[frascati]

Lacking any more sophisiticated measuring equipment right now than a Radio Shack SPL meter, is there any formula that can be used with SPL measurements to "rough in" a different sensitivity tweeter replacement in an old two way speaker? I'm assuming that at the same input level, the tweeter is going to be much lower SPL than the mid/woofer, but hoping that that difference is linear enough at different levels to be able to apply a formula.


It's presently got a protection cap and single 2ohm resistor in series with the tweeter. Along with the new tweeter I plan on replacing the series resistor with a circuit from the following table
http://reviews.ebay.co.uk/Tweeter-At...00000004030106


Do I understand the theory correctly here in that if I were to simply put a 23ohm resistor parallel on the present 2ohm series circuit, the level and FR of the tweeter would remain unchanged but the load on the amplifier would be more 8 ohm normalized?


If an SPL meter is not useful, is level matching pretty much always done by ear?

 

[reecew]

Why do you need to measure. The best way would to have a buddy with you, you sit in your lp playing various types of music/movies and have him adjust it till your happy. In the end what if you don't like the flat rersponse? Just ear tune it you'll be happier that way!

 

[frascati]

I understand that these sort of preferences are controversial as heck. But I'm a big fan of measurement. (tho only beginning to acquire the tools for it)


Accomodation/acclimation seems to be an underrated phenomemon where our senses are concerned. Perhaps no more so than in audio. It's so easy to get used to a certain level of sugar, salt, treble, bass, relative attenuation, etc. It's highly likely that a more sensitive tweeter replacing a less sensitive one is going sound 'better' than the old one. It's only over time, and being honest with one's self that a correctly tuned system is actually better, more accurate, than the 'crisper', saltier, sweeter, stuff you've acclimated to over time.


For those reasons, I mistrust my own perceptions in the short term (ie, while dynamically sitting in a chair, getting up and adjusting, sitting down again, getting up and adjusting, etc). I'd much rather make some attempt at establishing a 'baseline' equalized attenuation and then very carefully make small adjustments from there. For that matter, there is such a range of genre to use while 'voicing' speakers that I couldn't imagine where to begin. I'm frequently adjusting bass and treble between reggae and classical and female vocal. Not that any genre 'needs' it more than another, but according to a preference in the way I listen to certain types of music. Same way that we prefer more salt or sugar on some foods and not others I guess. So I'd strongly perfer to have a 'flat' speaker to begin with.


Can this particular job be done with an SPL meter?

 

[kbgl]



Quote:
Originally Posted by frascati /forum/post/18241585


Lacking any more sophisiticated measuring equipment right now than a Radio Shack SPL meter, is there any formula that can be used with SPL measurements to "rough in" a different sensitivity tweeter replacement in an old two way speaker? I'm assuming that at the same input level, the tweeter is going to be much lower SPL than the mid/woofer, but hoping that that difference is linear enough at different levels to be able to apply a formula.


It's presently got a protection cap and single 2ohm resistor in series with the tweeter. Along with the new tweeter I plan on replacing the series resistor with a circuit from the following table
http://reviews.ebay.co.uk/Tweeter-At...00000004030106


Do I understand the theory correctly here in that if I were to simply put a 23ohm resistor parallel on the present 2ohm series circuit, the level and FR of the tweeter would remain unchanged but the load on the amplifier would be more 8 ohm normalized?


If an SPL meter is not useful, is level matching pretty much always done by ear?

Adding the 23 ohm resistor would change the x-over slightly. Since you are changing the tweeter, there's a very good chance you need to change the x-over anyway. With test tones and your SPL meter, you can figure it out. Start by adjusting the resistors for the "theoretical" padding that you need, then play with the cap value, or new x-over components.

 

[reecew]

Go for it man i would, if its what makes YOU happy. Id give the SPL meter a fair chance and see how it does compared to what you think you like at that point in time. Let me know how it turns out! Ill be starting a new speaker build any day now and using an adjustable L-pad with it.

 

[frascati]

Can anyone suggest any standard practice for determining the tweeter/mid spl levels using the meter? Something along the lines of when the woofer is producing 80 db at a certain test tone the tweeter level using the same volume setting and test tone should be approx (exactly?) X db? Or is this simply not possible without graphing the frequency response?


And from the link I posted.

Quote:
Many people in the audio field do not like any variable units in the signal path, although they are very useful for testing purposes and can help

determining the values required for a high quality fixed resistor network to be used permanently in place of the ‘L’ pad

Personally I don't see what the admonishments against using adjustable l-pads amount to. I noted this in more than half of the threads here and on other forums concerning passive crossover level adjustments.

Sure, the wiper and windings can get corroded/degraded. But there are already a few 'variable units' in the signal path of 95 percent of the equipment we're using for feeding these speakers anyway, no? What's the problem?


For me, I'd just prefer the simplicity of the fixed resistor circuit so that I won't even be tempted to fiddle with it once I'm satisfied with the level.

 

[kbgl]



Quote:
Originally Posted by frascati /forum/post/18243133


Can anyone suggest any standard practice for determining the tweeter/mid spl levels using the meter? Something along the lines of when the woofer is producing 80 db at a certain test tone the tweeter level using the same volume setting and test tone should be approx (exactly?) X db? Or is this simply not possible without graphing the frequency response?


And from the link I posted.


Personally I don't see what the admonishments against using adjustable l-pads amount to. I noted this in more than half of the threads here and on other forums concerning passive crossover level adjustments.

Sure, the wiper and windings can get corroded/degraded. But there are already a few 'variable units' in the signal path of 95 percent of the equipment we're using for feeding these speakers anyway, no? What's the problem?


For me, I'd just prefer the simplicity of the fixed resistor circuit so that I won't even be tempted to fiddle with it once I'm satisfied with the level.

To compare levels, you need a warble tone of about 6k for the tweeter, and one of about 800 hz for the woofer. Adjust till the SPL is the same. If you have more tones, you can double check with other frequencies. Just stay an octave or more away from the x-over point. At the x-over point, both drivers should measure the same, and usually be about 3 to 6 dB lower than the frequencies on either side of the x-over. With both drivers playing, the spl should be equal too the rest at the x-over frequency.


Fixed resistors let you make more precise adjustments. Half a dB matters more than you might think.

 

[frascati]

kbgl thank you. Can I clarify a couple of things?

Quote:
To compare levels, you need a warble tone of about 6k for the tweeter, and one of about 800 hz for the woofer.

Is the second a 'warble' tone as well?

Use the 'A' rating option on the meter?

Place the meter about a meter directly in front of each driver when making the measurement?

Quote:
Adjust till the SPL is the same. If you have more tones, you can double check with other frequencies. Just stay an octave or more away from the x-over point. At the x-over point, both drivers should measure the same, and usually be about 3 to 6 dB lower than the frequencies on either side of the x-over. With both drivers playing, the spl should be equal too the rest at the x-over frequency.

The spl should be the same for each driver at crossover point but vary somewhat until they are well separated at about 800 and 6K?

 

[kbgl]



Quote:
Originally Posted by frascati /forum/post/18246572


kbgl thank you. Can I clarify a couple of things?




Is the second a 'warble' tone as well?

Use the 'A' rating option on the meter?

Place the meter about a meter directly in front of each driver when making the measurement?




The spl should be the same for each driver at crossover point but vary somewhat until they are well separated at about 800 and 6K?

Warble for both. I believe the C setting is used. This should get you close, but you may still want to adjust a few dB one way or the other.