diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Full Range (http://www.diyaudio.com/forums/full-range/)
-   -   SB acoustics SB12MNRX25-4 "full range" Build thread (http://www.diyaudio.com/forums/full-range/216030-sb-acoustics-sb12mnrx25-4-full-range-build-thread.html)

wintermute 12th July 2012 12:06 PM

SB acoustics SB12MNRX25-4 "full range" Build thread
 
2 Attachment(s)
So my 4 1/2 year old was complaining that the old computer I had designated as hers didn't have any speakers. I foolishly suggested that maybe we could make some for her. She was very excited at that idea, having seen daddy working on his speakers a couple of years earlier.

I wanted it to be fairly simple so full range seemed a good idea. Small as well so I was looking at 3" drivers. The choices available to me just didn't seem to be cutting it though. Then I stumbled across the SB accoustics SBMNRX25-4

It was bigger than I was intending to use, It is not really a full range, but it models VERY well in a small (2.5L) vented enclosure, and I figure I can probably tame the top end breakup if needed. At a later date it can be upgraded to a two way, but for starters it should suffice.

I've also wanted to do a three way for a long time and it gives me the opportunity to evaluate a potential driver for the mid as well ;)

Some pictures of the drivers are attached. I'm very impressed with the build quality.

I've started this thread to keep track of the build. Hopefully it won't take as long as my MTM's!!

Tony.

wintermute 12th July 2012 12:25 PM

4 Attachment(s)
So the first thing I did was to re-purpose my coil winding jig as a rig for holding the speaker in free air to measure the T/S params. I did the usual of using my Wallin Jig II and speaker workshop, and was VERY disappointed with the results.

The resonant frequency was way off (even after some breakin) and the T/S params were so far out that it just modelled horribly (except in a sealed enclosure, which was pretty good).

after an initial few days of being dejected, I decided something was amiss and started to look at other software for measuring the T/S params. I settled on REW and got results very similar to the manufacturers specs.

The first graph is the speaker-workshop measurement of the impedance for each woofer. The T/S params calculated at a QTS of about 0.58 :eek:

The second graph shows the impedance plot as measured with REW (I'm not sure what caused the glitch on the impedance peak but it was there for noth drivers, oddly it shifted down the curve with added mass).

Note that the S/W graph was pre-breakin, and the REW was post breakin, however post breakin S/W was still showing the resonant frequency as 80Hz.

The table shows the measured T/S params and the average of the two drivers.

Tony.

edit: note that I'm not sure what is going wrong with speakerworkshop. I did a comparison measurement of my 10" Vifa's in their 30 Hz tuned reflex enclosure, and both REW and speaker workshop gave very similar impedance curves, it is only these SB drivers that seem to vary significantly with the two measurement techniques. I suspect that SW's MLS is not working well with the extremely light cone of the SB 4" driver...

wintermute 12th July 2012 12:42 PM

3 Attachment(s)
The plan is to use these speakers with a low powered chip amp (max 5W / Channel). I'm going to go vented to get *some* bass extension. They model very nicely with the measured parameters in a 2.5L box tuned to 70Hz. I will however need to roll off the amp below 70 Hz as they will be destroyed by any significant bass below about 60Hz in that enclosure. Note that they can be made to go much lower (in theory) with a bigger box and different tuning, but the tradeoff is in much sloppier step response.

I think that should be quite adequate for computer speakers. My MTM's have an F3 of about 80 Hz, and whilst a bit bass shy have been playing very happily without the assistance of the 10" vifa's that are awaiting my completion of my active crossover project for the last couple of years! :rolleyes:

Attached are the modeled SPL and step response graphs. I'm particularly impressed with the step response!

As can be seen in the third attachment even at 5W the excursion below 60Hz is high (but could probably get away with 40Hz)... below that however things might start to get nasty.

I'm not going to have a chance to do anything on this project this weekend, but I will hopefully finalize a box design over the next week. I've done a little modeling with BDS (for baffle diffraction) and it looks pretty good.

I'm going to try something new (for me) after a discussion with Dave (planet10) and do a slot for the vent instead of a tube. I've yet to come fully to grips with the construction details of that however.

Tony.

tuxedocivic 12th July 2012 01:25 PM

Thanks for documenting. Interesting having this driver brought into light. Will be interesting to hear your thoughts.

