Help me correct bass

[Love4sound]

Then, set the gains on the output channels of the minidsp to match the individual sub trims. I expect the measurements to then match.



I wouldn't give up on the minidsp until I can establish that it's working as intended. YMMV.

So after few experiments with delays this is what I was able to achieve. The first image has two measurements. The green is pure direct graph of both subs with 0 delays in the mini dsp. The red graph is with the front sub delay at 1.5 and the second sub at 7.5 which I randomly choose after running multiple combinations. The second image is the measurements taken after applying EQ in the mini dsp. 3rd is after applying 5 dB gain for individual frequencies with dips

 

[gp_freak]

@Love4sound btw, do you notice any perceivable drop in bass output when running via Minidsp even when bypassing filters ? I was reading somewhere that when minidsp is in the loop, there is a output drop of -6 DB.

 

[Love4sound]

@Love4sound btw, do you notice any perceivable drop in bass output when running via Minidsp even when bypassing filters ? I was reading somewhere that when minidsp is in the loop, there is a output drop of -6 DB.

I haven’t calibrated and started listening yet. And also for now listening at very low volume levels

 

[ankitbhargava]

I have ordered mini dsp and a umic. Found this in YouTube and I feel it’s the best guide from A-Z in calibration of multiple subs and integrating it with ur mains using rew and mini dsp. Totally 4 parts and he explains it clearly taking live measurements

This is the best to refer to however I think its best to be referred after watching the video from Home Theatre Guru. Would make it much easier to understand this one.

 

[Love4sound]

This is the best to refer to however I think its best to be referred after watching the video from Home Theatre Guru. Would make it much easier to understand this one.

This guy himself thanks them in his video and it’s from there where he got the idea. But in terms of clear instructions this guy has nailed it. From installation to settings he got everything covered.

 

[liverpool_for_life]

The red graph is with the front sub delay at 1.5 and the second sub at 7.5 which I randomly choose after running multiple combinations.

It is redundant to apply individual delays for the 2 subs. The delay on one sub is relative to the other (as long as they have the same distance setting in the AVR and elsewhere). So, add delays to the closer sub in 1 ms increments (you shouldn't typically have to use delays much beyond the distance difference between the subs) and look at the measurements.

As you do this, you should see a delay setting beyond which the combined response gets progressively worse (with either more or less delay). That is your ideal delay. Any PEQ should happen on the basis of the response for that delay setting. Feel free to post measurements if you need assistance.

 

[Love4sound]

It is redundant to apply individual delays for the 2 subs. The delay on one sub is relative to the other (as long as they have the same distance setting in the AVR and elsewhere). So, add delays to the closer sub in 1 ms increments (you shouldn't typically have to use delays much beyond the distance difference between the subs) and look at the measurements.

As you do this, you should see a delay setting beyond which the combined response gets progressively worse (with either more or less delay). That is your ideal delay. Any PEQ should happen on the basis of the response for that delay setting. Feel free to post measurements if you need assistance.

Oh ok so using two delays is wrong? But if u see the PD graph it’s almost identical to the avr pd graph and also after eq the response seems good.

 

[liverpool_for_life]

Oh ok so using two delays is wrong?

Not wrong, just unnecessary. To use your example, a delay of 1.5 ms on sub 1 and 7.5 ms on sub 2 is the same as no delay on sub 1 and 6 ms on sub 2.

But if u see the PD graph it’s almost identical to the avr pd graph and also after eq the response seems good.

My preference is to get the best possible response before EQ. And to avoid boosting, if possible. YMMV.

 

[Love4sound]

Not wrong, just unnecessary. To use your example, a delay of 1.5 ms on sub 1 and 7.5 ms on sub 2 is the same as no delay on sub 1 and 6 ms on sub 2.



My preference is to get the best possible response before EQ. And to avoid boosting, if possible. YMMV.

Ok I’ll try with 0 and 6 later

 

[liverpool_for_life]

Ok I’ll try with 0 and 6 later

Do experiment with other values as I described above. 6 ms may or may not be the optimum delay for sub 2. Thankfully, you have measurements to guide you.

 

[Love4sound]

Do experiment with other values as I described above. 6 ms may or may not be the optimum delay for sub 2. Thankfully, you have measurements to guide you.

I tried the below combos. Any reason u feel 6 is not ideal? Hope u saw the measurements in the previous post

sub 1 front right
0-10
Sub 2 rear left
0-10
BOTH SUBS
Sub 1 sub 2
0-10 and 0-10

 

[liverpool_for_life]

I tried the below combos. Any reason u feel 6 is not ideal?

It seems to me to be odd that a delay of 1.5 ms on sub 1 and 7.5 ms on sub 2 produced the best response. I'd expect that any combination where the sub 2 was delayed by 6 ms more than sub 2 ought to produce the same results.

If it were me, I'd simply zero out the delay for sub 1 and manipulate the delay on sub 2 to figure out the ideal delay.

 

[Love4sound]

It seems to me to be odd that a delay of 1.5 ms on sub 1 and 7.5 ms on sub 2 produced the best response. I'd expect that any combination where the sub 2 was delayed by 6 ms more than sub 2 ought to produce the same results.

If it were me, I'd simply zero out the delay for sub 1 and manipulate the delay on sub 2 to figure out the ideal delay.

