PDA

View Full Version : Sierra-2 output measurements in my room



mikesiskav
10-07-2017, 10:57 PM
Another thread here got me interested in doing some output tests on my Sierra-2s in my room. I have 5 Sierra-2s but I just tested the front right channel along with my 2 JL F112 subs. All running off my XMC-1 with Dirac and XPA-5 amp. I measured it with Room EQ Wizard and my Umik-1 microphone. Here's the results:

https://s1.postimg.org/1y5lv0isrz/Compression_Levels.png

I didn't see any compression until the 95dB sweep (orange line). That was at -8dB on my preamp. You can see a little compression above 10khz and a little more around 150-400hz. I have a wide suckout in my room at the 150-400hz area so that's partially to blame. My subs are 5dB hot and I'm glad to see no compression at all and extension down to 15hz. When you start seeing compression, it means you're getting pretty close to the max output of the speaker, at least at those frequencies. But this is just with a single speaker. If you're watching a movie or listening to music, you're going to have multiple speakers playing simultaneously, which will increase the volume.

Since I have neighbors on the other side of the wall, I never listen louder than -13dB (purple line). Most of the time, it's more like -18dB (pink line). It's nice to see I still have some headroom available.

Edit: Forget to mention, the mic was placed on a tripod by my couch about 9ft away from the speaker.

New Measurement, the colors don't match the original picture.

https://i.postimg.cc/Jh99n32b/Sierra-2-and-JL-F112.png

davef
10-09-2017, 06:01 PM
Another thread here got me interested in doing some output tests on my Sierra-2s in my room. I have 5 Sierra-2s but I just tested the front right channel along with my 2 JL F112 subs. All running off my XMC-1 with Dirac and XPA-5 amp. I measured it with Room EQ Wizard and my Umik-1 microphone. Here's the results:

https://s1.postimg.org/1y5lv0isrz/Compression_Levels.png

I didn't see any compression until the 95dB sweep (orange line). That was at -8dB on my preamp. You can see a little compression above 10khz and a little more around 150-400hz. I have a wide suckout in my room at the 150-400hz area so that's partially to blame. My subs are 5dB hot and I'm glad to see no compression at all and extension down to 15hz. When you start seeing compression, it means you're getting pretty close to the max output of the speaker, at least at those frequencies. But this is just with a single speaker. If you're watching a movie or listening to music, you're going to have multiple speakers playing simultaneously, which will increase the volume.

Since I have neighbors on the other side of the wall, I never listen louder than -13dB (purple line). Most of the time, it's more like -18dB (pink line). It's nice to see I still have some headroom available.

Edit: Forget to mention, the mic was placed on a tripod by my couch about 9ft away from the speaker.

Thanks Mike -- as I mentioned in the other post - these are ridiculously impressive in-room response measurements, and your even using a 5dB scale.

Interesting that you are seeing a slight bit of compression at the higher tweeter frequencies. I would actually expect the opposite from the tweeter, with signs of compression starting in the lower ranges of the tweeter.

Keep in mind, microphone elements also compress - in fact, mics have a much more difficult time at higher frequencies compared to mids and lows. What mic did you use for this? Still, this is quite good considering the levels you are reaching at your mic distance with only a single speaker.

mikesiskav
10-09-2017, 06:44 PM
Thanks Mike -- as I mentioned in the other post - these are ridiculously impressive in-room response measurements, and your even using a 5dB scale.

Interesting that you are seeing a slight bit of compression at the higher tweeter frequencies. I would actually expect the opposite from the tweeter, with signs of compression starting in the lower ranges of the tweeter.

Keep in mind, microphone elements also compress - in fact, mics have a much more difficult time at higher frequencies compared to mids and lows. What mic did you use for this? Still, this is quite good considering the levels you are reaching at your mic distance with only a single speaker.

Thanks Dave. I did these measurements with Dirac engaged, so that's the reason for the nearly flat response. I should probably redo them without Dirac. I'm also using 1/6th smoothing.

