SevenSixTwo - InDIYana 2018 Coax Design

[jhollander]

I did take ~20 degree off axis measurements of the coax as you recommended, but I haven't imported those into PCD yet to see how they look. I assume those 20 deg files do not get combined with nearfield measurements?

For the off axis files I treat them like the on-axis. The mid gets combined with the near field and the tweeter gets trimmed, then minimum phase extracted on both. Use the same offsets and listening distance as your on axis files when you set up a second sim.

IMO you are hoping the off axis fixes the dips...

[Kevin K.]

I was thinking the same thing Craig said when I saw that this morning but didn't want to rain on the parade Keith. There's always the active option and in this case it's looking like a really good option. Does the contest require passive crossovers?

[JavadS]

I took some foam lining out of the vented cabinets to see if my Fb would rise back to 31hz. I got excited when removing the foam from the top of the port and the ceiling in that chamber measured an Fb of 32hz while I held the driver in place by hand. Unfortunately the moment I put the screws back in place for the real seal the Fb dropped right back down to 27.8 Hz. I'm having trouble modeling a vented cabinet with my port geometry in BassBox that ALSO has an Fb of 28Hz. BB insists my Fb should be 31Hz. I am probably making a mountain from a mole hill, but again, it bugs me when I can't get the simulation tools to match reality. Guess I'm putting....

Keith I feel your frustration, this is why I prefer round ports as it makes it easy for me to make changes, making a round port longer or shorter is quite trivial, and you can see the lengths I went to to make it easy to do in my Strafi build, to this day I can still easily change port tuning from 25-50hz by simply cutting some new tubes and installing them.

Javad

[JavadS]

Boy that tweeter response is going to be a bugger to work with.

It is, and it was, ask me how I know! =) I'm quite happy with how my horn loaded coax is measuring and sounding, but I had to get creative. That was one thing that struck me taking on this project, that how many bad coaxes there are on the market, it's quite clear these pro driver style coaxes are meant to be used with an active crossover as it was trivial to get a great sounding speaker with miniDSP, the response and roll off of both the mid and tweeter leave a lot to be desired compared to the ease of implementation with the typical hifi driver and dome tweeter.

Javad

[KEtheredge87]

I was thinking the same thing Craig said when I saw that this morning but didn't want to rain on the parade Keith. There's always the active option and in this case it's looking like a really good option. Does the contest require passive crossovers?

Hey Kevin, While I scoured the official contest rules portion of the InDIYana 2018 thread, I didn't see anything explicitly stating no actives. However, post #63 has Ben saying no to Actives. http://techtalk.parts-express.com/fo...97#post1338097

I'll be interested to see what I can pull together for a passive arrangement here. I had an idea that this coax thing would be a tough nut to crack, but it remains to be seen just how tough. Judging from your response and Javad's, I might have gotten a jawbreaker instead of a nut! Oh well... this is how we learn and enjoy the challenge!

[isaeagle4031]

Yeah, no active for the contest/theme portion. Just takes way to long for set-up/tear down.

[JavadS]

Hey Kevin, While I scoured the official contest rules portion of the InDIYana 2018 thread, I didn't see anything explicitly stating no actives. However, post #63 has Ben saying no to Actives. http://techtalk.parts-express.com/fo...97#post1338097

I'll be interested to see what I can pull together for a passive arrangement here. I had an idea that this coax thing would be a tough nut to crack, but it remains to be seen just how tough. Judging from your response and Javad's, I might have gotten a jawbreaker instead of a nut! Oh well... this is how we learn and enjoy the challenge!

Keith you can do it especially with the modeling tools these days, but it's going to take more time and require more complex topology than the average HiFi speaker IME, but you can do it and you'll figure it out.

A lot of the stuff you're taking on is intimidating the first time you do it, but once you get past it, it becomes second nature and you go on to grapple with tougher concepts and techniques, I'm still learning every day I touch this stuff!

You're dong a great job working to do this right, that's great, at the same time don't let perfection be the enemy of good, take some measurements,load them in the modeling program, see what you get, measure it, do it again, tweak and improve. I'd suggest don't get too bogged down with off axis measurements or blending and combining measurements just yet, take bites you can swallow and build off of that. You can design a very nice speaker with on axis gated 5ms measurements and not worrying too much about measuring accurately below 300hz, it's a great start and you can build from there. Also add Xsim to your modeling arsenal as it's very easy to work with and an intuitive Windows interface.

BTW you are an incredibly gracious poster and interact with everyone so well, always a pleasure to see your humble curiosity to learn and improve more, keep it up bud!

[Kevin K.]

+1 to what Javad said. As for taking bites you can swallow, my suggestion would be to first focus on the tweeter since it will probably need some contouring and learn the different ways to influence it's response. A higher part count may be in order to whip it into shape but it can certainly be done. And FWIW, I'm just as green as you at this so I'm not trying to sound like I'm a know it all, just sharing from my experience of a couple of builds. The Omni-Mic is your friend and it's fun to sit down in front of a driver with a bunch of crossover components and play with all the different ways you can influence it especially when you can see and hear the differences every time a component is changed. Good Luck!

