Here is the close woofer and then pulled out like a foot or so, these were done in room, can't really go outside right now due to Canada fire smoke. In regards to those concerned with the dip at ~600, it goes away as you pull out. Curiously the resonance problem area increases quite a bit., perhaps it is related to the height of the speaker or do to the mic being to able to pick up more of the port output when placed further away. I will try playing with elevation to see if that changes anything.
That is very interesting. I have not compared the data from these drivers before.
When I compared my NF scan of an SB34NRX75-6 to an outdoor ground plane scan, I found a dip at 600 Hz. Since it was not present in the far field, I assumed it was a measurement anomaly of the NF measurement. For a 12" woofer with an Sd of 500 cm^2, the upper limit for valid NF scans is 430 Hz (according to the Keele method)... so I just assumed the dip was an anomaly caused by measuring at too high of a frequency... But you information gives me something to think about...
Doing proper woofer and port NF measurements, you shoul see something like this:
Abowe is a Coral X-VII 3-way system. It is identical to yours wrt to port (mis)placement - in a corner near 2 boundaries. Modal resonance of the cabinet escaping through port is very strong, only 1 dB lower than main port resonance. The box is smaller - internal height 665 mm - which corresponds to half-wave resonance of 258 Hz. In real system resonance frequency is lowered by box stuffing and other factors, coming in at 226 Hz.
![x-vii(1)[1].JPG](https://www.diyaudio.com/community/attachments/x-vii-1-1-jpg.1194801/ "x-vii(1)[1].JPG")
Unwilling to optimize the port or thinking back-firing one would be unmarketable, Coral simply done away with it, switching to closed box for the next iteration of this model, DX-Seven.
Abowe is a Coral X-VII 3-way system. It is identical to yours wrt to port (mis)placement - in a corner near 2 boundaries. Modal resonance of the cabinet escaping through port is very strong, only 1 dB lower than main port resonance. The box is smaller - internal height 665 mm - which corresponds to half-wave resonance of 258 Hz. In real system resonance frequency is lowered by box stuffing and other factors, coming in at 226 Hz.
Unwilling to optimize the port or thinking back-firing one would be unmarketable, Coral simply done away with it, switching to closed box for the next iteration of this model, DX-Seven.
Hi there,
Please be patient & understanding with this post > I will explain.
Like you, I use a high quality transistor amplifier capable of high current. This not only means high Damping Factor, >200
but also means that the amp has very low output impedance.
Also like you, I have always been of the opinion that true active drive of a speaker meant no passive components are used.
I recently had a near heated debate regarding this on another 'thread'.
It turns out that there is a current 'school of thought/theory' that suggests there is a case for very specific passive components.
It goes like this:
Because our amplifiers have very low output impedance, the speaker terminals effectively see a 'short circuit', and this causes
the Back EMF Currents generated by the movement of the voice-coil itself no way of discharge through some form of resistance.
It says that this 'shorting' of Back EMF forces this energy back into the physical properties of the driver causing distortion.
They also say the amount of distortion encountered varies from driver to driver (model).
With the above, I would say that your situation is a perfect test case to see if any of this (re. bass) is true.
The attachment here is what I suggest. Also, because your woofer is 4 ohms, the inductor is not overly large or expensive.
(From what they say, you should expect some MAGICAL improvement)
can we get back to the actual problem, things are going way off topic with some really weird suggestions.
seems the issue is related to the cabinet dimensions and port location based off what ppl have suggested and what experiments have yielded. it's not the amp or baffle edge diffraction for instance. I mostly find those suggestions misguided. No one here is obligated to help me but I don't think it's out of line to think some suggestions are just way off the mark.
other than changing internal height and/or port location I'm not sure what can be done at this point.