The mic I'm using is the Umik-1 and I'm using a calibration file for it. The distance from the mic to the speaker was 9 ft.

What you said about the tweeter compressing near the lower range makes sense to me since there's more excursion there and that's actually what I was expecting. But based on my measurements I don't see any compression at all between 2-4khz, which is very impressive. Now that I think about it, I know that Dirac was boosting the frequencies above 10khz so that may be why I'm seeing compression in that range.

N Boros
10-12-2017, 12:40 PM
Nice measurements. I have Sierra 2s as well, so it is nice to see how much they can do in terms of these long term output compression plots. I'm curious though, where did you set the crossover for the subs?

mikesiskav
10-12-2017, 06:34 PM
Nice measurements. I have Sierra 2s as well, so it is nice to see how much they can do in terms of these long term output compression plots. I'm curious though, where did you set the crossover for the subs?

Thanks N Boros. My crossover is set to 80hz. That dip at 55hz is room induced. I might be able to fix it by experimenting with the subwoofers placement but I just haven't gotten around to it. The dip is fairly narrow so it doesn't bother me too much.

N Boros
10-20-2017, 09:04 AM
Does Room EQ Wizard have a way of measuring short term peaks to see where compression would set in?

I'm just curious because it seems like if someone was trying to play a movie at reference volume, that is 85 dB sustained and as much as 105 dB peaks. The 85 dB sustained appears to be fine in your room, but you started running into compression on the 95 dB sweep. That 95 dB sweep was more of a sustained measurement, than a short term burst though.

mikesiskav
10-20-2017, 12:39 PM
Does Room EQ Wizard have a way of measuring short term peaks to see where compression would set in?

I'm just curious because it seems like if someone was trying to play a movie at reference volume, that is 85 dB sustained and as much as 105 dB peaks. The 85 dB sustained appears to be fine in your room, but you started running into compression on the 95 dB sweep. That 95 dB sweep was more of a sustained measurement, than a short term burst though.

I failed to mention it in my original post, but these measurements were done with a fairly quick sweep (about 2 seconds in duration). These would probably be considered short term compression sweeps. I was worried that longer, sustained sweeps at high volume could damage my Sierra-2's and I didn't want to take that chance. I'm going to take some more measurements this weekend with Dirac disabled. I know for a fact that Dirac is boosting some frequencies and that is affecting the results.

N Boros
10-20-2017, 01:33 PM
I failed to mention it in my original post, but these measurements were done with a fairly quick sweep (about 2 seconds in duration). These would probably be considered short term compression sweeps. I was worried that longer, sustained sweeps at high volume could damage my Sierra-2's and I didn't want to take that chance. I'm going to take some more measurements this weekend with Dirac disabled. I know for a fact that Dirac is boosting some frequencies and that is affecting the results.

I could be wrong, but I think that short term peaks of 105 dB for reference volume are of shorter duration than 2 seconds. Probably a small fraction of a second.

davef
10-20-2017, 02:26 PM
I could be wrong, but I think that short term peaks of 105 dB for reference volume are of shorter duration than 2 seconds. Probably a small fraction of a second.

This is correct, typically for measurement purposes, the duration of short term peaks is typically 30ms or shorter. However, there is really no set standard for this. I would consider a 2 second burst to be sustained, not short term. I am not really sure RoomEQ Wizard is appropriate software for calculating short term compression measurements.

mikesiskav
10-20-2017, 03:09 PM
I'll take a look at the options in REW this weekend. I believe the option I selected was the shortest possible. Then again, I'm not an expert in the software.

mikesiskav
10-21-2017, 07:26 PM
I took some new measurements today, this time with Dirac turned off. And I have to say, the results are impressive. Doing the same 70-95dB sweeps in 5dB increments, I see almost no compression at all with the Sierra-2 at this level. I probably could have done a sweep at 100dB but I decided not to since it was already very loud and I didn't want to take any chances of breaking something. This is really impressive performance from a single bookshelf speaker. Again, this is measured at 9ft from the speaker.

