passive HPF circuit, 40Hz, 4th order - are these values correct?

[Compyfox]

I'd love to create a passive HPF (or lowcut) circuit for a project I'm working on. I just need confirmation if I'm on the right track or not.

My planned cutoff frequency is 40-50Hz, ideally 40Hz, but I don't know if that resistor value is existing. Also planned is at least a second order (12dB/Oct) filter, ideally fourth order (24dB/Oct).

I know that in order to get a steeper cutoff, I need to put the modules in series (up to 4 times). Maximum voltage used: a standard CD signal or guitar signal.


THIS IS FOR TESTING PURPOSES ONLY - I do not plan anything greater with it.


These are the values I've calculated:

R= resistor, C = capacitor

40Hz cutoff:
R = 390 Ohm
C = 10uF (micro farads)

50Hz cutoff
R = 330 Ohm
C = 10uF


Anyone can confirm that?
And if so, I hope the resistor values exist in real life. Though I'm sure I have some leftover 330 Ohms at my disposal.


Thanks in advance.

[skitchy]

If you dont have the right resistor values or they dont exist, you can make them up yourself.
Use multiple resistors in series if you want to add the values ie R!+R2=resistance. So a 1k and a 2k in series would be the same as a 3k by itself.
Using 2 resistors in parallel gives you (R1+R2)/2. So using the same example (1k+2k)/2=1.5k
Using more than 2 resistors in parallel also works, but the formula gets more complicated at this point.
Search for RC filter calculator, and parallel resistor calculator - there are loads of web pages where you just fill in the values and the calculations are done for you.
Hope that helps. Let us know how your project goes - we need more DIYers on KVR

[Compyfox]

Hm... I'm lucky in this case. Looks like there are 330 and 390 ohm ones. Apparently the math tool I used spit out these values - which seem to be normal available ones.

The question still remains... Are these values correct?
Or in other words: did someone build passive LowCuts/HPF at a similar frequency already?

[xtp]

Hi, heres a quick easy tool to experiment with - shareware.

Its only crippled by the size of the work area, so if you have a largish circuit you have to get creative or use another spice program like Kicad.

http://www.5spice.com/

you can build your passive filter in a few minutes and have it run a test which you can print out if you want.

-----------------------------------------------

i just drew it and ran it, 40Hz and 47Hz was the output. 50Hz was at -2.8db or thereabouts.

[mahaya]

You should be aware that if you cascade two or four filters you'll get a total of -6 or -12 dB at the -3 db frequency of a single stage. To keep the -3 dB freuquency you need to move your filter frequecy to sqrt(2^0.5 - 1) or sqrt(2^0.25 -1) of your target frequency. Also you will have a very slow roll off.

So although you'll get the 24 dB/octave by cascading four stages the cut off will take place at a higher frequency and the filter will also affect the pass band.

Another thing is that each following stage loads the previous stage.


Chris

[earlevel]

THIS IS FOR TESTING PURPOSES ONLY - I do not plan anything greater with it.

Then build it and test it!

But pay attention to what Chris said. I don't know what you're using it for, but loading and corner frequencies characteristics are among the reasons people use active filters for high orders...Don't forget component tolerance is a factor as well...

[xtp]

Then build it and test it!

I think this really sums it up. When it comes to audio in the end the ear wins and quite often what looks good on a mock up does not always sound good.

With filters I have found it useful to replace the resistor with a pot and then put it on the test board a few weeks.

Once I find I am using a single frequency I measure the resistance and replace the pot with a resistor.

I find it easier than pulling resistors in and out and easier to twiddle while playing, and less messy than a rotary switch.

[felix80]

As per xtp suggestion, use variable resistor i.e. potentiometer or preset rather than making use of fixed valued resistor. Check the output varying resistor value, if it is working fine at the particular value than measure that value and replace with fixed one.

surface mount assembly