I'm sure I'll say it again before long
The wet fart line WAS pretty funny, though!
Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Ampeg is mixing cabs!
- Thread starter wintremute
- Start date
The wet fart line WAS pretty funny, though!
I'm pretty sure many folks who think they would know one or two things about science do the following mistake.
If a cabinet phase response shows a phase shift of 90 degree at 100Hz they think the signal transfere along the cab was delayed by 2.5msec which equals 1/4 of time cycle of 100Hz.
So they (may) think in practice a sine wave of 100Hz was delayed by 2.5msec due to 90 degree phase shift.
They also may belief that two identical signals transfered along two different cabs that show different phase will cause little delay between the two sources caused by phase deviation.
In real world practice (and also in science) the mechanism are quite different.
If a sine wave was transfered to a cab that would show a steady 90 degree phase shift means that the sine wave was turned into cosine wave. This means if the input signal to the cab is a sine then the cab does (immediatelly) output a cosine waveform without any delaying time versus the input.
If a second cab would cause (steady) 180 degree phase shift at just the same frequency then the second cab would (immediatelly) output a reverse sine wave versus the sine input (no delay involved).
As neither of the cabs outputs the signal delayed means there is still 0 delay between the cabs although phase differs a very lot
If a cabinet phase response shows a phase shift of 90 degree at 100Hz they think the signal transfere along the cab was delayed by 2.5msec which equals 1/4 of time cycle of 100Hz.
So they (may) think in practice a sine wave of 100Hz was delayed by 2.5msec due to 90 degree phase shift.
They also may belief that two identical signals transfered along two different cabs that show different phase will cause little delay between the two sources caused by phase deviation.
In real world practice (and also in science) the mechanism are quite different.
If a sine wave was transfered to a cab that would show a steady 90 degree phase shift means that the sine wave was turned into cosine wave. This means if the input signal to the cab is a sine then the cab does (immediatelly) output a cosine waveform without any delaying time versus the input.
If a second cab would cause (steady) 180 degree phase shift at just the same frequency then the second cab would (immediatelly) output a reverse sine wave versus the sine input (no delay involved).
As neither of the cabs outputs the signal delayed means there is still 0 delay between the cabs although phase differs a very lot
Jeez, that's pretty close. CL400 is like the Led Zep of Talkbass quotes
Maybe we need a "JimmyM Says" quotes thread. Would make for excellent reading!
Or it could be a source of embarrassment and humiliation. Which I guess would be excellent for everyone else
LOL. You know, I didn't think of that.Or it could be a source of embarrassment and humiliation. Which I guess would be excellent for everyone else
I'm sure some of your "friends" would show up and say it was a plot to sell more Ampeg or something like that.
Last edited:
- Oct 16, 2014
- 6,942
- 29,552
- Disclosures
- Endorsing Artist: Langstrom Carrot Farms
Hey, there's a thread a few doors down from this one that is titled "Don't Expect the Impossible from a 15 inch driver". Should I go in there and post "that's why you need to use a 410 with it."?
Oh God, YES!!!Hey, there's a thread a few doors down from this one that is titled "Don't Expect the Impossible from a 15 inch driver". Should I go in there and post "that's why you need to use a 410 with it."?
Is there anything wrong with describing the totality of phase shifts of any one cabinet as its phase response?
phase response is the curve that shows phase shift versus frequency.
Without the curve it's impossible to predict group delay.
For a pure singular phase shift that referes (of course) to a distinct frequency point it's impossible to predict delay if the phase response (the function of the curve) is unknown.
That's nothing but pure math cause group delay is the derivative of phase response
If phase shift was 90 degree and stays steady from >DC to infinite then delay equals 0
If phase shift was 180 degree and stays steady from >DC to infinite then delay equals 0
If phase shift was 0 degree and stays steady from >DC to infinite then delay equals 0
If phase shift was 63,769 degree and stays steady from >DC to infinite then delay equals 0
Last edited:
Anyone who considers phase shift at 100Hz reveals themselves to not know as much as they think they do. Aside from reverse polarity phase sourced destructive interference doesn't occur in the lows, where wavelengths are measured in feet or meters. They occur in the midrange and highs, where wavelengths are measured in inches or centimeters. The same situation applies to group delay. For it to be audible the delay would have to be so high that the speaker design is totally flawed to begin with, so it doesn't need to be considered. To say group delay is unpredictable is a different situation. Even the most rudimentary speaker modeling software will predict it with a very high degree of accuracy. Of course that same software also shows phase response, including in the midrange where it matters.
Exactly the range where group delay does not matter any more cause the phase response flattens out just to show "more" steady numbers versus frequency.
Except for a passive 2-way crossover network that splits the frequency range which then might cause group delay artefacts if not proper designed.
edit, just to be honest, it's hard to argue with science versus your "wavelength" issue claim at mids and highs. As phase deviation between different MI bass cabs at mid bandwidth is rather small versus deviation at the lowish range, I have got no idea what kind of wisdom you try to tell us.
If there was really a phase issue with different driver size at mids like you claim, exatly THAN it was absolutely impossible to design proper crossover networks.
You may want read some basics of 2nd and 4th order filter transfer functions just to learn one or two things about drivers and cabs and crossover networks
Bill was right in this regard but did not mention that it's only true if there is only "one" source.
He also did not mention that (rather) high delay at the lowish bandwidth tells some essential things about cab design respectively damping factor of the cab system
Even science math can't predict delay if there is nothing known but phase shift vers a singular frequency point.
Without the mathematical function for phase response it's not possible to predict delay. However, if you read on TB forum carefully enough you will notice exactly an opposite claim!
Last edited:
In the meanwhile I'm reading and posting on TB forum since a couple of years now.
Let me do a short summary.
Some folks do claim a bad sound cause there was phase issues at the lowish range, while others claim the evil of bad sound at mids and/or treble.
Some think there is delay due to phase deviation that does mush to the room, while others claim cancelation for doing just the same artefact.
If everybody was right so how could any seroius science serve anybodys personal "cup of tea"? No way to explain all of experienced "claims" with only one science.
Will this make any sense?
IMHO No!
We could "create" an alternate science which might be "tuned" to anybodys needs.
Let me do a short summary.
Some folks do claim a bad sound cause there was phase issues at the lowish range, while others claim the evil of bad sound at mids and/or treble.
Some think there is delay due to phase deviation that does mush to the room, while others claim cancelation for doing just the same artefact.
If everybody was right so how could any seroius science serve anybodys personal "cup of tea"? No way to explain all of experienced "claims" with only one science.
Will this make any sense?
IMHO No!
We could "create" an alternate science which might be "tuned" to anybodys needs.
Last edited:
From Bill Fitzmaurice:
Pretty much. The assumption is often made that tens, twelves and fifteens each have a characteristic tone and/or frequency response, so using one from column A and one from column B will give you a 'best of both worlds' result. That assumption is just plain wrong. There is no difference in either tone or frequency response that can be attributed to driver size in and of itself. Where the average musical instrument driver is concerned there's actually very little difference between the three sizes. But there are differences in the phase responses of different drivers and cabs, even different cabs loaded with the same driver, and when you mix and match them the results are at best unpredictable, the worst case being the cabs will detract from each other as much as they augment each other. So if you must take on item from column A and one from column B do so it a Chinese restaurant, not a music store.
Bill was absolutely right on this that even same drivers cause different phase if loaded into different cabs.
However, he didn't tell ya in this regard that phase at mids and treble is dominated by the driver, not the cab design, so if the drivers was matching then there is only the lows (and little to none at low mids) of total bandwidth remaining where phase does NOT match.
If different cabs show different phase (and every different cabs with different f3 will show different phase) then there is some chance that group delay and step response and pulse response don't much. However he didn't tell ya that different cabs (different f3) but matching filter alignment (identical filter coefficients) yet show some deviation in phase response.
Any different cabs with diffrent f3 although designed to accomplish as well as possible in any regard yet show (in general) some difference in phase response.
While Bill themselves addressed the phase response "issue" even to same drivers loaded into different cabs, so phase response at mids and treble was of no interest but lows, so here you are with an opposite claim.
@Rick James said
Aside from reverse polarity phase sourced destructive interference doesn't occur in the lows, where wavelengths are measured in feet or meters. They occur in the midrange and highs, where wavelengths are measured in inches or centimeters.
and
Anyone who considers phase shift at 100Hz reveals themselves to not know as much as they think they do.
In Germany we tell it this way,
The ones who live in a house made of glass should take care to not throw stones against the walls
Pretty much. The assumption is often made that tens, twelves and fifteens each have a characteristic tone and/or frequency response, so using one from column A and one from column B will give you a 'best of both worlds' result. That assumption is just plain wrong. There is no difference in either tone or frequency response that can be attributed to driver size in and of itself. Where the average musical instrument driver is concerned there's actually very little difference between the three sizes. But there are differences in the phase responses of different drivers and cabs, even different cabs loaded with the same driver, and when you mix and match them the results are at best unpredictable, the worst case being the cabs will detract from each other as much as they augment each other. So if you must take on item from column A and one from column B do so it a Chinese restaurant, not a music store.
Bill was absolutely right on this that even same drivers cause different phase if loaded into different cabs.
However, he didn't tell ya in this regard that phase at mids and treble is dominated by the driver, not the cab design, so if the drivers was matching then there is only the lows (and little to none at low mids) of total bandwidth remaining where phase does NOT match.
If different cabs show different phase (and every different cabs with different f3 will show different phase) then there is some chance that group delay and step response and pulse response don't much. However he didn't tell ya that different cabs (different f3) but matching filter alignment (identical filter coefficients) yet show some deviation in phase response.
Any different cabs with diffrent f3 although designed to accomplish as well as possible in any regard yet show (in general) some difference in phase response.
While Bill themselves addressed the phase response "issue" even to same drivers loaded into different cabs, so phase response at mids and treble was of no interest but lows, so here you are with an opposite claim.
@Rick James said
Aside from reverse polarity phase sourced destructive interference doesn't occur in the lows, where wavelengths are measured in feet or meters. They occur in the midrange and highs, where wavelengths are measured in inches or centimeters.
and
Anyone who considers phase shift at 100Hz reveals themselves to not know as much as they think they do.
In Germany we tell it this way,
The ones who live in a house made of glass should take care to not throw stones against the walls
Last edited:
