Very basic modelling of W-Frame woofer a la Linkwitz

0 Members and 1 Guest are viewing this topic. Read 8696 times.

hogwild

  • Jr. Member
  • Posts: 19
Hi everyone:

I am a newbie, with little ability to do complex math, and even less knowledge about the hobby. Please be gentle.

I would like be able to use some woofers I already have in a Dipole W-frame configuration, similar to Linkwitz's. But since the woofers are a different size (10") than Linkwitz's and quite diff. T/S parameters, I wonder where someone like me might find help, or at least get started figuring out what dimensions to use for a W-frame, knowing the diff. speaker parameters.

I assume it is way too much trouble to be guided through it step by step, but where might I start with such minimal knowledge? Of course, I'm not aiming for perfection here, but these are good woofers, and I'd really like to use them.

Thanks in advance for any help.

sfdoddsy

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #1 on: 6 Jan 2014, 12:48 am »
What woofers do you have? Linkwitz uses 10" drivers in his W-frame.

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #2 on: 6 Jan 2014, 01:32 am »
Thanks for helping such a newbie!

These are 4  10" fiberglass woofers custom made for Gershman Acoustics (+1 in case one breaks). They are labelled as Gershman GA-10. I'm not sure, but they appear to be the same woofers used in Gershman 's "X-Cube" sub and "GAP 828" speaker, but without the metal phase ring.

Acc. to the documentation on Linkwitz's site, The Orion woofers definitely use 12" drivers. Hence my reason for asking help with some basic modelling.

Whoops! I only have some of the T/S parameters. I forgot I didn't do the added mass/known enclosure tests yet, just the initial tests. I'm going to try to do those in the next couple of days, so please, your patience would be appreciated.

I'm not sure how accurate these are, as I'm...you guessed it...new to driver testing as well. I used a Dayton WT3 to do it, making sure to set up everything carefully acc. to instructions.

Re: 5.055 Ohms
F(s)=43.07 Hz
Q(ts)=0.726
Q(es)=0.843
Q(ms)=5.194
L(e)=1.357 mH.














sfdoddsy

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #3 on: 6 Jan 2014, 03:09 am »
Ah, sorry. I guess you meant the woofers for the Phoenix.

A chap named Gainphile did a W frame with 10" drivers, he might be able to give you dimensions.

http://www.htguide.com/forum/showthread.php?36626-S14-4-Way-OB-using-Seas-and-Vifa-Zaph-drivers&p=557623&viewfull=1#post557623

I have a pair of Gershman Avantgardes using those same drivers (I assume). Interesting speakers.

From the looks of those T/S specs the FS is a little high and the Qts considerably higher than the Phoenix drivers. This ought to mean less EQ is required.


hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #4 on: 6 Jan 2014, 04:20 am »
Ah, sorry. I guess you meant the woofers for the Phoenix.

Yes, sorry, I should have specified that.

A chap named Gainphile did a W frame with 10" drivers, he might be able to give you dimensions.


http://www.htguide.com/forum/showthread.php?36626-S14-4-Way-OB-using-Seas-and-Vifa-Zaph-drivers&p=557623&viewfull=1#post557623

I'm guessing you are talking about the woofers from Gainphile's S9? If so, those are H-Frame, and it's really
W-Frame I'm wanting to build. Or is it another project you're thinking of?


I have a pair of Gershman Avantgardes using those same drivers (I assume). Interesting speakers.

Holy Cow! What are the odds on that? I don't run into many people with Gershmans.

From the looks of those T/S specs the FS is a little high and the Qts considerably higher than the Phoenix drivers.

Yes, that was what I noticed (with my limited knowledge).
This ought to mean less EQ is required.

Didn't know that. Thanks.

I'll be using a MiniDSP for active crossover duties, so that should make things a little easier/more convenient.

However,if I want to use a config similar to the Phoenix, I wonder if I will be forced to switch to one of Linkwitz's specified woofers.  I wanted to drive these with a DIY Gainclone LM3886 amp I'm building, but the resulting amp would have trouble driving a load that low. Two of those Gershman woofers in parallel would be 2.5 Ω nominal.

Is there any way around this that's practical?

