|
|
Post by thompson12 on Mar 20, 2023 19:37:11 GMT -5
 Both surrounds, Emotiva ERD-1 surround speakers, not sure whats going on with the peaks and wide dips Could this have anything to do with the bypole/dipole switches on the backs? Mitch |
|
|
|
Post by thompson12 on Mar 20, 2023 19:41:57 GMT -5
 They all look normal but the surrounds. Mitch |
|
|
|
Post by marcl on Mar 20, 2023 19:46:36 GMT -5
View AttachmentBoth surrounds, Emotiva ERD-1 surround speakers, not sure whats going on with the peaks and wide dips Could this have anything to do with the bypole/dipole switches on the backs? Mitch This is good. To start with, both speakers are negative polarity. So you'll want to compare them to all your other speakers. If they're all negative, that's fine. But they need to be the same. If the other speakers are positive, then you should switch the wires on the surrounds. As for bipole/dipole. Yes that could have something to do with what you see. The impulse response might be seeing odd reflections. Are they both set the same ... either bipole or dipole? |
|
|
|
Post by marcl on Mar 20, 2023 19:51:26 GMT -5
It kind of looks like they are all the same polarity, but the surrounds have extra peaks. Is that the way you see it? |
|
ttocs
Global Moderator 
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,146
|
Post by ttocs on Mar 20, 2023 20:00:33 GMT -5
It kind of looks like they are all the same polarity, but the surrounds have extra peaks. Is that the way you see it? The RS (Right Surround I assume) has less intensity on its first peak and 100% on the second peak. Stronger reflection? |
|
|
|
Post by marcl on Mar 20, 2023 20:10:24 GMT -5
It kind of looks like they are all the same polarity, but the surrounds have extra peaks. Is that the way you see it? The RS (Right Surround I assume) has less intensity on its first peak and 100% on the second peak. Stronger reflection? Maybe one is set to dipole and the other bipole. If they are adjacent to MLP I guess I'd say they should be dipole, but the out of phase wave could make for a weird impulse. In any case ... both the same switch setting and all speakers same polarity and aligned on the first pulse. |
|
|
|
Post by thompson12 on Mar 20, 2023 20:16:18 GMT -5
View AttachmentBoth surrounds, Emotiva ERD-1 surround speakers, not sure whats going on with the peaks and wide dips Could this have anything to do with the bypole/dipole switches on the backs? Mitch This is good. To start with, both speakers are negative polarity. So you'll want to compare them to all your other speakers. If they're all negative, that's fine. But they need to be the same. If the other speakers are positive, then you should switch the wires on the surrounds. As for bipole/dipole. Yes that could have something to do with what you see. The impulse response might be seeing odd reflections. Are they both set the same ... either bipole or dipole? They are all reversed polarity the switches are set the same I don't remember what settings. Mitch |
|
|
|
Post by thompson12 on Mar 20, 2023 20:19:48 GMT -5
It kind of looks like they are all the same polarity, but the surrounds have extra peaks. Is that the way you see it? The RS (Right Surround I assume) has less intensity on its first peak and 100% on the second peak. Stronger reflection? that's what I was seeing I was wondering what's causing the peaks to be like that. Mitch |
|
|
|
Post by thompson12 on Mar 20, 2023 20:29:41 GMT -5
 I just checked the subs quick the red and blue are revel 15"subs and the brown is a SVS 12" the other svs 12" would not power up for some reason maybe a fuse I hope. They line up good for no adjustments but the SVS has the fist peak low as well. Mitch |
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,146
|
Post by ttocs on Mar 20, 2023 20:44:54 GMT -5
They line up good for no adjustments but the SVS has the fist peak low as well. Mitch But this one looks more normal. By changing Phase, that relationship with the first and second peaks will change, altering the size of each. In that earlier image there is a lot of "distance" between the minor and major peaks vs being consecutive cycles. |
|
|
|
Post by thompson12 on Mar 20, 2023 20:56:25 GMT -5
Maybe this weekend I'll take the EDR's down and try different setting.
And I think I'll sum the subs again the way that OCA guy dose it for flatter response instead of summing them for maximum gain and see how that works out.
