Hi,
As i have both kind ( wide/narrow baffle) i can tell there is a difference but i'm not sure the pinpoint imaging is better than a more diffuse one:
with my own recording i never encountered such pinpoint imaging from 'true stereo' technique ( pair of mic even with XY).
In that sense wide baffle are not untrue to the image rendering. It works great for classical and 'classic' rock and roll ( Led Zepelin, Hendrix, things like that). But the 3 ways have other interesting pros which make them useful tool for other things with different styles too ( mid is cleaner, bass authority,...).
And i must say i've heard 'pinpoint' from inwalls... so very very wide baffle... so i wonder if it's not a diffraction issue more than a width issue?
Pinpoint can be nice feeling for multi miced and close to source recording as well as electronic music.
Maybe you had issue with location in your room too: my wide baffle always had been more touchy about location and require more step back to sum coherently in contrast to eg: 6,5"+1".
As i have both kind ( wide/narrow baffle) i can tell there is a difference but i'm not sure the pinpoint imaging is better than a more diffuse one:
with my own recording i never encountered such pinpoint imaging from 'true stereo' technique ( pair of mic even with XY).
In that sense wide baffle are not untrue to the image rendering. It works great for classical and 'classic' rock and roll ( Led Zepelin, Hendrix, things like that). But the 3 ways have other interesting pros which make them useful tool for other things with different styles too ( mid is cleaner, bass authority,...).
And i must say i've heard 'pinpoint' from inwalls... so very very wide baffle... so i wonder if it's not a diffraction issue more than a width issue?
Pinpoint can be nice feeling for multi miced and close to source recording as well as electronic music.
Maybe you had issue with location in your room too: my wide baffle always had been more touchy about location and require more step back to sum coherently in contrast to eg: 6,5"+1".
Last edited:
Yeah, just 5.1 outs from motherboard into amps, xover done on PC. Amiga is passive, it's just the kit. I actually won a little contest at PE for my Amiga's. First thing I ever veneered, first DIY speaker. They have a lot of sentimental value.
My active 3 way is honestly a hassle and I'm always scared the 3.5mm plugs will just slide out, they have once and sent loud buss into the mids. I feel like DIY active 3 way is not friendly to my income. I'd like to make a passive xover but my attempts have been poor and it seems like the speaker has other issues. I'd argue my complaints aren't so much for imaging, I can hear where things are, but the sound stage feels flat, not much depth to it. The Amigas comparatively have a wide stage that floats about in the air, in the song Kenotaph by Ben Lukas Boysen I feel like I can hear where he is hitting the snare in 3d space, there is left and right as well as front to back.
Last edited:
Nice prize!
One can't really compare speaker differences unless the amp is the same and passive vs passive.
I haven't heard the newer models but when I did hear the early MiniDSP products, I thought the speakers they designed to show them off sounded flat, lifeless, boring and dead, and the imaging was non-existent compared to analog stuffs.
In the OSMC thread, the team used DSP first but ultimately went with passive cause it sounded better.
One can't really compare speaker differences unless the amp is the same and passive vs passive.
I haven't heard the newer models but when I did hear the early MiniDSP products, I thought the speakers they designed to show them off sounded flat, lifeless, boring and dead, and the imaging was non-existent compared to analog stuffs.
In the OSMC thread, the team used DSP first but ultimately went with passive cause it sounded better.
That's how minidsp sounded to me as well, which is why I do it all in the computer. This active setup also prevents me from plugging in my audio interfaces and recording which is a huge bummer.
Your Amiga's looks gorgeous ( and your three ways are not bad looking either).
My large baffle three ways definitely doesn't lack depth of scene. The mid driver's volume is fully loaded with absorbent material ( looks a lot like denim ultratouch material but it wasn't it as the loudspeakers are older...) and i'm sure it play a role with mine about depth.
My large baffle three ways definitely doesn't lack depth of scene. The mid driver's volume is fully loaded with absorbent material ( looks a lot like denim ultratouch material but it wasn't it as the loudspeakers are older...) and i'm sure it play a role with mine about depth.
I just have some polyfill in the mid, and the walls are lined with some sort of thing yoga mat stuff. I did have a bmr tebm46 tossed on some cardboard mounted in there and honestly thought that sounded better than the dc130. I thought about having friend 3dprint a waveguide/flare for that mid but that's another can of worms. Perhaps this mid is not the best, idk.