I wonder if my McIntosh MC2105 would be better suited to driving a load that low. If so, my Gainclone could drive the mids/highs.

If not, does anyone know where one might find some of the woofers in Linkwitz's examples? They are all discontinued and very hard to find, at least in North America.

THANK YOU!
« Last Edit: 6 Jan 2014, 05:26 am by hogwild »

sfdoddsy

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #5 on: 6 Jan 2014, 08:38 am »
Gainphile's S14 speakers use Dayton RSS265s in a W baffle. That link was to the build thread.

I'm not sure how suitable the Gershman drivers are for dipole use, but it is certainly doable if you can get the dimensions of the W frame right.

You'll probably have to measure to get the EQ just right,

There is a simple primer for this on Gainphile's blog.

http://gainphile.blogspot.com.au/2011/01/measuring-open-baffle-loudspeakers.html

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #6 on: 6 Jan 2014, 11:14 pm »
sfdoddsy:

I saw that thanks. But I think that is a few light years above my level. Is there something else I could use to do a basic model, or do you know of another pre-existing project that uses woofers which are available now (as opposed to Linkwitz's Phoenix dipole?)  Any other ideas.


Thanks



sfdoddsy

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #7 on: 6 Jan 2014, 11:28 pm »
Here's the in-room response of my Gershmans with (I assume) those woofers. As you can see they plummet below 50Hz. They aren't likely to do much better in a W frame. A W-frame has resonances which affect the upper frequency as well, so depending on the range you wish to cover there may be better choices.

You might find it easier to start with a simple H frame dipole of similar dimensions to the Orion bass section. This thread will show the sort of EQ required for the dipole roll-off, and for the driver EQ:

http://www.diyaudio.com/forums/minidsp/174145-minidsp-linkwitz-orion-asp.html






Davey

  • Full Member
  • Posts: 1481
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #8 on: 6 Jan 2014, 11:50 pm »
You haven't mentioned it yet, but what is your objective in building W-frame enclosures vice H-frame?  If you're a new builder, I would select an H-frame construction first since it's a simpler concept.  The only reason to build a W-frame setup is if vibration is an issue and you need the force cancellation that configuration yields.

There aren't any good modeling utilities for these I'm afraid.  Yes, you can input a front/back cancellation "D" distance into a spreadsheet/utility and compute some numbers, but there's no substitute for actually building a prototype and measuring it.
The objective is to make "D" large enough so you don't need oodles of equalization, but small enough so the construction is fairly compact and you don't create significantly large cavities.

Cheers,

Dave.

matevana

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #9 on: 7 Jan 2014, 12:38 pm »
A good rule of thumb in building an H-Frame is to keep the baffle depth < 1/2 of the width and height. For example, if your H frame enclosure measures 15" W x 15" H, keep the baffle distance from the front and rear of the enclosure less than 7.5".  This assumes a symmetrical frame and helps quell any additional resonances you would then have to factor into the design....making the build more complicated. 

mcgsxr

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #10 on: 7 Jan 2014, 01:11 pm »
I have always been an experimenter when it came to building up OB speakers, but I did like to do some rudimentary planning before making sawdust.

I found the xls file at this link useful for a ballpark idea of what effect certain size compromises would make with a given woofer.

http://www.diysubwoofers.org/dipole/

In looking at the specs for your woofers the QTS is excellent, the FS good for OB - I wonder if you know the XMAX?  That may be the limiting factor for fun OB bass with those woofers.  Be careful applying power to them.

Good luck with it!

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #11 on: 8 Jan 2014, 05:22 pm »
Umm...holy cow..forget roll-off below 50 Hz. Is that chart really saying that the typical SPL is about 71dB or am I missing something. That woofer doesn't seem to practical if that's the case.




Here's the in-room response of my Gershmans with (I assume) those woofers. As you can see they plummet below 50Hz. They aren't likely to do much better in a W frame. A W-frame has resonances which affect the upper frequency as well, so depending on the range you wish to cover there may be better choices.

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #12 on: 8 Jan 2014, 05:23 pm »
mcgsxr:

Thanks for that. Being curious, I checked the nominal impedance of the drivers suggested on that page, and they are rated at 5.5 ohms. Not a big diff. between them and the Gershmans I have sitting here. Are there many folks out there driving ~5 ohm woofs in parallel effectively? I don't think I'll be able to achieve that with my LM3886-based Gainclone or even my McIntosh amp.

