rplace Posted January 17, 2021 Author Share Posted January 17, 2021 12 hours ago, The Dude said: Any ideas to try? If group delay jumps up at that/those same places try inverting phase. I'm no expert but I think #3 is the best if we only consider the three visually Quote Link to comment Share on other sites More sharing options...
Chris A Posted January 17, 2021 Share Posted January 17, 2021 Try adding an additional 1ms delay to the HF channel... Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 17, 2021 Share Posted January 17, 2021 3 hours ago, Chris A said: Try adding an additional 1ms delay to the HF channel... The hypex software only allows about 10 us feet (think that the unit of measurement). The third one had a inverted polarity with 36 mm of delay or 107 U.S., Before adding your recommended delay showing 5k and above at dang near dead Zero. Here is one with adding 10 U.S. or 4 mm with invert checked Then un inverting Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 17, 2021 Share Posted January 17, 2021 9 hours ago, rplace said: If group delay jumps up at that/those same places try inverting phase. Group delays before Chris's 1 ms recommendation After Chris's recommendation 9 hours ago, rplace said: I'm no expert but I think #3 is the best if we only consider the three visually My thoughts as well Quote Link to comment Share on other sites More sharing options...
rplace Posted January 18, 2021 Author Share Posted January 18, 2021 You need to really zoom in on the group delay. +/- 10ms or less...like 5 maybe. Go back and look at some of my posts. You want everything above 200 to be under 2ms. Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 18, 2021 Share Posted January 18, 2021 15 hours ago, rplace said: You need to really zoom in on the group delay. +/- 10ms or less...like 5 maybe. Go back and look at some of my posts. You want everything above 200 to be under 2ms. Approaching everything one at a time has been very helpful to understand how everyone has dialed in REW. This all is starting to make a little more since. Now to figure out the correct equation. Unfortunately i was hearing some obvious distortion in higher frequencies from one of the drivers. Quote Link to comment Share on other sites More sharing options...
rplace Posted January 18, 2021 Author Share Posted January 18, 2021 I don't really know why. I just know to try. If you xo point is around that "problem" 400-700. Invert phase or add some delay. You also need to add minimum phase to the GD plot to help figure delay. 1 Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 23, 2021 Share Posted January 23, 2021 Wow what a difference a new diaphragm does. There still some more work to do, but if we are attempting under 2 m, then that is pretty close to me. Quote Link to comment Share on other sites More sharing options...
rplace Posted January 23, 2021 Author Share Posted January 23, 2021 Group delay looks good. Try 1ms of delay to get rid of that sharp cut to the right on the spectrogram at 802. Add minimum phase to GD plot. I'll look for the "how to" instructions if you don't know how or hopefully @Chris A will chime in. Quote Link to comment Share on other sites More sharing options...
Chris A Posted January 23, 2021 Share Posted January 23, 2021 1 hour ago, rplace said: Try 1ms of delay to get rid of that sharp cut to the right on the spectrogram at 802. Dude, If you could move the HF section back (or the LF forward) 13.5 inches relative to each other, that would help to solve your delay problem. I know that you said that you were out of delay with the FusionAmps, but physical delay of the HF section could help you get there. What are you using in terms of crossover filter order? If you're using something like 4th order or higher, I'd recommend using first order filters instead. That will get you most of the needed delay back. What kind of bass bin section are you using that has more than 10 feet of delay at 800 Hz? That's about double the delay of the K-horn bass bin, and the K-horn bass bin only goes up to ~400 Hz...pushing it, at that. Chris 1 Quote Link to comment Share on other sites More sharing options...
rplace Posted January 23, 2021 Author Share Posted January 23, 2021 38 minutes ago, Chris A said: I know that you said that you were out of delay with the FusionAmps Did not realize that. Was I right in theory how to get rid of that sharp cut to the right? Quote Link to comment Share on other sites More sharing options...
Chris A Posted January 23, 2021 Share Posted January 23, 2021 Probably. Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 23, 2021 Share Posted January 23, 2021 Sorry if I miss worded it, but I am not out of delay (at least I don't think I am), I'm just not able to fine tune it to a 1m as it appears hypex only allows 3m increments (I think). I can say the lost I can add is 11 U.S., which I assume is feet. Oh now think of it, I have convert meters/feet to speed of sound as time, is that correct?Chris, I am using a diy Belle bass bin. I've tried crossover at 1st, 2nd and 4th order at around 420, 450, and I believe 500. I believe it is 2nd order 420 hp and 2nd order 470 lp, but will need to double check.Sent from my SM-G986U using Tapatalk Quote Link to comment Share on other sites More sharing options...
Chris A Posted January 23, 2021 Share Posted January 23, 2021 I think that my FusionAmp FA122s provided microsecond-level delay adjustment (10 -3 milliseconds), so I think you can just add another 1 ms of delay to what you've got on the HF channels and take another measurement. The FusionAmp manual says they have 19.2 milliseconds of total delay per channel. Chris 2 Quote Link to comment Share on other sites More sharing options...
Khornukopia Posted January 23, 2021 Share Posted January 23, 2021 4 hours ago, The Dude said: I'm just not able to fine tune it to a 1m as it appears hypex only allows 3m increments (I think). I am not familiar with the Hypex system, so I don't have any helpful advice about that at this time, but it seems that mixing feet, meters, microseconds and milliseconds may lead to some confusion. I would pick just one for all the calculations and settings. 1 Quote Link to comment Share on other sites More sharing options...
Khornukopia Posted January 23, 2021 Share Posted January 23, 2021 At one time here on the forum there was a JuBelle project with a K-510 horn that listed a time delay of 1.9 milliseconds. That could be a good reference point. 1 Quote Link to comment Share on other sites More sharing options...
PrestonTom Posted January 24, 2021 Share Posted January 24, 2021 4 hours ago, Khornukopia said: At one time here on the forum there was a JuBelle project with a K-510 horn that listed a time delay of 1.9 milliseconds. That could be a good reference point. The time alignment is one of the more important issues. There is no reason to guess. HolmImpulse, REW, and other packages (and these are free) will give you a reasonably precise value. Yes, you do need a microphone, but these are not expensive (and certainly worth the investment). In terms of the phase spectrum: time alignment and a driver wired out of phase are big issues and easily correctible. 1 Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 24, 2021 Share Posted January 24, 2021 According to Google 1 MM = 1MS. If so, than 11 U.S. = 3.8 MM or 3.8 MS of delay. I also noted I incorrectly designed my crossover as I was supposed to design it with a q adjustment of .71. This has now been changed as you can see, my shift is further down to the crossover point instead of the 700-800 hz region. Some settings I found along the way for the K402/jubscala settings. Current crossover should be 4th order LR lowpass at 420, .71 q and Highpass 470, .71q. There is a slight bump between 1k and 1.6k. Sent from my SM-G986U using Tapatalk Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 24, 2021 Share Posted January 24, 2021 Now what should be considered 2nd order with lp at 450 with q of .5 and hp at 450 with q of .5. There always seems to be that shift around 700-800. Now knowing that the crossover was 4th order at the 420/470. I am going to give that a try next. Chris had mentioned 1st order, is the a Butterworth versus a LR? Quote Link to comment Share on other sites More sharing options...
The Dude Posted January 24, 2021 Share Posted January 24, 2021 Here is the 420/470, note not much change other than the shift that used to be at the mid/crossover point is now further down the line. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.