Noob Question: What's the preferred method of attenuating a tweeter to the woofer?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey, guys

I have a 6 ohm tweeter and an 8 ohm woofer and was curious as to what was you preferred method of attenuating the tweeter to match the woofer?

(Side note: I did a search but didn't quite understand all that was said, as such putting it into simple laymans terms would be greatly appreciated)
 
Moderator
Joined 2011
I have a 6 ohm tweeter and an 8 ohm woofer and was curious as to what was you preferred method of attenuating the tweeter to match the woofer?

An L-pad. It's two variable resistors connected together, with about the same total impedance as the driver.
When you adjust the knob, one resistor goes up, while the other resistor goes down.
It connects between the crossover and the driver. You can also use two resistors of the right values instead.
Variable L-pad: Shavano Music Online - Using L-Pads
Fixed resistors: L-Pad Driver Attenuation Circuit) Designer / Calculator Help
 
Last edited:
You attenuate a tweeter's sensitivity so it matches that of a woofer, so that their sound pressure level, expressed in decibels, are the same, regardless of their impedance (ohms). The impedance of the individual drivers, along with the desired attenuation in decibels are used to figure out the values of the resistors in the L-pad. In other words, you need to figure out the difference in sensitivity of the drivers, and how much padding down you need to do.
 
So if I'm to understand you correctly, so long as the spl levels of the two drivers match up, the impedance between the two don't (hypothetically) matter? The reason I ask of this is because the drivers I'm looking at are of different impedances, however, they have the same levels is spl. At that point, do the differences in impedances still matter?

(Thanks for the continued help in advance guys, it's appreciated)
 
At this point it becomes a little more technical ! As the impedance Z has an imaginary part :eek::eek: then there's the amplifier, constant voltage type which tries to keep up with the request of current. The critical band is the bass region where the max phase rotation before breakdown is 35° ( plus and minus), referred to solid state devices and to the lowest Z accepted, often not under 4 Ω
 
Last edited:
So if I'm to understand you correctly, so long as the spl levels of the two drivers match up, the impedance between the two don't (hypothetically) matter?
Yes

The reason I ask of this is because the drivers I'm looking at are of different impedances, however, they have the same levels is spl. At that point, do the differences in impedances still matter?
You can match drivers with different impedance, you still need to check the final impedance (drivers + crossover) over the full audio band, and decide if it is an easy load for the amp or not.

A word of caution on drivers with the same sensitivity (same SPL number). For a tweeter the SPL you see on the spec is the SPL you will get. For a midwoofer this is not true because of the baffle step effect: depending on the baffle width, there is a frequency where gradually the output decreases as the frequency goes down, up to a 6dB of loss. This is because as the frequency lowers, the wave length increases and at a certain point the baffle is "tiny" compared to the wave length and the wave wraps around the cabinet, and only half of the output is in front of the speaker as the other half is behind. In order to get a balanced SPL from bass to mid you need to account for this loss (partially or totally), and set the total SPL to a value in between (depending on the speaker location one can apply a total baffle step compensation, i.e. 6dB, or a smaller value, e.g. 3-4dB).
All this talk to say that if you have say a tweeter and a woofer both rated at 90dB, once installed in a box with an appropriate filter, the SPL of the woofer will be 84-87dB, and won't match the tweeter anymore, so at the end you have to attenuate also the tweeter.

Apart the l-pad, you can attenuate a tweeter also by using a single appropriate resistor in front of the tweeter filter.

Ralf
 
There are 3 standard ways to pad a driver, 2 of which have already been mentioned:

1. L-pad which maintains the shape of the driver's FR

2. series resistance before the filter which tends to give an upward tilt to the response

3. series resistance after the filter which tends to give a downward tilt to the response

Numbers 2 and 3 will also result in increases to the impedance which can frequently be a useful goal; are perhaps a little easier to adjust when fine-tuning (changing 1 value instead of 2); and can also be combined together to achieve the desired response target.
 
Moderator
Joined 2011
the drivers I'm looking at are of different impedances, however, they have the same levels is spl.
At that point, do the differences in impedances still matter?

Most crossovers have the filter sections for each driver all in parallel, driven by a voltage source (the amplifier).
Each section is then electrically independent, and the driver impedance does not affect the other filter sections.
You could disconnect entirely one of the filter sections, and the others would not be affected electrically at all.
 
Last edited:
Perhaps as an extension of my question - if the two drivers are of the same impedance, yet have different degrees of spl sensitivities, how would one be able to match them so as to have them attenuated to the same level (ex: a tweeter and woofer of the same impedance, yet the woofer has higher spl levels, how would they be attenuated and matched to the same level) Or does this not matter as long as the impedances are the same?

Thanks in advance for the continued help guys. Much love
 
diyAudio Moderator
Joined 2008
Paid Member
If the tweeter wants less power then it needs less current and that equates to a higher impedance. You can arrange it (using a series and a parallel resistor, an L-pad) so that you have 8 ohms and lower tweeter level but in this case you'll pass more current than you need, the extra current is deliberately wasted.

It's a choice, and the amp will respond the way it responds, which will typically be no different. Most amps present a low source impedance and will be fine either way.
 
It's generally considered a no no to pad down the woofer, so if your woofer has higher SPL than your tweeter (after baffle step loss is accounted for, usually about 6dB) then you are kind of out of luck.

