Using an SPL Meter to adjust audio

Joined
1 June 2005
Messages
948
Location
Happy Little Tree, Indiana
Per the recommendation of prime members, I bought a Radio Shack SPL meter. I wanted some advice on how to better use this. I have an Alpine Head Unit, and a kenwood 9-band EQ with 9-band EQ, and a parametric EQ. I have an JL 300/4 amp. Link for the amp...http://mobile.jlaudio.com/pdfs/7111.pdf

see pic...

attachment.php


I take it I need to start as flat as possible. I will level all setting on the head unit and the EQ. So, on the JL, it has a crossover with adjustments from 50 to 500 Hz, 1x and 10x frequency range, and a 12db and 24db filter slope.

1. How would you recommend starting out with the amp settings?

I am downloading some test tones of various frequencies at a 1/3 octave step. The SPL meter has a range of 60db to 120db. So, do I take a test tone at, for example, 300Hz, and adjust the db of that tone to a particular level, and then adjust all sounds at a particular frequency, to the same level?

This is crazy confusing for me! :biggrin:

Basically, I want to get better sound, but don't want to sacrifice 10 hours of my life just to find it sounds like crap! :smile:

Thanks for your help!
 
SPL meters are used with equalizers and pink noise generators to get a flat frequency response in a particular room, usually recording studios. You don't need this shit to tune your equalizer in a car. Just play with the sliders until it sounds good to you. Everyone's ear "hears" sounds differently due to our individual anatomy. The curve you end up setting on the eq using tones or pink noise and an SPL might sound awful to you. Just adjust the damn thing manually.

I see you have an Alpine head unit. Does it have the time correction feature? That will make a big difference when set correctly because it will properly balance the sound stage for your position in the car. RTFM! :biggrin:
 
Now that is what I wanted to hear, no pun intended. The more I look at this thing, it seems about worthless, and a lot of work to get something going.
 
djdrock said:
Now that is what I wanted to hear, no pun intended. The more I look at this thing, it seems about worthless, and a lot of work to get something going.

You can have fun with the SPL meter and record the volume of your farts after eating different foods and then regularly PM the results to the Prime members that recommended the silly thing to begin with. Or you can just go back to Radio Shack and get a refund.:smile:

By the way did you see that I updated my original post asking about your head unit's time feature?
 
Hugh said:
...I see you have an Alpine head unit. Does it have the time correction feature? That will make a big difference when set correctly because it will properly balance the sound stage for your position in the car. RTFM! :biggrin:
Agree with you wholeheartedly on the initial comment re set-up Hugh

On the Time Alignment feature I don't know that you'll get a whole lot out of a 2 channel set-up - however I have my tweets & mids bi-amped with separate channels to each component & with 4 way TA the result is staggering. Of course only works for the driver (i.e. moi!) but what do I care about my passenger? :biggrin:
(to be honest only have the audio on very low if at all with a passenger in car anyway)
 
You really need an RTA (real time analyzer) to do this right. Even after tuning the system flat it can sound like crap. A good sound shop can tune it for you and get you pretty flat which I am not sure if that Radio Shack SPL meter is suited very well for doing so.

I think you have the sub in the front right? You can start off with a pretty high crossover frequency on the sub ~200 Hz. It will put less stress on your mids and make it easier to tune IMO. But I agree with Hugh, tune it to how you like it. Everyone's ears and preferences are different.
 
Hugh said:
Just play with the sliders until it sounds good to you.

Definitely don't listen to Hugh. :)

For anyone other than someone who truly knows what he is doing and has experience with tuning systems, doing the correct thing is difficult to do without test gear.

The fact that people hear differently doesn't mean that you should randomly start messing with what the recording engineer has done, the objective is to tune a system to perform correctly first and foremost... minor tweaks to adjust to one's tastes and hearing sensitivites can then be done but they too shouldn't be done until one is well familar with the system.

Djdrock, give me a list of all the equipment that you have installed in the car and I will guide you. Assuming you have 6.5" speakers in the door and a 10" sub up front... set it all flat and set the X-over at 1X not 10X. I would start the x-over frequency at about 100 Hz, and a slope of 12/db Octave.

Start running through your test frequencies, and get a pen and paper and jot what you get down. Use C Weighing on the meter and "slow" response, and bring up the volume to 75-80 dB and start running through it all just changing the frequency and seeing what you get for numbers. Be patient... the meter will fluctuate... especially with bass. This is normal. Try to see what you get as an average.

Once we know where the peaks and dips are, we try to flatten it as much as possible just by adjusting crossover frequency and slope. What is left can then be EQ'd, to a certain extent.

