A difference a couple db over a very narrow frequency band is not audible. A loss of a couple db is a significant portion of your port output.
I assume you saved the graphs if you went to all the work of measuring at varying levels just to find out if your port was acting properly. Please post them, pics or it didn't happen.
You can't turn it up past xmax, that's the WHOLE POINT of the high pass filter. If you turn it up and push it well past xmax, why even use a high pass filter in the first place?
Xmax on the driver shown is 6mm. Why would you turn it up to over twice that excursion? You are not making any sense and I think you are seriously backpedalling here.
You specifically said "False, and ignorant. A high pass filter will have no effect on port velocity at any given frequency and output level."
I proved you wrong (again).
The point of a high pass filter is so you can turn up the volume so the 2nd excursion bump hits xmax without damaging the driver with the 1st really bad excursion bump. If you didn't have the high pass filter and turned the volume down so the first really bad excursion peak stayed below 6mm then the WHOLE PASSBAND would be very quiet.
At this point it doesn't look like you understand what high pass filters are for or what effect they have.
As I mentioned before, velocity on OP's design was pushing past 16 m/s without a high pass filter. If OP doesn't HAVE an appropriate high pass filter, a high pass filter probably won't be used. This design is already not loud enough and it's possible and entirely probably that this driver will be pushed well past xmax, adding a LOT more velocity than shown (well past 16 m/s).
And I'll address this separately so it doesn't get lost in the mix.
I did NOT lower the volume. I said very clearly the ONLY difference between the two sets of graphs is a high pass filter is added. You can even see the voltage input right on the graphs. You can also see that the higher freqeuncies are at exactly the same level, the only thing that changed is the output level, excursion and velocity right around tuning.
A pipe organ sounding at 16 hz is not a sharp narrow bandwidth signal. There will be some output below 16 hz. Even if there wasn't, the designs being shown are tuned around 16 hz, and the hpf shown is what it takes to protect a driver AT it's tuning frequency.
At this point you've called me ignorant, claimed this was a bad design, you've been proven wrong about velocity issues, the purpose of a high pass filter, the severity of resonances, Allison effect and omni speakers and so on and so on. Shall we continue?
Last edited:
Seriously, the personal insults are really getting out of hand. It doesn't bother me, but argue with technical information, not insults.
This is perhaps the best example of clueless that I've ever seen . . . and pretty much explains why just about everything you claim to have "proven" is wrong.
There is nothing below the 16.35 Hz fundamental of the lowest note of the 32" stop. NOTHING. All 25 other notes in the rank (assuming that it maps to the pedal, and that the pedal is typical of small-medium church organs) are similarly defined (you could look up the fundamental pitch of each to the nearest cent if you cared). All their harmonics are higher (duh), none lower. The lowest frequency the speaker sees is the fundamental of the lowest note sounding (which is by all odds rarely that 32' pipe).
When you put that silly filter in it does a wonderful job of reducing excursion that isn't happening, and in the process lowers the level of the fundamental of the sounding note that is. Raise the level up to restore the volume of that note and your filter has changed nothing (except giving less "protection" from excursion that wasn't going to happen anyway).
Until you understand that . . . (unlikely, since you already know everything there is to know) . . . any further "conversation" on the subject is pointless.
Even a 16 hz sine wave is going to have a wider bandwidth of 16.0 to 16.0 hz when played on a speaker and measured. A pipe organ is not a pure 16.0 tone with no overhang on either side. Not that this matters.
I'm NOT saying there are harmonics lower than 16 hz, simply that a 16 hz tone has a wider bandwidth than 16.0 hz.
If you play a 16 hz note on a 16 hz tuned box the excursion absolutely WILL be happening, what are you even talking about? YOU CAN'T RAISE THE LEVEL PAST XMAX IF YOU WANT TO STAY INSIDE THE LINEAR REGION, THAT'S THE WHOLE POINT OF THE FILTER.
I understand you don't want to admit you are wrong, but you can just stop instead. What you are saying doesn't make any sense at all. High pass filters are to protect the driver at and below tuning. If there was no excursion at or below tuning they would not be required. If you want to get around using one you have to tune LOWER than the lowest note played or keep the volume way way down.
To make this really really clear for you, if a 16 hz note is played on a 16 hz tuned box without a high pass filter you need to keep the volume down or the driver is going to be destroyed.
So far I've proved that almost every single thing you have said so far is wrong, and I did it in a technical and scientific manner with supporting evidence. The only real rebuttal you have provided is a constant stream of personal insults. Like I said, I don't mind the insults but can you try a little harder to counter with actual technical information? I'm spending actual time and effort on this, you are not.
"best example of clueless I've ever seen, ignorant, sense of humor exceeds your intelligence, you already know all there is to know, you made stupid assumption(s), ... even modestly clever people ..."
That's just a small sample from the last 10 posts, every time I show technical evidence that you are wrong your only rebuttal is personal insult. I am blunt and I am unforgiving in pointing out your constant stream of mistakes, but come on man, throw in a little technical information every once in awhile to make this interesting.
I'm NOT saying there are harmonics lower than 16 hz, simply that a 16 hz tone has a wider bandwidth than 16.0 hz.
If you play a 16 hz note on a 16 hz tuned box the excursion absolutely WILL be happening, what are you even talking about? YOU CAN'T RAISE THE LEVEL PAST XMAX IF YOU WANT TO STAY INSIDE THE LINEAR REGION, THAT'S THE WHOLE POINT OF THE FILTER.
I understand you don't want to admit you are wrong, but you can just stop instead. What you are saying doesn't make any sense at all. High pass filters are to protect the driver at and below tuning. If there was no excursion at or below tuning they would not be required. If you want to get around using one you have to tune LOWER than the lowest note played or keep the volume way way down.
To make this really really clear for you, if a 16 hz note is played on a 16 hz tuned box without a high pass filter you need to keep the volume down or the driver is going to be destroyed.
So far I've proved that almost every single thing you have said so far is wrong, and I did it in a technical and scientific manner with supporting evidence. The only real rebuttal you have provided is a constant stream of personal insults. Like I said, I don't mind the insults but can you try a little harder to counter with actual technical information? I'm spending actual time and effort on this, you are not.
"best example of clueless I've ever seen, ignorant, sense of humor exceeds your intelligence, you already know all there is to know, you made stupid assumption(s), ... even modestly clever people ..."
That's just a small sample from the last 10 posts, every time I show technical evidence that you are wrong your only rebuttal is personal insult. I am blunt and I am unforgiving in pointing out your constant stream of mistakes, but come on man, throw in a little technical information every once in awhile to make this interesting.
Last edited:
This is the image that was posted before, a 10 second frequency/amplitude analysis of a pipe organ. Clearly the 16 hz note has a LOT wider bandwidth than 16.0 to 16.0 hz. It looks like it extends down to 14 hz at the 10 db down point. Even at 3 db down it's 2 or 3 hz wide, extending below and above 16 hz.
Again, not that this matters much, but a hpf is required, and I showed you a proper hpf and what happens when you apply it.
Again, not that this matters much, but a hpf is required, and I showed you a proper hpf and what happens when you apply it.
Last edited:
With [13] 16' and [2] 32', this makes 32 Hz a serious 'Voice of God' presentation: OHS Database: Instrument Details
GM
No, it doesn't, and it isn't. What you are looking at is not the "bandwidth" of the signal, but the width of the sampling filter in the spectrum analyzer . . . the workings of which you don't understand any better than you understand organs or acoustics (which is, in turn, far far less than you imagine that you do).
No, it isn't and you didn't. Repeatedly saying that you've done something which you haven't doesn't make it so that you did. All it does is further demonstrate that "know it all" and "idiot" are often synonyms . . .
Come on now fellas, would it kill us to be a little more kind?
For the record, I'm able to play that sample out here in the garage, where I have no high pass on my subs, at quite ridiculous volumes. The garage subs are dual 10" bass reflex, and are tuned to around 21 hz. They're relatively low displacement Peerless *antiques* driven from a fairly powerful receiver that'll do a bit over 100w. It has no problem clacking the coils during the 1812 overture, however the organ music is fairly safe, unless I turn it up to the point where I'm looking for earplugs to protect myself from screaming trompette ranks.
All of this proves nothing of course, and what we really need is someone with an RTA and a buddy who can record samples with all the stops out and all the pedals mashed, but in the meantime, can we please not kill one another over potential issues that, in either case, could be solved with relative ease?
For the record, I'm able to play that sample out here in the garage, where I have no high pass on my subs, at quite ridiculous volumes. The garage subs are dual 10" bass reflex, and are tuned to around 21 hz. They're relatively low displacement Peerless *antiques* driven from a fairly powerful receiver that'll do a bit over 100w. It has no problem clacking the coils during the 1812 overture, however the organ music is fairly safe, unless I turn it up to the point where I'm looking for earplugs to protect myself from screaming trompette ranks.
