SVS AS-EQ1 Thoughts... - Page 206

Make sure. I can't have my 2 front subs each reach my MLP at even SPL ( separately) without turning up the gain on the right one significantly. Room boundaries just make it that way or I will overdriveit he right one by making it work harder. I gain match them ( take close readings) instead and as a pair they work together to hit my MLP at the SPL I want.

You have hit on one of the difficulties in calibrating two subs. By measuring the sub output with the SPL close as you describe, you are close to checking whether the subs are "gain matched", rather than "level matched". (The better gain-matching procedure is to place the sub at a spot in the middle of the room, set its output to 75dB with the SPL up close, and repeat the procedure with the second sub. It's important that both measurements be taken with the sub in the same spot in the room.)

Once the subs are gain-matched, you can measure each sub's level at the MLP. Depending on room placement, you are likely to get different readings for each sub, sometimes significantly different. That's when trying different placements can get the levels much closer together. Once they are close, then run the Audyssey calibration, but don't change the sub gain settings at the beginning of the calibration--let Audyssey determine the final trim setting to balance the subs' output.

Edit: based on your last post, I think we are following the same procedure.

Why is that, Jerry, if the mic is an inch from the driver? I may have a go at this gain matching malarkey myself...

As I understand it, Keith, the interaction between the sub and the room creates sound waves that influence the mic's response regardless of how close the mic is to the driver. Placing each sub in exactly the same spot in the room ensures that each sub's measurement is subject to similar room response influences. When gain-matching the subs, the gains should be set such that the combined output measures within the acceptable range when running the Audyssey calibration (75dB +/- 3dB), and no further gain adjustment should be made at the beginning of the Audyssey calibration.

I started gain-matching when I installed a third sub after reading discussions on the topic by Craig John and Mark Seaton. IIRC, Dr Hsu of Hsu Research also advocates this procedure. With all three subs gain-matched, I am protected from over-driving any single sub when listening to content with intense bass passages.

Thanks Jerry. I didn't realise that (the bit I highlighted).  But when you move the subs back to their proper positions, won't the result now be wrong, thanks to room gain, boundary reinforcement etc?  Or is that where you just let Audyssey do its thing?

I think I am pretty much gain matched anyway but I may well have a go at this, just for OCD-ness :)

Gain matching is the most recommended option for multi-subs. As I understand the process, firstly, place the SUB1 in the middle of the room, place SPL meter 1-2 feet from the subwoofer cone, generate sub test tone from the AVR to get the meter to read 75dB (adjust the gain knob accordingly). Repeat the process for SUB2, SUB3, and etc...

Now, with EQ1, there is a step for LEVEL matching. I usually do this step by placing the SUBS in their final location. With EQ1 mic in listening position, I allow EQ1 to generate test tone and I adjust the gain knob accordingly to get the EQ1 software to read as close as possible to 0dB ( AVR volume set to reference level).

My question would be, how do you gain-match with EQ1? I believe the EQ1 is doing level-match.

 

Gain-matching simply means that the subs are outputting the same absolute energy level, regardless of room placement.  You are correct, Audyssey will then level-match the subs as part of its calibration process.  In a perfect world where the subs were carefully placed, the resulting Audyssey trim levels would be equal, preserving the balance between the subs.  However, we rarely achieve perfect placement, so the Audyssey trim levels are likely to be somewhat different (hopefully not a huge difference). 

 

I believe Craig John recommends taking the difference between the resulting Audyssey trim levels, halving the difference, and then adding the value to one sub and subtracting it from the other, thereby maintaining the gain-match.  In my case, the resulting Audyssey trims differ by only .5dB, so this adjustment is unnecessary.

 

I know this could be described as a fringe-OCD process, but if you think about it, maintaining a balance between (or among) the subs makes a good bit of sense. 

Thanks Jerry. Got it.

 

Fringe-OCD isn't as good as the full-blown Total-OCD mode, but I guess it will do ;)

For those of you posting graphs where a null is eliminated/reduced by the AS-EQ1, are you measuring afterwards to see if there is really a difference?

Here is my first try at using the AS-EQ1. My AVR is a Pioneer Elite VSX-33 with MCACC and I am running a dual sub setup with a SVS PB12 plus and stacked (daisychained) SB12nsds acting as a single second sub.

While it does sound better, my ears and SPL meter are telling me that the null around 36Hz is still there. I have tried all possible placments of the subs and changing the room is not an option. After calling SVS, they told me that running the mic pass through, using auto EQ assist or using the sat channel was no longer needed. I just simply level matched the subs (using SVS software), ran the AS-EQ1 then ran Mcacc as told to me by Ed, Doug and Jack. I also have to run the stacked SBs about 6-7db hotter (after calibration)to match the perceived output of the PB12plus due to the layout of my room which opens to the rest of the apartment

filAS-EQ1 Output Certificate.pdf 204k .pdf file

Nulls cannot be equalized out. Lessened, yes. But only sub placement and treatments can reduce/eliminate nulls.