[Billet]

You may find that they sound better than they measure. I don't think ruler flat graphs are a strength of coaxes, but they do have some other advantages in their court. I suggest starting with a simple crossover that gets the levels close and do some listening.

The drivers probably need to break in a little (or a lot) also. I think pro drivers are stiffer and may take longer to break in.

[KEtheredge87]

Thanks Kevin K. JavadS and Billet. I was always taught to play nicely with others as I grew up Making friends is a lot more fun than making folks mad anyway.

I agree the tweeter is going to need some serious attention. The last XO simulation I did with some FRDs traced off the spec sheets was 3rd order on the tweeter with two series notch filters and some parallel resistors on that driver alone. My goal at this point is to get things reasonably flat from a simulation perspective then order whatever parts I don't already have. I've been building a small stock of XO parts from those unfinished SuperBee two way speakers, so I'll just keep adding to the parts bin as needed.Then it's just as you've said. Listen, measure, tweak, repeat until happy!

As for X-sim, yes... I like that program too. I've not played with it quite as much as I have with PCD, but I like getting a circuit layout when I'm closer to being done, as well as getting checks on estimated power through components like resistors. Javad, I also liked your recent Pretty Persuasions post about heat sinks on some of your resistors. I may need to borrow that page from your playbook, as I'm unsure how much power will need to be siphoned off to bring the coax driver down to the efficiency of the Estoeric woofer. Did you use a laser / infrared thermometer to check those surface temps?

Oh, lastly... on the topic of driver break-in. I did hook these coax drivers up and run the midrange cone section through a sine sweep loop, going up and down from 80-100 Hz. These things only have a 2mm Xmax rating, so there wasn't a lot visually going on to confirm break-in behavior, but I gave them 8+ hours on decent volume to get the suspension moving as much as possible. Hopefully that was enough!

[JavadS]

As for X-sim, yes... I like that program too. I've not played with it quite as much as I have with PCD, but I like getting a circuit layout when I'm closer to being done, as well as getting checks on estimated power through components like resistors. Javad, I also liked your recent Pretty Persuasions post about heat sinks on some of your resistors. I may need to borrow that page from your playbook, as I'm unsure how much power will need to be siphoned off to bring the coax driver down to the efficiency of the Estoeric woofer. Did you use a laser / infrared thermometer to check those surface temps?

Keep in mind you can do all our modeling in Xsim too, you just load the FRD and ZMA files and add your components to the circuit. Yes I use a laser temp gun, works great, nice to model but nothing like some real world data.

Oh, lastly... on the topic of driver break-in. I did hook these coax drivers up and run the midrange cone section through a sine sweep loop, going up and down from 80-100 Hz. These things only have a 2mm Xmax rating, so there wasn't a lot visually going on to confirm break-in behavior, but I gave them 8+ hours on decent volume to get the suspension moving as much as possible. Hopefully that was enough!

I can't say my 6HX150 broke in at all, I have them crossed over around 250hz and the cone doesn't move, but I flex my driver suspensions real good when I take them out of the box and that may be all it needed.

Also most important part...have fun! If it's not fun or you're getting frustrated, take a break and clear your head, always helps to come back fresh =)

Javad

[Wolf]

I have no doubt you'll do fine, Keith! One thing I tend to do on horns is to comp the Fs. If you get the impedance peaks out of the horn, it tends to smooth out, and then the FR bumps might not be so drastic.

Have fun!
Wolf

[KEtheredge87]

Hi Guys,

I've been fiddling for a few days and I think I have arrived at a solution that's good enough to build and test. As expected, the coax gave me plenty of challenges to address. The biggest challenge actually has been the area right around 2 kHz, as that's where I've been targeting the XO for the mid cone and HF horn sections. When I put my best foot forward in terms of matching response curve to a target response in PCD, I ended up with an impedance minima around 1.8 ohms and a BIG electric phase spike up toward +60 deg. It took an extra day or two of fiddling around to get back to a passable response that kept system impedance around 5 ohms and stayed within +/- 30 degrees of electric phase. Without further ado, here's some specs and pictures!

SevenSixTwo - Initial XO Prototype
XO Points - 560 Hz and 2060 Hz
Woofer to Mid - LR2 symmetric slopes
Mid to HF - Asymmetric slopes, LR4 arrangement (HF response matched LR6 response slope, Mid response went for LR4 slope)
Total parts count: 29 or 30 (might have missed a series resistor in the count somewhere)

I'll be ordering parts for these tonight so I can start playing around with them soon. Hopefully I have enough space inside my Coax chamber for all these XO parts. Only 3 or 4 of these belong to the woofer!

Let the games begin! (3-panel pictures here... individual sections next post)

[KEtheredge87]

Individual Sections for Prototype SevenSixTwo Coaxials XO

[Wolf]

Very nicely done, Keith!! Looks like the tweeter did need a hammer, but you have it nailed down.

Later,
Wolf