All of this proves nothing of course, and what we really need is someone with an RTA and a buddy who can record samples with all the stops out and all the pedals mashed, but in the meantime, can we please not kill one another over potential issues that, in either case, could be solved with relative ease?
FFT analysis of a dynamically changing signal, like music, gives just rough estimate. Theoretically it is invalid.
Thanks DrDyna.
I should be using Audactiy a little more.
Nice choice of organ music. Michael Murray is a bit laid back in some ways like tempo but the recordings and overall presentation are quite interesting.
I should be using Audactiy a little more.
Nice choice of organ music. Michael Murray is a bit laid back in some ways like tempo but the recordings and overall presentation are quite interesting.
I remember reading that John Ergle took mics up to the pipe mouths to get the effect he heard in the Methuen Hall onto his recording.
Talking about a long wire run!
Talking about a long wire run!
On the bandwidth issue, I think it's important to note that the pipes, the wind that drives them, and all the environmental factors probably do add up to low notes that don't perfectly track a single fundamental frequency, but rather wag around a little bit, especially if they're played with vibrato.
We can't live in the theory world when we're talking about a physical instrument that most likely refuses to comply at every available opportunity.
We can't live in the theory world when we're talking about a physical instrument that most likely refuses to comply at every available opportunity.
I think GM, DrDyna, and dewardh have all pointed out that this application is sensitive to many effects that DIY designers generally overlook or do not consider important.
Generally people have not really heard clean low end bass. I have done demonstrations where this was most apparent. The results of the demonstration was that almost half of the people preferred the sound of distorted bass! It sounds louder and fuller.
But it is not an accurate reproduction of the incoming signal.
And I'll throw my hat into the ring about port compression.
There's one fact. A simulation program will not hear port compression.
A discerning listener will.
If you have the hearing acuity to tune an instrument you will hear these effects of box and port resonances, and port compression.
I know this because I hear all these effects.
And quite a few people that I have worked with do to. Ones that work in audio daily and ones that are simply acute listeners.
Generally people have not really heard clean low end bass. I have done demonstrations where this was most apparent. The results of the demonstration was that almost half of the people preferred the sound of distorted bass! It sounds louder and fuller.
But it is not an accurate reproduction of the incoming signal.
And I'll throw my hat into the ring about port compression.
There's one fact. A simulation program will not hear port compression.
A discerning listener will.
If you have the hearing acuity to tune an instrument you will hear these effects of box and port resonances, and port compression.
I know this because I hear all these effects.
And quite a few people that I have worked with do to. Ones that work in audio daily and ones that are simply acute listeners.
One thing not discussed so far is the driver of choice and it's contribution of distortion to the signal. Sorry if I missed it.
You can find quite a few independent driver tests here:
Data-Bass
And tested system of driver and enclosure here:
Data-Bass
Why am I posting this?
Distortion masks clarity of reproduction. In the intended use of this thread getting the sound as clean as possible means you will hear more of the organ sound and less of the drivers sound.
Here are two examples.
Data-Bass
Static Graphs screen grab is below.
And something different:
Data-Bass
Again from the static graphs section.
Differences?
A simple rule of thumb in bass reproduction is that below 10% distortion people don't usually notice it.
Take a look at the figures and see what you can learn from this comparative example.
There are quite a few nuggets of information available in these charts.
One is the power input measured in volts input.
Another is sound pressure level generated for a given power input.
I'll throw in a monkey wrench. Factor in cost.
There are other drivers and systems measured on this website. I know Josh and he is very careful to make accurate well constructed measurements that are both repeatable and useful as comparative measurements.
And these are not SIMS!
These are actual functioning systems.
You can find quite a few independent driver tests here:
Data-Bass
And tested system of driver and enclosure here:
Data-Bass
Why am I posting this?
Distortion masks clarity of reproduction. In the intended use of this thread getting the sound as clean as possible means you will hear more of the organ sound and less of the drivers sound.
Here are two examples.
Data-Bass
Static Graphs screen grab is below.
And something different:
Data-Bass
Again from the static graphs section.
Differences?
A simple rule of thumb in bass reproduction is that below 10% distortion people don't usually notice it.
Take a look at the figures and see what you can learn from this comparative example.
There are quite a few nuggets of information available in these charts.
One is the power input measured in volts input.
Another is sound pressure level generated for a given power input.
I'll throw in a monkey wrench. Factor in cost.
There are other drivers and systems measured on this website. I know Josh and he is very careful to make accurate well constructed measurements that are both repeatable and useful as comparative measurements.
And these are not SIMS!
These are actual functioning systems.
- Status
- Not open for further replies.