Hi there,
I'm trying to lower the output of a tweeter by 5dB in total. The issue is the tweeter impedance is 4ohms (2.8 min) and the woofer being matched to is 8ohm. I want to increase the min impedance so the box can be rated nominally "8 ohm"
What I want to know is how the response would be affected if I wanted to add resistance to the attentuation circuit? Eg. I lower the output by 2dB with an attenuation circuit - 0.8R in series, 15.5R in parallel (worked out using the nominal 4R impedance - but should I be working the figures out from 2.8R min!?).
If I then want to add 4ohm series resistance to the circuit in order to further lower tweeter output 3dB for 5dB total attenuation can I just add this value to the series resistance? The calculators I've seen warn against the resistance values deviating the overall resistance but how does this actually affect response?
My idea was simply to use a 2dB L-pad and then add 4ohms in series to further add 3dB of attenuation to make 5dB drop total. Is this not possible?
Or is there another way to equalize the impedance of the tweeter circuit to a nominal 8ohm that I am unaware of?
Sorry for the long post, look forward to your replies!!
I'm trying to lower the output of a tweeter by 5dB in total. The issue is the tweeter impedance is 4ohms (2.8 min) and the woofer being matched to is 8ohm. I want to increase the min impedance so the box can be rated nominally "8 ohm"
What I want to know is how the response would be affected if I wanted to add resistance to the attentuation circuit? Eg. I lower the output by 2dB with an attenuation circuit - 0.8R in series, 15.5R in parallel (worked out using the nominal 4R impedance - but should I be working the figures out from 2.8R min!?).
If I then want to add 4ohm series resistance to the circuit in order to further lower tweeter output 3dB for 5dB total attenuation can I just add this value to the series resistance? The calculators I've seen warn against the resistance values deviating the overall resistance but how does this actually affect response?
My idea was simply to use a 2dB L-pad and then add 4ohms in series to further add 3dB of attenuation to make 5dB drop total. Is this not possible?
Or is there another way to equalize the impedance of the tweeter circuit to a nominal 8ohm that I am unaware of?
Sorry for the long post, look forward to your replies!!
Hi,
Why not just use a 3.1Ohm series resistor? This will give you 5dB attenuation if your tweeter has an impedance of 4Ohm and will raise your total impedance to 7.1Ohm. Quite close to your design goal. Just make sure the resistor can handle the power since it will dissipate almost halve of the power provided by your amp.
Good luck!
Why not just use a 3.1Ohm series resistor? This will give you 5dB attenuation if your tweeter has an impedance of 4Ohm and will raise your total impedance to 7.1Ohm. Quite close to your design goal. Just make sure the resistor can handle the power since it will dissipate almost halve of the power provided by your amp.
Good luck!
Is this an existing speaker system or something you're designing? If there's already a passive crossover, adding series resistance only will lower the crossover frequency to the tweeter proportionally. Otherwise if it's something you're in the process of designing, series resistance will alter the damping of the tweeter, but if the crossover frequency is well above resonance, it will have minimal effect on the final response.
Mike
Mike
Hi and thanks for the replies!
So the simplest way might be just to add a series resistance as Mark suggested and then treat the circuit as an 8ohm as far as the crossover goes.
Michael, could you tell me more about how the tweeter damping is affected. In this case the crossover is around 2khz, resonance is around 800hz so should be well clear of that.
I was under the impression that adding series resistance without changing the inductor and capacitor values would alter the Q of the crossover slope rather than the crossover frequency?
This is a small speaker I'm designing just for fun!
So the simplest way might be just to add a series resistance as Mark suggested and then treat the circuit as an 8ohm as far as the crossover goes.
Michael, could you tell me more about how the tweeter damping is affected. In this case the crossover is around 2khz, resonance is around 800hz so should be well clear of that.
I was under the impression that adding series resistance without changing the inductor and capacitor values would alter the Q of the crossover slope rather than the crossover frequency?
This is a small speaker I'm designing just for fun!
Hi chucky12,
Have you ever tried to simulate the filter with a simple program like AADE filter design (free program)? This will show you the slopes / attenuation quite nicely.
As source resistor use 0.1Ohm and as termination resistor 4Ohm (for the AADE sim)
Take a look at the voltage effective gain and set the parameters to Freq1 = 10. | Freq 2 = 100.K | db1 = 10. | db2 = -60.
Have you ever tried to simulate the filter with a simple program like AADE filter design (free program)? This will show you the slopes / attenuation quite nicely.
As source resistor use 0.1Ohm and as termination resistor 4Ohm (for the AADE sim)
Take a look at the voltage effective gain and set the parameters to Freq1 = 10. | Freq 2 = 100.K | db1 = 10. | db2 = -60.
Yes, for a second order electrical circuit, Michael Bean is describing a first order circuit.
Many speakers dip close to (or below) 4 ohms. Even if you add just an ohm or two, you'll probably get away with it.
Hi Allen, sorry you're right I should have been more clear there's a 2nd order filter involved!
Mark, I'll check out that program, thanks!
In order to correctly calculate the Q of my filters then I guess I should be using the accurately measured impedance around the crossover frequency for both drivers? Maybe somewhere in between the min and nominal figures?
Mark, I'll check out that program, thanks!
In order to correctly calculate the Q of my filters then I guess I should be using the accurately measured impedance around the crossover frequency for both drivers? Maybe somewhere in between the min and nominal figures?
It really depends on how far you want to go... whether you want to be able to fully simulate a response using measured data, or you will be designing by hand.
Using the impedance at the crossover point should get you close, i.e. to within tweaking distance. If the impedance has any gross deviation within an octave of the crossover point then results will be less predictable.
Using the impedance at the crossover point should get you close, i.e. to within tweaking distance. If the impedance has any gross deviation within an octave of the crossover point then results will be less predictable.
Thanks again for the help everyone. As far as I can tell the impedance of the tweeter rises from around 3ohms at crossover to a maximum around 4ohms around 20khz.
Unless my maths is wrong then adding a series resistor will drop the level up to 5dB at crossover but only 4dB ish up top. I suppose a 1dB upward slope over the entire range isn't that bad though!
Kahooli, I've not heard of a pi-pad or t-pad but I'll look into it thanks!
Unless my maths is wrong then adding a series resistor will drop the level up to 5dB at crossover but only 4dB ish up top. I suppose a 1dB upward slope over the entire range isn't that bad though!
Kahooli, I've not heard of a pi-pad or t-pad but I'll look into it thanks!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.