Hi guys, after building TQWP, bassreflex and closed cabinets in hornresp, I met delayed bass. According to Aldoshina's researching in 1985, there is acceptable Group Delay values that human ear can't detect: 40-80 hz: 15 ms 80-160 hz: 3 ms 160-1 KHz: 2.5 ms >1 KHz: 1.5 ms What do you think about value - can we increase 40-80 hz range up to 25 ms without blurred, trailing bass? Or there is another factors of delayed bass...?
... I cannot reply with a humongous amount of scientific knowledge about this, but personal preference would be to keep the design at or less than ~3-5ms/200hz ~10-12ms/40hz ~20-30ms/20hz. It's very tricky to get bass 100% "just right", there's always some room for improvement.
I've never heard a setup that sounded "perfect" on everything, but then again others might have found the holy grail.
I've never heard a setup that sounded "perfect" on everything, but then again others might have found the holy grail.
While doing some research on group delay for the eBook in my sig, this was about all I could find: Some observations on group delay.
Mitchba:
"Another interesting fact to ponder is that group delay
accumulates throughout the entire analog recording chain, due
to the limited bandwidth of each mic, preamp, amp, recording
medium, etc."
This is my conclusion as well, even if your system has really good values on group delay you will eventually arrive at a point of deminishing returns, not because of fidelity of the sound reproduction, but because of lack of fidelity in the recording chain and/or studio. At a certain point many will become more selective regarding recording material because the quality of the recording does not fit well with the abilities of the system used for reproduction, and does not relate to the performance of the musicians involved. And also the choices of the technician in the studio and his/her familiarity with the equipment also plays a role in this. So if your system reaches a certain level of quality, to achieve even higher levels of enjoyment of the system and the music it is sometimes better to add a bit of delay, and maybe even reeverb at certain frequencies.
"Another interesting fact to ponder is that group delay
accumulates throughout the entire analog recording chain, due
to the limited bandwidth of each mic, preamp, amp, recording
medium, etc."
This is my conclusion as well, even if your system has really good values on group delay you will eventually arrive at a point of deminishing returns, not because of fidelity of the sound reproduction, but because of lack of fidelity in the recording chain and/or studio. At a certain point many will become more selective regarding recording material because the quality of the recording does not fit well with the abilities of the system used for reproduction, and does not relate to the performance of the musicians involved. And also the choices of the technician in the studio and his/her familiarity with the equipment also plays a role in this. So if your system reaches a certain level of quality, to achieve even higher levels of enjoyment of the system and the music it is sometimes better to add a bit of delay, and maybe even reeverb at certain frequencies.
"...Group delay (GD) can be thought of as related to the time elapsed between a signal of a specific frequency applied to the driver and the cone's
attempt to recreate that stimulus, as compared to the next adjacent frequency. (And the next -ad infinum.)
This delay is a function of the phase of the system at those frequencies..."
A couple of years ago I set to try to understand this better. I used a JBL 2242, driven by a 2kw AE Techron DC coupled amplifier, a laser for cone position, my TEF measurment platform with a mic on one channel and the laser on the other and finally the amp output, the mic and the laser on three channels of a four channel scope.
I need to set this up again and go further but the takeaway is the speakers coil/cone assembly follows the drive voltage with no delay until the frequency is high enough that the voice coil inductance comes into play. What is late is the pressure wave from the cone as measured by the mic, its peak is right about 90 degrees behind the cone excursion peak. For the record the mic was less than 0.250" from the dust cap and the laser focused on the dust cap right next to it.
Since phase is relative to something elses clock like the original signal but delay is measured in real time, this 90 degree phase lag becomes (significant) larger in real time as frequency decreases. At 20Hz it's 25% of 50ms even though the cone is following the amp exactly.
This leaves me to wonder, since a periodical signal obviously has a time component, in the acoustical world, it also has size. So, how much of what we see in the measurment of group delay in a very low frequency system is a natural phenomenon, not an electrical / acoustical problem?
Barry.
attempt to recreate that stimulus, as compared to the next adjacent frequency. (And the next -ad infinum.)
This delay is a function of the phase of the system at those frequencies..."
A couple of years ago I set to try to understand this better. I used a JBL 2242, driven by a 2kw AE Techron DC coupled amplifier, a laser for cone position, my TEF measurment platform with a mic on one channel and the laser on the other and finally the amp output, the mic and the laser on three channels of a four channel scope.
I need to set this up again and go further but the takeaway is the speakers coil/cone assembly follows the drive voltage with no delay until the frequency is high enough that the voice coil inductance comes into play. What is late is the pressure wave from the cone as measured by the mic, its peak is right about 90 degrees behind the cone excursion peak. For the record the mic was less than 0.250" from the dust cap and the laser focused on the dust cap right next to it.
Since phase is relative to something elses clock like the original signal but delay is measured in real time, this 90 degree phase lag becomes (significant) larger in real time as frequency decreases. At 20Hz it's 25% of 50ms even though the cone is following the amp exactly.
This leaves me to wonder, since a periodical signal obviously has a time component, in the acoustical world, it also has size. So, how much of what we see in the measurment of group delay in a very low frequency system is a natural phenomenon, not an electrical / acoustical problem?
Barry.
I just stumbled across you post and I find this very interesting! It verifies what I had determined a while back and got into an argument about - that the peak of the pressure occurs when the driver is moving fastest, which is at the mid-point of it's excursion in either direction. At the peaks the driver is not moving so it isn't creating any pressure. It didn't occur to me though that this doesn't coincide with a condenser mic's reading so the recording/playback chain puts everything 90 degrees out of phase! Would applying a digital bandpass phase adjustment of 90 degrees have any effect on this? Seems a ribbon mic. used during recording would also solve this problem. It gets more complicated when playing very low frequencies in a small space, because then the pressure peak does start to correspond more closely with the peak excursions of the driver. So we have to look at the entire recording chain and the properties of the speaker and the playback environment it is in.
Last edited: