MBOWI 3-Way Phase and Measurement Question

[Swerd]

I never should have let this take so long. I'm having real trouble leaving the house.

This thing ... it's not letting me go.

No DIYer ever said it better than that .

I think I used about 25' of the 1000' I had on hand.

Better to have it and not need it ...

By the time you use the first 100' of that 1000' spool, you'll be scanning Craig's List for more. Just in case you run low.

 

[everettT]

No DIYer ever said it better than that .
By the time you use the first 100' of that 1000' spool, you'll be scanning Craig's List for more. Just in case you run low.

He is still chasing that reverse null...

 

[Alex2507]

reverse null

Wow, whatever it is that you started ... ain't nobody got time for that right now.

I do plan to make myself still not understand that but on a deeper level.

Power response and Reverse Null, what do they show? - Techtalk Speaker Building, Audio, Video Discussion Forum

Want a second or third opinion about your speaker cabinet design or other audio related problem? Post your question or comment on the Technical Discussion Board. Hundreds of technicians, engineers, and hobbyists, nationwide read and discuss electronics related questions each week. We welcome...

techtalk.parts-express.com

There's a work training thing I gotta go to today. I brought the mic and mixing board home last night. And the class is in NH. This is what day drinking was invented for. I'm ready for Matt's REW video and they want me 2 states up to build scaffolds for free ... ugh.

Gotsta git

 

[Swerd]

Reverse Null is one of those audio DIY terms that can cause people to scratch their heads. It means, if you reverse the polarity on one of the drivers, the crossover frequency shows up as a deep trough, a null.

The examples below show the frequency response of a 3-way speaker, my old JBL L100A speakers with Dennis Murphy's crossover designed for it. The first graph shows the frequency response with all the drivers wired as intended by the design. You can see the two crossover points, at about 950 Hz (where the blue and red lines cross) and at 5000 Hz (where the red and green lines cross).

And here is the same speaker but with reversed polarity for the mid range driver. That driver becomes out-of-phase with both the woofer and tweeter, making deep troughs (nulls) in the frequency response where the two crossover points are. You can also see how symmetrical the roll-off & roll-in slopes are.

Dennis's crossover design software allows him to view all this on the computer without actually changing any wires. Click on the "Reverse Null" button and you can see the result. His software uses a dual-duplex (?) sound card, so he can hear the result and see the frequency response curve at the same time.

When I first built my CAOW1 speakers (2-way speakers with 2nd order crossovers, where the tweeter is supposed to be wired in opposite polarity to the mid-woofer) I goofed as I built it, ending up with the tweeter's polarity wrong. Dennis heard it in about 1 second. I, of course, heard it wrong from it's first sound, and assumed I had done it right. Dennis said many DIYers make that mistake. Wiring the drivers to the crossover board is one of the last steps, and is often done late at night as you're trying to finish.

I distinctly remember a DIY speaker builder's meeting at Dennis's house, where one guy showed off his DIY speakers made with very expensive SEAS Excel drivers. He also swore he could hear the effects of exotic speaker wire brands on his speakers. But he never noticed that he had a deep Reverse Null at the crossover point (about 2200 Hz) because he mistakenly wired his tweeter wrong .

 

[Swerd]

Here's another example of a Reverse Null frequency response from another of Dennis's 3-way designs, the Philharmonic 3.

First, the on-axis response with correct wiring

And further down on that page, the Reverse Null. Dennis said this:

"The following plot shows the “reverse null” of the Philharmonic 3, which uses a “Linkwitz-Riley 4th Order” crossover. For this type of crossover, the drivers should be precisely in phase with each other where they overlap at the relevant crossover frequency. If this goal has been achieved, then reversing the connections to the midrange driver will throw it fully out of phase with the woofer and tweeter, and a deep null should develop at the crossover frequencies as the output between the woofer and the midrange, and the midrange and the tweeter cancel. As the plot demonstrates, this in fact occurs at the crossover frequencies of about 640 Hz and 2800 Hz."​

 

[everettT]

Here's another example of a Reverse Null frequency response from another of Dennis's 3-way designs, the Philharmonic 3.

First, the on-axis response with correct wiring
View attachment 32650

And further down on that page, the Reverse Null. Dennis said this:

"The following plot shows the “reverse null” of the Philharmonic 3, which uses a “Linkwitz-Riley 4th Order” crossover. For this type of crossover, the drivers should be precisely in phase with each other where they overlap at the relevant crossover frequency. If this goal has been achieved, then reversing the connections to the midrange driver will throw it fully out of phase with the woofer and tweeter, and a deep null should develop at the crossover frequencies as the output between the woofer and the midrange, and the midrange and the tweeter cancel. As the plot demonstrates, this in fact occurs at the crossover frequencies of about 640 Hz and 2800 Hz."​

View attachment 32651

Thanks for taking the time to write the explanation. At the beginning of the thread I'd mentioned it to make sure the polarity was correct, which IIRC he had one in phase at the time.