Yeah even I was surprised but thats what the measurements show. I also boosted gain for sub 1 by 3 and sub 2 by 5 to get a matching response to avr PD graph

 

[liverpool_for_life]

I also boosted gain for sub 1 by 3 and sub 2 by 5 to get a matching response to avr PD graph

Let's start over. Level match the 2 subs at your MLP. Set the gains on the minidsp for each output channel at whatever values that allows them to be level matched. Zero out the delay on sub1 and post measurements with increasing delays (1 ms increments) for sub 2. Don't go too far beyond the point where the combined response keeps worsening.

 

[Love4sound]

Let's start over. Level match the 2 subs at your MLP. Set the gains on the minidsp for each output channel at whatever values that allows them to be level matched. Zero out the delay on sub1 and post measurements with increasing delays (1 ms increments) for sub 2. Don't go too far beyond the point where the combined response keeps worsening.

Ok will try this in configuration 2 so that the current graph isn’t disturbed.will do it later and update the response

 

[Love4sound]

Let's start over. Level match the 2 subs at your MLP. Set the gains on the minidsp for each output channel at whatever values that allows them to be level matched. Zero out the delay on sub1 and post measurements with increasing delays (1 ms increments) for sub 2. Don't go too far beyond the point where the combined response keeps worsening.

For some reasons I didn’t get the same response as shared in the above graph. Wen I measured there where massive dips. So I reset everything and loaded yesterday’s graph and settings and still same issues. Don’t know why as there was no changes made in room. So I tried taking new measurements and set a house curve but ended in vain. So as per ur suggestion started from first. Kept 1st sub at 0 and tested delays in 2 nd sub. 4 gave the best results which is in the 1st image(red graph) massive dip in 96 and moderate dip in 43 as well. Set house curve target at 70 which gave the best results and shown in 2 image which is the response after applying the house cure through mini dsp. Then ran audyssey. Sub response with audysey is shown in image 3. Then I took measurements to check if it’s integrated with the mains. 4th image is the response of LCR crossed over at 80. Green is the FR response and blue is the center. Not sure why those two nulls are present. Even with a single sub I didn’t have nulls at that frequencies. Also not sure why response is different some times I run measurements. For example I run a measurement for sub and immediately after few seconds I run the same it gives two different graphs. Also same happens when I turn off and on mini dsp. Audyssey is setting the gain for sub at -1.5 which is very high. So lot of scenes are boomy and don’t feel the bass to be smooth. I want to increase the gain in sub so audysey can set it at somewhere near -7 or -8. Now that subs are level matched at 75bd how to overcome this? How can I level match it in a way that audysey sets it at -7 or -8?

 

[liverpool_for_life]

Green is the FR response and blue is the center.

What is Green the FR response of?

How can I level match it in a way that audysey sets it at -7 or -8?

You can increase the level of your level match, from 75 to 78 dB. I don't believe that will fix your perceived boominess, however.

Looking at your results, I'd skip Audyssey for the subs. The minidsp response is pretty respectable (especially if you didn't boost anything). HST, given where a large portion of your un EQ'd curve resides, I'd still be interested to see how things look if the target curve were set to 72dB instead of 70.

 

[Love4sound]

What is Green the FR response of?



You can increase the level of your level match, from 75 to 78 dB. I don't believe that will fix your perceived boominess, however.

Looking at your results, I'd skip Audyssey for the subs. The minidsp response is pretty respectable (especially if you didn't boost anything). HST, given where a large portion of your un EQ'd curve resides, I'd still be interested to see how things look if the target curve were set to 72dB instead of 70.

Green is the Front right and left response crossed over at 80. If I set target curve at 72 there is a lot of nulls. 70 gave the best results.

 

[liverpool_for_life]

70 gave the best results.

Okay, cool.

Green is the Front right and left response crossed over at 80.

For HT, I'd focus on integrating the center with the subs.

a. Set the subwoofer distance in the sub 1 pre-out of your AVR to be the average of your 2 subwoofer distances from the MLP.
b. Confirm that the center and sub have phase integration issues (dips that aren't there in either center or sub alone appear in the combined response).
c. Adjust the subwoofer distance (1 ft at a time) and measure the combined response to see if you get an improvement in/around the xover region.
d. Leave the subwoofer distance at whatever produces the best combined response for the center. You shouldn't have to go past an overall change in distance of 7 ft to get the best response.

Enjoy!

 

[Love4sound]

Okay, cool.



For HT, I'd focus on integrating the center with the subs.

a. Set the subwoofer distance in the sub 1 pre-out of your AVR to be the average of your 2 subwoofer distances from the MLP.
b. Confirm that the center and sub have phase integration issues (dips that aren't there in either center or sub alone appear in the combined response).
c. Adjust the subwoofer distance (1 ft at a time) and measure the combined response to see if you get an improvement in/around the xover region.
d. Leave the subwoofer distance at whatever produces the best combined response for the center. You shouldn't have to go past an overall change in distance of 7 ft to get the best response.

Enjoy!

In the same graph u will find the Center response as well.The 4th image blue graph is center response crossed over at 80. Audyssey set delay at 18.2 feet and that’s what gave the best results with no phase issues