I managed to get rigged up for measuring these things. Had to get a fresh install of Win 10 to resurrect the old gifted laptop. I did a lot of reading on how to take the measurements and unfortunately I'm never going to get a good environment for doing so. But I took some near field and gated and smoothed to try to iron out potential reflections.
Also learned that I have no means to do the impedance tests and so I've ordered a DATS. I thought it would be easier for me than creating a rig that I don't really understand.
The mid range seemed to measure well. The tweeter really rolls off after 10khz. I ought to have disconnected the rear one and retested. Dont know whether that's a cancellation thing from the dipole/omni. Shall revisit that.
Bare drivers below. I did the sweep from 700hz for the tweeter and i think 70 for the mid. Maybe a mistake. S1 is the mid
Having said that loading these new curves into XSim with the values of crossover components I am using presently shows it not to be too horrendous.
The above schematic is exactly as I am listening to it now, ow it was before I measured anything. I add two 4 ohm tweeters to get me the correct impedance and therefore values for the hi pass. I mute one it because Xsim will sum the response of them assuming they are both front firing, which obviously they aren't. Now to play with the software and see what I should be aiming for.!
Also learned that I have no means to do the impedance tests and so I've ordered a DATS. I thought it would be easier for me than creating a rig that I don't really understand.
The mid range seemed to measure well. The tweeter really rolls off after 10khz. I ought to have disconnected the rear one and retested. Dont know whether that's a cancellation thing from the dipole/omni. Shall revisit that.
Bare drivers below. I did the sweep from 700hz for the tweeter and i think 70 for the mid. Maybe a mistake. S1 is the mid
Having said that loading these new curves into XSim with the values of crossover components I am using presently shows it not to be too horrendous.
The above schematic is exactly as I am listening to it now, ow it was before I measured anything. I add two 4 ohm tweeters to get me the correct impedance and therefore values for the hi pass. I mute one it because Xsim will sum the response of them assuming they are both front firing, which obviously they aren't. Now to play with the software and see what I should be aiming for.!
And a re hashed XO with the ZMA loaded in and an attempt at some baffle step compensation.
Only one tweeter shown as the files are for them both, and the zobel removed from the mid as the measured files had that in place during the sweeps . Ignore the 4.7uf shunt on the bass, it was there as a test zobel with the 0.47r as XRK has it on his design but it is 2nd order here so not needed I don't think . Should have removed it before the screenshot
Only one tweeter shown as the files are for them both, and the zobel removed from the mid as the measured files had that in place during the sweeps . Ignore the 4.7uf shunt on the bass, it was there as a test zobel with the 0.47r as XRK has it on his design but it is 2nd order here so not needed I don't think . Should have removed it before the screenshot
Last edited:
A bit of an update. Here is the latest Xsim layout of the crossover I was last listening to, and mighty fine they sounded too. L6 and C8 are a notch I shall out.
The reverse nulls aren't great however.
How does the phase look? Not too many wraps? I admit to not being too aware of the issues of phase.
So I took to Xsim once again and tried to get some better nulls. Also to try a 4th order on the tweeters as I'm pretty sure they protest at high levels. I also added a series/parallel thing over the series attenuating R for the tweeter and seemed to have lifted it right at the top a bit where it seemed to be rolling off. Not sure what this is called and dont quite understand where I get the 'extra' from?!
And the nulls now are much better
So i think I'll have a bit of a play with the XO again and I'm getting some better aircored inductors.
The reverse nulls aren't great however.
How does the phase look? Not too many wraps? I admit to not being too aware of the issues of phase.
So I took to Xsim once again and tried to get some better nulls. Also to try a 4th order on the tweeters as I'm pretty sure they protest at high levels. I also added a series/parallel thing over the series attenuating R for the tweeter and seemed to have lifted it right at the top a bit where it seemed to be rolling off. Not sure what this is called and dont quite understand where I get the 'extra' from?!
And the nulls now are much better
So i think I'll have a bit of a play with the XO again and I'm getting some better aircored inductors.
I also afforded myself some time......quite a lot actually to improve the aesthetics of the bare boxes. Won some Walnut veneer in eBay for a song so I set about popping my veneering cherry a few weeks ago.
