W5-1138SMF Specs

[stephenmarklay]

I am considering using this little sub for my soundbar. I have been looking at the very cool Anarchy 7” but after thinking about it, it is just not the best one for this project.

The W5-1138SMF is a better fit but I am wondering if the published specs are sufficient to calculate the volume and passive radiator needs (2 x ND140 5.25” look like a good match.). I am opting for a PR since the box volume is oddly shaped and getting a long enough port would require some elbows.

I have more volume (almost .60 ft^3) than the Voxel uses so I want to see if I can use that volume or if I need to look otherwise.

Thank you in advance

[Chris Roemer]

That sub and a pair of the ND140 PRs w/19g added mass (on each one) should be adequate and reach down to 40Hz (in a 0.6cf box).
A 2" port (that's 13" long w/elbow - nothing wrong w/elbows) should make it down to 30Hz.

[stephenmarklay]

That sub and a pair of the ND140 PRs w/19g added mass (on each one) should be adequate and reach down to 40Hz (in a 0.6cf box).
A 2" port (that's 13" long w/elbow - nothing wrong w/elbows) should make it down to 30Hz.

I think I read that elbows can contribute to chuffing. But really I doubt I am going to drive it really hard.
My box volume is shallow but wide (soundbar) so a 13” port is no problem with an elbow.

Thank you so much Chris. I greatly appreciate it.

I need to get a copy of WINLSD so I can actually calculate and play with this and learn.

Have a great evening.

[tomzarbo]

I love to use passive radiators when I can, but I like Chris's 2" x 13" long vent sim in this case. The extra bottom end would be nice to have, and that driver is a little monster that can take some abuse. They have grey colored electrical conduit tubing with very shallow radiused elbows if you're worried about sharp bends, but normal pvc or abs plastic really isn't usually an issue in practice.

http://www.linearteam.org/

TomZ

[tomzarbo]

Here is a link to the pro version that I think Chris prefers... https://softfamous.com/winisd-pro/
kind of amazing, you can adjust box size on the fly and see what it does to response.

TomZ

[stephenmarklay]

Thank you Tom for the sub insights and the links.

I am with you, Chris’s sim shows that little sub getting really low. It should be about perfect for my needs.

[stephenmarklay]

I ordered it up today along with the precision port 2”. Thank you all for the input and help.

[stephenmarklay]

Does anyone have a winisd file for the TB W5-1138SMF. There is an older version in the database has more information than I can find on the SMF version, I am also wondering if the measured data matches the published data?

[Chris Roemer]

Most ISD users learn how to enter their own drivers: "New Project", > (then) "New"; set "Auto calc parms" (at bottom) to "ON" (check), then enter parms:
Qes, Qms, (do NOT enter Qts - let it autocalculate it), Fs, Vas (be aware if you're entering it in cu.ft. or liters), Re, Le, Sd (use "cm^2), Xmax, and Pe.

Errs get generated by entering too many parms, 'cause the other numbers won't EXACTLY match out to all the specified decimal places. Just enter what I listed above. You NEED Qes, Qms, Fs, and Vas to model a box response. If you're lacking others (like Le, Pe or Xmax) your model will still be valid but OTHER graphs won't be possible. Still, the graphs you CAN generate WILL be correct.

I've not used THIS W5, but believe most users have successfully designed w/TB's data (so it must be pretty close).
AFAIK, the NON-neo vers. of this driver (sm"F") is close enough to the Neo vers. that box models are pretty much nearly the same.

(I've got over 2000 drivers in my DB, so I'm not prone to try newer (the newest!) vers. 'cause (in the past) the driver DBes weren't compatible between versions.)

[stephenmarklay]

Most ISD users learn how to enter their own drivers: "New Project", > (then) "New"; set "Auto calc parms" (at bottom) to "ON" (check), then enter parms:
Qes, Qms, (do NOT enter Qts - let it autocalculate it), Fs, Vas (be aware if you're entering it in cu.ft. or liters), Re, Le, Sd (use "cm^2), Xmax, and Pe.

Errs get generated by entering too many parms, 'cause the other numbers won't EXACTLY match out to all the specified decimal places. Just enter what I listed above. You NEED Qes, Qms, Fs, and Vas to model a box response. If you're lacking others (like Le, Pe or Xmax) your model will still be valid but OTHER graphs won't be possible. Still, the graphs you CAN generate WILL be correct.

I've not used THIS W5, but believe most users have successfully designed w/TB's data (so it must be pretty close).
AFAIK, the NON-neo vers. of this driver (sm"F") is close enough to the Neo vers. that box models are pretty much nearly the same.

(I've got over 2000 drivers in my DB, so I'm not prone to try newer (the newest!) vers. 'cause (in the past) the driver DBes weren't compatible between versions.)

