high pass filter

A capacitor.

As Kr4 said, a cap will do the trick BUT you will need to know the impedence of your bookshelf speakers at 80hz and whether you want a shallow rolloff(6db per octave) or a sharper rolloff(12 or 18db per octave).

Also, the quality of the main cap will make a BIG difference in mid and upper octave transparancy but a plain non polorized electrolytic will also do the job.

It would help if we knew the brand/model of your bookshelf speakers...

Cheers

they are cambridge audio s30's. i would like to try both 6db and 12db rolloffs. i have taken the speakers apart and i'm assuming that i could probably just put a cap on the midrange's + terminal to the lead feeding into the crossover?? i metered the midrange and it is showing 3.8ohms. would i use that reading to find what cap size i need?

1. You need to know the impedance, not the DC resistance, and you need to know it across the relevant range from about 200Hz down.

2. For 6dB rolloff, only a cap is needed. For 12dB, you will need a cap and a coil.

Kal

how would i measure that Kal? thanks for the help

Let's make this easy, for a ~4 ohm impedence at or around 80hz, you will need the following:

6db - 500uF cap
12db - 350uF cap, 11mH shunt coil

For ~8 ohm impedence:
6db - 250uF cap
12db - 175uF cap, 22mH shunt coil

These values will get you in the ballpark and most likely be just fine and fine tuning the values in relation to the exact impedence at 80hz will yield better results but in most likelyhood, your room acoustics will swamp any differences an exact impedence match would reveal.

Use ~100 volt non-polorized caps in the main feed and build an external crossover box...

Cheers

I really wonder if those numbers will be sufficient. Impedance varies greatly near resonance and is affected by enclosure loading. The rated 'nominal' impedance is estimated at 1Kohms but, in the bass, the difference can be huge.

You can measure it best with an impedance bridge or with an AC multimeter.

Kal

"Impedance varies greatly near resonance and is affected by enclosure loading. The rated 'nominal' impedance is estimated at 1Kohms but, in the bass, the difference can be huge." -Kr4

The impedance peak(s) at system resonance will cause a simple series capacitor or textbook L-C filter to exhibit non-textbook behavior. The problem is, the response is insufficiently suppressed at the resonant peak(s). Throwing a bigger capacitor at the problem just results in a deeper frequency response "saddle" above the peak.

The highpass filter should be optimized for that particular driver's frequency response and impedance curve, and doing so is largely a matter of trial and error, often calling for additional components in the circuit. Not saying it isn't worth taking a shot at, but this is one of those things that's more complicated than it appears at first glance, and the component cost to really do it right can be rather daunting.

A small, high-quality series capacitor in between the preamp and power amp is often a more elegant and cost-effective solution. Find out the power amp's input impedance, and calculate the capacitor value for the frequency you want using that number. You'll have to do some soldering involving RCA jacks and plugs, or balanced connectors, but you'll come out way ahead on parts cost for equivalent sound quality.

Imho, ime, ymmv, etc.

Duke
dealer/manufacturer

Hey Duke, I was just going to suggest that the OP consider using a line level x-over instead of the high level route which at the 80hz frequency can be rather daunting.

Harrison Labs makes two such line level x-overs called the FMOD or the PFMOD and will be much cheaper although the quality of their line level x-over's leaves a little to be desired. Ask me how I know... :-)

http://www.hlabs.com/products/crossover/

Cheers

Yes, line level would be easier and there are off-the-shelf solutions, as Nmusicman has suggested.

BTW, what I meant to say in my earlier post was:

Impedance varies greatly near resonance and is affected by enclosure loading. The rated 'nominal' impedance is estimated at

1000Hz

but, in the bass, the difference can be huge.