Question about dual sub, rew and allignment

[sigbergaudio]

The one that puts out 2db more?

Could be either one, we can't see it when it is playing alone.

 

[Pieter1267]

Green is running both with right sub inverted. Purple is running both with left sub inverted. Orange is both with none inverted.

 

[Flaesh]

with what? winzip demands paid subscription.

https://7-zip.org/ for example

 

[Pieter1267]

https://7-zip.org/ for example

 

[sigbergaudio]

View attachment 421305

Green is running both with right sub inverted. Purple is running both with left sub inverted. Orange is both with none inverted.

That is ..very strange. Are you inverting in MiniDSP? And have you felt the drivers to check that both are actually playing?

 

[Pieter1267]

That is ..very strange. Are you inverting in MiniDSP?

yes sir. I also think if it was polarity, they wouldnt look so alike when measured individually.

 

[sigbergaudio]

yes sir.

have you felt the drivers to check that both are actually playing?

 

[sigbergaudio]

I guess we have to call Ghostbusters.

 

[Pieter1267]

have you felt the drivers to check that both are actually playing?

They are. If they dont, the line would be flat. The measurement is higher.

But i also suspect something strange with the windows or rew. The spl gauge sometimes gives the weirdest readings. Ghostbusters indeed.

Tomorrow more testing. Ill go a push the no1 sub back a bit more away from the corner. I cant walk and use a wheelchair so i cant really do a subwoofer crawl. Well, actually you would think wheelchair guys could actually crawl.

We

Will

Get

To

The

Bottom

Of

This!

 

[Keith_W]

One look at the .mdat and the answer is pretty clear. @Flaesh was right all along. I was so confident that it would be a phase issue that it was the very first thing I looked at.

Blue: L sub sans miniDSP
Purple: R sub sans miniDSP

The subs are 160 degrees out-of-phase with each other. The result:

Broad area of phase cancellation when the two subs play together. The highlighted curve (in dark purple with yellow glow) is the two subs playing together, and the lighter coloured curves are individual subs.

Now I will show you some proof that Trace Arithmetic A+B predicts the measured result. In red, we have L+R subs added together with REW. In purple, we have your measured result of both subs playing together. You can see the two curves look almost the same.

Now look what happens if we invert the polarity of one sub, and add it to the other sub. The result is shown in green. I am comparing it to the previous graph, with both subs un-inverted (red). Ta-daaah!! Looks much better, doesn't it?

BTW I am pretty certain that the graph you posted here is a complete BS result. It is impossible for you to invert the polarity of one sub and have the sum of both subs measure exactly the same as having both subs un-inverted.

It also appears that you are measuring in a very noisy environment. If you want to see for yourself, go to the waterfall window and extend the time window to 1000ms. The noise is the flat area of the waterfall. Click through each measurement and you can see the noise floor changing between all the measurements.

 

[Pieter1267]

One look at the .mdat and the answer is pretty clear. @Flaesh was right all along. I was so confident that it would be a phase issue that it was the very first thing I looked at.

View attachment 421337
Blue: L sub sans miniDSP
Purple: R sub sans miniDSP

The subs are 160 degrees out-of-phase with each other. The result:

View attachment 421338

Broad area of phase cancellation when the two subs play together. The highlighted curve (in dark purple with yellow glow) is the two subs playing together, and the lighter coloured curves are individual subs.

View attachment 421339
Now I will show you some proof that Trace Arithmetic A+B predicts the measured result. In red, we have L+R subs added together with REW. In purple, we have your measured result of both subs playing together. You can see the two curves look almost the same.

View attachment 421340

Now look what happens if we invert the polarity of one sub, and add it to the other sub. The result is shown in green. I am comparing it to the previous graph, with both subs un-inverted (red). Ta-daaah!! Looks much better, doesn't it?

BTW I am pretty certain that the graph you posted here is a complete BS result. It is impossible for you to invert the polarity of one sub and have the sum of both subs measure exactly the same as having both subs un-inverted.

It also appears that you are measuring in a very noisy environment. If you want to see for yourself, go to the waterfall window and extend the time window to 1000ms. The noise is the flat area of the waterfall. Click through each measurement and you can see the noise floor changing between all the measurements.

Thanks Keith.


I decided to look some more into REW manual and then check the windows settings. I saw so many strange readings in REW. It looked almost like REW was hacked or something.

For one, there was an enhancement on that was connected to the UMI1 input channel. REW warns about that in the manual ""windows these days, has a lot of nosie cancelling stuff for microphone inputs''. A-ha.

Also, I cleaned up the soundcard and other settings.

Then I ran again and now it already looks much more how it should be. See above. The red line is both subs. Much closer and above 40hz it overtakes at some point - as it should.

But ill continue too with your remarks for tomorrow. Thanks for the time and effort!

 

[Keith_W]

Very simply, you have made an error when you inverted the polarity of your subwoofer. You failed to invert it, when you thought that you did. Like I said, the graph you posted earlier is impossible. I would be looking into that if I were you. Once you fix it, the result will look like my last graph.