Quite a pleasurable experience. I used contact adhesive. Seemed to go pretty well but I do have a couple bits that have lifted slightly and also leaving them in the sun after you've applied linseed oil to hasten the drying time is a bad idea. Slight shrinkage between 2 of the leaves. All hidden with some dark wax ! I was keen to get a really glossy finish but my desire to build the actual speakers overtook my desire for french polishing!
I got some proper isolation and had a couple of slabs of granite cut as bases. I added auxillary unfiltered outputs on the top of the cabinet. I've realised though I can't feed a 2 way ' top section ' from 2 wires though! Doh. Need to rethink that. Was trying to neaten any external wiring for SWMBO.!
And because I'm impatient and the tops will be a while yet until I best decide how to mount it all tidily, I added a cheap pair of those JVC speakers. I was surprised to find there is no filter in these at all. No wonder they get upset with anything approaching bass. And again due to impatience I just put a bipolar 16uf (4ohm speaker) on them to make them useable. They sound pretty good! Imaging is good.
More to follow on due course.
Any comments on the phase and my tweeter 'thing' much appreciated
Quite a pleasurable experience. I used contact adhesive. Seemed to go pretty well but I do have a couple bits that have lifted slightly and also leaving them in the sun after you've applied linseed oil to hasten the drying time is a bad idea. Slight shrinkage between 2 of the leaves. All hidden with some dark wax ! I was keen to get a really glossy finish but my desire to build the actual speakers overtook my desire for french polishing!
I got some proper isolation and had a couple of slabs of granite cut as bases. I added auxillary unfiltered outputs on the top of the cabinet. I've realised though I can't feed a 2 way ' top section ' from 2 wires though! Doh. Need to rethink that. Was trying to neaten any external wiring for SWMBO.!
And because I'm impatient and the tops will be a while yet until I best decide how to mount it all tidily, I added a cheap pair of those JVC speakers. I was surprised to find there is no filter in these at all. No wonder they get upset with anything approaching bass. And again due to impatience I just put a bipolar 16uf (4ohm speaker) on them to make them useable. They sound pretty good! Imaging is good.
More to follow on due course.
Any comments on the phase and my tweeter 'thing' much appreciated
Attachments
Finally some measurements!
Initial one proved something amiss and that was due to incorrect tweeter polarity. I have a lot to learn in REW so bare with me. The IR right window is something im never too sure where to set.
below with tweeter swap polarity. Still a bit of a lack of output between 2k and 8k so I shall address that. Maybe a combo of the mid lowpass and the tweeter.
Big nulls at about 100hz and 200hz. I assume this is the room and maybe floor bounce. They arent there when measured close to the woofer driver. See below;
And some off axis. These were quite surprising. For the better.
15 degrees (ish)!
45degrees
60 degrees
And the full 90 degrees off
I will learn to do overlays 'normalised' to make viewing this lot easier.
Initial one proved something amiss and that was due to incorrect tweeter polarity. I have a lot to learn in REW so bare with me. The IR right window is something im never too sure where to set.
below with tweeter swap polarity. Still a bit of a lack of output between 2k and 8k so I shall address that. Maybe a combo of the mid lowpass and the tweeter.
Big nulls at about 100hz and 200hz. I assume this is the room and maybe floor bounce. They arent there when measured close to the woofer driver. See below;
And some off axis. These were quite surprising. For the better.
15 degrees (ish)!
45degrees
60 degrees
And the full 90 degrees off
I will learn to do overlays 'normalised' to make viewing this lot easier.
Changed the series damping resistor from 7.5 to 5.6R, added 1.8uf in paralled over the existing 1.5uf at C5 (first pole of tweeter HP series C)
impedance sweep
impedance looks a bit funky but dont know if it matters.
Not sure if any of the sub 300Hz is believable as I'm not sure about gated measurements. This is in the room so most of the dips and crests are probably reflections?
impedance sweep
impedance looks a bit funky but dont know if it matters.
Not sure if any of the sub 300Hz is believable as I'm not sure about gated measurements. This is in the room so most of the dips and crests are probably reflections?