Everything looks to be the same between configurations aside from the phase. All the transient info with the linear phase filters just sounds smoothed out, like nothing really has an attack. Imaging feels better with linear phase.
FWIW my delays are 0.12ms for the tweeters and 0.10 for the mid driver. I admit I'm not sure what the best method for calculating delays is, I just converted distance to time and watched the RTA and did sweeps to make sure there weren't issues showing up.
I really gotta grab a DSP unit, the processing delay and not having access to my interfaces ASIO driver is basically killing my ability to write. Seems like kind of a struggle to find a unit that will give me the performance I get running filters in the computer. Lots of noise complaints.
I really gotta grab a DSP unit, the processing delay and not having access to my interfaces ASIO driver is basically killing my ability to write. Seems like kind of a struggle to find a unit that will give me the performance I get running filters in the computer. Lots of noise complaints.
Last edited:
I played with some different filters and ran into some stuff I don't really know to interpret.
I have three filter sets here, LR24 Min phase, LR24 Min Phase, and 48db Min phase. They are all rehpase generated filters with the exact same delay of 16,384 samples. The driver delay settings are identical across them all.
I noticed different responses on axis with all the filters. I expected that with 24vs 48db filters, but not the two 24db filters which would be minimum vs. linear phase.
(also rew's stock image export settings are about as bad as this forums quote system)
Here is impulse response, no idea what to make of it.
Here are the different phase measurements.
I have three filter sets here, LR24 Min phase, LR24 Min Phase, and 48db Min phase. They are all rehpase generated filters with the exact same delay of 16,384 samples. The driver delay settings are identical across them all.
I noticed different responses on axis with all the filters. I expected that with 24vs 48db filters, but not the two 24db filters which would be minimum vs. linear phase.
(also rew's stock image export settings are about as bad as this forums quote system)
Here is impulse response, no idea what to make of it.
Here are the different phase measurements.
Thank you, it was useful to read your comment. I'm still learning about filters, but I was intrigued by your observation of the different axis responses with all the filters, even with two 24dB filters.
No problem, although not sure what value there is there though. I didn't really present the impulse info right. Surprised I didn't get any comments on any of it though. I did make a typo, the yellow trace is linear phase not minimum.
This projects probably not the most interesting to update, but I did end up testing the DX25 tweeter after routing the appropriate terminal holes. I had them lying around so why not. I think out of the three I tried on this speaker I prefer these the most.
I did a rudimentary dispersion test but simply sitting on axis to one speaker and moving off, keeping a keen ear on when the tweeters tonality changed. I wanted to see which tweeter to my ear, cast the widest sound. I think I did this because I was bored and I find dispersion to be one of the more confusing aspects to correlate what the data says with my ears and seem to have a lot of difficulty getting a stable image and some aspects of stereo, or just my speakers bothers me. For instance I find the fact that moving to the left or right tends to make one speaker immediately dominant.
I've been pretty curious about tweeter loading for awhile, and there was a neat thread at ASR showing how different plate terminations lead to different effects.
D27 Peerless - Terrible, seemed to totally disappear once you were 30 degrees off. Dome is loaded deep so that makes sense to me somehow.
ND25fw Dayton - Very smooth transitions from on to off, disappears around 45 degrees
DX25 Peerless - Not as smooth transition but still audible very far off axis.
Another user here suggested the DX25 and D27 sound similar and I'm gonna disagree there, totally different IME.
I did a rudimentary dispersion test but simply sitting on axis to one speaker and moving off, keeping a keen ear on when the tweeters tonality changed. I wanted to see which tweeter to my ear, cast the widest sound. I think I did this because I was bored and I find dispersion to be one of the more confusing aspects to correlate what the data says with my ears and seem to have a lot of difficulty getting a stable image and some aspects of stereo, or just my speakers bothers me. For instance I find the fact that moving to the left or right tends to make one speaker immediately dominant.
I've been pretty curious about tweeter loading for awhile, and there was a neat thread at ASR showing how different plate terminations lead to different effects.
D27 Peerless - Terrible, seemed to totally disappear once you were 30 degrees off. Dome is loaded deep so that makes sense to me somehow.
ND25fw Dayton - Very smooth transitions from on to off, disappears around 45 degrees
DX25 Peerless - Not as smooth transition but still audible very far off axis.
Another user here suggested the DX25 and D27 sound similar and I'm gonna disagree there, totally different IME.
I got VBS 10.2 up and running, much better than my 3 way in everything but bass. Gonna take the woofer from the 3 way and make a new box with matching dimensions to the VBS 10.2 and use them as sealed subs. That controlled directivity really is something, incredible speakers so far. Guess that's the end of the story for the 3 way speakers.
Your impulse Y axis being % is good, but your X axis resolution should be msec not usec as per. You're looking primarily for the first reflection which you can see in my measurement at 4.5msec and gate that out
Honeymoon phase with VBS ended and honestly, I think they might be a bit too narrow for my tastes. There's a distinct lack of expansion outward that my brain expects that doesn't happen. I hooked my 3 ways back up and was happy to hear that mid range expansion again. Although, my xover is just not there. I'm not going about it right atm, I wanted to get out and get a better set of data, both for left and right side of the speaker. I'll have to wait until winter passes to do that.
