Horn driver impedance

[dirkwright]

I'm designing a midbass tractrix horn using a B&C 12PE32 driver. I need to design the crossover for it also, of course. I've looked at some reference material and I can't seem to find out how the horn affects the driver's impedance curve as provided by the manufacturer. Maybe it doesn't affect it all. Obviously, I need this information in order to design a proper crossover. The back chamber will be sealed, so I won't have the two impedance peaks of a vented enclosure. Do the different horn expansion shapes have different effects on the driver impedance?

I also need to design better crossovers for my midrange and tweeter horns. I just bought the DE10 with matching ME10 tweeter horn. I have an Edgar tractrix midrange horn for the Dynaudio D54 driver.

Thanks in advance for any help.

[Brett]

Model in Hornresp and you'll get to see the Z curve as one option.

[AllenB]

As unfortunate as it may seem, measuring it is your best bet, but it really isn't that hard. It takes a couple of resistors, some clip leads and a computer.

[dirkwright]

Originally Posted by Brett

Model in Hornresp and you'll get to see the Z curve as one option.

OK, thanks. I did that, and this B&C driver on a 1/4 size 60 Hz tractrix horn still shows the typical rising impedance with frequency due to voice coil inductance, and the peak at Fs as well. So, in order to design a passive crossover, I still need to install compensation filters also.

[dirkwright]

Originally Posted by AllenB

As unfortunate as it may seem, measuring it is your best bet, but it really isn't that hard. It takes a couple of resistors, some clip leads and a computer.

Yeah, thanks. My main question I need to answer myself is "how does the driver electrical impedance change when the driver is installed in a horn?". I can see from Hornresp that the curve for the B&C 12" driver is basically the same shape but I don't know if the values have changed. I guess I really need WT3 or equivalent. I can't design a passive crossover unless I know the impedance right around the crossover frequency, now can I?

I'm also wondering how the driver's resonance affects the frequency response near the horn's cutoff frequency. Particularly on a tractrix flare. If tractrix horns are particularly sensitive to this, then that may explain the poor bass performance people have reported for them.

[AllenB]

Originally Posted by dirkwright

I can't design a passive crossover unless I know the impedance right around the crossover frequency, now can I?

Correct, but it only matters what it is now and not how it changed. Am I missing something?

[John Sheerin]

You need to know the impedance of all your drivers at all frequencies, not just the crossover frequencies, just like you need to know the frequency responses of the drivers in the cabinet. Measuring all this is the only way to go. Simulation is to get you something close to what you want to work with. Then you design and build your cabinet, build it and measure the drivers in it. Then you design a crossover.

[dirkwright]

Originally Posted by AllenB

Correct, but it only matters what it is now and not how it changed. Am I missing something?

No, I'm just curious about this, that's all. I guess "everything affects everything" is a good way to think about it.

[dirkwright]

Originally Posted by John Sheerin

You need to know the impedance of all your drivers at all frequencies, not just the crossover frequencies, just like you need to know the frequency responses of the drivers in the cabinet. Measuring all this is the only way to go. Simulation is to get you something close to what you want to work with. Then you design and build your cabinet, build it and measure the drivers in it. Then you design a crossover.

Thanks. Time to buy some test gear.

[AllenB]

For what it's worth, if you were designing without measuring and created an impedance rise compensation filter based on the raw impedance, I doubt it would change much in the box and I'd leave it at that.