[ajimboe]

hello, i have a dip problem when both left and right channel plays. wondering if can help with this problem.

my setup
-2 way active front, door and a pillars
- 12” sub behind passenger seat firing forward. (i need space at back for my equipment)

suzuki wagon blind van.

crossovers
tweeter crossed at 3.5khz-20khz
midbass from 70hz- 3.3khz
sub 20hz-60hz

mic used dayton audio imm-6 and audiotool app on android. ( yes is not perfect accurate but its what is available to me).

below are individual measurements of left and right channels. yes not perfect linear but is ok for me. target curve linear from 20khz until about 200hz then rises up.

and below are the measurements when both left and right channel plays.

L+R FULL are when all drivers are in phase( i think). quite linear down until 200-180-160hz where it dips.

L+R NOT PHASE i switched one of the midbass 180° and shows more significant dips and peaks. dip at 80hz, 250hz and throughout the midrange not quite as linear. so i assume the L+R FULL measurement are in phase.

the mic was placed at listening height, centered at the headrest roughly where my ears would be. moving the mic left and right few inches does not help with the dip. but putting the mic at passenger side gets rid of the dip.

how to fix something like this? or maybe i am doing something wrong with the timings at the dsp.

thank you

 

[My_Rav4]

Why do the rear windows locations have foam?

 

[ajimboe]

hi My_Rav4,

its a little van with no windows at the back. it also has a layer of sound deadening under the foam too.

if i didnt do that the rear of the car would have been all flat metal surface reflecting sound back up front. i also guess it helps reduce vibrations from the sub and quite down outside noises.

thanks

 

[INFRNL]

I just watched a video that talks about this a bit.
here
Hopefully it will help. Unfortunately i don't remember exactly where he talks about it.

 

[ajimboe]

I just watched a video that talks about this a bit.
here
Hopefully it will help. Unfortunately i don't remember exactly where he talks about it.

thank you

I just watched a video that talks about this a bit.
here
Hopefully it will help. Unfortunately i don't remember exactly where he talks about it.

Ive just opened the link and noticed its one of peter's videos. i am quite familiar with his content, however he doesnt really like to specify what he does once both L & R are playing. at around 13:00 he does mention about the timings of the drivers, but then says to read or research about it else where. I assume that i have to play around with the time allignment on each drivers so that they cancel out the phase issue at 180hz.

 

[Picassotheimpaler]

hello, i have a dip problem when both left and right channel plays. wondering if can help with this problem.

my setup
-2 way active front, door and a pillars
- 12” sub behind passenger seat firing forward. (i need space at back for my equipment) View attachment 327654
suzuki wagon blind van.

crossovers
tweeter crossed at 3.5khz-20khz
midbass from 70hz- 3.3khz
sub 20hz-60hz

mic used dayton audio imm-6 and audiotool app on android. ( yes is not perfect accurate but its what is available to me).

below are individual measurements of left and right channels. yes not perfect linear but is ok for me. target curve linear from 20khz until about 200hz then rises up.

View attachment 327645 View attachment 327646
and below are the measurements when both left and right channel plays. View attachment 327647 View attachment 327648 L+R FULL are when all drivers are in phase( i think). quite linear down until 200-180-160hz where it dips.

L+R NOT PHASE i switched one of the midbass 180° and shows more significant dips and peaks. dip at 80hz, 250hz and throughout the midrange not quite as linear. so i assume the L+R FULL measurement are in phase.

the mic was placed at listening height, centered at the headrest roughly where my ears would be. moving the mic left and right few inches does not help with the dip. but putting the mic at passenger side gets rid of the dip.

how to fix something like this? or maybe i am doing something wrong with the timings at the dsp.

thank you

If the dip is only in combined response, it's a phase issue. See how your issue when combined gives 2 dips? There is a main suckout, and a second that is a multiple of the main. You have a comb filter in the combined response.
It may be due to your measuring technique, but much more likely that you either have your time alignment off by quite a bit, or you have reflection issues. If you are sure your driver's are in phase and your time alignment is spot on by either tape measure or IR measure than your problem requires a couple of all pass filters.

 

[INFRNL]

thank you



Ive just opened the link and noticed its one of peter's videos. i am quite familiar with his content, however he doesnt really like to specify what he does once both L & R are playing. at around 13:00 he does mention about the timings of the drivers, but then says to read or research about it else where. I assume that i have to play around with the time allignment on each drivers so that they cancel out the phase issue at 180hz.

I could have sworn he got into a little more detail, but you are correct; he doesn't give out exactly what you need.

hopefully picasso's advise will help as well. I haven't made it this far yet, so I have no clue. I haven't gotten into DSPs yet but I will when I get my project truck done.
I hope you get something figured out, or at least get it to sound how you want.

 

[Fulcrum]

Its a reflection from a boundary, most probably the roof of the car - a 180Hz coincides with about 70-75cm and is quite common in home audio. Whatever you do it will be there as its tied to the wave of the affected loudpeaker and has little to do with delay (little as the two doors are further apart so any cancellation would be lower in frequency. To verify if its this - add good amount of absorption (fiberglass insulation would do better as it absorbs better the lower spectrum) or some hard diffuser to scatter the wave and see if it changes.

 

[lithium]

I dont see the dip on the individual L and R measurements, or am I missing it?. Are you using a correlated pink noise track for the L+R measurement?

can you hear the dip at 180Hz? you should notice the sound stage pull to one side or a drop in response.

