Thanks, Markus.
Can you repeat at ~22.5° off-axis, please? Is that up or down or both in combination, i.e., 7 measurements in each direction?
Just to be sure I'm understanding your approach, the total span of vertical movement of the mic is ~30"?
Can you repeat at ~22.5° off-axis, please? Is that up or down or both in combination, i.e., 7 measurements in each direction?
Just to be sure I'm understanding your approach, the total span of vertical movement of the mic is ~30"?
We see 35cm up and down, so the overall coverage is around 40°. Is a bigger angle of any practical interest?
Best, Markus
Best, Markus
markus76 said:
No, that's fine, and comparable. Arctan 35/100 = 19.29°, equivalent to the dark blue curves in my separate up and down plots.
The real question is how it behaves on the recommended 22.5° horizontal listening axis....
This was an average of the response of my abbey across it's listening axis. I see no sign of the null found on axis.
Mind you, this is in room with no windowing to remove room artifacts. This just uses the averaging technique.
Oops, thats the center's measurement, that has a modification to the crossover to improve axial response. That was averaged across a 45 degree range I believe.
Below are the L and R speakers, measurements taken in room at two meters, speakers in place, so not moved to the middle of the room.
Again, lot's of room in them responses.
Below are the L and R speakers, measurements taken in room at two meters, speakers in place, so not moved to the middle of the room.
Again, lot's of room in them responses.
pjpoes said:
We wouldn't expect to see any null on-axis vertically.
The question is whether nulls appear off-axis there, and if so, at what angles....
It looks like a driver with a few dB of peaking from 700-1kHz might be desirable here.
Eminence magnum series drivers (discontinued) all had a little peaking here with fairly clean rolloffs above (all things being equal).
I'll try those with my woofer-> OS waveguide project. I have 15HOs anyway.
Eminence magnum series drivers (discontinued) all had a little peaking here with fairly clean rolloffs above (all things being equal).
I'll try those with my woofer-> OS waveguide project. I have 15HOs anyway.
markus76 said:
I imagine those sound different. 😉
The ear is pretty sensitive in the roller coaster ride area of the response.
ZilchLab said:
I think you misunderstood. The null that is found on axis isn't showing up in those off axis averaged response. It's horizontal axis not vertical though.
pjpoes said:
Yes, I believe we understand that; it's the reason Earl recommends they neither be listened to nor EQ's on-axis horizontally.
We're looking for the vertical nulls at the horizontal listening axis here.
badman said:
Yes, but absent more off-axis measurements, I'd be hesitant to attribute that "dip" to driver interference, though I wouldn't be much surprised to discover that it in fact was a null edge....
15 measurements on a vertical straight line 1m in front of the Nathan on 22.5°, each 5cm apart. Window time is 3ms, nearest boundary was >1m away:
Up - "Keep your seat belt fastened while seated. Turbulence can strike at any time.":
Down:
Center tweeter:
Center tweeter/woofer:
Center woofer:
Best, Markus
An externally hosted image should be here but it was not working when we last tested it.
Up - "Keep your seat belt fastened while seated. Turbulence can strike at any time.":
An externally hosted image should be here but it was not working when we last tested it.
Down:
An externally hosted image should be here but it was not working when we last tested it.
Center tweeter:
An externally hosted image should be here but it was not working when we last tested it.
Center tweeter/woofer:
An externally hosted image should be here but it was not working when we last tested it.
Center woofer:
An externally hosted image should be here but it was not working when we last tested it.
Best, Markus
What do you mean by "Center tweeter/woofer"?
Looks a little like the phase between the woofer and tweeter might be off a little bit as the null should be a little further "up". The resistor in series with the large LP cap has a big effect on this phase and will shift this null up and down with almost no change in any other aspect. I'm just finishing a couple pair of Nathans that I'll test this week and see if they show the same thing and what I can do to improve it. I'll let you know.
Thanks for that data by the way. Its good to see since, as I said, it's very difficult for me to do this at 22.5 degrees.
Thats about a 25 dB scale?
Looks a little like the phase between the woofer and tweeter might be off a little bit as the null should be a little further "up". The resistor in series with the large LP cap has a big effect on this phase and will shift this null up and down with almost no change in any other aspect. I'm just finishing a couple pair of Nathans that I'll test this week and see if they show the same thing and what I can do to improve it. I'll let you know.
Thanks for that data by the way. Its good to see since, as I said, it's very difficult for me to do this at 22.5 degrees.
Thats about a 25 dB scale?
markus76 said:"It's Oh So Quiet" 🙂
Do we know Nathan's crossover frequency?
I was off doing a redo, 'cause when I tried to measure the driver offset, I discovered that the wrong amp was on, and thus I was measuring a different speaker on the test stand, a Karma Indignia, if anybody cares.
Here's the driver offset, the tweeter impulse is red, the woofer, black. If this is a legit way of getting it, the time differential says the tweeter acoustic center is 0.66" ahead of the woofer's at 22.5°, tilting the plane of equidistance down, which makes about zero sense, as the two drivers are connected in phase:
Attachments
gedlee said:
Yes, 1dB increments.
I just slided the mic stand up and down. That's not "by the book" but error in measurement should be minimal.
Here's a picture that shows "Center tweeter/woofer", "Center tweeter" and "Center woofer":
An externally hosted image should be here but it was not working when we last tested it.
Best, Markus
- Status
- Not open for further replies.