tuning with DSP before ordering crossover components?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I had a strange thought. Could I build a 2-way speaker and effectively calculate, and 'tune', the crossover on an iNuke before ordering the components? I'd hook the tweeter up to one channel and the woofer up to the other and use each channel's DSP to 'tune the crossover'. Then, I could just order XO components to 'match' my DSP settings. Seems a heck of a lot easier/cheaper for those of us that don't have gobs of components laying around.
 
I have an Inuke (not with DSP), however I do use active EQ (in J River Media Center); also have an old MiniDSP. I think your idea to "model" a desired EQ at first makes good sense. Do you have/plan to use measurements (e.g. for frequency) to fine-tune your settings? Then, if you must have a passive x-over, you can use a designer software or whatever method you choose.
 
Hi kc, the most common function of a simulation program does what you want, find passive parts values prior to purchasing. It has been done in this forum before with great success. The measurements of the raw drivers in an enclosure are prerequisite.
 
I think that's a very viable option, but to honest the question that pops into my mind is - if you have a dsp and multi-channel amp laying around, why do you intend to go back to passive crossovers anyways? Sorry, am just a big proponent of active amplification as it solves so many problems...
 
Measurements

Hi

If you measure the frequency response of the drivers in the box you want to use them in.

You also have to take in to account the impedance rise of the drivers, and the impedance hump at the resonance frequency of the drivers. You probably have to make some impedance correction circuits (RC) for those.

Then it could work

Best regards

Uwe
 
In theory, this can work.

A 2nd order crossover before or after the amplifier ought to make no difference - the driver will receive the same signal.
In theory.

In practice, you must design a passive crossover around the impedance curve of the driver(s) in question. An active crossover will apply a perfect slope to a given driver - the amplifier shouldn't care what the impedance curve of the driver(s) looks like, it'll still apply the chosen slope.
In order to make sure the passive crossover is the same as the active one, you'll need to flatten the impedance curve in the area of interest, or you'll get all sorts of interesting effects that you really don't want.

For instance, if you apply a highpass filter to a tweeter, at its resonant frequency it'll have a peak in its response where the impedance curve peaks. As a result, the frequency response curve won't follow the crossover slope any more. An LCR filter can be used to flatten the tweeter's impedance peak.
Woofers have a similar problem with inductance - read up on Zobel filters for more information.

Another consideration is that when you add an inductor, it adds a series resistance which will alter the crossover away from the ideal active crossover. It'll also alter the effective Qes (thus Qts) of the driver which needs to be considered in the cabinet design.

Rod Elliott has a really good article here - Passive Crossover Network Design
On the full-scale design of a passive crossovers. What it takes to do them properly.

So yes, it is possible to do this, but there are things to sort out at both ends (ie, alter the level in DSP to account for series resistance, flatten the impedance curve for the passive crossover) before you can reliably jump between the two.

Chris
 
  • Like
Reactions: 1 user
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.