The goal isn't to have a ruler flat response when you are finished. The human ear will not hear that as very good actually... but a flat response is what you want to get the gear dialed in correctly... I will show you how to adjust after you get a fairly flat response. Can you reply to me with the #'s?

Your main area of concern should actually be the bass region, because that is where the cabin will most interact with the sound. The mids and highs are not as deeply affected and with decent placement, should be fairly flat to begin with. So don't get too worked up with the higher frequencies yet.

This may be a bit time consuming, but if you want a good result then its a must-do. Its time consuming because you don't have access to the right gear thats all. But it doesn't matter. We can still get it done.

By the way this is not a hobby for me but my line of work.
 
Last edited:
I forgot to mention, use the meter correctly. Hold it at about ear level with the mike pointing straight up, don't point it towards any speaker. Do you have the digital or analog model?
 
D'Ecosse said:
Agree with you wholeheartedly on the initial comment re set-up Hugh

On the Time Alignment feature I don't know that you'll get a whole lot out of a 2 channel set-up - however I have my tweets & mids bi-amped with separate channels to each component & with 4 way TA the result is staggering. Of course only works for the driver (i.e. moi!) but what do I care about my passenger? :biggrin:
(to be honest only have the audio on very low if at all with a passenger in car anyway)

No, that particular Alpine unit has no time correction feature.

Hugh said:
You can have fun with the SPL meter and record the volume of your farts after eating different foods and then regularly PM the results to the Prime members that recommended the silly thing to begin with. Or you can just go back to Radio Shack and get a refund.:smile:

If I fart on the SPL, do you think Radio Shack will let me return it? :)


Malibu Rapper said:
You really need an RTA (real time analyzer) to do this right. Even after tuning the system flat it can sound like crap. A good sound shop can tune it for you and get you pretty flat which I am not sure if that Radio Shack SPL meter is suited very well for doing so.

I think you have the sub in the front right? You can start off with a pretty high crossover frequency on the sub ~200 Hz. It will put less stress on your mids and make it easier to tune IMO. But I agree with Hugh, tune it to how you like it. Everyone's ears and preferences are different.

Yes, an RTA would be nice. And yes, I have a JL sub in the front.


TURBO2GO said:
Definitely don't listen to Hugh. :)

For anyone other than someone who truly knows what he is doing and has experience with tuning systems, doing the correct thing is difficult to do.

The fact that people hear differently doesn't mean that you should randomly start messing with what the recording engineer has done, the objective is to tune a system to perform correctly first and foremost... minor tweaks to adjust to one's tastes and hearing sensitivites can then be done but they too shouldn't be done until one is well familar with the system.

Djdrock, give me a list of all the equipment that you have installed in the car and I will guide you. Assuming you have 6.5" speakers in the door and a 10" sub up front... set it all flat and set the X-over at 1X not 10X. I would start the x-over frequency at about 100 Hz, and a slope of 12/db Octave.

Start running through your test frequencies, and get a pen and paper and jot what you get down. Use C Weighing on the meter and "slow" response, and bring up the volume to 75-80 dB and start running through it all just changing the frequency and seeing what you get for numbers. Be patient... the meter will fluctuate... especially with bass. This is normal. Try to see what you get as an average.

Once we know where the peaks and dips are, we try to flatten it as much as possible just by adjusting crossover frequency and slope. What is left can then be EQ'd, to a certain extent.

The goal isn't to have a ruler flat response when you are finished. The human ear will not hear that as very good actually... but a flat response is what you want to get the gear dialed in correctly... I will show you how to adjust after you get a fairly flat response. Can you reply to me with the #'s?

Your main area of concern should actually be the bass region, because that is where the cabin will most interact with the sound. The mids and highs are not as deeply affected and with decent placement, should be fairly flat to begin with. So don't get too worked up with the higher frequencies yet.

This may be a bit time consuming, but if you want a good result then its a must-do. Its time consuming because you don't have access to the right gear thats all. But it doesn't matter. We can still get it done.

By the way this is not a hobby for me but my line of work.

I guess what I don't get, is that the more reading I do, the more I think I need an RTA. I totally agree with your statements regarding sound. I am a bit of a nut when it comes to my soundsystem, as I mentioned in another thread. Hell, I bought a top-of-the-line SACD player when they first came out, I have a great system at home, and to my ear, it needs absolutely no adjustment to EQ, bass, treble etc...My point, I am not one of those that likes to max out bass and treble. Most people think that this sounds good. I do not.

