getting my Valencia project wrapped up and wondering the thoughts regarding time delay between horn and woofer in bi amp application. using a pa2 for xo duty.
where do you measure from the horn lip, or the compression driver? or don't bother?
thks
Printable View
getting my Valencia project wrapped up and wondering the thoughts regarding time delay between horn and woofer in bi amp application. using a pa2 for xo duty.
where do you measure from the horn lip, or the compression driver? or don't bother?
thks
The Acoustic Center of a driver is usually somewhere around the vertical plane of the voice-coil.
So, measure the distance between the 2 voice coils.
:)
Quote:
Originally Posted by Earl K
thk you Earl!
For extra points use an FFT measurement app and measure the arrival time for the woofer and horn driver individually if you want to get it down to a few mS.
Hi trip,
The answer to your question is compression driver, the way Earl mentioned. However the don't bother may also apply (i'll be back later to explain this, no time now).
In the meantime i suggest you read the short and easy to understand paragraph titled "Setting the delay time" on page 2 of the attached document for some background info. See you later.
Richard
Attachment 88127
RE your "or don't bother?"
According to Eargle there's a far more important issue in routine SR than delay. And he questions the audible nature of time offsets, which may be perceived as such but are something else or end up being "buried" under a more serious problem.
"More fundamentally, we must ask the question of just how audible time offsets really are. Most of the immediate audibility of variable time delay in portions of a multi-way system does not result from the effects of relative time delay itself, but rather from amplitude variations in the crossover region due to delay-dependent reinforcements and cancellations between drivers." (...)
"The relatively long HF horns provide most of the delay offset in sound reinforcement systems; this indicates that the LF sections should be delayed with respect to them for proper alignment. These effects are generally inaudible, and their correction in routine sound reinforcement work is not worth the trouble. A far more important consideration is that of maintaining smooth amplitude response through the transition between adjacent elements." (John Eargle, Handbook of Sound System Design, Dividing Networks/Component Matching, Time Domain Performance of Systems, P. 139, 141)
So one could get the delay right and still have to deal with that other more important crossover region issue... Regards,
Richard
yeahQuote:
Originally Posted by RMC
that all makes much sense Richard! thank you for the explain and read!
i moved some delays around live on my pa2 while listening and could hear some of that good and bad cancelling going on.
That unit is for time alignment of multiple speaker sound sources to a main stage system such as fill speakers 200 feet away from the main system speakers.Quote:
Originally Posted by RMC
The original post is about time alignment within the speaker cabinet between the drivers being at different
depths in the cabinet.
It makes a difference and it worth the effort since you have the tool to do it with.
You will be applying delay to the woofer to move it back in time to align with sound leaving the horn.
A few ways to come up with the setting.....
- measure what the voice coil to voice distance is and be sorta close keep in mind
the actual acoustic delay can be different from the physical off set.
- do a null test, for the first one you just need a sine wave tone generator and your ears.
- set up one speaker bi-amped and then with tone generator set to the frequency that you are crossing over at set the volume of the woofer and the horn to be the same, it does not need to be crazy loud!
- now reverse the polarity on either the woofer or the high frequency driver so the drivers are operating in reverse polarity from each other.
- setting straight back from the speaker a few feet directly on axis to the speaker with the PA2 in your hand or using the control app starting a 0 delay on the woofer start slowing increasing the delay time till you get to the point where the sound level cancels out the most as you keep adding delay from that point the sound level will start to increase reduce the delay back till you get back to most cancellation, you could use a sound level meter to display level change as well.
-at the point of the most cancellation that is the offset delay setting.
As Riley mentioned an FFT measurement with something like SMARRT or the shareware program REW will get you even closer.
Using a measurement program the set up is the same only with pink noise and SMARRT you can watch the transfer function display in real time and see the null notch change with the delay settings.
Using REW that program uses a tone sweep and then will display the transfer function response, so you need to do multiple test sweeps as you make adjustments to see what the changes look like.
With those programs you can also look at the phase response of the system, make sure you wire the driver back to normal polarity before looking closely at the phase response.
Changing the crossover cutoff slopes, going higher on the high frequency driver maybe lower on the woofer and then under lapping the crossover points some, 800hz on the high pass for the high frequency and 750hz or 700hz on the low pass for the woofer.
Set the high frequency and slide the woofer low pass around and watch the frequency response and phase response.
Don't forget to set a low frequency high pass filter on the woofer and again don't forget to
wire the speaker back to normal polarity!!!
Have fun!!!
The Rolls delay unit was NOT presented as the solution to his problem.
But rather for "... read the short and easy to understand paragraph titled "Setting the delay time" on page 2 of the attached document for some background info." i.e. basic general speaker delay principle, since he seemed less familiar with delay notion. This, as post indicates, pending another reply from me that would follow.
According to his reply the OP appears to have understood that.