Are notch filter calculators accurate?

[Taperwood]

How accurate are the online software calculators for figuring out the components for a parallel notch filter? I have found more than a couple on the net. I'll be working on the 2 to 8Khz range.

I'm trying to tone down the highs on my Frugel-horn Fostex Fe108EZ's. I want to attenuate the range from 2Khz to 8Khz at least 3dB. I would need to buy high-quality capacitors, so buying and trying experimentation would get very expensive very fast.

Any advice or pointers would be greatly appreciated. Thanks.

Doug

[kristleifur]

The calculations are pretty straightforward, so the attenuation you'll get from the filter should be reasonably well represented in the calculator AFAIK. (I'm no expert.)

Here's a calculator you might not have found:
http://lautsprechershop.de/tools/t_p...ngkreis_en.htm

[Cal Weldon]

Are you planning a series or parallel notch?

[Taperwood]

Kristleifur,

Yes, I did miss that one. Thank you. Using this calculator, I think I can shape the filter slope exactly how I want it. Just to be sure I understand this correctly, you use the C and L values to shape the slope and the R value for amount of attenuation? Is this correct, or am I off?

My one question now is how do I figure the amount of attenuation using this calculator? Sorry if this is a dumb question, but I honestly do not know.

Originally posted by Cal Weldon
Are you planning a series or parallel notch?

Cal,

I'm doing a parallel. According to Fostex's specs on this driver, the impedance slope does not rise at all at higher frequencies. It appears to be 9-10 ohms all the way out. So I don't seem to need to fix that. As I asked above, my one question now is what value of resistance lowers the dB by how much. I read somewhere recently that resistance of 1 ohm reduces efficiency by about 1 dB, but I don't know how accurate that is as this was in relation to a discussion of speaker wires, and we all know how muddy those can get.

Doug

[holdent]

I think kristleifur got you only part of the way there. The tool you're looking for is a notch filter (the site linked called it a "Frequency Response Leveller") and can be found directly at http://lautsprechershop.de/tools/ind...ngkreis_en.htm. I like this calculater a lot becuase it shows the attentuation at various frequencies and let's you check what specific inductor, resistor, and capacitor values will do. It helped demonstrate to me the potential problems of having components (particularly inductors) that are not of relatively close tolerances. For example inductors 10% out of spec will create an audible change over 3 db at and around the target frequencies.

[Taperwood]

Holdent,

Thank you. I probably would have found this one eventually, as it's on the same site, but thanks for pointing it out. It looks exactly what I need. Also, thank you for the head's up on inductors being so out of spec. I should be able to measure them, and I plan to use pretty high quality resistors and caps, so hopefully it will work out without too much fussing.

Doug

[tinitus]

Sounds like the one you are looking fore is "Frequency response leveller" at Strassacker - it shows attenuatin in db

Nice tools, but I also tweak by ear - cap and resistor are easily changed

[kristleifur]

Note that you can simulate a notch filter in software before building it, if you've got the attenuation values of the hardware filter.

If you're proficient in Javascript, you can look at the script that runs the filter calculator on lautsprechershop.

Here's an Excel spreadsheet I cooked up from the formula in the Lautsprechershop filter:
http://nemendur.ru.is/kristleifur04/...calc-ROUGH.xls

I don't understand the math anymore I don't have time to re-figure it out, but you could be able to make use of it nonetheless. I used this to make a software notch for the Cyburgs Needle, and that sounds fine - the math seemed to work OK

[Taperwood]

I started playing around with the calculator on Strassacker tonight. Really neat stuff. It's amazing how you can customize the slope at the lower and upper ends. Luckily, I have pages of readings on these speakers via my trusty SPL meter. I hope to order up a selection of components soon and get to tweaking again. I sure would like to finalize this project though .

I can't say enough how thankful I am for everyone's help here. Just in the last two weeks, I'm finally feeling that my thinking regarding electricity and sound is starting to gel. It's starting to make sense to me. Not fully yet, but much closer, and since I am so terrible at math I really do need something like the Strassacker calculators for help.

Doug