On an slightly unrelated note I've been working on some mixes and with my two eq filters at the resonance translation is good. I've been having problems with the fact that the speakers image poorly. I swapped the tweeter with an nd25fw and raised xover to 2500hz and they sound much better.
i think the biggest hurdle is piece of mind. the resonance isn't really a problem 90% of the time but when you do hear it, it sucks. I'm very much leaning towards rebuilding the cabinets and fixing other problems with the build. I'd like to follow the osmc construction plans. The overall cabinet seems well thought out and may eliminate my issues. the fact that I have an xt25 and wg300 ready to go makes the idea more tempting.
I have one last idea that I'm sure someoneay point out an issue with. I've considered adding a piece on the top brace that covers a good portion of the braces holes. this should be super simple to implement. I have no idea if this could help but I'll try anything at this point.
other than changing internal height and/or port location I'm not sure what can be done at this point.
On an slightly unrelated note I've been working on some mixes and with my two eq filters at the resonance translation is good. I've been having problems with the fact that the speakers image poorly. I swapped the tweeter with an nd25fw and raised xover to 2500hz and they sound much better.
i think the biggest hurdle is piece of mind. the resonance isn't really a problem 90% of the time but when you do hear it, it sucks. I'm very much leaning towards rebuilding the cabinets and fixing other problems with the build. I'd like to follow the osmc construction plans. The overall cabinet seems well thought out and may eliminate my issues. the fact that I have an xt25 and wg300 ready to go makes the idea more tempting.
I have one last idea that I'm sure someoneay point out an issue with. I've considered adding a piece on the top brace that covers a good portion of the braces holes. this should be super simple to implement. I have no idea if this could help but I'll try anything at this point.
Before you do anything else, I'd suggest cutting a big hunk of polyfill, adding it to the box [ by taking the woofers out] and then listening again.
Or
Simply closing off the port with a blanking piece and ditto.
Just a random thought and what I sometimes do because I usually find sealed boxes much easier
Or
Simply closing off the port with a blanking piece and ditto.
Just a random thought and what I sometimes do because I usually find sealed boxes much easier
Hi,
yeah its easy and practical way to think about it, analyze what driver load impedance is and how driver distortion mechanisms affect current in the circuit. Basically one can evaluate how much driver motor distortion mechanisms in electrical domain turn into acoustic domain by analyzing the current, analyzing the circuit impedance. Series impedance does not actually reduce driver distortion because the distortion mechanisms are inherent in the motor physical properties but it would affect current, and by reducing current we can reduce how much of the distortion turns into acoustic domain.
Its debatable how audible any of this is with the playback level expected and depends on the driver motor quality, although with little reasoning cone breakup peak is probably the place where distortion gets audible if ever as it is effectively an amplifier in acoustic domain, high Q peak on-axis, and thus most important place to consider passive network for a active system I think.
See this:
https://purifi-audio.com/blog/app-notes-2/low-distortion-filter-for-ptt6-5x04-naa-11
Usually there is low pass filter below the worst breakup frequency to prevent nasty sound, also some kind of a notch EQ to reduce the on axis peak in acoustic domain, but if the filters do not increase driver load impedance as well then the distortion products would be left unattenuated. Distortion products beyond low pass filter frequency get less masking from the main signal. Furthermore a cone breakup will amplify the motor born distortion (it amplifies anything, any current through the voice coil) and could be +10db. If this this less masked, high Q peak amplified distortion is not audible then none of the driver motor distortion is 🙂
Problem is that we can attenuate/filter the main signal in multiple ways to knock out a cone resonance peak while only the series impedance method affects driver motor distortion as well, so probably should use that one if distortion is a concern. Just be conscious about it and you are fine, deal with it if necessary, its only few passive parts.
Here is nice article for basis how it works out: https://www.edn.com/loudspeaker-operation-the-superiority-of-current-drive-over-voltage-drive/
Pretend not seeing all the current drive / voltage drive (amplifiers) fight of, its irrelevant, it is about impedance in series with a driver.
Last edited:
As far as the polyfill suggestion, been there, done that. Tried various materials with no real effect.
Looking back through the thread, I do not see a near field scan of the port... or did I miss it?
If the 200 Hz problem is not coming through the cone, it must be coming through either the port, or the cabinet walls... a NF scan of the port will either confirm or eliminate the port as the source of the problem.
j.