I noticed that Geert Meddens page lists his "Contrapunt" woofer as using two Peerless 831857 models, which might be a little easier to find. Those are discontinued too, but I wonder...does anyone know where I might find some?



matevana:

Thank you for that. I still don't know which dimensions to begin to experiment with, because as I said, I'm a complete newbie. Any (ballpark) suggestions on that front?


A good rule of thumb in building an H-Frame is to keep the baffle depth < 1/2 of the width and height. For example, if your H frame enclosure measures 15" W x 15" H, keep the baffle distance from the front and rear of the enclosure less than 7.5".  This assumes a symmetrical frame and helps quell any additional resonances you would then have to factor into the design....making the build more complicated.



scorpion

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #13 on: 8 Jan 2014, 10:50 pm »
First, just go and read the MJK classic: http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf .
Dimension like the depth is set according to crossover frequency and frequency resonance. So 7.5 inches is no missfortune
and would be equal for any loudspeaker dimension just to fit MJK:s approximations. Which by the way are exellent.

/Erling


hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #14 on: 9 Jan 2014, 12:12 am »
scorpion:

I have seen your posts here while lurking many times before. I know you are very experienced and appreciate this quality of expertise in advice.

But let me restate what I wrote in my OP

"I am a newbie, with little ability to do complex math, and even less knowledge about the hobby..."

So let me rephrase the question instead now. Does anyone know an existing project that I can just do (stock) with woofers that are still commercially available. I don't understand complex math, I don't understand most of the science, but nevertheless would still like to build something. If the woofers in Linkwitz's Phoenix woofers were still available, I'd use them, but they are no longer available.

Thanks


matevana

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #15 on: 9 Jan 2014, 05:37 pm »
matevana:

Thank you for that. I still don't know which dimensions to begin to experiment with, because as I said, I'm a complete newbie. Any (ballpark) suggestions on that front?

Honestly, just go buy one of the Way Basics cubes for 20 bucks and throw in a baffle. It's an inexpensive test for one of your woofers and these dimensions generally work well with most 10".  For testing purposes you can easily extend any of the dimensions with cardboard, but you may find this not necessary once you hear them.

http://www.waybasics.com/shop/storage-cube-natural.html

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #16 on: 10 Jan 2014, 01:00 am »
matevana:

Thanks for that suggestion. Yes, I had seen your posting elsewhere on this and thought..."Hey, that would make a test baffle easily and quickly".

I think I will do that...as I continue to look for some toroids for my Gainclone project.

Thanks

hogwild

  • Jr. Member
  • Posts: 19
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #17 on: 13 Jan 2014, 11:31 pm »
Hey Matevana:

For your H-frames, are you using the woofers wired in series or parallel?


gab

  • Full Member
  • Posts: 626
Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #18 on: 14 Jan 2014, 12:03 am »
Honestly, just go buy one of the Way Basics cubes for 20 bucks and throw in a baffle. It's an inexpensive test for one of your woofers and these dimensions generally work well with most 10".  For testing purposes you can easily extend any of the dimensions with cardboard, but you may find this not necessary once you hear them.

http://www.waybasics.com/shop/storage-cube-natural.html

or maybe this?

http://www.diysoundgroup.com/flat-packs-1/speaker-flat-packs/67cuft-cube-flat-pack.html

matevana

Re: Very basic modelling of W-Frame woofer a la Linkwitz
« Reply #19 on: 14 Jan 2014, 08:23 pm »
Hey Matevana:

For your H-frames, are you using the woofers wired in series or parallel?

Hey Hogwild,

For the Hestia Forte which uses a total of four 10" Woofers, they are actually wired series/parallel and produce a monaural signal. Each set of two woofers are wired in series and then joined together in parallel so the amp sees an 8 ohm load  The x/o frequency is low enough that the woofers are largely non directional so there is no collapsing of the overall stereo image. This can also be advantageous with an amp that is run in bridged mode. Sometimes the amp's damping factor is enhanced (8 ohms bridged vs. 4 ohms per side) and at times I have found placement to be less fussy.