I'm currently just running first order filter for the mid to tweet, 2nd for the woofer to mid. Mostly doing this because currently it's the only thing that sounds good to me, which says I'm probably not getting phase right or understanding it. All my combo's of 2nd and 4th order and time delays have given me a phasey sound in some manner, like they're close but something is always off. I can never get the mid transition measurements to look as good as the first order filters either. I know first order are ill advised but it seems to measure well and sound good, and distortion doesn't even show up at the peak levels I use the speakers at. I mean if there any issue there? DC130 might be a bit bright around 4k though. How does one get such coherent sound of these first order filters with higher order?🙂
I'm currently just running first order filter for the mid to tweet, 2nd for the woofer to mid. Mostly doing this because currently it's the only thing that sounds good to me, which says I'm probably not getting phase right or understanding it. All my combo's of 2nd and 4th order and time delays have given me a phasey sound in some manner, like they're close but something is always off. I can never get the mid transition measurements to look as good as the first order filters either. I know first order are ill advised but it seems to measure well and sound good, and distortion doesn't even show up at the peak levels I use the speakers at. I mean if there any issue there? DC130 might be a bit bright around 4k though. How does one get such coherent sound of these first order filters with higher order?🙂
Last edited:
I've been wanting to rebuild these, if anything because the construction issues drive me insane to look at, and I got some really nice veneer that I'd rather not waste on something that isn't at least square. I think I can do better work.
I wanted to introduce some simplicity in my life and just run the speaker passive. I'd like to simple match the original Pit Viper speakers dimensions and driver placement but make the speaker shorter but I'm wondering if the xover would need changed. When I attempted to use the Pit viper xover I had a big dip in the 400hz range. I wasn't sure if this was due to the offset drivers or making the speaker short. The original pit viper is 15.5"x 34" and mine was like 14"x27" or something. Do you guys think I would run into the same problem again if I shortened the cabinet height?
I wanted to introduce some simplicity in my life and just run the speaker passive. I'd like to simple match the original Pit Viper speakers dimensions and driver placement but make the speaker shorter but I'm wondering if the xover would need changed. When I attempted to use the Pit viper xover I had a big dip in the 400hz range. I wasn't sure if this was due to the offset drivers or making the speaker short. The original pit viper is 15.5"x 34" and mine was like 14"x27" or something. Do you guys think I would run into the same problem again if I shortened the cabinet height?
The fear that the axial response varies with the details of the baffle arrangement and makes any new crossover potentially impossible, is a little toward the extreme end of things (if you take my meaning). Baffles tend to average more than that. Issues might even be attended to in global EQ.
The correct way to reduce the interference of issues like this is to apply acoustic measures. The simplest one, whenever you're not sure, is to round the edges.
The correct way to reduce the interference of issues like this is to apply acoustic measures. The simplest one, whenever you're not sure, is to round the edges.
I would tend to agree but then again, no one has figured out why my speakers had such a weird low mid response, I mean that high pass for the mid looks totally wrong to me. Are you suggesting rounding the edges would bring that low mid back because I don't think it would.
I'm not saying much, I just don't believe there's much reason not to make small variations in the baffle based on what might happen in the crossover.
Rounding the edges is not an indepth acoustic goal in itself, it's just something you do when you have nothing left to do.
What do you believe your acoustic issues have to do with your crossover, putting aside the EQ aspect of that crossover?
Rounding the edges is not an indepth acoustic goal in itself, it's just something you do when you have nothing left to do.
What do you believe your acoustic issues have to do with your crossover, putting aside the EQ aspect of that crossover?
Not my xover, the Pit Viper stock xover.
I'm not sure what you're asking, the speaker with that xover shouldn't be exhibiting that low mid dip with the passive filters. I'm trying to figure out why that was there. Baffle dimension changes were one idea. I'm throwing stuff out since no one seems to know why it's there.
The dip at 400Hz is partially acoustic in nature, but it appears it can be partially compensated using EQ.
Let's draw a line between the cross itself, and the EQ which is done directly by the crossover. The EQ can be taken over by global EQ and this is where I'd begin. Until you know the answer, there's not much point messing with the crossover itself.
I would try EQing as much as possible using care not to push too far. Your other tools include investigating the room interaction, and adjusting the part of the cross which is not the EQ, ie the driver sharing part.
Let's draw a line between the cross itself, and the EQ which is done directly by the crossover. The EQ can be taken over by global EQ and this is where I'd begin. Until you know the answer, there's not much point messing with the crossover itself.
I would try EQing as much as possible using care not to push too far. Your other tools include investigating the room interaction, and adjusting the part of the cross which is not the EQ, ie the driver sharing part.
So the answer is really "you're on your own, good luck, no one really knows why the dip is there".
Welp.
FWIW those are outdoor measurements and it's there indoor as well just as intense. The amount of EQ required to bring it back in is excessive. I'll give someone $50 if they can give me a single concrete answer about ANY issue I've stated in this thread with these speakers.
Welp.
FWIW those are outdoor measurements and it's there indoor as well just as intense. The amount of EQ required to bring it back in is excessive. I'll give someone $50 if they can give me a single concrete answer about ANY issue I've stated in this thread with these speakers.
Of course not, but you do have to look around a little if you want to find the exact answers.
Do you want the easy way? The easy way is to begin by EQing all that..
What is the problem with EQ?
I'd prefer to get the speaker right at the xover first.
Another user has sim'd the speaker and suggest the dip is intentional, possibly to alleviate floor reflections. He posted a modified xover that appears to resolve the dip. It's a very cheap mod so I may try that and report back.
Another user has sim'd the speaker and suggest the dip is intentional, possibly to alleviate floor reflections. He posted a modified xover that appears to resolve the dip. It's a very cheap mod so I may try that and report back.