Regarding equipment, I have the Kenwood EQ, the Alpine head unit, JL amp, JL speakers and sub. Model numbers for the JL stuff, I don't know except the JL amp is a 300/4, and it is about 1 1/2 years old.

What I don't get is how the SPL meter is supposed to help me adjust the EQ. It just does not make any sense to me. It measures db level, not noise, nor does it measure frequency. I would love to mess with it for a bit, but I have a feeling that without the proper equipment, this is going to be a bit of a waste of time. However, if you want to walk me through some of this, I can always give it a try. The system really does sound good. I feel that it can be tweaked better. It is the hardest system I have every tried to tune. Combining an amp with multiple settings, a head unit, an EQ with different settings, it is all a little mind boggling.
 
Hugh said:
SPL meters are used with equalizers and pink noise generators to get a flat frequency response in a particular room, usually recording studios. You don't need this shit to tune your equalizer in a car. Just play with the sliders until it sounds good to you. Everyone's ear "hears" sounds differently due to our individual anatomy. The curve you end up setting on the eq using tones or pink noise and an SPL might sound awful to you. Just adjust the damn thing manually.

I see you have an Alpine head unit. Does it have the time correction feature? That will make a big difference when set correctly because it will properly balance the sound stage for your position in the car. RTFM! :biggrin:

AMEN. You dont need one of those very unaccurate spl meters, if your that concered go to a high end audio shop around pay 100 bucks and they will tune it with a very expensive spl/rta-sq meter.
 
I had an Audio Control EQL in the trunk and used my Laptop with TrueAudio.com 1/6 oct. RTA and a Behringer mike. I have had two different sets of speakers in the car (Boston Pros & MB Q) and with both had HUGE spikes at 4.8k (almost 18db). Supriseingly the Bass looked fairly flat right off the bat (Image Dynamics IDQ10) with a gentle rise from 40hz to 22hz where it began the drop off.

I think what Turbo was trying to point out, that Hugh ignored, is that if you can get somewhat close to flat then you can tweek from there to what sounds good. Flat actually sounds like a Bose - no highs and no lows. :biggrin: After you get "flat" you right down the positions so you have a baseline to always go back to for different adjustments.

One thing the Radio Shack SPL is good for is setting L & R gains on the amp if they are separate and gains for speakers & sub. Personally I run my sub about 3db hot (higher) than the door separates.

Or you can just follow Hugh's line of thinking and make the sliders look like a smiley face.
 
Last edited:
What I don't get is how the SPL meter is supposed to help me adjust the EQ. It just does not make any sense to me. It measures db level, not noise, nor does it measure frequency. I would love to mess with it for a bit, but I have a feeling that without the proper equipment, this is going to be a bit of a waste of time. However, if you want to walk me through some of this, I can always give it a try. The system really does sound good. I feel that it can be tweaked better. It is the hardest system I have every tried to tune. Combining an amp with multiple settings, a head unit, an EQ with different settings, it is all a little mind boggling.

Turbo2go gave good advice. I even agree with his crossover frequency choice as a good place to start. If you have a choice, I would choose 24 db/oct or 18 db/oct. over 12 db/oct crossover rates.

The way you use the meter to adjust the eq:

You will use your test tones to take measurements. I like to run through the entire range. You should have pencil and paper to write it down. You also need to know what frequency the test tone is.

So prepare a table with the test tone frequency and write down the dB level.

Set the volume so the meter reads around 80.

After you complete the pass you are searching for peaks and valleys. For example if at 2500 hz the response is +7 db you know you have a peak. You would then use your EQ to CUT at that frequency. I would not cut the full amount at first. If the response is a dip - you'll adjust the EQ to BOOST at that frequency.

This will take a bit of time and you'll probably have more variation below 500 hz because of the effect of car acoustics.

I DO NOT think you should just adjust a ten band eq by ear. It's a stupid excuse to say everyone hears differently and adjust by your own ear. Why else do speaker designers even bother to tune their systems if it didn't matter.

I think you can do this in 30-60 minutes.
 
djdrock said:
It measures db level, not noise, nor does it measure frequency. I would love to mess with it for a bit, but I have a feeling that without the proper equipment, this is going to be a bit of a waste of time.

First off, you're right it measures db. So what you do is get a test disk that has several sine waves. For example, lets say for have a CD that has a 20, 35, 50, 70, 100, 150, 250, 500, 1k, 2.5k 5k 8k, 14k tones. Track 1 of the CD is 20hz... while it is playing the RS SPL shows 80db. Then 35hz shows 82db, 50 shows 80db, 70 hz shows 88db, etc... if you graph the whole thing out... then you have basically the same thing as a RTA but a much slower process.
 

