Incoming unit #2 has the original NPE caps, but the pots have been "upgraded" with L-pads:
Measurements on tweeter axis and midline, no attenuation, i.e., as if the mid and tweeter pots have been removed:
Added the 25-Ohm resistors across the drivers as recommended by RoyC at CSP, then put on the grille:
The results of the resistor addition at L-pad max setting is substantially the same as earlier found with AR4x: there's a small amount of attenuation across the bandwidth, and differentially a bit more at highpass and just above as the filter frequency moves slightly upward.
Putting on the grille swamps all of that with substantial attenuation in the high frequencies, and the mid must be attenuated to balance and flatten the response:
Mid pot is intermittent on unit #1, but here's a first look at the summed mid/twt response in comparison to #2:
[Yeah, I know what to do....
]
Incoming unit #1, all original and early, w/front terminal mid and tweeter, pots now bypassed but remaining in-circuit, i.e., both "Max:"
Individual driver and summed responses comparing Unit #s 1 & 2 per legend; #2 with 25 Ohms in parallel with both mid and twt, no grilles:
Divot in midrange response is precisely at 1.8 kHz (well, 1.7996520, actually). Tweeter response is "compromised" by comparision, but still sums well on-axis. Again, ~4 dB mid driver attenuation required for maximally flat. #1 mid highpass filter is a bit lower, perhaps an aging capacitor issue.
Polars:
Inboard is toward the mid driver:
Outboard is edgeward from the tweeter:
Perhaps the conversion of the 1.8 kHz interference notch into a reinforcement off-axis outboard affords a clue as to its source?
L-pad attenuation on the mid driver axis:
It'll take some surgery for a direct comparison to the stock pot, but I removed the 25-Ohm resistor and normalized for one setting here, Black, max. The L-pad behaves like L-pads do, without introducing frequency shifts or other significant anomalies.
Surgery completed, here's 16-Ohm pot attenuation in 2-Ohm increments for comparison:
And an 8-Ohm L-pad without a resistor in parallel with the driver under identical conditions:
Reflected impedance analysis here:
http://www.audiokarma.org/forums/showthread.php?p=2561858#post2561858
Measurements on tweeter axis and midline, no attenuation, i.e., as if the mid and tweeter pots have been removed:
Added the 25-Ohm resistors across the drivers as recommended by RoyC at CSP, then put on the grille:
The results of the resistor addition at L-pad max setting is substantially the same as earlier found with AR4x: there's a small amount of attenuation across the bandwidth, and differentially a bit more at highpass and just above as the filter frequency moves slightly upward.
Putting on the grille swamps all of that with substantial attenuation in the high frequencies, and the mid must be attenuated to balance and flatten the response:
Mid pot is intermittent on unit #1, but here's a first look at the summed mid/twt response in comparison to #2:
[Yeah, I know what to do....
]Incoming unit #1, all original and early, w/front terminal mid and tweeter, pots now bypassed but remaining in-circuit, i.e., both "Max:"
Individual driver and summed responses comparing Unit #s 1 & 2 per legend; #2 with 25 Ohms in parallel with both mid and twt, no grilles:
Divot in midrange response is precisely at 1.8 kHz (well, 1.7996520, actually). Tweeter response is "compromised" by comparision, but still sums well on-axis. Again, ~4 dB mid driver attenuation required for maximally flat. #1 mid highpass filter is a bit lower, perhaps an aging capacitor issue.
Polars:
Inboard is toward the mid driver:
Outboard is edgeward from the tweeter:
Perhaps the conversion of the 1.8 kHz interference notch into a reinforcement off-axis outboard affords a clue as to its source?
L-pad attenuation on the mid driver axis:
It'll take some surgery for a direct comparison to the stock pot, but I removed the 25-Ohm resistor and normalized for one setting here, Black, max. The L-pad behaves like L-pads do, without introducing frequency shifts or other significant anomalies.
Surgery completed, here's 16-Ohm pot attenuation in 2-Ohm increments for comparison:
And an 8-Ohm L-pad without a resistor in parallel with the driver under identical conditions:
Reflected impedance analysis here:
http://www.audiokarma.org/forums/showthread.php?p=2561858#post2561858
Attachments
Last edited:
Yes, a definite lurker thread!
