Hello folks,
been lurking for a while and am a novice speaker builder (but I stay at the Holiday Inn Express frequently and am a quick learner).
When you have a driver rated for 25-300 Hz for example, and let's say it's flat from 60-250 Hz again just for example, I could see an argument to be made for using the peak of the flat region as a 0 db reference and simply looking for where you cross the -3db line as you approach 25 Hz.
Hoever, in a real world application (HT setup for example crossed at 80 Hz), do your need to adjust your 0 db reference? WHat if I have really nice left right front speakers and would like to cross to the sub at 40 or 60 Hz instead?
I.e. should the F3 rating only be used in context of the frequency band where you will actually be driving the sub?
Back to the real world, my first DIY project was to build an appox 3 cu ft cube subwoofer for a 12 inch well known driver. Only I had forgotten I was too cheap to actually buy the driver several years earlier and had bought a more run of the mill 10 inch driver instead on sale. A quick redo of the front side of the enclosure and I had my first working DIY speaker. My Carver TFM-15 bridged or not drives it quite well and it sounds decent enough (I've got no baseline to compare it to though).
Today, a few years later I went back into WinISD and calculated the sub has an F3 of around 80 Hz. Well, I thought I was shooting for some magical 20HZ (or at least in the low 30s). This got me thinking though about what the F3 0db reference is and whether the sub I built isn't actually "better" than the design software suggests.
As a follow-up I am in the process of redoing the sub in a ported enclosure with a target F3 of around 30 Hz.
Not trying to start a vented vs closed discussion, just looking for a check on whether or not I understand F3.
Thanks for reading
been lurking for a while and am a novice speaker builder (but I stay at the Holiday Inn Express frequently and am a quick learner).
When you have a driver rated for 25-300 Hz for example, and let's say it's flat from 60-250 Hz again just for example, I could see an argument to be made for using the peak of the flat region as a 0 db reference and simply looking for where you cross the -3db line as you approach 25 Hz.
Hoever, in a real world application (HT setup for example crossed at 80 Hz), do your need to adjust your 0 db reference? WHat if I have really nice left right front speakers and would like to cross to the sub at 40 or 60 Hz instead?
I.e. should the F3 rating only be used in context of the frequency band where you will actually be driving the sub?
Back to the real world, my first DIY project was to build an appox 3 cu ft cube subwoofer for a 12 inch well known driver. Only I had forgotten I was too cheap to actually buy the driver several years earlier and had bought a more run of the mill 10 inch driver instead on sale. A quick redo of the front side of the enclosure and I had my first working DIY speaker. My Carver TFM-15 bridged or not drives it quite well and it sounds decent enough (I've got no baseline to compare it to though).
Today, a few years later I went back into WinISD and calculated the sub has an F3 of around 80 Hz. Well, I thought I was shooting for some magical 20HZ (or at least in the low 30s). This got me thinking though about what the F3 0db reference is and whether the sub I built isn't actually "better" than the design software suggests.
As a follow-up I am in the process of redoing the sub in a ported enclosure with a target F3 of around 30 Hz.
Not trying to start a vented vs closed discussion, just looking for a check on whether or not I understand F3.
Thanks for reading
Thanks for replay. Not sure I get it 100% yet though.
Is "the low frequency portion "a fuzzy definition"?
It looks like 300 Hz is at the 0 db reference level and the F3 values are determined by that. Does that mean I should assume there is a low pass filter at 300 Hz in this example? Continuing down this path, if I had a low pass filter at 100 Hz, the F3 values would change dramatically to about 60 Hz in the low graph (assuming at 100 Hz it was about +3 db).
Am I on to something or visiting Linus n left field?
Is "the low frequency portion "a fuzzy definition"?
It looks like 300 Hz is at the 0 db reference level and the F3 values are determined by that. Does that mean I should assume there is a low pass filter at 300 Hz in this example? Continuing down this path, if I had a low pass filter at 100 Hz, the F3 values would change dramatically to about 60 Hz in the low graph (assuming at 100 Hz it was about +3 db).
Am I on to something or visiting Linus n left field?
In a maximally flat classic Theile Small alignment, F3 is the frequency that is 3 dB down from the flat portion.
Classic TS alignments often end up with a box size undesirably large, or a LF corner undesirably high, so one can use different alignments.
The two examples posted in post #2 are not maximally flat, the upper example shows a "bump" of about 1.5 dB above the 0 dB "flat" response point.
The lower example has a 3 dB "bump" over the 0 dB "flat" response reference.
Neither simulation use low pass filters, the upper roll off is a representation of the idealized mass rolloff, in reality the response would not look like that, there would be peaks and dips.
A sealed speaker in general rolls off at only 12 dB per octave below F3, equalization can be used to flatten response below F3, as excursion goes up relatively slowly below F3
A ported cabinet in general rolls off at 24 dB per octave, and excursion rises very rapidly, there is generally very little usable response below F3.
Classic TS alignments often end up with a box size undesirably large, or a LF corner undesirably high, so one can use different alignments.
The two examples posted in post #2 are not maximally flat, the upper example shows a "bump" of about 1.5 dB above the 0 dB "flat" response point.
The lower example has a 3 dB "bump" over the 0 dB "flat" response reference.
Neither simulation use low pass filters, the upper roll off is a representation of the idealized mass rolloff, in reality the response would not look like that, there would be peaks and dips.
A sealed speaker in general rolls off at only 12 dB per octave below F3, equalization can be used to flatten response below F3, as excursion goes up relatively slowly below F3
A ported cabinet in general rolls off at 24 dB per octave, and excursion rises very rapidly, there is generally very little usable response below F3.
Hibernator,
Do you mean the upper usable frequency or the -3dB low point Art is explaining?
I'm under the impression that you want to know how low you should cross your mains to the sub, correct?
Most smaller subs will play higher then the '80hz' crossover that is used a lot.
Art said it, there are peaks and dips if you cross over too high.
If your mains can produce 55hz, then cross over there, a sub is to reproduce the lower frequency's otherwise hard to play.
If your mains can't go lower then 100hz, better to use two 'stereo'-subs or at least place the sub centred.
Maybe look at the -3dB low roll-off and look for the same at the upper side, I'm not sure of this.
Do you mean the upper usable frequency or the -3dB low point Art is explaining?
I'm under the impression that you want to know how low you should cross your mains to the sub, correct?
Most smaller subs will play higher then the '80hz' crossover that is used a lot.
Art said it, there are peaks and dips if you cross over too high.
If your mains can produce 55hz, then cross over there, a sub is to reproduce the lower frequency's otherwise hard to play.
If your mains can't go lower then 100hz, better to use two 'stereo'-subs or at least place the sub centred.
Maybe look at the -3dB low roll-off and look for the same at the upper side, I'm not sure of this.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.