DiyMobileAudio.com Car Stereo Forum banner

1 - 19 of 19 Posts

·
Registered
Joined
·
28 Posts
Discussion Starter · #1 ·
Ok so I finally got around to taking a computer out into the garage and setting up TrueRTA with my behringer mic. It's very cool to do sweeps and make crossover/eq adjustments and comparing the resulting graphs.

However, even with a 31-band EQ, the chances that your dips/peaks that need correcting are exact values on your EQ (50,100,1600,5000,etc) are rather slim.

For example, let's say I have a severe dip at 4912hz. On my P99rs the closest frequency to 4800 on my eq is 5000. So I start to bump up 5000 knowing that it probably won't achieve what I'm after. I do another sweep and there is and frequencies near 4912 (4850, 5000, etc) are starting to rise. Soon I will have peaks where I don't want them in trying to remove the dip at 4912. Do you see the problem here?

So I guess I'm just looking for input, techniques, and suggestions. I really understand now how some people can say that the crossover settings are more helpful as a tuning tool than the EQ. I wish on my P99rs I could set which frequencies to boost/cut, and be able to choose 31 of them. And also choose a Q value for each. Now that would rock!


-------------------------


On a side note, what do you think they are trying to convey here:

Phase adjustment
When the cross-over point value for filters on
both sides is set to –12 dB/oct., the phase is
reversed 180 degrees at the filter cut-off frequency.
In this case, reversing the phase assures
improved sound continuity.
From the P99rs manual
 

·
Registered
Joined
·
2,780 Posts
RTA helps you see what you are hearing. The rest is up to you on how you want to apply that. You may not be able to get the exact frequencies, but you can get close. Also, remember a graphic eq is just a parametric with a locked Q value, so frequencies adjacent to the one being boosted/cut are also affected to an extent.

On the pioneer quote, a 12db crossover slope puts the signal 180* out of phase. You need to flip the phase of one set of speakers so they play in phase with the other speakers.
 

·
Inappropriate Thoughts
Joined
·
4,005 Posts
The only advice I can offer is to be extremely careful when boosting frequencies to correct vehicle abnormalities. Any boost, 3 decibels or more, could result in that particular speaker's amplifier being driven into clipping. Even following the theory of doubling amplifier RMS power for every 3 decibels of boost can net undesirable results.

My method of tuning via parametric equalization is to cut versus boost for the reason previously stated!
 

·
Registered
Joined
·
28 Posts
Discussion Starter · #4 ·
12db crossover slope puts the signal 180* out of phase.
I guess I just don't understand why or how they are related. I understand how TA affects phase, but not a 12db slope.

Thanks for the replies so far!

Yes, one thing that was great about using the RTA, like I said, was SEEING what I'd been hearing. It REALLY helped me understand more about how the EQ and crossover settings affect the sound. It was the first time I'd ever done that.
 

·
Registered
Joined
·
28 Posts
Discussion Starter · #5 ·
Any boost, 3 decibels or more, could result in that particular speaker's amplifier being driven into clipping. Even following the theory of doubling amplifier RMS power for every 3 decibels of boost can net undesirable results.
I thought about that too. I try not to boost too far on any frequency, and I try to do more cutting than anything else. Thanks.. that confirms my thoughts about clipping with EQ.

Also....
graphic eq is just a parametric with a locked Q value
Yeah I could really see that on the RTA. That Q value is pretty narrow though. It barely affected surrounding frequencies, but I could see it.



Another thing about the RTA... I was REALLY surprised how accurate the sweeps were. As in.. if I clicked quick sweep a bunch of times, without changing any settings, and saving each one to memory and then comparing the results, the differences were extremely slight. I guess I expected more variance for some reason. I did them all at night in the garage so it was really quiet.


I remember reading about IMPRINT and seeing that it does 500 points of corrections along the frequency spectrum! I wish I could have an EQ like that. Anyone ever done a before and after RTA comparison of imprint? I'm tempted to hook up my 9887 and see how much it actually smooths out the curve.
 

·
Inappropriate Thoughts
Joined
·
4,005 Posts
I guess I just don't understand why or how they are related. I understand how TA affects phase, but not a 12db slope.
It's just the simple theory of the order of the crossover network. You can read everything you never wanted to know about crossovers at the following:
Crossovers
Passive Crossovers
Electronic Crossover

Thanks for the replies so far!
You're welcome.

Yes, one thing that was great about using the RTA, like I said, was SEEING what I'd been hearing. It REALLY helped me understand more about how the EQ and crossover settings affect the sound. It was the first time I'd ever done that.
Contrary to what the "trust your ears" crowd will say, the RTA is one of the few ways you can actually visualize what you are hearing. The reason you can't trust your ears is due to the fact that they are connected to your brain. You'd be amazed at how the brain corrects for what you hear!

Going further, I can't always trust my ears because they vary with my allergy symptoms and fluid build-up in my ears. Let's say I work on a friend's install with a sinus infection, I can still use the RTA to help me tune the setup to be listenable versus tuning it to sound good to my congested ears.
 

·
Registered
Joined
·
5,097 Posts
Just curious, what are your xo points? How much of a dip do you have at 4912hz?
 

·
Registered
Joined
·
283 Posts
Ok so I finally got around to taking a computer out into the garage and setting up TrueRTA with my behringer mic. It's very cool to do sweeps and make crossover/eq adjustments and comparing the resulting graphs.

However, even with a 31-band EQ, the chances that your dips/peaks that need correcting are exact values on your EQ (50,100,1600,5000,etc) are rather slim.

