I've recently finished building the Audax HT front channels. After auditioning the speakers I found them to be a little bright. I tested them using a pink noise generator and spectrum analyzer and sure enough there was a marked increase in amplitude above 2.6kHz which happens to be the crossover frequency. I believe I followed the design plans to the letter but I can't understand why there is such a difference in level between the woofers and the tweeter.
Anyone have any advice on how I could tweak these speakers to flatten out the response a little bit?
Some construction notes on some possible reasons:
- I used 1" MDF throughout (as opposed to the 3/4" back, sides, top and bottom called for in the design). I did compensate for the additional length and width to keep the internal dimensions constant at 19 x 11 x 8
- I did not use any additional damping other than the eggcrate foam called for in the design.
- I used 12 ga wire to connect the woofers and I used 18 ga wire to connect the tweeter
- I sized the large (inductor to accomodate certain connections in the crossover, I trimmed about 5 or 6 inches off the inductor end to keep connections clean
- Due to speaker sensitivity, the design calls for a 100 watt/channel+ amp to drive them and I did all my tests with a fairly cheap ONKYO amp at 70W/channel. Could this have an impact on the tweeter being so much louder than the woofer mains?
Other than that, the response is fairly flat before and after 2.6kHz, it's just the differences in levels between the two ranges that's getting me here. Any and all help is appreciated.
MikeA
Anyone have any advice on how I could tweak these speakers to flatten out the response a little bit?
Some construction notes on some possible reasons:
- I used 1" MDF throughout (as opposed to the 3/4" back, sides, top and bottom called for in the design). I did compensate for the additional length and width to keep the internal dimensions constant at 19 x 11 x 8
- I did not use any additional damping other than the eggcrate foam called for in the design.
- I used 12 ga wire to connect the woofers and I used 18 ga wire to connect the tweeter
- I sized the large (inductor to accomodate certain connections in the crossover, I trimmed about 5 or 6 inches off the inductor end to keep connections clean
- Due to speaker sensitivity, the design calls for a 100 watt/channel+ amp to drive them and I did all my tests with a fairly cheap ONKYO amp at 70W/channel. Could this have an impact on the tweeter being so much louder than the woofer mains?
Other than that, the response is fairly flat before and after 2.6kHz, it's just the differences in levels between the two ranges that's getting me here. Any and all help is appreciated.
MikeA
Multiple L-pads in series
Does the attenuation of an L-pad mimic the dynamics of resistors when used in series (ie summing of resistance in series)?
For example, if I wanted to pad the tweeter by an additional 2 or 3 db could I use the following network
.. ------- R1 --------- R2 ---------- +
. | |
. | |
.. ------- R3---------- R4 ---------- -
Where say R1 + R3 have attenuation of 3 db and R2 + R4 provide attenuation of 2 db .. would I get a total attenuation network on the driver of 5db?.... would this affect overall impedance at all differently from the use of a single l-pad (which, my understanding is, theoretically should not affect the overall impedance of the network). Or, am I better off just replaceing the 2 resistors I have with the equivalent single attenuation network. I'm hesitant to do this becuase in my novice zest, I glued in the crossover board before really testing the crossover(!).. I know, I know... it was a caffeine deficiency moment...
Thanks again for your help..
Does the attenuation of an L-pad mimic the dynamics of resistors when used in series (ie summing of resistance in series)?
For example, if I wanted to pad the tweeter by an additional 2 or 3 db could I use the following network
.. ------- R1 --------- R2 ---------- +
. | |
. | |
.. ------- R3---------- R4 ---------- -
Where say R1 + R3 have attenuation of 3 db and R2 + R4 provide attenuation of 2 db .. would I get a total attenuation network on the driver of 5db?.... would this affect overall impedance at all differently from the use of a single l-pad (which, my understanding is, theoretically should not affect the overall impedance of the network). Or, am I better off just replaceing the 2 resistors I have with the equivalent single attenuation network. I'm hesitant to do this becuase in my novice zest, I glued in the crossover board before really testing the crossover(!).. I know, I know... it was a caffeine deficiency moment...
Thanks again for your help..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.