This is part of the design process: you want to match drivers that are going to work properly together, and one aspect of that means that you have to select a tweeter, or lets call it the driver(s) that handle the higher frequencies (so a mid and a tweeter in a 3-way) that are going to have high enough SPL's to match with the woofer SPL levels after bsc is accounted for.

Don't get too caught up in driver impedance matching - it's the summed FR that you are trying to design to a flat line not the impedance curve, as long as your amp can handle where ever the minimum impedance level(s) are at.
 
So...essentially what we're looking for is equivalent spl sensitivities/levels for all of the drivers to be matched equally, regardless of the impedances of the individual drivers? (Ex: say a tweeter at 94db at 6 ohm and a woofer at 94db at 8 ohm - that scenario would be fine considering that they are equal spl levels?)

Thanks in advance for the continued help guys, it's greatly appreciated. Much love
 
You should read how the measurement is done.
I haven't done any, but the procedure is easy: you fix the driver to a baffle or leave it naked; you "feed" it with a sine tone centered at 1 kHz
:confused:
You place on axys at 1 meter a microphone; all the system is calibrated, so you get 1 W output from DUT ( device under test ) by feeding it with 2,83 V
The SPL that the microphone indicates is what the sensitivity data is.
So it's referred to the distance unit and to power or voltage ( dB/W/m or dB/V/m ).
To let a 4Ω DUT work at 1 W you need " only" 2 V.

Another thing is the FR ( frequency response ) graph. The DUT is feed with a variable frequency source - it's called a "sweep" then the response is put in graph...

Different from sensivity is the efficiency - but similar :p

So, in the end, you'll find that combining drivers having different Z's, the one with the lower Z will sound louder ( ok, you rescued the data before and you knew it ;) ) very often ...
 
ReDress, I have a suspicion that you have yet to understand baffle step loss.

Read these:

True Audio Speaker Topics: Spatial Loading
True Audio TechTopics: Diffraction Loss

So if your speaker is operating in a more or less 2pi environment, ie. some distance from most boundaries, your 94dB woofer will be down 6dB at its lower frequencies, at about 88dB. This level is what determines your speaker sensitivity and is the level that you need to match your tweeter to. This doesn't mean that you must select a 88dB tweeter. The 94dB one in your example will be fine after you just bring it's level down to 88dB to match the woofer with one of the padding resistor methods previously described.

In point of fact, you are probably better off starting with a tweeter that has higher SPL levels than the 88dB in this example so that you have some flexibility to make adjustments. Let's say that your speaker placement has to be closer to some boundaries so that the real world baffle step loss might only be about 3 or 4dB. So now the woofer level is going to be closer to 90 or 91dB and you will now need a tweeter capable of matching those levels. The 94dB tweeter again still works with some padding. A 91dB tweeter will probably work as long as your xo doesn't tilt the FR too much in a downward direction. The 88dB tweeter would definitely not work in this situation.

Notice again, that I'm totally unconcerned with individual driver impedances in these examples as long as your amp is going to be happy with whatever the minimums are going to be.

Simulating the speaker before buying your drivers is always a good idea. If you don't know how to do that many here will probably suggest that you start with a kit or proven design, which isn't a bad idea. I get the feeling you want to learn how to do it yourself, in which case you need to start working through the Speaker Building Bible.
 
diyAudio Moderator
Joined 2008
Paid Member
So...essentially what we're looking for is equivalent spl sensitivities/levels for all of the drivers to be matched equally, regardless of the impedances of the individual drivers? (Ex: say a tweeter at 94db at 6 ohm and a woofer at 94db at 8 ohm - that scenario would be fine considering that they are equal spl levels?)

Thanks in advance for the continued help guys, it's greatly appreciated. Much love
Yes, in a nutshell.

Ok, so there are other things but you have to get this far first ;)
 
There are 3 standard ways to pad a driver, 2 of which have already been mentioned:

1. L-pad which maintains the shape of the driver's FR

2. series resistance before the filter which tends to give an upward tilt to the response

3. series resistance after the filter which tends to give a downward tilt to the response

Numbers 2 and 3 will also result in increases to the impedance which can frequently be a useful goal; are perhaps a little easier to adjust when fine-tuning (changing 1 value instead of 2); and can also be combined together to achieve the desired response target.

Your sentiments aren't quite accurate above. A series resistor before the xover does not really impart any tilt to the response, just as a shunt resistor imparts practically no tilt when placed across the tweeter. An L-pad can actually tilt the response down due to the series resistor post-network.

Later,
Wolf
 
diyAudio Moderator
Joined 2008
Paid Member
It depends too. A series resistance before or after a capacitor filter makes no difference.

A series resistor before the driver, or a parallel one each make the impedance more resistive. As with the L-pad combination, the impedance will be somewhere between the tweeter impedance and a resistor, except that with series resistance there will be some response modification independent of the preceding filter.
 
It depends too. A series resistance before or after a capacitor filter makes no difference.

A series resistor before the driver, or a parallel one each make the impedance more resistive. As with the L-pad combination, the impedance will be somewhere between the tweeter impedance and a resistor, except that with series resistance there will be some response modification independent of the preceding filter.

Agreed- series filters without shunts involved, make the position of the components irrelevant.

I disagree with your statement above I made bold. The series resistance added to the reactive impedance will only make the level of impedance higher and will not change the amount of variation inherent in the driver's impedance. It will not be more resistive with only a series resistor element.
There has to be a paralleled resistor to make the impedance delta less than naturally inherent in the driver.

Later,
Wolf
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.