Mitch
|
|
|
|
Post by thompson12 on Mar 20, 2023 21:00:48 GMT -5
They line up good for no adjustments but the SVS has the fist peak low as well. Mitch But this one looks more normal. By changing Phase, that relationship with the first and second peaks will change, altering the size of each. In that earlier image there is a lot of "distance" between the minor and major peaks vs being consecutive cycles. I'm gonna watch this video a couple times and try it and see how it works out. m.youtube.com/watch?v=ga2eOwJRtXoMitch |
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,146
|
Post by ttocs on Mar 20, 2023 22:18:47 GMT -5
Here's a little bit of fun to show what happens with Phase change on a sub with variable Phase adjustment. It shows Delay in action, but also notice the flip in polarity when the main 100% peak goes from positive to negative. And then there's the pretty Spirograph to look at.  |
|
|
|
Post by marcl on Mar 21, 2023 9:35:03 GMT -5
But this one looks more normal. By changing Phase, that relationship with the first and second peaks will change, altering the size of each. In that earlier image there is a lot of "distance" between the minor and major peaks vs being consecutive cycles. I'm gonna watch this video a couple times and try it and see how it works out. m.youtube.com/watch?v=ga2eOwJRtXoMitch + ttocs I just watched the video and it looked brilliant. Wow, look at those features ... now I understand, very simple. UNTIL ... the result was he needed to dial in -5.xxx delay to subwoofer 2! He went all the way through optimizing everything and never talked about the fact that you can't put in a negative value for delay ... you can't have the sub start playing before the sound gets there. The only way to accomplish this would be to physically move subwoofer 2 five feet closer to the MLP. Am I missing something? |
|
KeithL
Administrator
Posts: 10,256
|
Post by KeithL on Mar 21, 2023 9:49:09 GMT -5
It's also worth mentioning that the "phase adjustment" is going to act quite differently on different subs. For example, if a sub has an internal DSP processor, there is the possibility of applying a pure physical delay to the signal. But the designers may have chosen not to do it that way. However, if the sub has analog crossover circuitry, or no internal crossover, then it is using some sort of analog phase shift network. And there are lots of different types of analog phase shift networks - which will each have different effects on phase and delay. However, with virtually all of the commonly used variations, the amount of phase shift will vary with frequency. (The general expectation is that most of them will be used "to adjust the sub to be in phase with your main speakers at the crossover point".) Remember that a "pure time delay", which would be ideal for adjusting for distance, will have different phase shifts at different frequencies. Whereas a "pure phase shift", which shifts all frequencies by the same amount of phase angle, would result in different amounts of delay at different frequencies. (A specific time delay - in milliseconds, only corresponds to a specific phase shift - in degrees, at a single specific frequency.) And, of course, on most subs, the "Phase" control is usually somewhere in-between. (My point is simply that you probably shouldn't assume exactly what "adjusting the phase on a sub" is really going to do until you measure it.) (My second point is that, in this regard, every sub is going to act differently, often very differently, so you shouldn't necessarily expect even similar results with different subs.) They line up good for no adjustments but the SVS has the fist peak low as well. Mitch But this one looks more normal. By changing Phase, that relationship with the first and second peaks will change, altering the size of each. In that earlier image there is a lot of "distance" between the minor and major peaks vs being consecutive cycles. |
|
KeithL
Administrator
Posts: 10,256
|
Post by KeithL on Mar 21, 2023 9:54:42 GMT -5
I didn't watch the video... But your comment reminded me of a short science fiction story I read a very long time ago... It was about a fellow who discovered an adjustable electronic delay line in his fancy new radio... He figured out that he could listen to news broadcasts of the future by setting the value on the delay line to less than zero...   + ttocs I just watched the video and it looked brilliant. Wow, look at those features ... now I understand, very simple. UNTIL ... the result was he needed to dial in -5.xxx delay to subwoofer 2! He went all the way through optimizing everything and never talked about the fact that you can't put in a negative value for delay ... you can't have the sub start playing before the sound gets there. The only way to accomplish this would be to physically move subwoofer 2 five feet closer to the MLP. Am I missing something? |
|
|
|
Post by marcl on Mar 21, 2023 10:09:35 GMT -5
It's also worth mentioning that the "phase adjustment" is going to act quite differently on different subs. For example, if a sub has an internal DSP processor, there is the possibility of applying a pure physical delay to the signal. But the designers may have chosen not to do it that way. However, if the sub has analog crossover circuitry, or no internal crossover, then it is using some sort of analog phase shift network. And there are lots of different types of analog phase shift networks - which will each have different effects on phase and delay. However, with virtually all of the commonly used variations, the amount of phase shift will vary with frequency. (The general expectation is that most of them will be used "to adjust the sub to be in phase with your main speakers at the crossover point".) Remember that a "pure time delay", which would be ideal for adjusting for distance, will have different phase shifts at different frequencies. Whereas a "pure phase shift", which shifts all frequencies by the same amount of phase angle, would result in different amounts of delay at different frequencies. (A specific time delay - in milliseconds, only corresponds to a specific phase shift - in degrees, at a single specific frequency.) And, of course, on most subs, the "Phase" control is usually somewhere