Minimum acceptable impedance

Ron_E

I've simmed a crossover for a 3 way and can't get around a dip in impedance that drops down to 3 ohms around the crossover point between the woofer and mid. It's a combination of a 4 ohm mid with barely enough sensitivity that causes the problem The crossover points are 800 and 4K. Is a drop to 3 ohms tolerable for a 4 ohm capable amp?

Zin


Zin Phase


FR Sim


Crossover


Ron

cssaudio1

That looks fine to me Ron.  Where you dip to 3 ohms, the phase angle is nearly 0 and there won't be a ton of power at 1kHz anyway.  Any 4 ohm capable amp should have no problem with that.

Dan

Ron_E

Thanks Dan.  I'll order the crossover parts today.

Ron

PWRRYD

I concur,

jhollander

Agree. I assumed you tried swapping the mid high pass and low pass circuits?

I wouldn't mind playing with the response if the FRDs and ZMAs are transportable from Soundeasy.   I'm between projects and am itching to start another.

Ron_E

I did swap the high and low filters on the mid around. It moved the impedance drop to the 4K crossover point. Thanks for the offer to play with the design but I need to get the parts ordered so I can finish them up.

Ron

jhollander

This transfer function is close to what you have shown and is at 4 ohms.

The 15 uF cap to ground in the mid circuit is what's pulling the impedance down.

Wolf

I wondered if it was a circuit resonance...

jhollander

There's definitely something with the circuit as I nearly matched Ron's impedance in the 1K to 4K range before connecting a dummy driver

Ron_E

You get an A+. 

Original FR and modified crossover FR


Zin


Zin Phase


Did you have some method to the improved version or was it just trial and error with a simulation? I haven't put actual DCR values in the sim for the inductors but they won't change the final FR much. It looks good.

Thanks,
Ron

jhollander

When you build a circuit in XSim you can isolate the components to see what's dropping the impedance. 

After that it was tweaking the components to get the same XSim dummy frequency response as your original x-o.  By matching the two frequency responses the transfer functions should be pretty close.

Ron_E

I corrected the DCR values and dropped the 2.7 ohm resistor to 2.5 and I'm happy.



Ron

Ron_E

jhollander said:

When you build a circuit in XSim you can isolate the components to see what's dropping the impedance. 

After that it was tweaking the components to get the same XSim dummy frequency response as your original x-o.  By matching the two frequency responses the transfer functions should be pretty close.

I'd say that worked pretty well.

Thanks again,
Ron

greywarden

I made a little calculator for the inductance of the load, basically you multiply the tangent of the angle by the resistance, which is -11.186 (5R @ -20*) vs 0 (3R @ 0*, -4.6 if you change it to -1*) but I have no idea what that really means. Any engineers have any input? 

Ron_E

I bread boarded the crossover and took some measurements. They're a little off compared to the sim but acceptable. The correlation looks pretty spot on between the sim and the actual response from 800hz up which is where the woofer and mid cross. I'll hold off final judgement until I can listen to the pair. I did try dropping the pad on the mid but of course that dropped the impedance so I think I'm going to call it good. A little extra bass will probably be just fine with the new owners.

Zin and Phase


FR


Ron