 

[Alex2507]

At the beginning of the thread I'd mentioned it to make sure the polarity was correct, which IIRC he had one in phase at the time.

That's right, I did do that but the trick is that I was only sure about how the tweet was wired. Anything could have been going on at the low pass and band pass filters. There was an issue with one of the binding posts being flipped so the red was on the left. I corrected that but had not really looked at the xo's to verify adherence to the schematics. And it measured better with REW the way I had it "wired wrong" and it sounded better.

When I decided to really go in for the bottom end redo, the speakers got taken to a DIY meet where I'm pretty sure we fried a couple of resistors with this massive Aragon amp and then all that's wrong with the world got righted.

I just demo-ed them for de-niece and think she might be loosening her purse strings for speakers. Another part of this hobby I enjoy is spending other people's ching. Life is too short for TV speakers. What's funny is that the 3-ways sound great in the room at any volume, really it's great but at volume outside the room, the bass is overpowering. It's like my doorway acts as the port in a TL but what's coming out ain't pretty. It was for demo purposes only ... for my neice and my neighbors I guess.

the drivers should be precisely in phase with each other where they overlap at the relevant crossover frequency.

Just another variable that adds to the difficulty of designing xo's. Like, can you imagine looking over FR graphs until you spot the phase angles being in alignment at a couple of odd frequencies? God bless him for his efforts.

By the time you use the first 100' of that 1000' spool, you'll be scanning Craig's List for more.

I scan CL regularly and I probably wouldn't be able to resist a roll of 12 or even 10 ga. Maybe not 1000 feet of it because it gets expensive quick as you step up the gauge but I'd drop another $100 on heavier wire without flinching. I also still scan for amps on like 3 forums, CL and Audiogon. Don't judge.

 

[everettT]

That's right, I did do that but the trick is that I was only sure about how the tweet was wired but actually, anything could have been going on at the low pass and band pass filters. There was an issue with one of the binding posts being flipped so the red was on the left. I corrected that but had not really looked at the xo's to verify adherence to the schematics. And it measured better with REW the way I had it "wired wrong" and it sounded better.

When I decided to really go in for the bottom end redo, the speakers got taken to a DIY meet where I'm pretty sure we fried a couple of resistors with this massive Aragon amp and then all that's wrong with the world got righted.

I just demo-ed them for de-niece and think she might be loosening her purse strings for speakers. Another part of this hobby I enjoy is spending other people's ching. Life is too short for TV speakers. What's funny is that the 3-ways sound great in the room at any volume, really it's great but at volume outside the room, the bass is overpowering. It's like my doorway acts as the port in a TL but what's coming out ain't pretty. It was for demo purposes only ... for my neice and my neighbors I guess.



Just another variable that adds to the difficulty of designing xo's. Like, can you imagine looking over FR graphs until you spot the phase angles being in alignment at a couple of odd frequencies? God bless him for his efforts.



I scan CL regularly and I probably wouldn't be able to resist a roll of 12 or even 10 ga. Maybe not 1000 feet of it because it gets expensive quick as you step up the gauge but I'd drop another $100 on heavier wire without flinching. I also still scan for amps on like 3 forums, CL and Audiogon. Don't judge.

I'm sure you're getting old enough to settle on a life mate amp.

 

[Swerd]

What's funny is that the 3-ways sound great in the room at any volume, really it's great but at volume outside the room…

I've noticed that about several of Dennis's designs, Salks, Phils, or DIY. They sound great at low or high volume. Other speakers I've owned had to be turned up to high volume before they could sound good.

 

[Alex2507]

I've noticed that about several of Dennis's designs, Salks, Phils, or DIY. They sound great at low or high volume. Other speakers I've owned had to be turned up to high volume before they could sound good.

Mine get played low all the time. Now that I dragged one of the DIY subs out of the corner, the bass response lets speakers and subs blend well enough without the sonic punctuation of the bottom end boom produced by the lowest frequencies. I was gonna hide wires yesterday but it was more fun interacting with other living beings. Imagine that.

I'm sure you're getting old enough to settle on a life mate amp.

I saw a 3 channel Audio by Van Alstine for a grand. I wonder if it was like this for people in those cultures where you basically bought women. "I like the more expensive model but I can only afford the slightly used model from the home with no smokers and no pets".