For example, let's say I have a severe dip at 4912hz. On my P99rs the closest frequency to 4800 on my eq is 5000. So I start to bump up 5000 knowing that it probably won't achieve what I'm after. I do another sweep and there is and frequencies near 4912 (4850, 5000, etc) are starting to rise. Soon I will have peaks where I don't want them in trying to remove the dip at 4912. Do you see the problem here?

So I guess I'm just looking for input, techniques, and suggestions. I really understand now how some people can say that the crossover settings are more helpful as a tuning tool than the EQ. I wish on my P99rs I could set which frequencies to boost/cut, and be able to choose 31 of them. And also choose a Q value for each. Now that would rock!


-------------------------


On a side note, what do you think they are trying to convey here:



From the P99rs manual
If you have 31 band EQ as your tuning tool, you can use 1/3 octave RTA. That should match what you see in the RTA graph with EQ frequency band.

If you have full parametric EQ (1Hz precision, 0.1 Q step, 0.1db gain step) that's another story...
 

·
Registered
Joined
·
4,372 Posts
Contrary to what the "trust your ears" crowd will say, the RTA is one of the few ways you can actually visualize what you are hearing. The reason you can't trust your ears is due to the fact that they are connected to your brain. You'd be amazed at how the brain corrects for what you hear!

Going further, I can't always trust my ears because they vary with my allergy symptoms and fluid build-up in my ears. Let's say I work on a friend's install with a sinus infection, I can still use the RTA to help me tune the setup to be listenable versus tuning it to sound good to my congested ears.
No offence but an out of calibration mic will yield poor results too.... Not to mention how the mic were placed when a RTA is made.....
 

·
Registered
Joined
·
5,234 Posts
No offence but an out of calibration mic will yield poor results too.... Not to mention how the mic were placed when a RTA is made.....
I think it's assumed here that the mic is calibrated and placed in the correct spot.:rolleyes:
 

·
Registered
Joined
·
28 Posts
Discussion Starter · #12 ·
If you have 31 band EQ as your tuning tool, you can use 1/3 octave RTA. That should match what you see in the RTA graph with EQ frequency band.

If you have full parametric EQ (1Hz precision, 0.1 Q step, 0.1db gain step) that's another story...
Good to know! Thank you
 

·
Registered
Joined
·
6,373 Posts
On a side note, what do you think they are trying to convey here:



From the P99rs manual
Crossovers change the phase, the steeper the slope the more they change it. To simplify you can get out of phase at the crossover overlap and it will cancel. Usually only an issue at the crossover points when multiple speakers (like mid and sub) are playing the same sounds out of phase. It can be a tuning tool, you may want them to cancel some. If you ignore it you might wonder what happened to your midbass after changing your slope.
 

·
Registered
Joined
·
5,097 Posts
Never did get my answer, but if you could give your xo points and maybe some pics for the install you may get better help in identifying the problem.

Assuming that you measured a summed for L+R signal, is the dip being caused by both L&R set of drivers? or is one side weaker than the other by 6db resulting in a 5db loss at this point.......big difference between the two scenarios.
 

·
Registered
Joined
·
2,780 Posts
Also, play a single pair of speakers and watch the rta. my midbass speakers are crossed at 200Hz/24db slopes, they are clearly playing things at 500Hz. Gapping the crossover points may help if you're dealing with an acoustic phase issue.
 

·
Registered
Joined
·
632 Posts
Crossovers change the phase, the steeper the slope the more they change it. To simplify you can get out of phase at the crossover overlap and it will cancel. Usually only an issue at the crossover points when multiple speakers (like mid and sub) are playing the same sounds out of phase. It can be a tuning tool, you may want them to cancel some. If you ignore it you might wonder what happened to your midbass after changing your slope.
Passive x-overs will.... active don´t... correct me if i'm wrong...
 

·
Registered
Joined
·
283 Posts
Passive x-overs will.... active don´t... correct me if i'm wrong...
Nah,
both active & passive change the phase at a rate of 90 degree per 6dB.
So, you need 24dB/octave to make it back to normal (90 x 24/6)=360=0 degree.
Or 48dB, or... (just calculate from the formula)
 

·
Registered
Joined
·
28 Posts
Discussion Starter · #18 ·
Never did get my answer, but if you could give your xo points and maybe some pics for the install you may get better help in identifying the problem.

Assuming that you measured a summed for L+R signal, is the dip being caused by both L&R set of drivers? or is one side weaker than the other by 6db resulting in a 5db loss at this point.......big difference between the two scenarios.
Yes you did get your answer, scroll up dude. It was just an example. The whole thread was only an example, I'm not trying to solve a specific problem, only asking about RTAs eliminating dips and peaks. LOL
 

·
Premium Member
Joined
·
6,214 Posts
Good read: http://www.diymobileaudio.com/forum/how-articles-provided-our-members/81137-rta-binaural-microphones-their-application.html - please read spatial averaging carefully...

For your system, look at the RTA, use your EQ and do your best to flatten any peaks.
Play with your EQ to boost the dip up some (as stated less than 3dBs) and check if you get any change in the response - no change? It's a phase problem so put your boost back to 0 and move to your next problem. A phase problem cannot be resolved with an EQ.

The EQ in your P99 is Graphical = fixed Q. It's a really good all-in-one HU that has separate L/R GEQ ; but if you need more EQ power, you need to add an external processor.

Kelvin
 
1 - 19 of 19 Posts
Top