 

[ajimboe]

If the dip is only in combined response, it's a phase issue. See how your issue when combined gives 2 dips? There is a main suckout, and a second that is a multiple of the main. You have a comb filter in the combined response.
It may be due to your measuring technique, but much more likely that you either have your time alignment off by quite a bit, or you have reflection issues. If you are sure your driver's are in phase and your time alignment is spot on by either tape measure or IR measure than your problem requires a couple of all pass filters.

thank you for replying.

Yes I agree. I have two dips, one big one at 180hz then a little peaky at 250hz and the other dip at around 400hz.

I am sure that both midbass drivers are in phase right now.

I measured the time alligntment from head position at driver seat to the center of the speaker cone on each of them. input the measurements in tracerite.com to get them in ms.

 

[ajimboe]

I dont see the dip on the individual L and R measurements, or am I missing it?. Are you using a correlated pink noise track for the L+R measurement?

can you hear the dip at 180Hz? you should notice the sound stage pull to one side or a drop in response.

Yes there is no dip on individual measurement at 180hz. however it appears once both L and R are playing. (I added text on the image L+R FULL). I used both uncorelated and corelated pink noise and both have same response with the dip being visible.

actually to be honest I used a frequency generator and set to 180hz, it plays it ok. played a sine wave from 100 to 500hz and didnt notice it reduce in db at 180hz from my ears. but why does my readings show a dip. could it possibly be a mic location problem at listening position.

 

[ajimboe]

thank you for the reply guys. i am fairly new to this car audio thing for maybe about 10 months and a lot more for me to learn. i will try and make an update for you guys if necessary or if you are interested.

 

[Picassotheimpaler]

Yes there is no dip on individual measurement at 180hz. however it appears once both L and R are playing. (I added text on the image L+R FULL). I used both uncorelated and corelated pink noise and both have same response with the dip being visible.

actually to be honest I used a frequency generator and set to 180hz, it plays it ok. played a sine wave from 100 to 500hz and didnt notice it reduce in db at 180hz from my ears. but why does my readings show a dip. could it possibly be a mic location problem at listening position.

Could be mic location like you said.
What dsp are you using?

 

[lithium]

Guys, think about it for a second.

1. 180Hz sound wave is about 6.25 feet long. Mic placement just needs to be in the general vicinity of the head to measure response for the listener.
2. When measuring individual sides the null doesn't appear

So why is there a dip when measuring the combined response?

The answer is there is a phase difference between the left and right side at the microphone position (assuming a correlated pink noise track is being played).

If you're not using correlated pink noise the phase of the signal to each side is not the same. So in this case you're just measuring the sound file.

On top of that, when playing back tones (OP should try 1/3 octave pink noise as well perhaps) at 180 hz OP can't hear a null. If you can't hear it that there isn't a null. Meaning the measurement is incorrect.

 

[ajimboe]

Could be mic location like you said.
What dsp are you using?

yes could be mic location at driver side. when mic is placed on passenger side headrest area the 180hz dip is gone. so it is capable of playing 180hz. just not on driver side which is weird. im going to try play around switching the time allignment as if it is left hand drive. my car is right hand drive, so currently left tweeter is furthest so it has 0ms delay.

i am using dsp from a brand called codia, its a local brand here. my model is 450.6. has 6channels with 4x50watt amplified channel. 31 eq bands fully parametric. time allignment settings. just a basic dsp.

 

[Picassotheimpaler]

yes could be mic location at driver side. when mic is placed on passenger side headrest area the 180hz dip is gone. so it is capable of playing 180hz. just not on driver side which is weird. im going to try play around switching the time allignment as if it is left hand drive. my car is right hand drive, so currently left tweeter is furthest so it has 0ms delay.

i am using dsp from a brand called codia, its a local brand here. my model is 450.6. has 6channels with 4x50watt amplified channel. 31 eq bands fully parametric. time allignment settings. just a basic dsp.

So no AP filters I assume.
And you can't hear the dip when listening to band limited Pink noise or a tone gen?

 

[ajimboe]

sorry i just had a listen again and i do hear the dip at 180hz playing tone generator. move my head to the passenger side and it gets more audible. so the dip is there at driver side.

 

[ajimboe]

just an update maybe the solution for me. playing with the time allignment on the midbass helps. yes i should have known this was the issue and what per from pssound is talking about when the pair plays and he says things about timing both driver.

from my initial delay of 0.95ms on right midbass, adding small increments of 0.1 ms on right side midbass does help. until about adding +1ms so total 1.95ms the dip completely gone, but ruins the staging pulls a bit to left side and creates other dips in the midrange. just about fine tuning at this point maybe. idk just messing around now. i dont really know .

 

[lithium]

You should only set TA by distance. If you have REW you can do an impulse measurement to see how drivers line but but for 180hz you're going to get pretty close with just a tape measure.

If you can hear the null its likely a cabin mode related issue. This is the typical midbass - door mounting issue you'll find folks talk about. I'm surprised why its not showing up on microphone but that doesn't really matter, if you hear it than it exist.

 

[Picassotheimpaler]

You should only set TA by distance. If you have REW you can do an impulse measurement to see how drivers line but but for 180hz you're going to get pretty close with just a tape measure.

If you can hear the null its likely a cabin mode related issue. This is the typical midbass - door mounting issue you'll find folks talk about. I'm surprised why its not showing up on microphone but that doesn't really matter, if you hear it than it exist.

Odd that it wouldn't show in individual measurements if it was a cabin mode though.