BillFitzmaurice.info

The completed cab looks incredible. I always liked the Mesa Boogie Roadready cabs, which I'm sure you've seen b/c yours looks pretty identical. Beautiful yet bulletproof.

I'm very sorry to hear about your hand. I hope you have some therapy planned? If your hand is working well enough to work with wood, you can get it to work as good as it did before the accident. The fact that you can do some limited woodworking is VERY encouraging and a great sign. Get well soon.

miked, thanks from my heart for your good words and wishes. Coming from you they mean a lot to me.

Had the aforementioned shot in the neck, that seems to have accomplished approximately nothing. Primary care guy referred me to PT, and the bigtime neurosurgeon says surgery has at very best a 50% chance of achieving any improvement, so that's about it where therapy is concerned. Nobody is going to cut on my neck if it's just a coin flip. My belief is that any recovery is up to me and always was anyway, so I've started "playing" again, primarily bass guitar without amplification, to try to encourage that nerve to start firing again. Faith, prayer, force of will, and work. Few things can withstand that combination forever. Maybe that hand will never work properly again, but I would not bet on that .

Have spent much time in testing Box 1 of 2, and the chart below shows a synthesis of many, many test runs:

I do not have full confidence in the testing tools I'm using, but this is as accurate a set of frequency response measurements as can be obtained with these tools. The results shown do reflect what my ears hear.

As you'll see, this box falls well short of the low-end response shown in Bill's published curve. Since we fortunately have a fairly complete record of how the cab was built, with photographs at many stages, I hope the community (and Bill if he has time) can help me identify

a) what I did wrong in the LF part of the build
b) what I can do to achieve better low-end response from Box 2 of 2 if anything, since it is not done yet
c) what, if anything, can be done to improve the low-end response of Box 1 of 2.

The cabinet is certainly usable as-is, and with 15-to-18 dB boost at low frequencies does yield good strong bass, but my hope was to achieve good low-frequency response with more modest EQ (and hence amplifier power). The 3012LFs were of course broken in as prescribed in the build plans before being installed in the cabinet.

With that strong boost between say 30 Hz and 200 Hz, tapering as frequency increases, the cabinet's overall reproduction is to my ears genuinely outstanding: clean, precise, and controlled. The combination of big paper cone midrange and wooden horn with titanium-diaphragm ABS horn tweeter does an excellent job of reproducing acoustic stringed instruments and vocals, as well as the detailed sound of acoustic double bass (recordings by Bela Fleck and the Flecktones for example). And of course, again with the proper EQ, the cabinet excels at high-volume reproduction of music by shredders like Joe Satriani and Steve Morse.

So that's where the project stands at the moment. Have restarted work on Box 2 of 2 now that Box 1 of 2 is complete and tested. More as it becomes available, and once again, if anyone can help regarding improving the low-end response of Box 1 of 2, your advice and suggestions will be most welcome.

Thanks, and all the best for 2014 and beyond
--aeolos

what does your measurement rig consist of? how did you calibrate (for accurate spl measurements), and what environment did you test in?

EDIT::

Nvm, I found it a couple pages back. all that remains is what your actuall measurement procedure is (cab placement/micplacement).

Hi sine143,

Good questions, and thanks so much for your help, no fooling. Here's a link to a chart showing the results of the very first test run weeks ago:

On that chart is a description of the test gear and environment.

Here's a photograph of that highly sophisticated test environment : There's nothing in front of the cabinet for 10 meters or so, and at that point there are 2 trees (no leaves at this time of year) of 8 - 12" diameter, and 10 meters past those trees, a bunch of scrub bushes and so on. The hillside slopes down gently from where the cabinet is in the photo. We live in a semi-rural area so I can crank it for a while without anyone but my wife and daughter yelling at me. Our closest neighbor (whose property begins about 300 feet behind the cabinet) likes guns, so he has no grounds for complaint, since I'm all for people with guns having some idea of how to use them properly. You may fire when ready there bubba! Practice makes perfect and all that sort of thing, just point your weapon in a safe direction, and remember that them there bullets bounce eh.

The point of all the foregoing being that reflected sound has indeed some, but not in my opinion a whole lot, of influence on results obtained in this location. But I am certainly ready to receive instruction and correction!