If the 200 Hz problem is not coming through the cone, it must be coming through either the port, or the cabinet walls... a NF scan of the port will either confirm or eliminate the port as the source of the problem.
j.
You sure it's not in the woofer? I see a peak at the spot on my NF woofer data on post #121 (top of this page).
This was NF port measurement.
https://www.diyaudio.com/community/threads/help-with-3-way.400834/post-7390673
This was NF port measurement.
https://www.diyaudio.com/community/threads/help-with-3-way.400834/post-7390673
in post #121 the NF scan of the woofer shows a +1 dB peak at ~180 Hz. You also made a room response scan (good idea BTW) shown on the same plot, and this shows a +3 dB peak at 180 Hz. The fact that the measured peak is higher in the room curve than in the NF woofer curve tells me that it is not coming from the woofer... A small amount of it might be coming through the woofer cone, but most of the 180 Hz resonance is coming from somewhere else.
Looking at the scan of the port in post #9, it looks like there is a peak at ~ 180 Hz. This could be the source. But it is also possible that there is an acoustical resonance or structural resonance at 180, and it is just leaking out the port.
If you send the speaker a 180 Hz sine wave signal, and tune the frequency to the peak, you should be able to determine whether or not a panel is excessively vibrating. This can rule out or confirm a structural resonance. Or you can use a mechanics stethoscope as I mentioned earlier.
Looking at the scan of the port in post #9, it looks like there is a peak at ~ 180 Hz. This could be the source. But it is also possible that there is an acoustical resonance or structural resonance at 180, and it is just leaking out the port.
If you send the speaker a 180 Hz sine wave signal, and tune the frequency to the peak, you should be able to determine whether or not a panel is excessively vibrating. This can rule out or confirm a structural resonance. Or you can use a mechanics stethoscope as I mentioned earlier.
What would you qualify as excessive vibration? I'd argue almost every panel in this speaker feels quite 'lively' to me except for the baffle.
Can you clarify on the use of the stethoscope, my partner works in medicine so we do have some of those around.
Can you clarify on the use of the stethoscope, my partner works in medicine so we do have some of those around.
So just move the port to the outside of the box put a 90 on the baffle and fire the port downwards and listen
A medical stethoscope is way too sensitive for this work... you could damage your hearing using it on a speaker cabinet.
A mechanic's stethoscope is used to probe an engine, transmission, or other machine for unusual internal sounds.
Expect to pay about $10 at an auto parts store.
A structural resonance will radiate from the panel, and it will also leak out through the port.
A mechanic's stethoscope is used to probe an engine, transmission, or other machine for unusual internal sounds.
Expect to pay about $10 at an auto parts store.
Honestly, that does not sound good. On a well constructed speaker cabinet, you should feel a very faint vibration when the speaker is at normal SPL (80 - 90 dB). Of course the ideal situation would be no perceptible vibration, but our hands are fairly sensitive, and we will perceive a slight vibration before it becomes audible. But if the vibration is "quite lively" then it is probably quite audible.
A structural resonance will radiate from the panel, and it will also leak out through the port.
Medical scope noted, yes those are crazy sensitive and kinda scare me to use.
I'd like to attempt to tackle bracing the side panels, they feel and sound like they are radiating a lot with the problem frequencies.
What type of bracing is ideal? Panel to panel? strips of wood on the offending panels? Both? I'm limited in what can be connected to vertically by the brace behind the midrange. Better to use solid hardwoods or ply? I feel like the ply wood I have is not the greatest and might prone to splitting depending on the piece used.
I'd like to attempt to tackle bracing the side panels, they feel and sound like they are radiating a lot with the problem frequencies.
What type of bracing is ideal? Panel to panel? strips of wood on the offending panels? Both? I'm limited in what can be connected to vertically by the brace behind the midrange. Better to use solid hardwoods or ply? I feel like the ply wood I have is not the greatest and might prone to splitting depending on the piece used.