using an active high-pass filter to reduce distortion

[gordoncalder]

Hi,

I'm curious if anyone has experimented with putting an active high-pass filter set to around 30-40 Hz before, say, a pair of 2 or 2.5-way speakers. (I'm sure someone's done it). I'm imaging that it would reduce distortion higher in frequency, but of course I could be wrong.

I was hoping to find some sort of before-and-after distortion measurement out there on the internet just to see how significant the effect would be, but couldn't find anything. If you knew of any relevent links I'd appreciate it.

thanks,

Gordon

[dthomas]

Hi,

I'm curious if anyone has experimented with putting an active high-pass filter set to around 30-40 Hz before, say, a pair of 2 or 2.5-way speakers. (I'm sure someone's done it). I'm imaging that it would reduce distortion higher in frequency, but of course I could be wrong.

I was hoping to find some sort of before-and-after distortion measurement out there on the internet just to see how significant the effect would be, but couldn't find anything. If you knew of any relevent links I'd appreciate it.

thanks,

Gordon

I have done a couple of opamp based designs that used a 20Hz 4th order high pass filter on the front end of the tweeter high and woofer low pass network. I did not do distortion measurements but sure it helps and there is no real noticeable affect in most music genres. Of course it won't make a driver that strains at 40Hz sound better. But it does remove any potential low frequency stuff out that migh cause over excursion. I suspect if you really wanted to protect a smaller woofer you could go to as high as 40Hz or higher of course you will have some loss in low frequency content but it would protect the woofer and reduce distion.

Not sure what kind of links you are looking for but there are several good salen-key calculators for opamp circuits that help you determine proper resistor/capacitor values for opamp based filters. The circuit is simple and easy to DIY. You just need a couple of opamps and a 12 to 15V differential power supply. I like this spreadsheet/calculator it used to be located at the link below but the AOL Hometown site has been shut down you might try google his name in the address and see if he has a new site. I can send you a copy of the spreadsheet if you PM me.

http://members.aol.com/johnpomann/sp.../crossover.htm

[gordoncalder]

Hi Dave,

Thanks for the link. I found his site here... http://www.snippets.org/filters/crossover.htm
and there are digital filters as well (minidsp is one I was looking at). Either would do the trick though.

I was thinking that (for whatever sized woofer) if the active high-pass were just below the tuning frequency it would reduce distortion even if the driver were still within it's safe excursion range. I was interested in a before/after distortion measurement just to see the significance, and if it's worth the expense.

If I manage to get around to this over the next few months maybe I can post my own results.

[dthomas]

Hi Dave,

Thanks for the link. I found his site here... http://www.snippets.org/filters/crossover.htm
and there are digital filters as well (minidsp is one I was looking at). Either would do the trick though.

I was thinking that (for whatever sized woofer) if the active high-pass were just below the tuning frequency it would reduce distortion even if the driver were still within it's safe excursion range. I was interested in a before/after distortion measurement just to see the significance, and if it's worth the expense.

If I manage to get around to this over the next few months maybe I can post my own results.

I was pretty sure he had a site back up. I agree that there is benefit with high passing, problem is getting accurate measurements for most diyers myself included is difficult sub 200Hz so the accuracy of distortion measurements will be an issue. I don't thnk there is any quesiton that distortion will go down how dramatically is a good question as is how audible the reduction will be. Assuming the driver is not in overexcursion the reduced distortion at very low frequencies may not be very audible. The distortion caused by overexcursion will however be very audible.

[gordoncalder]

I was pretty sure he had a site back up. I agree that there is benefit with high passing, problem is getting accurate measurements for most diyers myself included is difficult sub 200Hz so the accuracy of distortion measurements will be an issue. I don't thnk there is any quesiton that distortion will go down how dramatically is a good question as is how audible the reduction will be. Assuming the driver is not in overexcursion the reduced distortion at very low frequencies may not be very audible. The distortion caused by overexcursion will however be very audible.

Now you've got me thinking about low-end distortion. I guess diy-measurement accuracy is limited by room effects as you say. But it should still be useful in terms of relative levels shouldn't it? (such as Mark K and Zaph post on their sites).

And how to reduce it would depend on what's causing it. If it is excursion related, then altering the port tuning, using an active high-pass, or going with a 2.5-way design might help. But if it's due to a motor or a driver-assembly issue then maybe those won't help at all. Or are those just much less significant than excursion in general?

[dthomas]

Now you've got me thinking about low-end distortion. I guess diy-measurement accuracy is limited by room effects as you say. But it should still be useful in terms of relative levels shouldn't it? (such as Mark K and Zaph post on their sites).

And how to reduce it would depend on what's causing it. If it is excursion related, then altering the port tuning, using an active high-pass, or going with a 2.5-way design might help. But if it's due to a motor or a driver-assembly issue then maybe those won't help at all. Or are those just much less significant than excursion in general?

Not sure what criteria they use when measuring distortion but I suspect it is a fairly large baffle and no box. But I hear you, think of this when using a woofer in an enclosure you have the induced distortion from the back wave hitting the cone or escaping the port tube plus you have added distortion created by the frame on the rear of the driver. I think it is a testamant to we are less sensitive to distortions at low frequencies. Or maybe we are just conditioned to hearing bass this way.

My Son plays bass I can tell when he is playing his acoustic bass non-amplfied it has a very pure sound that is never evident when he amplifies it. It sounds great amplified but it is never as pure as unamplified. I sat in and listened to him and his instructor a week or so back and they were playing together he on his acoustic bass and his instructor on acoustic guitar they were playing "TreeTop Flyer" the acoustic version on Stephen Stills solo album "Still Alone" all I kept wondering is how much you lose from recording to playback vs. hearing two guys play this live on acosutic guitars. We were in a room that was about 10' x 12" so acoustics were probably not the best but the sound was amazing.