i've been thinking about getting a dB meter so i can do some more accurate tuning on my car stereo myself, just wondering if anyone else has done much theirselves, and what methods/tools they used to do it.

i was thinking something like this (http://www.dse.com.au/cgi-bin/dse.storefront/43c0c85309079b96273fc0a87f9c06ed/Product/View/Q1461) to measure the level of certain frequencies and tune it reasonably flat (apparently slightly higher treble and bass sounds a bit better?)

I dunno, 126dB isnt all that much

I got one exactly the same havent actually used it yet, as I usually use an RTA.

beware they arent as accurate down low or up high. But still useful 100hz-10000hz id imagine.

To measure freq response it doesnt need to be loud at all, you would test it around 90 db or less.

I cheat and use my Professional PA gear to tune the frequency response for a car audio system using a audio analyser (RTA) and reference mics. :P
Downside with that is I have to take the whole rack with me which is a nuisance.

The DSE SPL meter has an accuracy of +/- 2db @ 114db, which normally means anything above 114db can be extremely inaccurate. Even a +/- 2db is deemed inaccurate whilst most calibrated gear are within 0.5db accuracy.

I want one of these though: http://www.ivie.com/ie33/

so +/-2dB is too innacurate for it to be helpful? i'd prefer if it was able to do below 100Hz aswell!

but yeah, not really fussed if it cant read over 125dB, as i'd turn it down to tune anyway, just wanna get an idea what the curve is like now

haha i'm a perfectionist :)

Don't forget though, for each 3dB increase, the sound will be twice as loud. So 2dB can make a difference.

Oh and you're not going to work out your curve with an SPL meter ;)

what i meant by "curve" was going thru a whole heap of tones and making up a graph, showing where there was peaks in certain frequencies... then using the EQ to try and flatten it out, or get it as even as possible.

kinda sucks that it is out by about 2db, coz like u said, even a little out will make a huge difference to actual volume!

remember it is only $50, maybe you should look at a TermLab? its only about 5 times the price but ive heard super accurate.

http://www.termpro.com/term-lab/index.asp

It's a lot more than 5 times the price dude! I think you got mixed up with the special introductory offer for the RTA upgrade :P

If it was really $250, I would be really concerned considering my reference mic alone is about that (and thats on the cheap side too!)

what i meant by "curve" was going thru a whole heap of tones and making up a graph, showing where there was peaks in certain frequencies... then using the EQ to try and flatten it out, or get it as even as possible.



Ahh I see, yes that will work to a degree, it'll be better then nothing i guess. It's only $50, go have some fun :P

Decent RTA's are around $4000

One of the CAASQ guys uses a small pc and two mics at 45 degrees to simulate ears. Forget the name of the software but I tuned my system on this and was very accurate. You need some other equipment though and it ends up being pretty expensive. but under $500 iirc.

Most systems that are untuned will have peaks of around 8db. So even if it's off by +/- 1 or 2 db, it will be better than nothing.

PM Megatron- he does the audio testing for the QLD division and should be able to give you a heads up on what to get and how much it will cost. May also be able to quote on some of it for you.

Cheers,
Bonnie

See, $4500 for equipment like that is a joke. You're better off just annoying some bloke at a car audio place every week. While with the $4500 you can put in a couple of nice sounding IDMAX subs :P

Figured it out...

Download a 10day trial version of SpectraRta. If you find a working crack PM me the link. (So I can report such illegal activities to the local authorities :liar: )

All you need is a mic. And a pre amp. then a pc. Accuracy determined by quality of mic and quality of line in on ur soundcard/mixer etc.

Using an $8 mic it seems ok. Will add a new mic to the upgrade list though.

I just loaded it in my room with a dodgy DJ mic and Pioneer Mixer, and it works a treat.

Will load it up onto the laptop and plug it into the car when I get a chance to see how she goes.

Just use Pink noise to test it out.

Have fun, just remember a flat RTA response doesnt necesarily equate to a good sounding car. You ears should always be the judge.

Peace

Tim

just remember a flat RTA response doesnt necesarily equate to a good sounding car. You ears should always be the judge.


I'm hearin ya....

I spent a whole weekend mucking around with test equipment to get a flat response only to find it utterly boring to listen to...

I mainly use a function generator and an oscilloscope with the SPL mic for the function gen sitting where the drivers head would be...

http://www.lavendergreenooaks.com/BEN/Test%20equip.jpg

I've found that my car has a resonant frequency of around 60-65Hz. I imagine most 3rd gen owners will have a similar resonant freq, unless they have a lot of sound deadening....

Yeh a good response is one with no big dips or spikes. Some aim for a seagull type graph that dips around 1khz (with slight dips around 20z and 20khz.

Others aim for a bit more character and definition in their sub-bass. You should clearly be able to define all changes in subbass. So some aim for a flat graph with peaking at 20 hz then flowing down all the way to 20 khz.

But flat is definetly boring, and undesirable.

Peace

Tim

I've got a problem ATM with my response....

I can get it perfect at a given volume, but if I change the volume on the H.U. up or down the response curve changes... Is it too much to ask for it to be perfect at all levels of volume?
I'm thinking its the cheap ass amps that I've got that's letting me down...