Crossover Clarification

[jheu02]

I have some nice older cabinets with walnut veneer that I want to put new drivers in to make a "vintage looking" 2 way set with new drivers. I want to try some of the Jamo (Philips) drivers that are on clearance at PE and after running the numbers for an 8" woofer have decided on either JAMO 22316 or Jamo 20418. Both woofers are 8 Ohm. I don't want to put a new front baffle on or alter the external appearance of these, so I realize some compromises will be made.

I'd like to match them to JAMO 20138 1" Soft Dome Shielded Tweeter 4 Ohm. If I do this I'll need to wire in series for a 12 Ohm speaker correct (parallel calculates out to 2-3 Ohm IIRC)?

My question involving the crossover is this: I want to keep it uber simple ala EPI T/E series and just have a capacitor for the high pass. When wiring the cap into the circuit, I'd wire from the neg woofer terminal thru the cap to the pos tweeter terminal correct? If so, do I still choose the cap value based on the driver's 4 Ohms or 12 Ohm since the signal's going thru the woofer?

My other thought is to not go all Jamo and use this cone tweeter:
Pioneer FD66AP45-13F. That would keep both drivers at 8 Ohms so I could wire parallel for a 4 Ohm system. Does anyone know what the Fs of this tweeter is? The reviews seem decent and it would drop right in to the existing cutout. Would it's 93dB be too high compared to the woofer's 87dB?

Thanks for all the help,
John

 

[Nat]

The rules you are using for paralleling speakers really only works with resistors. Speakers have reactance and reluctance as well as resistance -- reactance is the reverse voltage of the capacitors and reactance is the reverse voltage of the inductors (coils in the crossover if you use them, and the voice coils of the drivers). The reverse voltage diminishes the potential difference so less current flows, but not because of resistance. That's the reason for resistance and impedance being different. Reactance and reluctance are both frequency dependent which also makes them different from resistance.
I think you might do well to either study crossover design a bit, or use one of the many crossover calculators on the net to get you started on what values of components and what drivers to use -- rarely do calculators really get it right, so you may need to tinker a bit, but its a start.
Just an aside -- you have made sure your woofers will work in your enclosures, right?

 

[jheu02]

The rules you are using for paralleling speakers really only works with resistors. Speakers have reactance and reluctance as well as resistance -- reactance is the reverse voltage of the capacitors and reactance is the reverse voltage of the inductors (coils in the crossover if you use them, and the voice coils of the drivers). The reverse voltage diminishes the potential difference so less current flows, but not because of resistance. That's the reason for resistance and impedance being different. Reactance and reluctance are both frequency dependent which also makes them different from resistance.
I think you might do well to either study crossover design a bit, or use one of the many crossover calculators on the net to get you started on what values of components and what drivers to use -- rarely do calculators really get it right, so you may need to tinker a bit, but its a start.
Just an aside -- you have made sure your woofers will work in your enclosures, right?

I'll go back and re-read the sources I've been using, but I thought I had at least a basic understanding of crossovers. The online and book sources both state that when you run a parallel connection from the terminals to the individual drivers, the impedance is (simplified explanation) cut in half. i.e. 8 ohm woofer and 8 ohm tweeter, wired parallel will show a system impendance to the amp of 4 ohm, whereas, if run in series, it would become a 16 ohm system load as seen by the amp. Is this not correct?

The woofers I chose can both be used in the enclosure, that was my starting point. Then I went to choose a tweeter and figure the crossover point from there. Unfortunately, the tweeter I wanted to use is 4 ohm and the woofer 8ohm. I was wondering if it could be done when wiring in series. I know that it's recommended to use drivers all of the same impedance, so I guess I'll need to go with a different tweeter. I didn't want to put a lot of money into these in either very expensive drivers, or a multi-component crossover, as I have some EPI speakers and others where the entire crossover is just the capacitor to the tweeter (high pass). The woofer is allowed to just roll off naturally. That was my plan, though I could always go back in and add an inductor for the woofer if it doesn't roll off adequately. I'm just planning a 1st order cross.

Thanks for the reply and hope to get further clarification.

John

 

[Nat]

If you were running both the woofer and tweeter full range, the calculations you are using would be closer to actuality (even though individual drivers have frequency dependent impedances -- just check the resistance of an 8 ohm driver and you'll see that you get something more like 5.4 ohms since you are only testing resistance with DC). But once you start using driver in specific frequency ranges with caps and coils to roll them off, than its more like they are each operating by themselves in that specific range, rather than paralleled.

 

[jheu02]

If you were running both the woofer and tweeter full range, the calculations you are using would be closer to actuality (even though individual drivers have frequency dependent impedances -- just check the resistance of an 8 ohm driver and you'll see that you get something more like 5.4 ohms since you are only testing resistance with DC). But once you start using driver in specific frequency ranges with caps and coils to roll them off, than its more like they are each operating by themselves in that specific range, rather than paralleled.

OK, I follow what you're saying, I think. Let me ask the question this way then:

If I use an 8 ohm woofer and 8 ohm tweeter in parallel, really the only time they're presenting a 4 ohm load is at the crossover frequency. So if I wanted to, I COULD use a 4 ohm tweeter matched with an 8 ohm woofer. I'd base the tweeter's high pass cap value on the 4 ohm impendance, the woofers inductor value on the 8ohm impedance, there'd be a potentially problematic load on the amp around the crossover frequency, and the tweeter would behave like it's getting twice the power compared to the woofer, thus making it way more forward and unbalanced as opposed to the woofer.

Is that the real reason behind the recommendation to use drivers of the same impedence?

Thanks for helping me understand the reasoning behind the calculations.

 

[Nat]

First: I'm no expert. But I think that since the amp is seeing the drivers through the crossover components which have increasing impedance around the crossover frequencies so actually the amp is seeing two higher than the nominal resistances in parallel, so the total impedance is safe (though it can still be a difficult load for other reasons). I think the idea behind matching the impedances of drivers is simply to try to get them more or less equally efficient -- if you have to use resistor to pad down one driver, you are taking the amplifier's power and converting that amount to heat, which is a waste, and also has a bad effect on damping.