Testing autotuba responses?

Hello all.

I'm now the proud owner of one of Bill Fitzmaurice's wonderful horn-loaded enclosures :) One problem... the decibel meter I have is suspect.

It's a £50 digital run-of-the-mill job, and appears to have some serious inaccuracies below about 65Hz. Ie, the horn's output sounds flat, but the reading is at least 25dB down from what it should be!

Is there an easier way of doing this than a frequency generator, decibel meter and a pad of paper?! I haven't been able to find anything software-wise that seems to do the job. Either far, far too confusing, or otherwise unusable (ie, expensive).

On a related topic, would a 'vanilla' condenser microphone insert be accurate enough and/or calibratable for low, low bass frequencies? I'm looking at stuff from about 150Hz downwards (limit will probably be about 20Hz or so).

Thanks everyone :)



2004-10-11 5:21 pm
I'd suggest you find a car Audio shop who has an Audiocontrol or better Termlab SPL meter, and ask if you can use their meter to test it.

A Termlab would allow you to just play a sweep tone and it would give you a nice bar chart of SLP Vs Hz (1Hz divisions I think too). as the Termlab seems to be the industry standard now (in car audio) you would have something as a reference too.
Compare my measurements and see what you think. Bill's quarter-space estimates are rather enthusiastic in my mind. I don't know how much cabin gain improves this.

It's a good design but there isn't much output below the horn cutoff ~35-40Hz. (as expected -- you can't cheat the physics of it)

Since posting the above I got an email from an 'interested party' who pointed me to a message on another forum from Bill, who has his own version of events, in which Jim merely posted from Jim's home as he often visits with him. He feels he was banned without warning or explanation.

I find all of this somewhat suspect, but as there are 2 sides to every story, it seems only fair to mention it.
I tend to side with DIYaudio moderators on this.

Bill never gave out any accurate FR curves (by accurate I mean every testing condition, how could you achieve what is advertised... etc), couldn't back up anything IMHO, and timed after every Bill post, there's a Jim York post to back him up with big fanboyism. tiroth measurements proved this to my eyes, Bill plans are way overhyped. If he was really banned unfairly, he would have at least asked why, especially since he was probably making good business with DIYaudio. I think he doesn't care because he knows he was wrong on this.

At least he wasn't charging alot for his plans so it's not "that" unfair.
Well. :D now there's a little controversy going around...

The rubbishy Maplin meter I have is not going to make any conclusive results, so I'll put that aside for a moment (unless someone knows of a way to make a flat response mic I can connect to my pc, and a piece of software to do all the hard work?).

The autotuba is great, but like tiroth it's awkward to get the desired low-end response without messing around with equalisation. I did some quite unscientific tests (this whole thing was an advanced higher physics project after all...) and basically concluded the following:

1. The Autotuba is very sensitive to room placement. It likes corners and facing down to the floor, with (oddly enough) one edge of the mouth touching the floor, and the other about an inch off it. ie, leaning over precariously :smash:

2. The Autotuba is best for in-car use. It also suits smaller cars with shorter 'resonant chambers'... an equaliser would probably do a world of good compared to my rubbishy head unit's bass controls, but still.

2.5 The Autotuba works well at low volumes and excursions. I have built two of them (the first, rather shaky attempt was mine, the second rock-solid one was for a friend... 127dB at 40Hz, using ~100w input power, my one just rattled horribly) and found that they sound the best at lower volumes - I've found they have a tendency to 'boom' and in some cases make loud popping noises. All of which were quite disconcerting for the driver I paid $70 shipping to get! (well, two drivers technically but...)

3. The Autotuba doesn't work very well outside. It needs what I presume to be the horn effect of a car (with the front doors open... Mmm... I found a large node about 50m away, sitting around about 15Hz :bigeyes: ) to do its stuff properly.

4. The construction plans are quite vague... hence I ended up using about 300 screws during its construction. :hot: and about half a litre of PVA... yeah, maybe I'm a bit enthusiastic. When I say vague, I mean that it didn't give useful dimensions... it said what the overall sizes were, but didn't give any mention to angles, the length of individual pieces etc... hence it was an utter pain to try and put together. About 2 hours of planning lost... but it was okay in the end.

I agree that Bill really does have to answer for the inconsistent test results he provides.. even though I admit, different cabinets perform in different ways depending on the conditions, but a typical concrete-floored room of 4m by 4m, with plasterboard walls and carpet would make a nice benchmark. Some more hints as to get the best out of the designs would be nice too :D

Back on topic, anyhow - cheap cartridge microphone with a flat response, and software to do the hard work for me (graphs + sweeps)?