For my subwoofers, without EQ, there's a broad hump around 20-30hz. I'm reaching 115dB in this area with my two JL F112s before compression sets in.

https://s1.postimg.org/6ghbrll4yn/Compression_Levels_No_EQ.png

davef
10-23-2017, 10:54 PM
I took some new measurements today, this time with Dirac turned off. And I have to say, the results are impressive. Doing the same 70-95dB sweeps in 5dB increments, I see almost no compression at all with the Sierra-2 at this level. I probably could have done a sweep at 100dB but I decided not to since it was already very loud and I didn't want to take any chances of breaking something. This is really impressive performance from a single bookshelf speaker. Again, this is measured at 9ft from the speaker.

For my subwoofers, without EQ, there's a broad hump around 20-30hz. I'm reaching 115dB in this area with my two JL F112s before compression sets in.

https://s1.postimg.org/6ghbrll4yn/Compression_Levels_No_EQ.png

These look great - really not too much correction and, at your listening distance, we are seeing the SPL's I would expect. Looks like there was about 3-4dB of HF boost being applied when Dirac was enabled, which cuts the headroom of your amp in half at the frequencies.

mikesiskav
10-25-2017, 03:53 PM
Thanks Dave. I suspect the high frequency roll-off may be caused by my extensive room treatments. I've got a total of 18 acoustic panels and a large, thick rug on the floor.

davef
10-25-2017, 09:11 PM
Thanks Dave. I suspect the high frequency roll-off may be caused by my extensive room treatments. I've got a total of 18 acoustic panels and a large, thick rug on the floor.

Were you measuring with the mic directly on axis with the tweeter? If not, this would cause the rather mild HF roll off. In addition, all rooms have HF losses - and since room EQ wizard uses swept sine waves for measurements, you can't effectively isolate the speaker's response from the room response so you are measuring reflections as well, which bounce back to the mic out-of-phase from the direct sound which causes cancellations.

That said, this is a very mild rolloff and is probably not worth correcting for using Dirac (if possible to avoid EQ at the upper range).

Even without Dirac engaged, the speaker's + room response looks fantastic and the big dip you see at 180Hz is just floor bounce.

mikesiskav
10-26-2017, 03:04 PM
Were you measuring with the mic directly on axis with the tweeter? If not, this would cause the rather mild HF roll off. In addition, all rooms have HF losses - and since room EQ wizard uses swept sine waves for measurements, you can't effectively isolate the speaker's response from the room response so you are measuring reflections as well, which bounce back to the mic out-of-phase from the direct sound which causes cancellations.

That said, this is a very mild rolloff and is probably not worth correcting for using Dirac (if possible to avoid EQ at the upper range).

Even without Dirac engaged, the speaker's + room response looks fantastic and the big dip you see at 180Hz is just floor bounce.

Yes, the microphone was pointed straight ahead at my TV, although my speakers are toed in quite a bit. Any way to fix the floor bounce dip? Bass traps on the floor? :)

N Boros
10-27-2017, 06:06 AM
Yes, the microphone was pointed straight ahead at my TV, although my speakers are toed in quite a bit. Any way to fix the floor bounce dip? Bass traps on the floor? :)

Put a whole bunch of thick blankets and comforters, so that you don’t have to take down the panels to check. Maybe it will make a difference or maybe not. If it does make a difference, then you might be able to look into a thick carpet pad.

eyecatcher
10-27-2017, 12:09 PM
A lot of variables come into play mostly room related. Kinda looks like it is bass management related but you said its at 80. Make sure any listening mode is disabled. Do you see the bounce if you measure just the fronts or just the subs? If you do it's a bounce or reflection. If you don't, you may be able to adjust the phase first on each sub to get them to their loudest/smoothest combined output with the mains. Then try adjusting the distance on each sub independantly to see if that helps the dip. Also if the tip of mic is pointed straight at the speakers make sure you use the 0 degree axis microphone calibration file otherwise most of the time you'd use the 90 axis for room measurements with it pointed straight towards the ceiling.

