Hi,
I have a 4:th order LR filer crossed at 2khz. Looking at the HP section, a 1khz sinewave is distorted almoust to a squarewave. I use capacitors in the range of 10's of nF and resistors in the kohm range.
Any ideas?
/Jesper
I have a 4:th order LR filer crossed at 2khz. Looking at the HP section, a 1khz sinewave is distorted almoust to a squarewave. I use capacitors in the range of 10's of nF and resistors in the kohm range.
Any ideas?
/Jesper
dir you check again your wiring ? Are you sure you didn't mix up inverting and non-inverting inputs of opamps ? Is the NFB correctly applied ? How about supply voltage ? etc etc
At what input levels does this "phenomenon" occur ? What is the level of this distorted signal ?
Regards
Charles
At what input levels does this "phenomenon" occur ? What is the level of this distorted signal ?
Regards
Charles
Everything seems ok, the filter sounds good and measures good (aucustically), just got very confused when I got squarewaves in the stopband of the HP filter... It apperas when the out-signal levels is in the 200-100mV range and bellow (1V insignal).
/Jesper
/Jesper
And I guess you are running these of clean dual +/- 15V rails? Then if you are using adequate opamps and the wiring is correct the distortion should be pretty much at the limits of your measurement equipment. Is the transfer function indeed a 4th order L-R if you look at it with a spectrum analyser/tracking generator?
/Magnus
/Magnus
Yes, they are running from a clean +-15V. Everything works ok except the triangular shape (instead of sinus) starting to apperar approx 1 octave below x-over frequency in the HP (think I wrote squarewave earlier..).
I guess something is wrong with my measurment setup, not exactly uncommon.. =)
/Jesper
I guess something is wrong with my measurment setup, not exactly uncommon.. =)
/Jesper
Hello,
There are two versions of the Sallen and Key topology.
1. With unity gain buffers
2. With gain in the opamps -> stability is an issue
Which one are you using? Maybe you switched some components?
What is the amplitude of the “noise” you measure?
Good luck 🙂
\Jens
There are two versions of the Sallen and Key topology.
1. With unity gain buffers
2. With gain in the opamps -> stability is an issue
Which one are you using? Maybe you switched some components?
What is the amplitude of the “noise” you measure?
Good luck 🙂
\Jens
Thanks for the replies.
I'm using unity gain Sllen-key topology, it is stable.
Measurment setup is bench power supply (linear), Meteix signal generator and Tektronix 100Mhz oscilloscope.
The filter is a basically a 4:th order 2khz 2-way with an extra second order HP at 80hz (for matching with a sub). Actually the 2kLP is 3:th order (giving 4:t order acustically). There's also some bafflestep, notch and delaylinks. Op's are LM6172. Resistors are metal film, capacitors polypropylen. Byppassess are cheramics and tantal.
I experience the triangel-wave distortion in both the sub 80 and the sub 2k HP-filter stopband. Really weard..
Can't comment on the noise level at the moment, I'll do some more measuments later on today, when I get the time. I'll try with a HP 33120A generator too but the filter is input buffered so it the generator performance shouldn't make any significant differense in this regard, I think...
/Jesper
I'm using unity gain Sllen-key topology, it is stable.
Measurment setup is bench power supply (linear), Meteix signal generator and Tektronix 100Mhz oscilloscope.
The filter is a basically a 4:th order 2khz 2-way with an extra second order HP at 80hz (for matching with a sub). Actually the 2kLP is 3:th order (giving 4:t order acustically). There's also some bafflestep, notch and delaylinks. Op's are LM6172. Resistors are metal film, capacitors polypropylen. Byppassess are cheramics and tantal.
I experience the triangel-wave distortion in both the sub 80 and the sub 2k HP-filter stopband. Really weard..
Can't comment on the noise level at the moment, I'll do some more measuments later on today, when I get the time. I'll try with a HP 33120A generator too but the filter is input buffered so it the generator performance shouldn't make any significant differense in this regard, I think...
/Jesper
- Status
- Not open for further replies.