Attachments

  • rta.jpg
    rta.jpg
    40 KB · Views: 74
djdrock, I am sorry I thought that was clear. As I said previously you need the SPL meter and a test disc. Hofffam and CL65captain are right. This is basically a cheap RTA. I did not want to tell you to go out and buy a $3000 RTA when a $30 meter and a $10 disc will help you. The stereophile disc recommended earlier is a good one. This is easy to do, but you can also use a laptop with good software and a mike. Reading this requires some experience.

That meter, although is cheap and says Radio Shack on it, is actually a fairly accurate tool. I have tested it against more expensive meters that have done worse.

As for the first order or third order low pass (12 or 24), you need to measure right around the X-over frequency and see which one works best. I would measure the response of the system WITHOUT the sub first... see where you start to drop off (a drop of 6 dB or so)... that will give a good guide as to where the sub needs to start picking up.

I am curious about the results here myself having never measured an NSX.
 
The RS meter is widely used by audio hobbyists. It isn't lab quality but it is good enough for basic calibration.

If I were doing this:

1. I'd pick the crossover frequency (say 100 hz)
2. Measure response at 200 hz (one octave above)
3. Meauser response at 50 hz (on octave below)
4. Adjust the crossover level so that the response at 200hz and 50hz are equal.
5. This will establish the basic subwoofer level vs. door speakers
6. Now through the rest of the frequency range as described above and use the EQ to adjust peaks and valleys.
7. Remember the RS meter is least accurate at high frequencies. I would adjust the EQ for flattest response using the meter for frequencies below 7-10K. Adjust the top band by ear.
 
Okay guys, how about this as an idea...It seems that using a specrum analyzer would make more sense. The price of these are trememdous, so, how about using a laptop, with a microphone, and spectrum analyzing software???

Here is a link that I found...http://members.cox.net/interflex/

triangle.gif


Or http://www.trueaudio.com/rta_abt1.htm

rta_abt1_scrn1.gif


I am sure there is alot of software available, I have done very little research at this point. Of course, a good microphone is a must. I just wanted some input from to see if this would be a better solution.

Thanks!
 
LOL yesterday you were going to do what Hugh said and just play with the sliders because it all seemed like too much work... now you are willing to invest in a mike and software? :)

Of course the second is better if you are willing to spend the money and the time than the SPL-meter solution.

Having this will simplify it all and it will really allow you to see what's going on and tune correctly. Do keep me updated as to what you decide. I am curious about your results.

This is what my sound engineer at work uses:

http://www.siasoft.com/
 
For God's sake, just use the EQ sliders, the head unit's settings, and your ears to acheive a sound that you like. Who gives a flying fuck what the laptop thinks? You are making a huge mountain out of a molehill.

More importantly make sure your crossover points and rolloffs are correctly set either on the head unit or amplifier side before you do the final adjustments with your ears and fingers
 
TURBO2GO said:
LOL yesterday you were going to do what Hugh said and just play with the sliders because it all seemed like too much work... now you are willing to invest in a mike and software? :)

Of course the second is better if you are willing to spend the money and the time than the SPL-meter solution.

Having this will simplify it all and it will really allow you to see what's going on and tune correctly. Do keep me updated as to what you decide. I am curious about your results.

This is what my sound engineer at work uses:

http://www.siasoft.com/

Yeah, I change my mind a bunch right? :biggrin: Basically, using the SPL did not seem like a worthwhile venture, and spending thousands on an RTA was not in budget, so I wondered if using software might be a good idea. I am researching different software, and there are some free options out there. It may be a bit before I do this, as I am still waiting on my laptop to be shipped.

Hugh said:
For God's sake, just use the EQ sliders, the head unit's settings, and your ears to acheive a sound that you like. Who gives a flying fuck what the laptop thinks? You are making a huge mountain out of a molehill.

More importantly make sure your crossover points and rolloffs are correctly set either on the head unit or amplifier side before you do the final adjustments with your ears and fingers

I tried setting things like a smiley face on the EQ, and I really liked it. It sounded terrible, but looked cool! :tongue: Seriously, I am very curious to see what a "tuned" system sounds like. And I agree with you, at the end of the day, I am NOT going to listen to a stereo that sounds like crap just because I am told this is correct.
 
Unfortunately Hugh's viewpoint is rather common, because most people have no idea what correct sound or video even is. They have no frame of reference at all... and so who cares if there is a 10 db peak at 80 Hz and it booms like there's no tomorrow or if skin tones look like everyone has a sunburn. Its bright... and its loud... and for 30 seconds, almost impressive.