The file generated by AudioTool is opened with UltraEdit and cleaned up (replacing spaces with tabs as column delimiters, multiple tabs with single tabs, and so on), then imported into Excel 2013, where the data are used to generate a scatter plot. Sometimes the frequency and SPL values are out of synch for a few rows, yuck. Indeed the necessity of file cleanup is one of the reasons for my lack of confidence in the toolset: why cannot AudioTool guarantee that it will keep its rows and columns straight for cryin' out loud? Android is not a realtime OS but that phone has a dual-core processor, surely we can find a way?

Oh just saw your edit, but will leave my blather here, by your leave . Sometimes the phone was taped to a mic stand, sometimes it was held in my hand as I held it in front of the cabinet from the side. I did not observe any differences in results, so long as it was just my arm sticking into the sound field.

Also I am aware that many phones have a built-in high-pass filter in their audio circuitry to reduce wind noise and so on (one spec frequently cited is 24 dB/octave from 250 Hz on down). But the clincher for me is that the results match what I hear. The cabinet "turns on" somewhere around 45 Hz, but doesn't really get going until somewhere around 200 - 250 Hz. So without very substantial EQ like that shown in an earlier screenshot of the Behringer DSP UI, it produces what we used to call a "boxy" sound, heavy on the midrange and light on the low end. The published curve does show that sensitivity in the 20 Hz- 200 Hz range is lower than that for higher frequencies, but the cab I built displays a bigger difference than does the published curve. Removing the restrictor plate did make a significant improvement, but the cabinet did not seem to be tuned any lower with the plate in place.

Removing the plate made a big difference in the port noise, an important point at least to me. Now even when rendering a 30 Hz sine wave, port noise while still audible is reduced by perhaps 12 dB from its level with the plate in place.

Physics imposes hard limits on what can be done in a small space, given the many constraints on any portable speaker design intended for reproduction of full-range audio content. Maybe my expectations were unrealistic, or maybe I just hosed up the build. My hope now is that perhaps I can avoid hosage with Box 2 of 2, and better still find a way to beef up the bottom end reproduction on Box 1 of 2.

I'm learning as I go and am most grateful to you, sine143, and to everyone who shares their expertise in these forums.

Thanks and all the best
aeolos

i would attempt to place the cabinet flat on its back, with the mic suspended above it, far away from that ledge (or as far as you can get).

I suspect the current placement of the cab as you have it is lifting it ever so slightly out of 1/2 space, and as such the low end response (as measured) is affected.

I also suspect a low order *rumble filter* being in place on your phone (or tablet or whatever it is), affecting frequencies down low

Keep in mind that even for a perfectly *flat* cab, your ears begin to drastically lose sensitivity in the LF range, as per the equal loudness curve



As volume is increased, our ears become "more flat", so at low volumes, one has to boost bass quite a bit for a "flat sound" and as its turned up, the boost is required less and less. so while it may appear the cab "turns on" around 45 hz, this is likely due to BOTH a dip in cabinet sensitivity, and a dip in our ears sensitivity. I dont know what spl you are listening at, but if its around 90, you need approximatly 10 more db of bass at 40hz compared to spl at 250 hz for a speaker to SOUND flat (not *be* flat, but sound flat to our ears).

Hi sine143,

Your suggestion to test in real half-space is a good one, and I did and do want to test the cab that way. It is just so boogerin' cold out that a) it's hard work to stay out there long enough to get a full set of measurements and b) I wonder if the temperatures affect the results. The most recent set of measurements was taken at about 46° F and 66% relative humidity, and that was a nice balmy day as things have been going here recently. I hate being cold, makes all my metal parts contract or something. As a native Bostonian I recall when 40° F really did seem like a balmy day, everybody took their coats off and so on, but man oh man not any more. Give me sand and sweat any time over snow and frost in the beard. Bleeaahh.

But we do have a place for such a test, although it is closer to the street and the system will have to be turned up louder than it was for these tests. Input voltage ranged from 2.0 V average to 10-12 V average. That most recent set of curves was generated at 2.0 V and 104 dB average SPL when driven with pink noise. One nice thing about the Omni 15 is, it gets loud on not much power. Haven't taken it to the upper limit yet, that I calculate to be about 35 V (306 W less or more into 4 Ω).

Yessir, Fletcher-Munson! Thus the "loudness" button on, for example, the good old Marantz 1090 to which the cab is hooked up (through the Behringer 6000 DSP) here in the bedroom. It is a tribute to the Omni 15 design and its designer that (with proper EQ ) it can be used as a "hi-fi" speaker, and do a good job of it.

