Three different versions of the N212 network.
Printable View
Three different versions of the N212 network.
Evaluation of the three different versions of the conjugate circuit in parallel with the 066 high frequency dome radiator.
Three custom versions
From inspection it can be seen that two Dayton 30uF-250V Polypropylene Capacitors (Part Number 027-440) or two Solen 30.0uF-400Vdc Polypropylene Capacitors (Part Number 027-590) will suffice to arrive at a 15.0 uF biased capacitor pair.
Evaluation of the three different versions of the conjugate circuit in parallel with the 112 low frequency transducer.
The math supports the measurements - Le of the 112 is ~ 0.83 mH so the third conjugate is the one to go with.
0.83 mH / (10 ohms)^2 = 8.3 uF
Two Dayton 18uF-250V Polypropylene Capacitors (Part Number 027-434) or two Solen 16.0uF-400Vdc Polypropylene Capacitors (Part Number 027-578) will yield a 8.0 uF to 9.0 uF biased capacitor pair.
Evaluation of the conjugate circuits in parallel with the LE5-9 midrange transducer.
The L212 used a 4.0 uF with the 7.5 ohm resistor while the L250 used a 6.0 uF with the 7.5 ohm resistor. A 5.0 uF might end up "looking" the nicest. :p
In any case, two Dayton 8.2uF-250V Polypropylene Capacitors (Part Number 027-426) or two Solen 8.2uF-400Vdc Polypropylene Capacitors (Part Number 027-564) will yield a 4.1 uF biased capacitor pair.
This data shows that the final version of the N212 network is the version to use for biasing.
A 15.0 uF biased pair looks "good enough" for the HF conjugate although Solen does have the 33.0uF-400Vdc Polypropylene Capacitors (Part Number 027-592) available which will yield a 16.5 uF biased pair.
Thanks Giskard for sharing this data.
Anyone comment on a couple of points?
The extra ripples in the 066 magnitude seem odd to me. They are not insignificant and could play into response/phase given the typical low-order xover used with this model.
Is the charged capacitor all or nothing, i.e. will using this only in the series capacitors but not the shunt capacitors diminish the effect?
Thanks,
DavidF
Thanks for the schematics and the supporting graphs Giskard.
Cheaper caps to create a great crossover that need not be bypassed means more $ for transducers!
I was looking at Auricaps and I'm still in sticker shock.
This is a much better alternative.
True.Quote:
Originally Posted by DavidF
Some people might prefer the 3113B network with the 12 dB/octave slope on the 066.I'd approach it with an all or nothing attitude.Quote:
Originally Posted by DavidF
It sounds pretty damn good.Quote:
Originally Posted by hector.murray
As for bypassing, I'd still bypass the Daytons with the 0.01 uF Audiocap Thetas. The Solens aren't supposed to need bypassing when biased.
jblnut is running brand new biased networks in his 250Ti's (they use all Solen capacitors and inductors). You might want to check out his recent posts.
So add the 0.3millihenry choke in parallel behind the Lpad and leave out the conjugate or leave it in?
Take the conjugate out. You would need to use the whole 3113B network though and adjust the conjugate on the LF accordingly.
Giskard, I found this thread that you posted a biased L212 network.
Thanks for the info, be it a year later.
I understand the reason for the 9 volt batteries between the capacitor pairs, but what is the reason for the battery on the negative(black) input line, on the 112A circuit? And if I am looking at it right, there are 6 batteries used for each xover, correct?
I've already got most of the resistors and inductors, just have to buy the Solen caps and 3meg resistors and I'll be ready for the upgrade.
One battery per network.
You attach the negative lead of the battery to the negative input rail of the loudspeaker system. The positive lead of the battery goes to one side of all the 3.0 Mohm resistors. The other ends of the 3.0 Mohm resistors go to each of the capacitor pairs.
Thanks, that makes it crystal clear.