Don't forget that there are crossover components as well. Check out some Stereophile on-line speaker reviews,
they plot speaker system impedance and phase. Usually, each driver in the speaker system has some crossover
components between it and the amplifier, and the inputs to these driver circuits are all connected in parallel
for the amplifier to drive. Sometimes other ways are used to connect the components, though.
Last edited:
In the Beginning, God created Spice, and all was good.
Loudspeakers and crossovers were modeled in Spice before SPL traced based crossover design tools became freely available.
There are a few tutorials on how to use the speaker T/S parameters to create a decent Spice model. You can use LTSPICE with a simple amplifier model to get impedance vs. frequency, phase, even SPL vs. freq with your crossover. Good for early paper design work.... before you take out a mic and measure all of your speakers in the cabinet for detailed SPL data for final simulation.
=====================
Computer-Aided Electroacoustic Design with Spice W. Marshall Leach
users.ece.gatech.edu/mleach/papers/spice_electro.pdf
http://micka.de/org/en/download/spice-tsp_en.pdf
www.micka.de/org/download/spicetsp.asc.zip
Another article on modelling of loudspeaker impedance: ePanorama.net | Audio | Video | Circuits | Electronics Design
Measuring Loudspeaker Driver Parameters
Loudspeakers
Loudspeakers and crossovers were modeled in Spice before SPL traced based crossover design tools became freely available.
There are a few tutorials on how to use the speaker T/S parameters to create a decent Spice model. You can use LTSPICE with a simple amplifier model to get impedance vs. frequency, phase, even SPL vs. freq with your crossover. Good for early paper design work.... before you take out a mic and measure all of your speakers in the cabinet for detailed SPL data for final simulation.
=====================
Computer-Aided Electroacoustic Design with Spice W. Marshall Leach
users.ece.gatech.edu/mleach/papers/spice_electro.pdf
http://micka.de/org/en/download/spice-tsp_en.pdf
www.micka.de/org/download/spicetsp.asc.zip
Another article on modelling of loudspeaker impedance: ePanorama.net | Audio | Video | Circuits | Electronics Design
Measuring Loudspeaker Driver Parameters
Loudspeakers
3-way loudspeakers means 3 different drivers with high-pass or low-pass (or both - for midrange) filter on each driver, and all combined in parallel (forget for this moment the serial crossover). Total impedance is a bit misleading, because the lowest impedance is what realy is counting (is more important). So, if the mid is 4-ohm and woofer and tweeter are 8-ohm, than the whole loudspeaker should be 4-ohm nominal impedance.
For sure if you put 8-ohm 3 drivers (without filters) in parallel, than you have 2.67-ohm loudspeaker.
Thank goodness we have moved on quite a bit from there. Spice never could do a good job of modeling a loudspeaker.
Hi,
Put simply no, not in the slightest. 3 8 ohm drivers
in a 3 way speaker = an 8 ohm speaker, not 2.67.
Doesn't matter whether the x/o is series or parallel.
rgds, sreten.
In reality unless you screw up the x/o the speakers
impedance is determined by the low bass section.
Last edited:
Yes, in practice with a proper crossover this is true, but he didn't understand that there's a crossover network necessary,
so in his terms, he was right.
Last edited:
Hi,
No. Just your ability to understand the OP's questions.
The OP had already understood parallel impedances.
rgds, sreten.
Last edited:
Of course Sreten is right, while Gedlee is accepting the wrong OP assumption, (which is that speakers in a 3 way speaker are actually in parallel, forgetting the necessary crossover).
Giving the automatic answer to a clearly wrong question is not helping the OP at all.
He should have spent a little more electronic ink , first explaining the error and then providing the correct answer: that his 3 way speaker with 3 8 ohms elements, is an 8 ohms load to his amplifier ... and not 2.66 ohms
Giving the automatic answer to a clearly wrong question is not helping the OP at all.
He should have spent a little more electronic ink , first explaining the error and then providing the correct answer: that his 3 way speaker with 3 8 ohms elements, is an 8 ohms load to his amplifier ... and not 2.66 ohms
Oh well I just did the unthinkable, I just use a professional LCR meter to measure my own 3 Way 150W RMS / 8 Ohm speaker.
Test Frequency .. Ohm .. capacitance .. D parameter
100 kHz .. 16.555 .. 73nF .. 0.77
10 kHz .. 8.861 .. 7.127 uF .. 3.97
1 kHz .. 6.223 .. 107 uF .. 4.19
120 Hz .. 6.34 .. 576 uF .. 2.75
100 Hz .. 6.09 .. 493 uF .. 1.88
It seems that what I have here is a curve in which the dissipation factor gets it peak at 1 kHz to 10 kHz.