in-between. (My point is simply that you probably shouldn't assume exactly what "adjusting the phase on a sub" is really going to do until you measure it.) (My second point is that, in this regard, every sub is going to act differently, often very differently, so you shouldn't necessarily expect even similar results with different subs.) But this one looks more normal. By changing Phase, that relationship with the first and second peaks will change, altering the size of each. In that earlier image there is a lot of "distance" between the minor and major peaks vs being consecutive cycles. I was wondering how some subs do a variable phase without being digital. In this video, though, he's only talking about time delay applied either in a miniDSP or in a processor where the sub distances can be entered separately. There are three parts to the video ... first is taking multiple sub measurements around the MLP and using REW Cross Correlation Alignment and Vector Average to get a new version of each sub's response using the multiple measurements; second is to use the Alignment Tool to recursively Align Phase Slope at Cursor until the flattest combined response is achieved and the delay values are determined; third part is using REW EQ to apply filters to finish the job. The first two parts would be useful and then we could use Dirac to do the filtering. He says way down in the comments if the value comes out negative, just apply it to the other sub ... but it's a little more complicated than that. Still, the Cross Correlation Alignment and Vector Average as a better starting point for each sub will be interesting to try. |
|
KeithL
Administrator
Posts: 10,256
|
Post by KeithL on Mar 21, 2023 11:18:59 GMT -5
There are a whole bunch of different ways of doing "phase shifts" and "all pass delays" with analog circuitry. Unfortunately, unlike with modern DSP-based filters, they vary wildly in terms of what they actually do... And, especially with older or simpler subs, most of them only vaguely approximate doing what the markings on the dial say they do. They're not really intended or expected to be terribly precise beyond "getting the sub in phase with the mains at the crossover point". (Remember that, since things like the cabinet tuning, and the driver itself, also introduce phase shift, they're just "part of the overall response".) And, even more problematic, with most of them, both the range of frequencies they affect, and the actual amount of phase shift they give you, may vary significantly with the setting. So you should assume that, if you move that dial, even a little, the results may vary significantly, and in unpredictable ways. (If you're going to take careful measurements, you should either leave that dial set in one place, and mark where it's set, and take new measurements if and when you move it.) If I were doing it I would proceed this way... First set that dial manually to deliver the highest level, at and around the crossover point, at the listening position (setting the sub to be in phase with the mains at the crossover frequency). Then take your measurements and try to figure out what's going on at other frequencies around there. (Or, if I was using something like a MiniDSP, if the sub had a "LFE" option, and it bypassed the "Phase" adjustment as well, which it may or may not, I would do that.) ............................................ I was wondering how some subs do a variable phase without being digital. In this video, though, he's only talking about time delay applied either in a miniDSP or in a processor where the sub distances can be entered separately. There are three parts to the video ... first is taking multiple sub measurements around the MLP and using REW Cross Correlation Alignment and Vector Average to get a new version of each sub's response using the multiple measurements; second is to use the Alignment Tool to recursively Align Phase Slope at Cursor until the flattest combined response is achieved and the delay values are determined; third part is using REW EQ to apply filters to finish the job. The first two parts would be useful and then we could use Dirac to do the filtering. He says way down in the comments if the value comes out negative, just apply it to the other sub ... but it's a little more complicated than that. Still, the Cross Correlation Alignment and Vector Average as a better starting point for each sub will be interesting to try. |
|
|
|
Post by thompson12 on Mar 21, 2023 18:58:40 GMT -5
+ ttocs I just watched the video and it looked brilliant. Wow, look at those features ... now I understand, very simple. UNTIL ... the result was he needed to dial in -5.xxx delay to subwoofer 2! He went all the way through optimizing everything and never talked about the fact that you can't put in a negative value for delay ... you can't have the sub start playing before the sound gets there. The only way to accomplish this would be to physically move subwoofer 2 five feet closer to the MLP. Am I missing something? I thought the same thing when I heard him say enter the negative value. So I thought maybe just stop clicking before it goes negative or try different dip points to click on and see what kind of value it comes up with. Mitch |
|
|
|
Post by marcl on Mar 21, 2023 20:20:57 GMT -5
+ ttocs I just watched the video and it looked brilliant. Wow, look at those features ... now I understand, very simple. UNTIL ... the result was he needed to dial in -5.xxx delay to subwoofer 2! He went all the way through optimizing everything and never talked about the fact that you can't put in a negative value for delay ... you can't have the sub start playing before the sound gets there. The only way to accomplish this would be to physically move subwoofer 2 five feet closer to the MLP. Am I missing something? I thought the same thing when I heard him say enter the negative value. So I thought maybe just stop clicking before it goes negative or try different dip points to click on and see what kind of value it comes up with. Mitch What will happen is you either get a totally positive value or a totally negative value. If it goes negative, then you switch the positions of the two subs in the Alignment Tool and correct the other one ... it will be positive. |
|