Just looking at djdrock's EQ I can tell that he is, like most people would, having a difficult time adjusting the thing. And he already has some frame of reference via a nice home system.

You can't just do this by ear or "what sounds good to you" unless you have years of experience tuning. Even then, equipment helps much to set things right at the beginning.

My point to you Hugh is this... what sounds good to an untrained ear sounds good for a short period of time... with a particular music selection... it's only a matter of time until you realize something is not quite right... and so you are constantly fiddling. Having a system dialed in correctly will allow even the ordinary listener to appreciate the system.. and nuances and subtleties appear that were not even there before.

I have this argument many times with my clients. They say "I won't see the difference" or "I won't hear the difference" yet when they walk into my showroom, they do clearly see AND hear the difference almost immediately. These people were previously listening to what sounded good to them but they would never go back to that after they have experienced a good system. I can guarantee you that if djddrock's car was here and I tuned it for him, he would NOT want to go back to the "what sounds good to your ear" of before.
 
TURBO2GO said:
I can guarantee you that if djddrock's car was here and I tuned it for him, he would NOT want to go back to the "what sounds good to your ear" of before.

Drive on out to Indiana! :smile: If I told you how many hours I have invested in the service menu of my DLP TV, you would faint. I have adjusted each and every input, phew! The TV is 3 years old, and to this day, when I walk into a friends house with newer and "better" equipment, I almost faint with disgust. My point...well, you get it! :wink:

I downloaded a copy of the Alan Parsons Sound Check CD, not in mp3 format. I have read this is a fantastic disc to use. Input???

I need a microphone, but I do not want to spend a fortune. Suggestions?

Thanks again!
 
I have TrueAudio.com RTA and I love it... very simple and easy to use.

PartsExpress.com has the Behringer mike which is cheap. You will also need a pre-amp with a phantom power as an inbetween the mike and the laptop. M-Audio makes a great one that is $100. I also have a Soundblaster external sound card as the onboard card on my Dell laptop suck!

My HT is tuned and so is the car, so I might be tempted to sell the whole "kit" together if you are interested. This will keep me from fiddeling anymore with it. :smile:

Jim
 
Malibu Rapper said:
It is very common for a competitor to have 2 EQ settings. One for competition for RTA analysis and one for listening.

That's because many sanctioning bodies like IASCA mandate a flat RTA result, and deduct points for deviation on adjacent bands. Of course, if you take a flat 31 band RTA result and super-impose that over the transfer function of a human ear canal... it sounds like shit. Thus, programmable equalizers like the Phoenix Gold DEQ became more popular for the die hard SQ guys so they could use one setting for the RTA scoring and one for the rest of their day.


djdrock said:
I am sure there is alot of software available, I have done very little research at this point. Of course, a good microphone is a must. I just wanted some input from to see if this would be a better solution.

Thanks!

Without having a proper microphone and a calibration disk to go with it, the results using $69 PC RTA software on a Creative A/D converter... will be questionable at best. Using the USB bus with a Labtec microphone would fair even worse- particularly through the Windows Audio stack on XP. This is among the many reasons why pro grade PC loud speaker analyzers from companies like LinearX come with there own hardware cards with on-board processors, calibrated directional mics, etc.. and tend to be so expensive.

You also have to really know what you are doing with these types of tools and understand the physics of audio and the measuring environment. Among less than 300 installers in the country at the time, I took an advanced mobile audio class at the Installers Institute in FL as part of my MECP masters certification back when I was a kid... and now looking back years later I realize how extremely limited my scope of knowledge in regards to audio analysis was, even thou I felt having graduated to be pretty confident in my abilities. I currently spend much of my time working on audio Q/A for Microsoft including working directly with the hardware topology of consumer audio devices and their interoperability with the OS via both our user and kernel mode stack.

No different than suspension tuning, for the vast majority of car audio enthusiasts this is an area where good enough is good enough.. as Hugh mentioned. This isn't an area whereas you throw a mic on a latop with a 31 band PC RTA and do anything significantly meaningful with an in-dash 2/3 octave EQ in terms of spectrum analysis and correction of vehicle transfer functions. Proper vehicle setup, a good basic system lay-out, and excellent cross-over design needs to be in place first, before concerning ones self with all these other variables left on the table.
 
Last edited:
Ahhhhhhhhhhhhhhhhhhhhhh! Just when I think I am going to give it a shot, then I get blasted in the head more information that makes me think I should just drop it! :eek:
 
Back
Top