PDA

View Full Version : What does this capacitor do?



jfine
07-09-2011, 06:08 PM
The L150 and L150A have the same woofers, basically the same mid, and the tweeters are slightly different.

The 150 is crossed at 1000, 4000.

The 150A is crossed at 1100, 3700.

Look at the LF band on the 150, then on the 150A.

They both have the same inductor winding.

The 150 has 4 resistors, and a 36 uf cap!

The 150A has just a single resistor.

I don't think the 4 resistors on the 150 make too much difference(?), but is the 36 uf cap used to lower the crossover point? Or what is it for?

:confused:

Robh3606
07-09-2011, 07:37 PM
Looks like a zobel with a 50 ohm parallel resistor. If I get a chance I will fire up LEAP and see what the difference in the voltage drives are.

Rob:)

tomt
07-10-2011, 04:06 PM
Or what is it for?



keeps higher frez. out of the woofer ...

jfine
07-10-2011, 06:30 PM
keeps higher frez. out of the woofer ...

If you look at the attached diagram, look at the LF of the 150, and the 150A.

Same woofers, same inductor.

If the cap keeps high freq out of the woofer, then what keeps the high freq out of the woofer on the 150A?

:confused:

4313B
07-10-2011, 08:39 PM
The 36 uF capacitor works with the 5 ohm 20 watt resistor to counter the inductive reactance of the voice coil. Coupled with the trailing 50 ohm 20 watt shunt resistor it effectively provides the 2.5 mH inductor with a flatter impedance load.

L3 in the L150 coupled with the C4 and R8 conjugate does the same thing, it mitigates the motional impedance and reactive inductance of the LE5 resulting in a much flatter impedance curve for the actual passband network (C2, C3, L2). R5 and R6 provide a 3 dB passband attenuation. L3 is a special inductor in that it is of very small gauge wire resulting in an intended DCR of ~ 7.5 ohms.

And again for the 033, L4 (with a nominal DCR of 7.5 ohms) and C6 make up the conjugate that mitigates the motional impdance and reactive inductance providing a flatter impedance curve for the highpass (C4 and C1).

The L110, L150 and L250 networks are patterned after the very smooth L212 network. The slopes are all 6 dB per octave.

The L96, L112 and L150A networks are patterned after the 4313 network. The slopes are 6 dB per octave (LF to MF) and 12 dB per octave (MF to HF).

jfine
07-11-2011, 09:46 AM
4313b Thanks for explaining the mid and high circuits as well. Very helpful.

Is it safe to say that in addition to providing a flatter impedance load, the 36 uf (+ inductor + resistors I suppose) is also "tuning" the circuit? Or is that just an artifact (happy accident) of the single purpose of providing a flatter impedance load?

For example if I swap in a polypropylene cap in place of the electrolytic 36, not only does the tightness of the bass change, it also affects/attenuates certain frequencies below the cutoff point differently than the original cap. I realize I'm talking 2 different cap designs, but it does affect this circuit nonetheless besides impedance loading.

This LF filter network seems very complex, and the interaction between the inductor, cap, and resistor, seems to change if 1 of those components is altered. Not sure swapping the cap for polypropylene would be smart because of the lateral effects.

The fact that the 150A omitted it altogether is puzzling. So by removing the cap from the 150A, was the flatter load compromised for the sake of sonics? I can't think of another reason. I think the L96, L112 and L150A does not contain any NPE's, so must be.

On a side note, raising the slope for the MF-HF on the 150A, I wonder if they did that to help protect the tweeter from possible high wattage leaking thru. One set of L150's I have the 033's were blown and the associated lpad looked fried inside.

DavidF
07-11-2011, 10:56 PM
4313b Thanks for explaining the mid and high circuits as well. Very helpful.

Is it safe to say that in addition to providing a flatter impedance load, the 36 uf (+ inductor + resistors I suppose) is also "tuning" the circuit? Or is that just an artifact (happy accident) of the single purpose of providing a flatter impedance load?

For example if I swap in a polypropylene cap in place of the electrolytic 36, not only does the tightness of the bass change, it also affects/attenuates certain frequencies below the cutoff point differently than the original cap. I realize I'm talking 2 different cap designs, but it does affect this circuit nonetheless besides impedance loading.

This LF filter network seems very complex, and the interaction between the inductor, cap, and resistor, seems to change if 1 of those components is altered. Not sure swapping the cap for polypropylene would be smart because of the lateral effects.

The fact that the 150A omitted it altogether is puzzling. So by removing the cap from the 150A, was the flatter load compromised for the sake of sonics? I can't think of another reason. I think the L96, L112 and L150A does not contain any NPE's, so must be.

On a side note, raising the slope for the MF-HF on the 150A, I wonder if they did that to help protect the tweeter from possible high wattage leaking thru. One set of L150's I have the 033's were blown and the associated lpad looked fried inside.

I think, possibly, you are making more of the changes in the circuit more than was intended. Often times a refinement is brought through simplicity rather than complexity. Yes, the even impedance load has benefits both the predicted filter response and making the system a better load for the amplifier. However, consider that the addition of xover components may well provide a desired effect at the cost of an undesired effect somewhere else in the chain. Just trying to say because a component is lacking does not necessarily imply a retrograde design. Could be the zobel circuit was redundant in the updated circuit design. Could be that the acoustic response of the woofer combined with the inductor provided all the response filtering needed.

FWIW I had both circuits in use (L112 components, home made cabinet) and settled on the L112/L150A design.

4313B
07-12-2011, 10:05 AM
Is it safe to say that in addition to providing a flatter impedance load, the 36 uf (+ inductor + resistors I suppose) is also "tuning" the circuit?The conjugate can be used to affect the knee of the curve.
The fact that the 150A omitted it altogether is puzzling. So by removing the cap from the 150A, was the flatter load compromised for the sake of sonics?Economy. Price points are adhered to rather stringently so a penny here, a penny there. Mark Gander and I had a conversation about this very subject way back in the day. It's pretty cool when one can end up using the same network for three different loudspeaker systems. Even cooler is that David Smith used a completely different network design for the 4411.
On a side note, raising the slope for the MF-HF on the 150A, I wonder if they did that to help protect the tweeter from possible high wattage leaking thru. One set of L150's I have the 033's were blown and the associated lpad looked fried inside.Yes, JBL used the higher slope to help protect the 066 in the 4313. With the L96, L112, and L150A they went a step further and replaced the aluminum voice coil with the copper voice coil in the 044.

jfine
07-12-2011, 11:51 AM
Thanks to all for the invaluable info,

Excuse the slight thread drift, but I can't help thinking that the popularity of the L96, L112, L150A, etc., networks are mostly due to the absence of non-polarized Electrolytic Caps. Removing these NPE's from the signal path would certainly seem to give the kind of results I read about when upgrading to this network design, i.e., more mid-presence, less distortion, etc.,

From my cap swapping results on the L150, it seems that while electrolytic's are inferior, I actually prefer their sound in this case over metalized polypropylene. :dont-know: too much 60's/70's R&R I guess.