I want to use a pair of H0955's as a mid-woofer rolling off @12dB at around 500 Hz. I was going to add a notch to eliminate the 1st peak @ 4.5K, which
is no problem. But the question is I have is : there is also a peak @ 5.9K ; 7.1K & 8.7K that are almost as prominent. Other than adding 4 unique notches filters centered at each of these frequencies (using LSPCAD for software to design these notches).... is there a better way to handle these resonances?
The design calls for 2nd order low pass so I was not planning on bumping it up to 4th LR etc. without major rework.
Is there any advantage just to notch the 1st harmonic and leave the rest or?
Any input would be helpful.
is no problem. But the question is I have is : there is also a peak @ 5.9K ; 7.1K & 8.7K that are almost as prominent. Other than adding 4 unique notches filters centered at each of these frequencies (using LSPCAD for software to design these notches).... is there a better way to handle these resonances?
The design calls for 2nd order low pass so I was not planning on bumping it up to 4th LR etc. without major rework.
Is there any advantage just to notch the 1st harmonic and leave the rest or?
Any input would be helpful.
You don't need any Notch filters if you cross at 500Hz (12dB). You are at least 3 octaves away from the first peak (4K5).
However, I strongly suggest a Zobel and a BSC. The Zobel will prevent the woofer response from rising after 500Hz, which means the peaks will stay buried and not intrude into the tweeter. The BSC will level off the mids. Without the BSC, the vocals will be over emphasized. At the same time, the BSC will add to lower the peaks.
However, I strongly suggest a Zobel and a BSC. The Zobel will prevent the woofer response from rising after 500Hz, which means the peaks will stay buried and not intrude into the tweeter. The BSC will level off the mids. Without the BSC, the vocals will be over emphasized. At the same time, the BSC will add to lower the peaks.
Member
Joined 2003
Sometimes you can get away with a single notch with lower Q...a little higher R than normal in the LCR. Attenuation won't be as great as low R, but multiple peaks can be covered with wider bandwidth. Lower attenuation might be more compatible with your planned 2nd order rolloff. Easy to sim in LSPcad.
Hi Dan
A quick way to check is to run a sweep. Use a signal generator or something similar and sweep from 100Hz upwards at 1W. This will tell you how audible -24dB is at 4K5. If it is still disturbing you, you can then apply a broad notch across the peaks.
Are you having some issues with the highs in your speaker? If so, the problem may not be from the woofer peaks. It is more likely to be the crossover between the mid and the tweeter.
Mike
I am in the design phase of the project. I will be designing a 3-1/2 way with Seas 10" 26L & 8" 21L handling the bass with the L26 rolling off at 100 Hz while the 21L handles up to 450-500 Hz then it hand over to a Energy Veritas 3" aluminum dome (aka Veritas V2.8 CSM module- Mid/Tweeter in one assembly) that runs up to 2.5K. It then hands off to its partner 1" Alum dome.
I have the crossover schematic for the V1.8 & V2.8 to use as reference. I will be using the mid/tweeter crossovers and unite it with the 2 seas woofers
to create a hybrid system with better woofers than the original Veritas V2.8 & V1.8's.
I designed a broad notch filter using two notch filters spaces at 4K & 7K to create a broad overlapping notch. The only problem is the Q is so low that it affects the response all the way down to the upper woofer -mid crossover @ 500Hz. I guess I can generate an upper woofer 2nd order low pass with a higher Q to compensate for the effects of this notch (below 1K) of the dual broad low Q notch centered at 5.5K
.
I have the crossover schematic for the V1.8 & V2.8 to use as reference. I will be using the mid/tweeter crossovers and unite it with the 2 seas woofers
to create a hybrid system with better woofers than the original Veritas V2.8 & V1.8's.
I designed a broad notch filter using two notch filters spaces at 4K & 7K to create a broad overlapping notch. The only problem is the Q is so low that it affects the response all the way down to the upper woofer -mid crossover @ 500Hz. I guess I can generate an upper woofer 2nd order low pass with a higher Q to compensate for the effects of this notch (below 1K) of the dual broad low Q notch centered at 5.5K
.
Member
Joined 2003
Sounds like your notch filters may be too aggressive. The notch(s) should attenuate the peaks only enough to match your target rolloff.
Try using the optimizer in LSPcad to achieve your target rolloff, including the low-pass components as well as the notch filter. You may find you don't need the zobel and/or another cap at the end of the 2nd order filter.
Can you post a screen capture of your schematic, including the .5 woofer? You are using actual measurements in your boxes, correct?
Try using the optimizer in LSPcad to achieve your target rolloff, including the low-pass components as well as the notch filter. You may find you don't need the zobel and/or another cap at the end of the 2nd order filter.
Can you post a screen capture of your schematic, including the .5 woofer? You are using actual measurements in your boxes, correct?
notch filter
Sorry for the intrusion
Re reading the above I assume you guys are talking series Notch filter (parallel to driver ) --- my question is ---> at the frequency you want to drop 15 db isn't a series notch creating a short across the driver and so reduce the volume at that point and at the same time giving the amp a headache
If so ,is there a way to reduce that, or do I just ignore it and deal with the smoke
Thanks for all the info in all these threads
JanK
The more I learn the less I know
Sorry for the intrusion
Re reading the above I assume you guys are talking series Notch filter (parallel to driver ) --- my question is ---> at the frequency you want to drop 15 db isn't a series notch creating a short across the driver and so reduce the volume at that point and at the same time giving the amp a headache
If so ,is there a way to reduce that, or do I just ignore it and deal with the smoke
Thanks for all the info in all these threads
JanK
The more I learn the less I know
Member
Joined 2003
Yes, a series notch across the driver creates a "short" (not a complete short because of the series R in the LCR). The notch filter is effectively isolated from the amp by the low-pass inductor(s) between the amp and driver. Properly done, there is little impact on impedance seen by the source.
Along a similar line I have a driver with a little peak at 1800 . So it looks like I need a 6.2 ohm resistor, .42 mh inductor and a 16.68 uf capacitor . But my question is in this aplication would the quality of the capacitor make a big diference in the sound quality. Oh I could probable spend $ 50 on some film caps but would that sound a lot better than a cheap bipolar electrolic with a small film bypass.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.