Jump to content

Resistor value to attenuate the tweeter to match the woofer ?


Leo Veto

Recommended Posts

In my experience,the most important part of the crossover is finding the correct value resistor for the tweeter and I believe it is critical to within 5 decimal places past the ohm.For example: 3.3 ohm may be too loud while 3.4 ohm may be too soft and the actual value could be 3.38511 as I have discovered .Has anyone else discovered just how critical fine tuning the resistor is?

Link to comment
Share on other sites



Probably worth bearing in mind every single driver will have slightly different characteristics as well, and if you find extreme accuracy is required for your resistor value, then you should be doing the same sort of extreme measurements on each driver and find the resistance that's perfect for each driver.

Link to comment
Share on other sites

"L" pad attenuation as shown in the reference from  mwhouston from the previous post, is the best method of correcting  the sensitivity level of a tweeter. A series resistor reduces damping and is not effective in  reducing impedance variations. The  human ear can  just detect   pressure level changes of around 1dB  under controlled conditions at the mid-range frequencies and given the normal  manufacturing tolerances of speakers, a resistance variation of 0.1 ohm is unlikely to be noticed.

  • Like 2
Link to comment
Share on other sites



A valid question, Ittaku.  A - 3dB pad is a common value and the series resistor would be around 2.3 ohms for an 8 ohm impedance tweeter;  (the // resistor being 19.4 ohm). The xover network design should  be one with an inductor connected from the input to the 2.3 ohm resistor and the opposite terminal of the tweeter. This indicates the use of a series first order network or a parallel second order network to ensure that the tweeter is damped below the  crossover frequency. This is an effective  design method  and it is better than trying to tame the resonant frequency of a tweeter with an  LCR  network  that invariably will be out of whack given that the tweeters Fs will vary with temperature and ageing.

Edited by VanArn
Link to comment
Share on other sites

1 hour ago, VanArn said:

A valid question, Ittaku.  A - 3dB pad is a common value and the series resistor would be around 2.3 ohms for an 8 ohm impedance tweeter;  (the // resistor being 19.4 ohm). The xover network design should  be one with an inductor connected from the input to the 2.3 ohm resistor and the opposite terminal of the tweeter. This indicates the use of a series first order network or a parallel second order network to ensure that the tweeter is damped below the  crossover frequency. This is an effective  design method  and it is better than trying to tame the resonant frequency of a tweeter with an  LCR  network  that invariably will be out of whack given that the tweeters Fs will vary with temperature and ageing.

Thanks but you didn't answer my question ?

Link to comment
Share on other sites

 The point I wanted to make is that a L pad in conjunction with  the appropriate crossover network will allow electrical damping of a tweeters Fs  which is so often neglected by the majority of designers. The simple answer is that the series resistor in an L pad will be lower for the same attenuation level than a series only resistor. If for example a -6 dB pad  were required to adjust  a dome tweeter with a 92 dB sensitivity to a bass/mid speaker of 86 dB sens. then the pad alone would provide a damping factor of 1 in the working range of the tweeter ( and considerably more at the Fs of the tweeter). For this case, the // resistor  will equal the tweeters nominal impedance and the series resistor would be half of the nominal impedance. A series resistor, only would equal the tweeters' impedance and result in  twice the impedance being presented to the xover network, which  is an added complication.

  • Like 1
Link to comment
Share on other sites

On 13/02/2019 at 3:51 PM, Leo Veto said:

In my experience,the most important part of the crossover is finding the correct value resistor for the tweeter and I believe it is critical to within 5 decimal places past the ohm.For example: 3.3 ohm may be too loud while 3.4 ohm may be too soft and the actual value could be 3.38511 as I have discovered .Has anyone else discovered just how critical fine tuning the resistor is?

No, because it's nowhere near that sensitive. Besides, where are you going to get that exact a value? I have an excellent LCR meter that will measure to microohms but it's a $20k unit and you would have to buy hundreds of resistors and measure them in a temp controlled room to get that sort of accuracy.

  • Like 2
Link to comment
Share on other sites



On 13/02/2019 at 3:51 PM, Leo Veto said:

Has anyone else discovered

I recently changed some Jantzen Superior-Z caps to their (superior!) Alumen caps, and found the purer sound of the tweeter to need more volume. I changed the attenuation resistors to 0.2 Ohms lower, and got a better balance. This indicates to me that any artifacts/distortions of a lesser capacitor relates to diminished tolerance to the listener, but that's a separate issue. The 0.2 Ohms difference (just as a series resistor, not an L-pad) made a clearly audible difference for me.

 

Of course, if you rob Peter you pay Paul; in this case there is less energy going to the mid and woofer, so the change can be in sound colour, not just tweeter volume.

 

My point being that in this, and many other cases, I have heard a difference in attenuation changes as small as 0.2 to 0.5 Ohms. Smaller than that, I would be exercising my corrugator supercilii muscles.

Link to comment
Share on other sites

I think I have noticed similar to what the OP is saying but not to the fine decimal resistance points he is claiming. For example, without using an L-pad arrangement changing from 3.2 to 3.3 or 3.4 ohm there seem to be an change in SPL. However, I put this down to what I understand that incremental changes of 0.1 ohm do not result in similar or even incremental dB change, it can jump more than you think and that is why it seems to be audible. For example, jumping by 0.1phm doesn’t result in an even 1dB decrease, it might be more or less.

 

Edited by Al.M
Link to comment
Share on other sites

Guest scumbag
On 13/02/2019 at 2:51 PM, Leo Veto said:

In my experience,the most important part of the crossover is finding the correct value resistor for the tweeter and I believe it is critical to within 5 decimal places past the ohm.For example: 3.3 ohm may be too loud while 3.4 ohm may be too soft and the actual value could be 3.38511 as I have discovered .Has anyone else discovered just how critical fine tuning the resistor is?

Do you test the sensitivity and impedance of your drivers to those tolerances? 

Edited by scumbag
Link to comment
Share on other sites

On 17/02/2019 at 12:17 AM, Al.M said:

I think I have noticed similar to what the OP is saying but not to the fine decimal resistance points he is claiming. For example, without using an L-pad arrangement changing from 3.2 to 3.3 or 3.4 ohm there seem to be an change in SPL. However, I put this down to what I understand that incremental changes of 0.1 ohm do not result in similar or even incremental dB change, it can jump more than you think and that is why it seems to be audible. For example, jumping by 0.1phm doesn’t result in an even 1dB decrease, it might be more or less.

Be aware that changing a resistor in series can change the crossover behaviour and alter the phase response at or close to the crossover point. That is likely having just as much if not more effect than the subtle change in amplitude.

  • Like 1
Link to comment
Share on other sites

1 hour ago, Ittaku said:

Be aware that changing a resistor in series can change the crossover behaviour and alter the phase response at or close to the crossover point. That is likely having just as much if not more effect than the subtle change in amplitude.

Good point, and that issue is also likely to what the OP may be hearing in small changes to the resistor value.

Link to comment
Share on other sites



Guest scumbag
3 minutes ago, Al.M said:

Good point, and that issue is also likely to what the OP may be hearing in small changes to the resistor value.

Which begs the question, is the op measuring the frequency response or is he tuning the design by ear? 

Link to comment
Share on other sites

On 13/02/2019 at 3:51 PM, Leo Veto said:

Has anyone else discovered just how critical

Definitely.  Driver levels, down fractions of a dB are very important in a speaker design.

On 13/02/2019 at 3:51 PM, Leo Veto said:

fine tuning the resistor is?

Resistance to 5 decimal places is fantasy on multiple levels (that you need it, that you can measure it, that the ambient temperature does drift is unacceptably, etc.)

 

Sure, the tweeter level to fractions of a dB is quite important..... but show me the circuit diagram where the resistor value to get such tight dB tolerances is so sensitive (?!)

Link to comment
Share on other sites

I thought the resistor value of 3.37724 was extremely close,but after living with and listening to the speakers for close to a week I  knew the tweeter was still too loud.After more series/parallel combinations I have arrived at 3.41339 and have lived with this now for over 2 weeks.This was the Eureka value.

Link to comment
Share on other sites



On 16/02/2019 at 6:41 PM, BioBrian said:

I recently changed some Jantzen Superior-Z caps to their (superior!) Alumen caps, and found the purer sound of the tweeter to need more volume. I changed the attenuation resistors to 0.2 Ohms lower, and got a better balance. This indicates to me that any artifacts/distortions of a lesser capacitor relates to diminished tolerance to the listener, but that's a separate issue. The 0.2 Ohms difference (just as a series resistor, not an L-pad) made a clearly audible difference for me.

 

Of course, if you rob Peter you pay Paul; in this case there is less energy going to the mid and woofer, so the change can be in sound colour, not just tweeter volume.

 

My point being that in this, and many other cases, I have heard a difference in attenuation changes as small as 0.2 to 0.5 Ohms. Smaller than that, I would be exercising my corrugator supercilii muscles.

Pleased to see others can detect  differences as low as 0.2 Ohms but that`s only a starting point.You can discern much lower than that.

Link to comment
Share on other sites

5 hours ago, Leo Veto said:

I don`t know but easily perceptible.

 

5 hours ago, Leo Veto said:

Pleased to see others can detect  differences as low as 0.2 Ohms but that`s only a starting point.You can discern much lower than that.

We cannot say that someone will be able to detect X ohms .... it depends on the circuit, and how much dB difference the specific resistance causes in that circuit.

 

Your circuit must be "strange", if 0.2ohm causes enough dB difference (>>0.1dB) to be audible)

Link to comment
Share on other sites

It isin`t strange It is a basic 2nd order LR 2 way network with 4 components and the resistor which I chose to the value needed by the tweeter to within 5 decimal places past the Ohm.After narrowing the value down to be between 3.3 and slightly higher,much trial and error brought me to the final and perfectly correct value of 3.41339 Ohms..As the tweeters were a matched pair bought from Parts Express,the only variable was the woofers and they were spot on which saved doing it all again for the second time.

Link to comment
Share on other sites



Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
To Top