mikesiskav
10-27-2017, 01:41 PM
Thanks for the tips guys. I'm going to play around with it some more this weekend. I'll measure the subs and speakers separately. Also, I just realized the distance from my speaker to the side wall is almost the same as the speaker to the floor which could be causing a dip in the FR.

davef
10-27-2017, 04:57 PM
Yes, the microphone was pointed straight ahead at my TV, although my speakers are toed in quite a bit. Any way to fix the floor bounce dip? Bass traps on the floor? :)

To reduce floor bounce you are going to need put down some extremely thick absorption -- about 3 feet thick. I have attached a pic of what we do at our production line testing area. Obviously, this isn't appropriate for a home environment ;)

Floor bounce is unavoidable and is usually a high q dip - meaning steep but of small bandwidth. Since the average listener typically hears at 1/3 octave smoothing (at best 1/6th octave) -- the dip looks bad in measurements but is likely not audible.

The simplest way to confirm if this is floor bounce is to take the measurement, mark the frequency of the steepest point of the notch, then raise (or lower) the speaker by a foot and then retake the measurement. Did the frequency of the dip change? If yes, definitely floor bounce.

Also note that if you have a solid ceiling and the speaker is halfway between the floor and the ceiling - the "bounce" will be worse.

I need to stress this though -- it all gets averaged into the overall room response and isn't something anyone should be concerned with.

N Boros
07-31-2018, 07:15 PM
I took some new measurements today, this time with Dirac turned off. And I have to say, the results are impressive. Doing the same 70-95dB sweeps in 5dB increments, I see almost no compression at all with the Sierra-2 at this level. I probably could have done a sweep at 100dB but I decided not to since it was already very loud and I didn't want to take any chances of breaking something. This is really impressive performance from a single bookshelf speaker. Again, this is measured at 9ft from the speaker.

For my subwoofers, without EQ, there's a broad hump around 20-30hz. I'm reaching 115dB in this area with my two JL F112s before compression sets in.

https://s1.postimg.org/6ghbrll4yn/Compression_Levels_No_EQ.png

Oh man. It looks like your awesome measurements are gone now.

mikesiskav
08-24-2018, 08:43 PM
Just noticed this now. I'll see if I can re-upload them somewhere. Thanks for the heads up.

mikesiskav
03-28-2019, 09:57 PM
Here's an updated measurement. Single Sierra-2 and single JL F112 with Dirac calibration. Measured from my couch around 9ft from the speakers. With my settings and calibration, 95dB is no problem for the Sierra-2.

https://i.postimg.cc/Jh99n32b/Sierra-2-and-JL-F112.png

theophile
03-31-2019, 08:05 AM
Here's an updated measurement. Single Sierra-2 and single JL F112 with Dirac calibration. Measured from my couch around 9ft from the speakers. With my settings and calibration, 95dB is no problem for the Sierra-2.

FR sweep looks great Mike...top to bottom, you have it all covered (2.2 or 2.? multi-channel setup)!! ;)

Ted

N Boros
04-02-2019, 09:29 AM
Here's an updated measurement. Single Sierra-2 and single JL F112 with Dirac calibration. Measured from my couch around 9ft from the speakers. With my settings and calibration, 95dB is no problem for the Sierra-2.

https://i.postimg.cc/Jh99n32b/Sierra-2-and-JL-F112.png

Those measurements look great. Thanks for putting them up! From what I recall, before you had or 3 or 4 dB boost in the subwoofer over what you have now. How do you like the difference in the sound?

mikesiskav
04-02-2019, 10:07 PM
FR sweep looks great Mike...top to bottom, you have it all covered (2.2 or 2.? multi-channel setup)!! ;)

Ted

Thanks very much. I'm running 5 x Sierra-2 and a JL F112.

mikesiskav
04-02-2019, 10:11 PM
Those measurements look great. Thanks for putting them up! From what I recall, before you had or 3 or 4 dB boost in the subwoofer over what you have now. How do you like the difference in the sound?

Thanks N Boros. I'm very happy with the way the system sounds right now. I find only a small amount of boost in the bass sounds best for my setup.