Have you considered sealed at all? Will you use the computer eq to correct the top end and baffle step? You could use it to correct the bottom end also.

wintermute 12th July 2012 10:12 PM

3 Attachment(s)
Hi tuxedocivic, I did think about sealed but in reality this driver is a midrange driver not a full range (I'm just trying to push the envelope a bit), and in a sealed enclosure it isn't going to go low enough. Attached are comparison graphs. Certainly from a power handling, and superior transient response perspective the sealed wins, and if I were to use this as a midrange in a multiway I'd definitly go sealed.

I was planning on doing BSC using passive line level correction as per Rod Elliots BSC article (see figure 3). as for taming the high end I hadn't thought about eq-ing, but was thinking I could use a notch filter.. This would most likely end up external to the cabinet though as its a small volume and I haven't allowed for and passive components in the volume calculations.

Tony.

tuxedocivic 13th July 2012 03:05 AM

If space is a concern, and the computer is the source, I'd strongly suggest using the onboard computer eq to accomplish BSC and notch filtering. I did this with my desktop setup and am very happy.

wintermute 19th July 2012 01:32 PM

2 Attachment(s)
I'll have a think about it. But it is supposed to be 4 1/2 year old friendly so I might stick to old school ;)

I haven't made a lot of progress but have what I think are my final dimentions for the baffle and internal volume. I modeled in BDS and am happy with the result. I'll have to draw up some proper plans. Whether I start cutting this weekend I'm not sure, as I also have some work to do on the car.

Baffle dimentions will more than likely be 134mm W X 204mm H
Internal dimentions 180mm H X 122mm W X 141mm D

The port will be (if I can construct it) at the bottom full width of 122mm 5mm high and 115mm long. tuning around 71Hz. and an F3 around 80Hz.

top bottom and baffle 12mm MDF sides 6mm MDF or ply (to keep the width down but still give the driver some breathing space) back either 6mm or 12mm.

BDS sim attached.

Tony.

wintermute 24th July 2012 03:48 AM

2 Attachment(s)
So I have almost got the plans for the box finished. Possibly will have some time tonight to finish them off.

They are pretty rough and not proper plans (too long since I did tech drawing and I can't remember how to do it properly), but will be enough for me to work out what I need to do.

I'll have to learn how to use sketchup some day, have played with it before but takes me ages to do anything at all.

I've attached the drawing of the front. It's not clear how it will go together but the sides will overlap the top and bottom (I'm still considering whether to make the top and bottom 6mm wider and rebate or just butt joint it).

The baffle will be a bit tricky. Rather than cut a slot in the baffle I'm planning to use the bottom of the box as the lower part of the port (so the bottom will stick out 12mm further. It does still require leaving 6 X 5mm on the sides at the bottom of the baffle though (2nd picture shows what I mean)

It's going to be interesting as I will be trying to achieve all this with nothing more than my jigsaw and router. Should be fun ;)

Tony.

wintermute 25th July 2012 03:46 AM

1 Attachment(s)
I "finished" off the plans last night. the scan is pretty ordinary but I have attached. I also downloaded a CAD program (creo elements direct express) last night and had a play. Will hopefully be able to get up to speed with it and do a better job of some plans ;) These have been uploaded here mainly as a record for myself.

Tony.

wintermute 28th July 2012 05:10 AM

ARRRGHHH
 
2 Attachment(s)
I just spent quite a bit of time cleaning up some old MDF I had (tricky as all I have is a jigsaw and a router). Trimmed it with the router to get it square, made front and rear baffles, and trimmed the width of top/bottom pieces.

I was about to separate the top and bottom panels and trim and decided to check the fit for the front and rear baffles.. BIG overlap... strange I thought. remeasured. Front 192mm and rear 204 mm, as I had planned. that is 12 mm... WTF. Then it dawned on me. I hadn't measured the MDF!! It was 16mm not 12mm DOH!

So it looks like I need to get some 12mm MDF as well as the 6mm. So these little speakers (at least this prototype any way. Will have 16mm front and rear baffle, 12mm top and bottom, and 6mm sides :rolleys:

Pic attached of my old MTM prototype re purposed as a router fence, and the cut pieces of MDF.

Tony.


All times are GMT. The time now is 05:46 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2