First (Noob) crossover project - upgrade klipsch RF-7

[okapi]

I am considering upgrading the crossover in my klipsch rf-7 speakers.

I thought this might be a good way to approach my first loudspeaker project as i currently do not have the facilities to do any type of woodworking.

Also, I have read on other forums that rf-7 owners have obtained a significant improvement with same spec but higher quality part substitution. i think this is the approach i would like to take initially (i am open to being convinced otherwise).

Capacitor replacement seems simple enough but i am unsure of how to replace the inductors. i assume that higher quality inductors will have a lower resistance and change the filter profile. would changing the inductors require a crossover redesign?

Here are some pics of the crossover:
1 2 3 4

Here is the schematic:


comments on capacitor/inductor brand selection would be appreciated as well.

i don't currently have a budget for this project but i am always looking for good value.

[roddyama]

Here is an air core inductor from Madisound (up in Madison, WI). If you look at the chart for the 16ga wound "Sidewinder"coils the 0.75mH coil has 0.29ohm DCR. I'm sure this will work fine in your application.

[R-Carpenter]

I am not sure how much improvement you will actually get from replacing the parts. They appear to be of a good quality.
Inductor replacement is the same as capacitors for as long as the value stays the same.
General understanding or believe is that higher gage inductors will perform better on the woofers, Such as 12 or 14 gage instead of 18. The difference may become apparent at higher volumes, IF.
Some people believe that copper or silver inductors such as this one
http://www.madisound.com/catalog/pro...products_id=20
will perform better on the tweeter.
You could also change resistors for the none inductive type such as Mills or Eagle.
I would do 1 speaker at a time and then blind listen with a help of a friend to see if there’s any difference at all.
If you are retaining same values, crossover doesn’t have to be re-designed. .75mhz is .75mhz anyway you look at it. Then changing caps, go for higher voltage cups such as this.

http://www.madisound.com/catalog/ind...cPath=404_5_11
Let us know how it worked out.

[okapi]

thank you for the replies.

my understanding is that changing out the iron core inductor in the low pass filter for an air core inductor will prevent distortion associated with core saturation. swapping in the inductor suggested by roddyama wouldn't change any other properties of the crossover because it has the same DCR (direct current resistance). seems like an easy choice.

the problem that i don't have a full handle on is the selection of lower guage inductors (as suggested by r-carpenter) as these inevitably have lower DCR. for example the inductor in the low pass filter has a DCR of .3 ohm, a lower guage inductor with the same inductance can have a DCR of about .05 ohm.

i am in the early stages of teaching myself how to model a crossover network so i am not sure of what effect a drop in DCR like the one mentioned above will have.

i know how to determine the impedance of the inductor as a function of frequency. i know how to calculate the corner frequency of a RC low pass filter. But i don't know how to generate an input output curve for a low pass filter with an inductor, let alone with a series resistor following the capacitor.

[roddyama]

Originally posted by okapi
the problem that i don't have a full handle on is the selection of lower guage inductors (as suggested by r-carpenter) as these inevitably have lower DCR. for example the inductor in the low pass filter has a DCR of .3 ohm, a lower guage inductor with the same inductance can have a DCR of about .05 ohm.

Just add a series resistor to make up the difference in the DCR. As an example, for the 0,125mH parallel coil shown in your schematic, start with the 20ga 0.15mH coil shown
here at the Madisound site. First get a rough count of the number of turns in the coil by counting the the turns across the width of the circumfrence and multiplying that by the number of layers. Than unwind about 17% of the turns and you'll be real close to 0.125mH. Or you can measure it if you have the means. For the resistance, just add a 0.3ohm 10W resistor in series with the other components in that parallel leg just before or just after the existing resistor.

For the 0.3mH coil, I'd go with the 19ga air core because it will be a bit easier to get a 0.2ohm resistor. Since this is just DCR, it can be placed inseries in that parallel leg with no effect on the response. Of course non-inductive resistors will be best but harder to get and more expensive.

I think you'll be pleased on how this will open up the sound of your speakers

[gtforme00]

As suggested, the only part worth playing with IMO is that steel core inductor. You have already indicated there is a certain amount of distortion associated with ferromagnetic core inductors. Core hysteresis will cause a certain amount of distortion, though the laminated steel core inductors used in this design will have low hysteresis at the frequencies in the low-pass section of this crossover. While distortion due to saturation is also possible, this would only occur at very high power levels (several hundred watts for most common inductor core sizes). I don't know if the distortion due to hysteresis will be audible in your application, but if you want to try any one thing, there is certainly the most possibility for improvement in this inductor. You may or may not want to add a series resistor to the inductor. Here are the reasons why.

Decreasing the resistance of the inductor will have two primary effects.

1. Altering the bass response of the woofers by changing the qts of the system. Typically this effect can be summed up as: lower series resistance, leaner (some would say tighter) bass; higher series resistance, more pronounced (some would say boomier) bass. Depending on your preference for bass alignments, you may like the more damped bass provided by a decrease in series resistance.

2. Altering the acoustic output across the range of the woofer by changing the resistive attenuation. This may manifest itself as a more prominent midrange due to the increased level of the woofer with respect to the tweeter. With such a small change in resistance as you are proposing, you will likely not notice the difference.

Changing the resistance of the inductor will not have an effect on the crossover frequency, as the filter effect is not a function of resistance (fixed regardless of frequency) but of impedance (changes with frequency).

The increasing impedance of an inductor with frequency is as you probably already know, the reason an inductor functions as a low pass filter. The impedance is the sum of the resistive and reactive components of the inductor at a given frequency.

At the 2200Hz crossover frequency, the impedance of the inductor will be:

Z=R+iX (ignore the i because we are only working with one complex impedance in this equation)

R = 0.30 Ohms
X = 2*pi*Frequency*Inductance
X = 2*pi*2200*0.75E-3
X = 10.37 Ohms

Z = 10.67 Ohms at 2200Hz

If you notice, the reactance of the inductor depends solely on the inductance value and frequency, thus the crossover filter frequency will rely solely on those values as well.

Note that in RLC circuits, the resistance value of the circuit determines the sharpness of the notch. In these circuits the sum of the inductor resistance and resistor value should be kept the same.

Added: The inductance of wirewound resistors is very small (see thread here for details) and should only be of any minor concern for large resistance values in series tweeter circuits.

Have fun with your project!
David

[R-Carpenter]

0H, here’s an idea. The changes that will be achieved from replacing these components will be minimal and probably not audible.
How about improving the cabinet itself first. May be some bracing?

[roddyama]

Originally posted by R-Carpenter
The changes that will be achieved from replacing these components will be minimal and probably not audible.

This is true. Some hearing aides have a very limited frequency response.

[okapi]

the responses have been great- exactly what i was looking for.

here is my tentative plan of action based on everyone's help.

all the changes i mention i will do one at a time so i can properly assess their effect.

i am going to change the inductor in the low pass filter with this inductor:

Goertz

for the following reasons:

1. the reduced DCR may result in a a little more midrange (and possibly bass??) - i find these speakers treble biased anyhow.

2. a possible increase in bass "tightness" (i find the bass on my speakers a little loose, i think this may be more amp/source related because i have heard both of my speakers [phase tech pc 3.1's] sound better with other upstream equipment, nevertheless i want to see if it will help).

3. possible reduction in distortion.

i may also try changing the resistor (with a non inductive type) and the cap in the low pass filter because their seems to be some difference of opinion on the possible outcome (none to some). i need to convince myself of the merits (or lack thereof) of boutique parts.

r-carpenter - your last comment has raised an issue that i would also like to consider. just this morning i had my hand on the cabinet assessing the vibration at high volume levels. obviously no vibration is better than some but i am not sure how much i can reasonably expect to reduce it. the cabinet already has two "windowpane" braces. i do have a vibration meter so i have the ability to quantitatively assess any changes.

along a similar line, when i had the crossover out i noticed that the foam dampening material was limited to a single large piece (2.5 to 3 inch thick) running from the bottom of the cabinet up to the base of the horn. the foam is not against the back of the cabinet but is up against the ports (2 of them), 6- 8 inches out from the back. This raises the question, could the speakers benefit from more or better placed stuffing? reducing the reflections would be the goal but i am also concerned about changing the tuning as I think i read somewhere on diyaudio that stuffing increases the cabinet volume.

once again, thanks for all the help. i will post the results and photos of the crossover mods here when i complete them.

[R-Carpenter]

Originally posted by roddyama

This is true. Some hearing aides have a very limited frequency response.

Don’t you doubt my hearing aid! The capacitors are Mundorf and the copper is oxygen free!