Huh, that ultratouch denim looks like I might need to grab some to use. I hate working with fibreglass.
Huh, that ultratouch denim looks like I might need to grab some to use. I hate working with fibreglass.
I don't really know Ultratouch specificaly - we have 'Isolant Metisse' in here... but if it is same product i know ( which i believe it is) it works wonder for mid/high but can be a bit too much dense for low end ( it's too much packed, not enough loose).
Otherwise yes it is another world than fiberglass: no hitching, no lung concern, recycled material...
Otherwise yes it is another world than fiberglass: no hitching, no lung concern, recycled material...
I'd love to quit this project, but I keep hoping it can be good.
Here is impedance of all drivers. I'm going to try and get some good outside measurements of the driver tomorrow. Other than that not really sure what course to take. I have them in another room so maybe that can help rule out if there's a bad resonance. Curiously the OSMC impedance plot for the woofer looks a lot worse than mine. These are from the cabinet with new braces, pretty much no change from before on this measurement it seems.
Here is impedance of all drivers. I'm going to try and get some good outside measurements of the driver tomorrow. Other than that not really sure what course to take. I have them in another room so maybe that can help rule out if there's a bad resonance. Curiously the OSMC impedance plot for the woofer looks a lot worse than mine. These are from the cabinet with new braces, pretty much no change from before on this measurement it seems.
Last edited:
If it was me, I'd rebuild the cabs. Copy OSMC aluminum corner brackets + bracing and damping plans. Consider two shorter front ports like Harbeth and Spendor + center align the drivers or rear port like OSMC. Work on passive crossover with people here 😀
A lot of work.
A lot of work.
If you can get electrical sweeps at each driver's terminals, at a consistent level, then someone could develop a copy of your existing crossover.
Most of us don't have good listening spaces [ in my case it's WAF and conflicts with "decor"] so the room does seem to be the defining factor in how speakers sound.
My loungeroom system has wide baffles and imaging is good at the sweet spot and lucky for us the sweet spot is reasonably wide
My loungeroom system has wide baffles and imaging is good at the sweet spot and lucky for us the sweet spot is reasonably wide
Attachments
Allen I'm not sure how to do that. I can get outside data and close woofer. one thing I ran into though is the mid to tweeter not translating in vcad. when it's flat in vcad I get a dip at xover point in real life. curiously when I modeled the pit viper crossover in vcad with my data, i got a dip but in real life there is no dip. I'm not sure why this disprency happens or if my frd measuremnts can be trusted.
Ok, then if your current active version of crossover is not satisfactory then there's no sense copying it into a passive version. I'd be happy to talk about the procedure, or to help when you decide to make the move.
Regarding your phase discrepancy I'd presume to look at the delay for each driver. Perhaps you can tweak your crossover delays back and forth to see?
Errors in delay could be produced by a USB mic, the measurement technique, incorrect use of REW (or other measurement program), deliberate incorrect but well intentioned adjustment in REW, using the simulator to treat a non-minimum phase curve as minimum phase, altering the distance settings in VC etc. etc.
Don't be put off by that list, getting phase correct shouldn't be difficult. The frd file is capable of properly holding the delay information.
Regarding your phase discrepancy I'd presume to look at the delay for each driver. Perhaps you can tweak your crossover delays back and forth to see?
Errors in delay could be produced by a USB mic, the measurement technique, incorrect use of REW (or other measurement program), deliberate incorrect but well intentioned adjustment in REW, using the simulator to treat a non-minimum phase curve as minimum phase, altering the distance settings in VC etc. etc.
Don't be put off by that list, getting phase correct shouldn't be difficult. The frd file is capable of properly holding the delay information.
Instead of measuring with a mic, you measure the electrical signal.
By simply plugging the RCAs of the crossover back into the sound card, REW will see the filter responses which we can use.
Some would prefer to measure at the speaker terminals directly, since this is the most relevant place for this. This is because it includes the various differences between the amps, such as level. (When you measure at the output of the amp you want to keep the level low enough for your sound card and check for DC offset of the amp and a common ground.)
You can also subtract your per driver filtered response from the raw in cabinet response (assuming volume and mic positions are the same for both). That should give you the electrical transfer function in software. I remember doing this in SpeakerWorkshop years ago. Same should be possible in REW etc...