It is possible to get a flat impedance curve with a resulting better sound quality using a Zobel Network across the driver terminals. Sometimes an Improved Zobel Network works best. Either way, ignore published formulas and use substitution banks of capacitors and resisters to find the optimal values. There's no easy way to find the best values, but it isn't that difficult (just time consuming) to keep switching values and measuring with Impedance Sweep in the Dayton DATS software until you have a flat line.
When working on speakers, all the Room EQ Wizard is ever going to do is help you find the best balance level between the drivers within your enclosures! So it's hardly worth considering in this context. In fact, your ears are probably all you need!
Now if you are after room correction and have the hardware, the Room EQ Wizard (as the name suggests) is fantstic.
When working on speakers, all the Room EQ Wizard is ever going to do is help you find the best balance level between the drivers within your enclosures! So it's hardly worth considering in this context. In fact, your ears are probably all you need!
Now if you are after room correction and have the hardware, the Room EQ Wizard (as the name suggests) is fantstic.
@PedroE I like to use XSim for this kind of work. You are absolutely right that online calculators are really rough approximations, but also, the Zobel has some effect on the low pass phase. I don't really care what happens 5 octaves above the crossover, but in the near distance a Zobel can be crucial to the low pass filter working as expected or not.
As I mentioned, you can sometimes bump the R or C here to improve the phase matching with the high pass filter as well. For all of this I find XSim, or whatever your favorite crossover simulator is, to be a great way to find the right values, and to see what helps or does not.
It's also very important to check the power requirement of Rz1/Rz2 and something like Xsim for that is critical. The more you lower the Z rise the more power your resistor(s) will need to dissipate.
As I mentioned, you can sometimes bump the R or C here to improve the phase matching with the high pass filter as well. For all of this I find XSim, or whatever your favorite crossover simulator is, to be a great way to find the right values, and to see what helps or does not.
It's also very important to check the power requirement of Rz1/Rz2 and something like Xsim for that is critical. The more you lower the Z rise the more power your resistor(s) will need to dissipate.
Perhaps I should check out XSim. I've heard that a flat impedance curve presented to the crossover allows it to perform at its best, and I have no reason to question that claim.
It is true, in absolute terms. I wrote about it in my blog here.
What I meant to say was that the Zobel is most useful through the crossover's transition region, and then a lot less useful further away. For instance, with a low pass filter at 100 Hz, by 1 kHz the effects of the impedance on the crossover function is less important because the impedance of the low pass filter section will often swamp out the driver's Le.
It's not unfair to call the Zobel a helper which makes the low pass filter easier to calculate.
Using a tool like XSim and having accurate impedance curves really helps both in choosing values for the Zobel and also determining if it even matters, or whether you need to use more advanced compensation. I've made some very fancy Zobel compensation circuits and then, using XSim, I realized that the output was nearly identical with and without it. 😆
At the end of the day what matters is your low pass filter performance is ideal, and you can achieve this with less than ideal Zobel component values.
To be clear, I am not advocating the elimination of Zobels at all, just saying that a tool like XSim should be used to evaluate it's worth on a case by case basis. Zobel's aren't free. The parts cost money and the circuit increases the overall current used.
🙂
Last edited: