# Crossover technical question

Discussion in 'Amps and Cabs [BG]' started by snoopyjc, Jan 22, 2005.

1. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
In investigating a simple 3db crossover for a tweeter in a bass guitar speaker I bought. If the tweeter is 8 ohms, and there is a 10 ohm resistor in series with it, and the capacitor in series with that, do I use 18 ohms as the impedence in the formula for computing the capacitor value, or do I use 8 ohms?

F = 0.159 / (C * R)

The builder claims that it's crossed over at 3500HZ, and there is a 3.3uf capacitor there. Measuring the audio output of the cabinet gives me a big dip at 4KHZ, which goes away if I short the capacitor.

Thanks!

--joe

2. ### Petebass

Dec 22, 2002
QLD Australia
What he's got there is a resistor to turn the level of the horn down. It should be in the signal chain before the capacitor. If so, then ignor it for calculating the cap value and use the tweeters 8 ohm nominal impedance.

Therefore your 3.3uF capacitor at 8 ohms gives a -6dB filter which kicks in at 6KHz.

I fairness to the builder, a 3.5K crodssover with a 6dB slope would be useless in most horns . At that frequency you'd need at least a 12dB rolloff. Upgrading this crossover shouldn't be to hard.

3. ### BillyB_from_LZSupporting Member

Sep 7, 2000
Chicago
Don't short the cap for very long or playing the cabinet very loud or you'll be out one tweeter real quick!

BTW, use the tweeter's impedance for crossover frequency and/or capacitor calculations.

For a 6 dB/octave crossover you want the capacitor's capacitive reactace (Xc) at the crossover frequency to equal the tweeter's impedance.

Xc = 1/[2*Pi*F*C) or F = 1/[2*Pi*C*Xc] or F = 1/[2*3.14*3.3x10-6*8] or 6 KHz

4. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
Actually, the resistor is AFTER the capacitor in the chain, but I don't think it really matters, does it, because they are effectively in series, right?

I mean, doesn't the capacitor act as a resistor whose value changes dependent on the frequency?

--joe

File size:
2.7 KB
Views:
50
5. ### LoveThatBass

Jun 28, 2004
Dallas/Ft. Worth Texas
The resistor being in the output to the horn is perfectly fine.
It limits the signal to the horn that is all.

6. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
I only had a signal generator on the input at 4KHZ with the capacitor shorted, so not to worry about me blowing the horn.

Ok, so can you please tell me how to fix it such that the "dip" is removed from the response? (Response curve attached).

--joe

File size:
14.4 KB
Views:
40
7. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
I found a very interesting (but complicated) article covering this topic:

http://sound.westhost.com/lr-passive.htm

(In order to read the resistor values, treat "R" as the decimal point.)

Things I've learned:

1. The "L" pad in the crossover is in the wrong place - as it's getting the full power of the amp across it. This is most likely a 100W L-pad, so it's going to get quite hot!

2. The 10 ohm resistor is acting to cut the signal to the tweeter, but should be replaced by an "L-pad" circuit (e.g. a series resistor, followed by a resistor parallel to the tweeter) in order to present the right impedance to the crossover. However, if I move the real L-pad to after the capacitor, then the resistor can probably stay!

3. The 3.3uf capacitor is crossing over at around 6KHZ, which is too high.

Now I'm going to redesign the whole crossover, moving the L-pad to after the capacitor, and changing the capacitor value.

--joe

8. ### BruceWane

Oct 31, 2002
Houston, TX
The problem you've got is that the tweeter you have probably will not handle frequencies below where it is currently crossed over. If you feed it stuff below its operating range, it'll blow very quickly. You have to have a tweeter that goes low enough to pick up right where your woofer starts to roll off. What you need to do is upgrade the entire tweeter system - i.e., the crossver and the l-pad and the tweeter itself.

You could keep your current tweeter and add a midrange, but that's more complicated and expensive than is necessary, since it is possible to get a tweeter that will do what you need.

This thread will probably be of interest to you..............

Note - PartsExpress lists the particular Peerless tweeter in this thread as "discontinued", but what they really mean is they no longer carry it. It's still available from other sources. Try www.madisound.com or www.speakercity.com ......this tweeter may be referred to as #811647 or HDT-100; One is a model number, the other is an "order id" number, but both refer to the same tweeter.

9. ### billfitzmauriceCommercial User

Sep 15, 2004
New Hampshire
Owner, Bill Fitzmaurice Loudspeaker Design
The dip has two possible sources. One is a crossover that's obviously too high; the other is that the drivers may be out of phase at the crossover point and their outputs are cancelling each other out. A combination of the two scenarios is also possible. Cure both by using a 4th order Linkwitz/Reilly highpass filter at 3kHz. Get the component values here:
http://www.lalena.com/audio/calculator/xover/

In pro-sound a 1st order HP filter is completely useless, and is only employed when the manufacturer is too ******* cheap to use a proper filter that actually works. It was set at 6kHz because any lower with the nearly non-existent protection of a single capacitor and the tweeter would fry for sure. As it is now the tweeter is reasonably safe, but you've lost a full octave of useable response out of it.

The Lpad in that figure is in the wrong place. It should be after the high pass filter. Whoever cobbled up this mess should get a job at the nearest butcher shop, where his talents can be properly appreciated.

Another tweeter may be required, but without a proper filter it won't work well either. Do the filter first before spending money on a tweeter you may not need.

10. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
The current tweeter in my cabinet is a QSL QS-T20 supertweeter. Specs here: http://www.partsexpress.com/pdf/264-476.pdf

Looks like I may not be able to move the crossover frequency down without blowing it. They recommend using a 5KHZ crossover for it at 12db/octave. That still leaves a big gap at 4KHZ.

--joe

11. ### billfitzmauriceCommercial User

Sep 15, 2004
New Hampshire
Owner, Bill Fitzmaurice Loudspeaker Design
According to the SPL chart that tweeter appears to have an fs in the vicinity of 2kHz, so with a 4th order filter it should run to 3kHz no problem. With the current 1st order filter you have on it the input at 1.5kHz is only attenuated by 12dB (1/16 power) plus that provided by the series resistor and Lpad. A 4th order filter at 3kHz would attenuate by 24dB at 1.5kHz (1/256 power), so even halving the corner frequency it would be far better protected. If you did blow the tweeter you'd still have the crossover that you'd need anyway for a better driver.

12. ### snoopyjc

Apr 26, 2002
Toms River, NJ, USA
Well, after all of my putzing around (and I did move the L-pad to after the capacitor), I find I like the sound of the cabinet WITHOUT the tweeter better than with it on!! I guess the cone drivers put out enough highs as is. With the tweeter on, the only thing I could hear coming out of it is plucking and string noise - certainly unneeded and undesirable!

Thanks for everybody's help!!
--joe http://www.simplegroovenj.com/