Would very much like to calibrate this test rig against a genuine test instrument like my old Ivie IE-30a, stolen long ago alas. With all due respect to AudioTool's author (and he is worthy of it, a good app), a one-man shop cannot have the resources to guarantee the correct function of its product on all the devices whereon it may be installed. Like you, I suspect the Note 2 audio hardware and software of skewing measurement results. The whole AudioTool/Note 2/IMM-6 combination cannot be expected to compare favorably to an IE-30a. Can anyone recommend a cheap-but-good real RTA?

Thanks and all the best
--aeolos

REW (Room eq wizard) is well priced (free).

http://www.hometheatershack.com/roomeq/

aeolos wrote:Your suggestion to test in real half-space is a good one, and I did and do want to test the cab that way.

You can't, unless you want to dig a hole in the ground deep enough to contain the cab, then back fill around it, then measure with a mic suspended at least 2 meters above it.
You can take a ground plane measurement, which gives a half-space result below the baffle step frequency, then put the cab on its back for a suspended mic measurement, which gives a half-space result above the baffle step frequency, then splice the two. Measurements must be taken from at least two meters away, preferably more with the suspended mic measurement of the high frequencies, otherwise the wave fronts will not be integrated at the mic position.

The most recent set of measurements was taken at about 46° F

The speaker will not work to spec below 60 degrees. Even if the daytime temperatures reach that the speaker must be kept warm overnight, or left in the sun all day to get fully up to temperature, as the damping insulates it against heat.

The speaker will not work to spec below 60 degrees.

Would Bill or someone else expand on this a bit; it's a brand new concept to me. Speakers do not perform to spec at moderately low temperatures? I learn something new every time I visit here, it seems.

Yep. I was listening to some cabs in my shop yesterday, the temp in the display/storage area was probably 35. The hf was very attenuated compared to when I originally listened to this setup a few days ago when it was close to 60. Sam asked me what I broke in the new cabs.

Philip Kelley wrote:The speaker will not work to spec below 60 degrees.

Would Bill or someone else expand on this a bit; it's a brand new concept to me. Speakers do not perform to spec at moderately low temperatures? I learn something new every time I visit here, it seems.

Driver suspensions stiffen up in the cold. If you want to get an accurate SPL measurement you need to do it with temps in the 70s. In the spring and fall when it can get cold at night I keep a cab I intend to measure the next day in the house the night before.

Aeolos,

How's the hand? I once went to sleep (slightly drunk) on a chair with very thin arm rests. My elbow was right on the thin arm rest and I then rested my head on my hand (OK... I was very drunk!). When I woke up the next morning BOTH of my hands were unusable (I had moved from one arm to the other in the middle of the night as I had a severe case of pins & needles). I could just about grip with my hand, but I could not straighten my hands. I went to A&E to find out what had happened and apparently I had crushed my radial nerve on the arm rests as I slept. I was told that the I may or may not get my hands working again Luckily after 6 weeks they started to come back to life and I was able to do things again.
Then 6 months after, I did it again this time whilst sleeping in my own bed! Again, the nerve came back to life after a few weeks luckily.

I am unsure by your description if this is similar to what happened to you, but from your description, I think you mention that you have had some nerve damage. I sincerely hope it will resolve itself like mine did - it did take a while, but I now suffer no after-effects (as I hope you will also)

And finally, good to see a fellow iNuke DSP user. I love my 2 units (3000DSP) - which is actually exactly what you have (the 6000 is just 2x3000s in bridge mode). According to Behringer's (well hidden) specs, you should get:



In the limiter section of the iNuke RC app, you should be able to double check what voltage the amp puts out when you have it hooked up to the PC. You can also get the app to do the work of conversion to wattage if you fill out the impedance value from the drop down box. One thing to note is that the voltage is peak, NOT RMS, so you need to divide the voltage by √2 (=1.414(=the 3dB crest factor of a sine wave )) - my 50V actually shows as 70.7V in the iNuke app. The wattage shown IS the RMS value though, just to add confusion!

Hope this helps. Also hope you get well soon. Finally, nice build. Loving the flightcasey look. Industrial speakers (although not industrial sounding, I trust )

you are suppossed to be limiting by peak voltage, not by RMS.