because is to close to back wall, but is my sweet spot for video ) ...I will prepare another for the front row (and I assume it will be better because is in the right place for best audio).

Suggestions will be apreciated ....Do I need a BFD???

[brucek]

I would say not at that position.

What's the waterfall look like (using the same scaling).

brucek

[salvasol]

brucek wrote:

I would say not at that position.

What's the waterfall look like (using the same scaling).

brucek

Here you go ...first is sub only and second sub+mains

[brucek]

There is still quite a bit of information present at 300ms in the 15Hz-40Hz area.

Does it drop below 45dB by 600msec?

brucek

[salvasol]

brucek wrote:

There is still quite a bit of information present at 300ms in the 15Hz-40Hz area.

Does it drop below 45dB by 600msec?

brucek

I will post new graph tonight ...I left my laptop at home, and is where I have REW .

[salvasol]

brucek wrote:

There is still quite a bit of information present at 300ms in the 15Hz-40Hz area.

Does it drop below 45dB by 600msec?

brucek

Sorry, I didn't have time and phone/internet service yesterday (burglars cut all phone wires to get into a neighbor business) ...here is the graph (I need to learn to interpret them )

Sub only ...


Sub with mains

[salvasol]

Here is the measurements for front sweet spot ...

Front row sweet spot setup...purple sub with mains and gold sub alone


Waterfall graph, sub with mains


Waterfall graph, sub alone

[brucek]

here is the graph (I need to learn to interpret them

With a waterfall you're basically looking at a response measurement with the time domain added . If you've selected a time range of 600msec, then it will be in 30 slices of time of 20msec (600ms/30) each after the sweep.

Slide the Slice control over to 0 and you're looking at your frequency response graph. Now move to slice 1, and you're looking at the sound after 20msec, and so on to 30 slices at 600msec. If you see portions of the sound still persisting (ringing) at 600msec with any level that you might actually hear above room noise (> 45dB), then you can see that you would like that decay time to be reduced if possible.

Take a look at one of your plots below. Look at between ~15Hz-30Hz. Even after 600msec (that's over a half a second) there is still signal at ~64dB. Try an experiment and play a sine wave with the REW generator at 30Hz at 64dB. You can hear it quite well. So, you can imagine how that ringing at 30Hz affects the sound.

One thing you have to be sure of is that the ringing signal is just that and not background noise in the room from a furnace or fridge etc.... Usually a test with the Spectrum anayser (set up to listen to noise) will tell you that, or simply using your own ears to be sure you measure in a quiet environment..

Anyway, you get the idea. One of the advantages of a parametric filter is that you can match the filters against those peaks and correct the decay quite a bit, if not completely...




The last set of plots look better though....

brucek

[Ricci]

What mains are you using and are they being crossed over? Seems strange that adding your mains in would be affecting the response and decay below 25hz so much if they are crossed at 80hz. Even if they are running full range, they would have to be very bass capable. I don't remember seeing that with my measurements.

The louder your measurements are the longer they will take to decay. You are hitting 100db in some spots. Try taking a waterfall at 85 or 75db. It'll likely look much better.

[salvasol]

Sorry for this newbie question ...(maybe it was already asked, but I'm lazy today to look for the answer . )

Do I need to re-calibrate the sound card and remove and add the SPL calibration file to the new REW version (4.11) when upgrading from 4.10???

I don't know if is my computer, or when I upgraded to ver. 4.11 I didn't do it right, or the steps I'm following to take measurements ...but I'm having some problems .

I have a Yamaha RX-V2700, Samson Servo 600 to power fronts and BKA1000 to power sonosub.

I turn everything on, set level (75db reading on SPL), take measurements fronts and sub, it works fine.

Then I turn off the Samson (instead of disconnecting the fronts) to measure the response of the sub alone ...but here is the problem, sometimes after doing this I get the "Low level error" (I just adjust the volume) and sometimes I don't get any sound from sub (not even when I adjust the volume to maximum, and I never get a low level error)

Am I going crazy??? ...Do I have some ghosts in my HT??? ...Or, What I'm doing wrong???

Please help ...I did some measurements of just my fronts yesterday, there's some response issues when I run the sub alone and when I add the fronts, but couldn't do the sub alone to compare it (Usually, I try to measure everything with the same set up, instead of combining new and old measurements).

[brucek]

Then I turn off the Samson (instead of disconnecting the fronts) to measure the response of the sub alone

When you do this, you must run through the Check Levels routines, etc., before taking another measurement, since the existing level setup enjoyed the energy from the fronts, that are no longer in the mix now.

When you load a new version of REW I always select the "Delete Settings and Shutdown".

brucek

[salvasol]

brucek wrote:

When you do this, you must run through the Check Levels routines, etc., before taking another measurement, since the existing level setup enjoyed the energy from the fronts, that is no longer in the mix now.

I did it, and didn't get any sound from sub ...I even adjusted volume to max and nothing
I also lowered the gain to minimum, and it was the same.

When you load a new version of REW I always select the "Delete Settings and Shutdown".

I will try this. I will start from scratch (remove REW from computer, then re-install everything again).

Thank You

[salvasol]

Ricci wrote:

What mains are you using and are they being crossed over? Seems strange that adding your mains in would be affecting the response and decay below 25hz so much if they are crossed at 80hz. Even if they are running full range, they would have to be very bass capable. I don't remember seeing that with my measurements...

I'm using JBL Stadium (36Hz-20KHz), I took some measurements but I will re-doit again (see previous posts), I compared response when used as small and large fronts, XO 80Hz ...I'm using them as large because is how I get a better response in the front row seat; but in the back row seat the response is better using them as small

This graphs is comparing YPAO set up (PEQ engaged green) and no PEQ (blue), front set as large


This is with front set as small (same PEQ engaged (red) and no PEQ (green))


This is comparing sub with large mains (blue), sub with small mains (green) and large fronts (red)


This compares large mains (red) and small mains (blue)




Because I'm a little crazy, I'm using two pair of speakers connected in parallel for each side ; and to be sure that has nothing to do with this responses I measured just one speakers on each side (see response below) ...I had to adjust the master volume to get the reading because level was low ...blue is one pair, red is two pair in parallel.


Do you think that the separate amp I'm using to power the fronts has something to do with this??? (Samson Servo 600 300WPC @ 4 ohms)

Sorry for not using the standard graph (15-200 and 45-105) ...I want to show a lower frequency response

[salvasol]

This are my final setup graphs ...

Back row sweet spot ...


Front row sweet spot (I was expecting a better response here)