Thank you Chris for the insights. I found a pretty good complete data set and created a new entry. Also thank you for the Qts tip and your comments about not having ALL of the parameters.



It all looks pretty “normal” I have been playing with the different alignments etc. It seems anyway I cut it, it looks best tuned to 35hz in a roughly 0.35 ft^3. using my .58 ft^3 of available space seems to just create a 2db +- hump no matter what the alignment. I might just have to block off some of the volume I have available for a flatter response. Obviously, I am not very versed and my results might be garbage

Although, the super boom box alignment while not flat, is about +-1.4 tuned to 30hz with my .58ft^3 with a 2” port at 14.72 “ in length. (+-1.4 from 27.5hz to 125hz)

[tomzarbo]

Thank you Chris for the insights. I found a pretty good complete data set and created a new entry. Also thank you for the Qts tip and your comments about not having ALL of the parameters.



It all looks pretty “normal” I have been playing with the different alignments etc. It seems anyway I cut it, it looks best tuned to 35hz in a roughly 0.35 ft^3. using my .58 ft^3 of available space seems to just create a 2db +- hump no matter what the alignment. I might just have to block off some of the volume I have available for a flatter response. Obviously, I am not very versed and my results might be garbage

Although, the super boom box alignment while not flat, is about +-1.4 tuned to 30hz with my .58ft^3 with a 2” port at 14.72 “ in length. (+-1.4 from 27.5hz to 125hz)

I can vouch for what Chris said about the neo and ferrite versions being nearly identical in useage. Specs may be off a bit from one to the other, but they both end up with nearly the same actual output frequency/box size whichever one you choose. When Neo went berserk cost-wise, TB wisely designed a ferrite version which is pretty much what I think everyone is using now unless they need light weight for some reason.

Anyway, usually when you increase the box size, the hump (which is approximately where the 'knee' is before low end rolloff begins) goes away the larger the box becomes. There can be a little mini hump the begins to show up on graphs when you really increase the box size on passive radiator boxes, but I don't think that's what you're speaking of.

Usually, as box size is reduced from what you could consider 'optimum' or basically flat until rolloff -- the hump will get more pronounced. Could you post up a few response graph pics of the smaller box and larger box models?

TomZ

[Chris Roemer]

I'd say all your observations are correct.

Notice though, that the larger box (.58cf) CAN get by w/a vent that's several inches shorter. This is "normal" that as a box gets smaller, the tuning vent needs to get longer. THAT's why small boxes, tuned low, sometimes are best tuned w/passive radiators.

Thing is, while your bigger box does have a bit of a "roller coaster" response before rolloff, IMO (since it's outside normal "music" range) any output down near 30Hz will most llikely be from LF-effects, and won't really (probably) matter one bit if they happen to be 1dB or 2 louder than the way they were "mixed". Also, you can right-click w/your pointer inside the little vented-box icon and slowly drag your mouse around and basically "draw" any curve that pleases you (the box vol. and Fb will continuously slide around). You can pick ANY vol./Fb you like, you're not limited to 2 or 3 "alignments" that WinISD might happen to show you. They're just ballpark starting points. Have fun.

[stephenmarklay]

Thank you both. You both have helped a ton. Chris, the drawing your own curve feature is great to see the “space” of solutions. I think you are also right that as soundbar, I doubt I will even notice 1-2db down in the 30-35hz range. Although it is nice to see what is possible.

Tom, I was able to take a picture of some curves. The curve with the highest peak is the full .58ft^3, tuned to 33hz, the next highest is .47ft^3 at 35hz and the flat curve is .36ft^3 at 36hz.

I am not sure which would perform best. The larger volume is a bit peaky but has a bit more output down low.

Any thoughts are much appreciated.

[BOBinGA]

Here is the WinISD file for the one I got back in 2104. You are not going to like it. Its way off the published specs. (Which is strange, because the W6 ferrite that I have was pretty close to spec.) Nonetheless, it still models quite close to what Chris calculated above. I get extension to 30 hz with a .5 cu ft box.

When you do you modeling, add in your intended low pass filter. This can greatly affect the results when you use an extended bass shelf modelled box and make it much flatter.

p.s. I had to change the file extension to .txt to get it past the filters. change it back to .wdr and it should read in WinISD.

[stephenmarklay]

Here is the WinISD file for the one I got back in 2104. You are not going to like it. Its way off the published specs. (Which is strange, because the W6 ferrite that I have was pretty close to spec.) Nonetheless, it still models quite close to what Chris calculated above. I get extension to 30 hz with a .5 cu ft box.

When you do you modeling, add in your intended low pass filter. This can greatly affect the results when you use an extended bass shelf modelled box and make it much flatter.

p.s. I had to change the file extension to .txt to get it past the filters. change it back to .wdr and it should read in WinISD.

Thank you! I will play wiith this today.