Test Frequency .. Ohm .. capacitance .. D parameter
100 kHz .. 16.555 .. 73nF .. 0.77
10 kHz .. 8.861 .. 7.127 uF .. 3.97
1 kHz .. 6.223 .. 107 uF .. 4.19
120 Hz .. 6.34 .. 576 uF .. 2.75
100 Hz .. 6.09 .. 493 uF .. 1.88
It seems that what I have here is a curve in which the dissipation factor gets it peak at 1 kHz to 10 kHz.
Last edited:
Of course Sreten is right about the serial crossover topology, the combined nominal impedance of the loudspeaker is the same as in the parallel topology, but to the nub eye the drivers in the serial crossover topology seems to be serially connected (in reality they are not). Sape specifically asked about the parallel conection of drivers and I just didn't want to confuse him.
Bass section drains the most energy from the amplifier, so the total nominal (or minimal) loudspeaker impedance is determined by the woofer, except in cases where the midrange is crossed at low frequency, than the loudspeaker impedance is determined by the midrange impedance (if the midrange has lower impedance than woofer).
I hope Sape has more clear picture now.
Last edited:
So if the 8 ohm drivers would be connected parallel(is there any other way to do it?)
WITHOUT crossover components, then total impedance would be fluctuating around 2.667ohms?
Otherwise with (parallel) crossover components only God knows the impedance?
Btw I am beginning to hate designing this f***** loudspeaker ,too much weird stuff.
I am now listening the middle range driver Tangband w5-704s in a cardboard box.
It is just that middle range driver desides basically everything,sensitivity, crosspoints etc. You just add "supertweeter" and "superwoofer" to midrange.
WITHOUT crossover components, then total impedance would be fluctuating around 2.667ohms?
Otherwise with (parallel) crossover components only God knows the impedance?
Btw I am beginning to hate designing this f***** loudspeaker ,too much weird stuff.
I am now listening the middle range driver Tangband w5-704s in a cardboard box.
It is just that middle range driver desides basically everything,sensitivity, crosspoints etc. You just add "supertweeter" and "superwoofer" to midrange.
Last edited:
As already said, loudspeaker impedance is usually not constant with frequency.
When a single impedance value is given for a loudspeaker, it is its nominal impedance. It's an average value, roughly estimated, not calculated.
Well this is generally undestood allright.
So you really can not calculate the fluctuating impedance with xo components combining drivers,you only can measure it?
I wonder why speaker manufacturers don't follow the apparently entirely sensible approach taken by Kiriakos? Perhaps they don't have access to the sophisticated and expensive equipment he must have used - or simply wouldn't understand how to use it?
ps - this post is tongue-in-cheek and alludes to the perfidy and/or desire of H-Fi manufacturers to over-complicate matters in order to confuse a buying public whose ultimate aim is most likely to be to listen to and enjoy sound at an acceptable cost - the KISS Principle doesn't suit anyone in this business
If .. without .. then , yes, but that is not the case.
Your 3 way cabinet does have a crossover so it's a moot point.
No, God and anybody else who knows that a crossover separates speaker components in the frequency domain so at no frequency they are simply in parallel.
You always have a crossover element in series with each speaker component so they are never "just in parallel".
And that's the concept you must clearly understand, using simplified "perfect" speakers, before introducing more complex concepts such as driver complex impedance.
It's absolutely useless trying to make the OP understand more advanced concepts before he understands the most basic ones.
say you crossover is @ 300Hz and 2.8kHz.
In the frequency range of 0Hz to 300Hz the Bass driver is the only significant load that the amplifier sees.
In the frequency range of 300Hz to 2.8kHz the Mid driver is the only significant load that the amplifier sees.
In the frequency range of 2.8kHz to 10MHz the Treble driver is the only significant load that the amplifier sees.
Over the full range the amplifier sees only one driver with it's filter as it's only signifiacnt load.
The other drivers do impose a load when outside their frequency range, but the load only becomes significant near the crossover frequency. Usually you find that the effective load the amplifier sees at each crossover frequency is slightly above the nominal speaker impedance.
It the crossover that makes all this possible.
Remove the crossover and drive each speaker as if it were a full range driver and the amplifier would indeed sees 2.67ohms (very roughly).
In the frequency range of 0Hz to 300Hz the Bass driver is the only significant load that the amplifier sees.
In the frequency range of 300Hz to 2.8kHz the Mid driver is the only significant load that the amplifier sees.
In the frequency range of 2.8kHz to 10MHz the Treble driver is the only significant load that the amplifier sees.
Over the full range the amplifier sees only one driver with it's filter as it's only signifiacnt load.
The other drivers do impose a load when outside their frequency range, but the load only becomes significant near the crossover frequency. Usually you find that the effective load the amplifier sees at each crossover frequency is slightly above the nominal speaker impedance.
It the crossover that makes all this possible.
Remove the crossover and drive each speaker as if it were a full range driver and the amplifier would indeed sees 2.67ohms (very roughly).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.