BTW, if you have a rectangular room (as opposed to some odd shape) and you can place the subs on OPPOSING walls a the midpoints, you can likely eliminate the nulls ... and peaks along with them.

Jeff

Would placement be along the short or long walls of the rectangular room? My listening area is setup along the long wall and the "room" opens up to the rest of the apartmet. Subs are placed in front on opposite sides of the mains at unequal distances from the Main Listening Position. This is after peforming the sub crawl with both subs and spending hours with an SPL meter and real traps CD. I even used music where the null was most noticable to to find the best possible placement. With the mains turned off, I would compare the output diferences between each bass note (or Hz level) close to the subs driver then compare the difference at the MLP.

I think the sub crawl is for only one sub. Too many variables with multiple subs. WRT placing dual subs, it doesn't matter if they are on the long wall or the short one. Your testing methodology is spot on for evaluating at the MLP.

Jeff

OK, I think I'm getting the hang of this thing now. Ran Calibration again this time eliminating the chair by my computer as a seating position. Bass sounds nice, tight and surprisingly even across the room. The null seems better too (at least for now) My only concern is that MCACC has my sub trim at +5 and I would like to add about 3db to the AVR sub trim level. I may also bump up the level of the SB12nsd a click or two each as well.

Hey guys just a process update on the sub bottoming situation. I did just as Keith said and switched the subs. The gains were swapped also and there was no issues with the bottoming noise. Just prior to that, I made sure I had the exact scene which makes th sub make the noise ( that scene is the one in xmen fc where magneto and professor x catches Emma frost in the Russian residence and magneto straps her with the metal from the bed. Just as soon as her diamond cracks that is where the sub bottomed). So when swapped neither sub made the noise. BUT when swapped everything seemed a little less intense than before. The bass didn't seem as strong and loud as before. Is there any reason for that no that we have diwcovered that the sub is not defective?I

+8dB for the sub trim is getting into dangerous territory IMO. I would raise the gain on the sub and run the EQ again. Ideally, I look for a sub trim setting of about -3dB or so myself - this then gives me the opportunity to run hot by 3dB if I want to and still have a trim at the 'ideal' 0dB. HST, I feel no need to run the sub hot but I have the headroom in hand if I want to.

 

How odd. Swapping the subs has fixed the problem?  Run Audyssey again and see what happens this time...

I don't have audyssey I have MCACC. Or do you mean the AS-EQ1?

Yes. In your system, that is "Audyssey."

Does anyone in here own a MCACC driven receiver? If so, could you pm me on how you did the gain matching process. Thanks.

 

Gain-matching is a procedure that is done BEFORE running any type of calibration, and is independent of the AVR.  Follow steps 1-8 described here:

 

http://www.avsforum.com/t/1141416/official-hsu-uls-15-thread/1980#post_21557746

Jerry, step 5 is what has me a bit puzzled. What volume should the AVR be at when adjusting the levels of the subs?

We are talking about the AVR's level-setting test tones. An AVR should automatically set master volume to reference (i.e "0") when outputting the test tones (at least my Denon does).

Follow these steps:

- Turn on the AVR's subwoofer test tone. The MV should automatically be set to zero. If not, set it to zero manually.
- Measure the first sub's output with the mic close to the cone (I would guess that it would be in the 80-90 dB range up close).
- Adjust the sub's gain to read 90 dB (I find this a good starting value).
- Remove first sub, replace it with the second sub, being careful that the sub's position is exactly the sme, and that the mic didn't move.
- turn on the sub test tone with MV=0.
- Adjust the second sub's gain to be 90 dB as well.
- Return the subs to the desired placement in the room (they are now gain-matched).
- Run the AS-EQ1 calibration.
- Note the recommended trim level at the end of the calibration. Set the sub channel trim setting in the AVR according to this recommended value.
- Done!

Ideally, the trim setting in the AVR should be +/- 3dB. If significantly higher or lower, then the 90 dB gain setting in step 3 above was too high or too low. Re-do the process using a higher or lower gain setting. For example, if the recommended AVR trim is -10, then try a lower gain setting in step 3, say 85 dB.

Does this make sense?

Yes Jerry, thanks again. It makes sense now and i shall try it this weekend and see what happens.Thanks

Anyone tried AS-EQ1 on Windows8? Will it work?

If it works on 7, it should work like any "7" app on 8.