Mixing Driver Impedance

[soda1]

I would like to replace a tweeter rated at 8 ohms with one rated at 4 ohms. The woofs and mids are rated at 8 ohms. Crossovers are simple first order. Is this ok and what will it do to the total impedance which now measures approx. 8 ohms?

My amp has taps for 4, 8, 16 ohms.

Soda1

[Chris Roemer]

> I would like to replace a tweeter rated at 8
> ohms with one rated at 4 ohms. The woofs and
> mids are rated at 8 ohms. Crossovers are
> simple first order. Is this ok and what will
> it do to the total impedance which now
> measures approx. 8 ohms?

> My amp has taps for 4, 8, 16 ohms.

> Soda1

Your system's "nominal impedance" will stay about the same. If you just have a cap to block bass from the tweeter, you'll have to parallel it with one of like value (or just swap it out with one of double the value), otherwise your tweeter will cross approx. one octave higher. Typically, a similar 4 ohm tweeter will end up being more efficient (3dB is possible), so you might end up having to attenuate it with about a 4 ohm resistor. If you put the resistor right next to the tweeter, you can leave the cap alone as you've just made your 4 ohm tweeter into a 3dB less efficient 8 ohm tweeter, and the cap will function as before (sort of).

[soda1]

> Your system's "nominal impedance"
> will stay about the same. If you just have a
> cap to block bass from the tweeter, you'll
> have to parallel it with one of like value
> (or just swap it out with one of double the
> value), otherwise your tweeter will cross
> approx. one octave higher. Typically, a
> similar 4 ohm tweeter will end up being more
> efficient (3dB is possible), so you might
> end up having to attenuate it with about a 4
> ohm resistor. If you put the resistor right
> next to the tweeter, you can leave the cap
> alone as you've just made your 4 ohm tweeter
> into a 3dB less efficient 8 ohm tweeter, and
> the cap will function as before (sort of).

Thanks for the quick reply, CJ!

I was infact planning to double the value of The capacitor from it's current value of 4 uf to 8 uf to maintain the same xover points....4750 dB.

However, since I am planning to install adujustable L-pads on all drivers, would it still be a good idea to install the 4 ohm fixed resistor on the tweet?

Soda1

[Chris Roemer]

> Thanks for the quick reply, CJ!

> I was infact planning to double the value of
> The capacitor from it's current value of 4
> uf to 8 uf to maintain the same xover
> points....4750 dB.

> However, since I am planning to install
> adujustable L-pads on all drivers, would it
> still be a good idea to install the 4 ohm
> fixed resistor on the tweet?

> Soda1

Either put a 4 ohm L-pad on the tweet with the 8uF cap, OR keep the 4uF cap and put the 4 ohm resistor in line with the tweet, and use an 8 ohm L-pad on it. No L-pad on the woofer BTW.

[soda1]

> Either put a 4 ohm L-pad on the tweet with
> the 8uF cap, OR keep the 4uF cap and put the
> 4 ohm resistor in line with the tweet, and
> use an 8 ohm L-pad on it. No L-pad on the
> woofer BTW.

Since I already have an 8 ohm adjustable L-Pad, I will probably go with the latter. I assume that a ceramic wire wound resistor rated at 10 watts would be sufficient...Is this correct? I will install the 4 ohm resistor between the driver and the L-Pad inline with the positive driver lead....Is this also correct?

Also, why no adjustable L-Pads on the woofs? I have used them in the past with no apparent problems. I use them to balance the left and right driver outputs with my sound level meter, and, also to turn the woofs off when balancing the other drivers.

[Chris Roemer]

> Since I already have an 8 ohm adjustable
> L-Pad, I will probably go with the latter. I
> assume that a ceramic wire wound resistor
> rated at 10 watts would be sufficient...Is
> this correct? I will install the 4 ohm
> resistor between the driver and the L-Pad
> inline with the positive driver lead....Is
> this also correct?

> Also, why no adjustable L-Pads on the woofs?
> I have used them in the past with no
> apparent problems. I use them to balance the
> left and right driver outputs with my sound
> level meter, and, also to turn the woofs off
> when balancing the other drivers.

You should have the tweeter nailed. How big of L-pads do you have? Many woofs can take 50, 100, or 200 RMS or more. I realize that you might not be attenuating them at high output levels. Mids and tweeters are most typically several dB more efficient than woofers, hence the L-pads. Also, a midrange has what's referred to as passband gain, so they normally need a tad more shelving than their SPL specs would indicate. And then there's BSC . . .

[winstonorcutt]

> Since I already have an 8 ohm adjustable
> L-Pad, I will probably go with the latter. I
> assume that a ceramic wire wound resistor
> rated at 10 watts would be sufficient...Is
> this correct? I will install the 4 ohm
> resistor between the driver and the L-Pad
> inline with the positive driver lead....Is
> this also correct?

> Also, why no adjustable L-Pads on the woofs?
> I have used them in the past with no
> apparent problems. I use them to balance the
> left and right driver outputs with my sound
> level meter, and, also to turn the woofs off
> when balancing the other drivers.

In a simple first-order parallel circuit the impedance of the tweeter branch (tweeter impedance plus resistor or l-pad) and capacitor value will affect the balance of the tweeter and midrange. The tweeter circuit is, in effect, a Zobel. One of the reasons first order networks are so unreasonably difficult to get right is that they are highly interactive. Your l-pad should provide some flexibility to get the balance right.

It is not considered good practice to insert resistive elements in the woofer circuit. They degrade the woofer/amplifier interface.

Have fun!

[paul_n_oneal]

> It is not considered good practice to insert
> resistive elements in the woofer circuit.
> They degrade the woofer/amplifier interface.

> Have fun!

I've never read this or heard this from anyone on this board before. From time to time, when needed, a resistor on the shunt side of a woofer can bring the FR Phase into alignment IMHO.

Paul

[winstonorcutt]

> I've never read this or heard this from
> anyone on this board before. From time to
> time, when needed, a resistor on the shunt
> side of a woofer can bring the FR Phase into
> alignment IMHO.

> Paul

Gee, Paul, it seems to me that it is covered in most texts I have read. Here's an example:

"Moderate resistive attenuation does not unduly affect the driver and crossover performance of the mid and high frequencies, but conversely is likely to significantly disturb an LF unit especially near the bass resonance, and sensitivity control must be ruled out for the bass unless a large auto-transformer is employed. It is therefore usual to direct couple the LF unit via the crossover, and to reserve any attenuation for the remaining drivers."

This is from Martin Colloms: High Performance Loudspeakers, Sixth Edition (2005) p. 269.

Series resistance (including cable resistance) degrades the amplifier's damping (control) of the LF unit.

Win

[paul_n_oneal]

The next time you have a problem aligning the phase of the woofer and midrange try a 0.5ohm to 1.0ohm on the shunt side of the capicitor of the woofer circuit. You can also model this and see the same thing.

Examples: Zobel's use resistors on the shunt side. A series LC/LCR on the woofer circuit can be used to tame peaks and induce the correct BSC.

It's a lot more complex than coming out and saying that resistors should not be used on woofer circuits. Indeed, I see them used all the time.

Paul

[winstonorcutt]

> The next time you have a problem aligning
> the phase of the woofer and midrange try a
> 0.5ohm to 1.0ohm on the shunt side of the
> capicitor of the woofer circuit. You can
> also model this and see the same thing.

> Examples: Zobel's use resistors on the shunt
> side. A series LC/LCR on the woofer circuit
> can be used to tame peaks and induce the
> correct BSC.

> It's a lot more complex than coming out and
> saying that resistors should not be used on
> woofer circuits. Indeed, I see them used all
> the time.

> Paul

I don't want to spend the rest of the week peeing on each others' shoes over what was intended to be a helpful comment and will make this my last contribution on this matter.

I am guilty of ambiguity. What I should have said specifically is that it is not considered good practice to insert resistive elements in series with woofers. Please accept my apologies.

[paul_n_oneal]

> I don't want to spend the rest of the week
> peeing on each others' shoes over what was
> intended to be a helpful comment and will
> make this my last contribution on this
> matter.

> I am guilty of ambiguity. What I should have
> said specifically is that it is not
> considered good practice to insert resistive
> elements in series with woofers. Please
> accept my apologies.

I can understand that, and I also didn't want to pee on each others' parade. I noticed on your last post that you said series and the light came on. I can agree that a series resistor used to pad a woofer down to match the mid or tweeter is a bad thing and shows poor design and choice of drivers. Thanks for clarifying.

Paul