 

[sigbergaudio]

Was it DIY subwoofers? Perhaps the safest way is to open and change the wires, instead of trying through MiniDSP

 

[Pieter1267]

Very simply, you have made an error when you inverted the polarity of your subwoofer. You failed to invert it, when you thought that you did. Like I said, the graph you posted earlier is impossible. I would be looking into that if I were you. Once you fix it, the result will look like my last graph.

One setting, okay, but both invert options not working in MINIDSP? When I tried a few days ago too? I checked and double checked the connections, the wiring. The fiddling with invertion should show some kind of response. It did not at all. The UMIK1 SPL gauge in REW, went bananas at times. I saw 20db when both subs went at the same setting that previously showed 74db.

Even if you might have a point, I am possitive windows is messing up or has messed up REW. Ill need to make a clean install for the microphone files, the soundcard settings etc and then run again but just changing and turning off the microphone enhancements - some kind of voice filter, made a big difference.

That said, i hope you are right.

Ill make new measurements sans minidsp to be sure. Then ill add minidsp again and try to invert. Hold on, i think the inuke6000dsp amplifier for the subs, also has a polarity switch option.

@sigbergaudio they are both DIY. One i made myself (the 2, R)). The other (the 1, L)was assembled. I removed the screws but couldnt get the driver out. The one i made myself i double checked. If Keith is right then perhaps i do need to open the assembled one to fix it. I actually really hope so.

Thanks for the effort guys. Really appreciated.

 

[sigbergaudio]

One setting, okay, but both invert options not working in MINIDSP? When I tried a few days ago too? I checked and double checked the connections, the wiring. The fiddling with invertion should show some kind of response. It did not at all. The UMIK1 SPL gauge in REW, went bananas at times. I saw 20db when both subs went at the same setting that previously showed 74db.

Even if you might have a point, I am possitive windows is messing up or has messed up REW. Ill need to make a clean install for the microphone files, the soundcard settings etc and then run again but just changing and turning off the microphone enhancements - some kind of voice filter, made a big difference.

That said, i hope you are right.

Ill make new measurements sans minidsp to be sure. Then ill add minidsp again and try to invert. Hold on, i think the inuke6000dsp amplifier for the subs, also has a polarity switch option.

@sigbergaudio they are both DIY. One i made myself (the 2, R)). The other (the 1, L)was assembled. I removed the screws but couldnt get the driver out. The one i made myself i double checked. If Keith is right then perhaps i do need to open the assembled one to fix it. I actually really hope so.

Thanks for the effort guys. Really appreciated.

If both are "wrong" it will also work, so you could still test switching on the one you can open. I would perhaps also test to put them both in the same position, so next to each other or stacked on top of each other and see what happens.

 

[Pieter1267]

If both are "wrong" it will also work, so you could still test switching on the one you can open. I would perhaps also test to put them both in the same position, so next to each other or stacked on top of each other and see what happens.

Ah, but me being in a wheelchair that isnt easy to do.

I made some new measurements, lower spl as my wife sleeps but I think Keith and @Flaesh are right.

Green is both. Blue is L sub inverted. Brown is R sub inverted.

 

[Pieter1267]

EDIT: AND ITS CONFIRMED!! THIS BIG BLACK BOY DIDNT PLAY NICE!!!!

I did the ole battery test and voila, this guy went backwards.

And its because the builder didnt use Speakon, but open terminals and somewhere along the way, someone changed the red screw where the black one shouldv been. I changed them around and saves us all a lot of hassle opening stuff up.

Pfew. That also means that the one I made myself, IS working well. Anyway, I also did the battery test on that one, the R in the measurements, and that guy is fine.

New measurement also confirms. Below, the upper bright purple line is both subs, the brown line is the R sub, the dark pruple line is the now corrected L sub.


Thanks guys! @Keith_W and especially @Flaesh I owe you guys BIGTIME!

Time for a beer methinks.

 

[sigbergaudio]

Never realized it was a passive sub. Then you didn't need to open it up of course, just reverse the speaker wires.

 

[Pieter1267]

Never realized it was a passive sub. Then you didn't need to open it up of course, just reverse the speaker wires.

Yup, I had an SVS SB 1000 pro and contemplated an upgrade, I read so many good things about the Dayton DIY sub, the drivers and these days, a second hand D class PA amplifier is very cheap on the ebays of this world. I actually got a 2nd hand Behringer Inuke 6000DSP (2x1600 watt @ 4 ohm!!) with that black sub, total € 325. I put the SVS on ebay and sold within a day for € 475.

Many people dont wanna bother with REW, MSO, DSP, etc so SVS realy is fine and great value for money. Me, I'm not so scared of a little setback. And I learned so much.

 

[Flaesh]

Sorry, no time yet. In short:

Which is what I wrote about earlier)
This may sound good: