The reason I used to cross domes this low is to keep the source location constant until below the more sensitive region...no problems there but one thing I didn't like about domes is the way they will glare somewhat, particularly when you begin to drive them.
The other thing I didn't like is the wide directivity, with its treble that is heavy once you get it flat on axis. This either means a waveguide or a driver that plays beyond breakup ie. a fullrange driver. It doesn't seem to bother everyone but it's a pet hate of mine.
This may (or not) be the issue. Things are getting hard to hear 45dB down and you're using a third order filter. I'd suggest it's one of the two things I mentioned above, including the possibility that the waveguide is being used too low in its current form.
The other thing I didn't like is the wide directivity, with its treble that is heavy once you get it flat on axis. This either means a waveguide or a driver that plays beyond breakup ie. a fullrange driver. It doesn't seem to bother everyone but it's a pet hate of mine.
No. The slopes should be what is needed to be using each driver correctly, and to hand over smoothly.
This may (or not) be the issue. Things are getting hard to hear 45dB down and you're using a third order filter. I'd suggest it's one of the two things I mentioned above, including the possibility that the waveguide is being used too low in its current form.
Just worked it out on a bit of paper and the current 2.4k crossover is 8.5db on the woofer and 23db on the tweeter.
The modeled 4th order acoustic at 1.9k is actually 2nd order acoustic (11.5db woofer and tweeter to be exact)
The 2k crossover that hurt my ears was 12db on the woofer and 11db on the tweeter.
You still struggling with this one, Mr. Marley? 
Here's my thoughts on using an 8" woofer without time alignment:
You use it pretty much full range in 30-60Litre reflex or closed box. Series filters become too problematic here.
You then get something quite musical like the well-liked Tekton M-Lore which uses a similar Eminence 8" vented unit.
Use a filter like one of these with the attenuation adjusted. Should be OK. You will need some bafflestep, but my 5kHz notch is pretty neat, IMO. Lines up phase quite well. BTW, the proud bass unit is a FEATURE, not a fault.
Here's my thoughts on using an 8" woofer without time alignment:
You use it pretty much full range in 30-60Litre reflex or closed box. Series filters become too problematic here.
An externally hosted image should be here but it was not working when we last tested it.
You then get something quite musical like the well-liked Tekton M-Lore which uses a similar Eminence 8" vented unit.
Use a filter like one of these with the attenuation adjusted. Should be OK. You will need some bafflestep, but my 5kHz notch is pretty neat, IMO. Lines up phase quite well. BTW, the proud bass unit is a FEATURE, not a fault.
Attachments
You still struggling with this one, Mr. Marley?
Not you again
I'm happy with the sound as they are but just want to be sure they're sounding as good as they can, without changing the drivers.
I've modeled a 2nd order acoustic crossover the crosses at 2k and looks pretty good. The phase angle goes to +56 at around 500Hz IIRC (haven't got the laptop handy) but impedance is around 9ohms at that point, so I assume will be ok? The tweeter is down -50db at Fs, so lets see if the data sheet is correct (that they can handle 12db at 2k)...
Last edited:
I thought you said there was something making your ears bleed!
Very difficult problem, 8" plus 1" tweeter (or 1.5" for that matter...).
Learn of something new called a Half-Zobel. A full Zobel is too powerful on rolloff on an 8" paper with any tweeter currently made. I guess yours is about 15R plus 8uF on a 0.9mH woofer.
This circuit crosses over at about 2.5kHz, but relies on some time alignment to get decent phase. You need to set the acoustic centre of the tweeter back about 3cms somehow.
If you are close to this, that's about as good as it gets IMO. Low crossover is really just swapping one form of distortion for another. That's how it works.
Very difficult problem, 8" plus 1" tweeter (or 1.5" for that matter...).
Learn of something new called a Half-Zobel. A full Zobel is too powerful on rolloff on an 8" paper with any tweeter currently made. I guess yours is about 15R plus 8uF on a 0.9mH woofer.
This circuit crosses over at about 2.5kHz, but relies on some time alignment to get decent phase. You need to set the acoustic centre of the tweeter back about 3cms somehow.
If you are close to this, that's about as good as it gets IMO. Low crossover is really just swapping one form of distortion for another. That's how it works.
Attachments
I thought you said there was something making your ears bleed!
That was when I tried crossing over at 2k at 11db. If I could get 3rd order to work at 2k, maybe that would be worth trying?
Not sure what's wrong with me today. Current crossover is 8.5db on woofer and 14.5 on tweeter.
Second order is ideal really, because third order gives up in phase something below, wait for it, 2kHz.
Remember the wise words of Lynn Olson:
You can't beat the laws of Physics. 2.5kHz is about IT IMO with an 8" bass.
I think you're right, but I wont know for sure if I don't try.
The fundamental sound of the driver size, their baffling and directivity is something you have limited control over with the crossover. In this case the directivity will narrow, then widen and narrow again with increasing frequency. You can choose your compromises but probably not 'fix' it with the crossover.
This is over and above driver selection (eg. replacing with like types that handle a particular range better).
Although I don't know how much of an effect your particular waveguide is offering, horn like devices tend to have issues at their lower end such as a mouth reflection, with issues like 'honk', and a standard caution would be to avoid using them low enough where waves would pass through them as if they are not there due to their size.
This is over and above driver selection (eg. replacing with like types that handle a particular range better).
Although I don't know how much of an effect your particular waveguide is offering, horn like devices tend to have issues at their lower end such as a mouth reflection, with issues like 'honk', and a standard caution would be to avoid using them low enough where waves would pass through them as if they are not there due to their size.
And here's the frequency response with the current 2.4k crossover:
This is 0, 15, 30, 45, and i've included the 60 degree off axis response.
Last edited:
I took the speakers outside and hoisted them of the floor to try and get a good measurement of the lower frequencies. The bottom of the woofer was 2.4mtrs off the ground and according to an online floor bounce calculator - If I set the gating to about 10ms I should get accurate results down to 45Hz. Here's what I got:
Why have I got a 6db bass boost? I was expecting to see a flat'ish response or maybe +/- a couple of db but not +6db! They don't sound bass heavy and i'm sure if I removed the 6db bump they'd sound terrible.
Why have I got a 6db bass boost? I was expecting to see a flat'ish response or maybe +/- a couple of db but not +6db! They don't sound bass heavy and i'm sure if I removed the 6db bump they'd sound terrible.
Last edited:
It seems my concerns were probably greater than required. The point about the acoustic nature of the speaker may be food for thought in the future. In the meantime yourself and Steve seem to be nailing this one quite well.
Are you still noticing a harshness? With some speakers a slight downward tilt to the treble is required.
Are you still noticing a harshness? With some speakers a slight downward tilt to the treble is required.
There's not even a hint of harshness now and I think they sound great. Problem is, I have the bug and want to build more speakers...
I'm still confused by the 6db bass hump - and due to the lack of response I assume everyone who has looked doesn't know why it's there either. I've got a pair of Mission M70 and some Philips speakers that came with a mini system. If I measure them and they have a large bass hump too, then I guess it must be a problem with the mic or calibration file?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.