Build your own 2x12" TH (The Kraken 212 TH)

[Djim]

Hi Oliver,

Quote: "The pressure chamber is also interesting for DIY-er. Under the premise that the chamber was once calibrated with a reference microphone, it is no problem, due to its small dimensions, to set the chamber up to calibrate any microphone in frequency response and level.................. Theoretically, for good sealed enclosure, a calibration of a microphone frequency response can be done without the comparison with a reference mic"

Theoretically that is correct but:

A) You still need a high quality reference to calibrate the pressure chamber
B) The driver needs to be powered with a precision of 0.010V (RMS) which involves expensive equipment that most people don't have.
C) Drivers do change over time so how often do you need to recalibrate?

(Btw the Monacor MCE 2000 uses a Panasonic capsule)

[tb46]

Hi Djim,

I just thought you might find it interesting (Post #217). For getting absolute values you are correct in your criticism, but, for relative values, i.e.: to determine the fall-off of a given microphone below 150 Hz, this would be a great tool to have, and it wouldn't be very critical at all.

Regards,

[weltersys]

Originally Posted by barco

Interesting discussion!

I started looking into solutions for this "problem" with uncalibrated microphones and ran into this SOFSCI SoundSweeper Acoustic Measurement Kit | Sounds of Science

They claim:
±2dB 10Hz to 19Hz
±1dB 20Hz to 20kHz
±1.5dB 20kHz to 24kHz

Which must be considered to be very good?

Good, but not quite as good as the B&K 4004 rated +/- 2dB from 10Hz to 40kHz .

That said, I generally use a much cheaper measurement mic.

At any rate, unless your measurement mic drops like a rock in the vicinity of the speaker F3 point (as some examples previously show do) probably not worth a lot of money to chase the ultimate.

Art

[Djim]

Hi Oliver,

Maybe it is interesting for developing a relative cheap calibration tool by integrating a small amp that is connected to a standard Digital to Analog converter based on USB connection, which gets driven by a simple sinus program. That way it can become an easy calibration tool for your mic (reference dB point) anywhere below 100Hz and give a mic response plot. The error rate could be made smaller than 1dB between 5Hz and 100Hz if a small correction eq is integrated in the software.

Question is, would there be a market for such tool that would cost around the 250 dollars?

[tb46]

Hi Djim,

That's a great idea - now, have at it :-). Seriously, that sounds like a marketable product, especially when integrated with an overall simple measurement system.

Regards,

[weltersys]

Originally Posted by Djim

Hi Oliver,

Maybe it is interesting for developing a relative cheap calibration tool by integrating a small amp that is connected to a standard Digital to Analog converter based on USB connection, which gets driven by a simple sinus program. That way it can become an easy calibration tool for your mic (reference dB point) anywhere below 100Hz and give a mic response plot. The error rate could be made smaller than 1dB between 5Hz and 100Hz if a small correction eq is integrated in the software.

Question is, would there be a market for such tool that would cost around the 250 dollars?

Good idea, but most would prefer to pay someone a fraction of that amount to make a deviation chart for their mic, or pay for a better quality mic that comes with a chart, so I think the idea has limited commercial appeal.

My favorite measurement tool was a Hall Engineering ATG-301, which consisted of a dB meter with a deviation chart and a sweepable finite-bandwidth pink noise generator with a 20-20K frequency range in bandwidths from 1 to 1/20th of an octave.

I probably only paid $250 for the set , unfortunately it was stolen (along with $150K of sound and lighting equipment) but would pay a lot more to have a similar tool again. Even with Smaart and RTAs, I really miss that hand sweepable pink nois generator, and can’t seem to find anything similar.

The ATG-301 was made in the 1980’s, 20 Hz was lower than almost any speaker’s response then, now I’d want a unit to be calibrated down to a few Hz...

Art

[Djim]

Hi Art,

That’s why I asked the question in the end. The difference with a supplied chart is that you still have no reference to adjust to in a real measuring setup. A 'normal' pistonphone is often used for calibrating to 94dB at the usual frequency of 250Hz (they cost more than 500 dollars). The instrument I described would be able to deliver any level between 80dB and 140dB at any frequency spot between 5Hz and 100Hz within accuracy of 1dB.

Practically, that means you can calibrate any measuring setup within the same level range you are going to measure. Do you think that is worth the extra investment?

[barco]

Originally Posted by weltersys

Good, but not quite as good as the B&K 4004 rated +/- 2dB from 10Hz to 40kHz .

That said, I generally use a much cheaper measurement mic.

At any rate, unless your measurement mic drops like a rock in the vicinity of the speaker F3 point (as some examples previously show do) probably not worth a lot of money to chase the ultimate.

Art

Well, I think we can argue if the B&K is really better. Above 24kHz yes. But between 20Hz and 20kHz it is worse (+/- 2 dB compared to +/-1dB for the sofsci). Between 10-20Hz they are the same. Sounds more useful to me.

Never the less, I had the opportunity to borrow a second ECM8000 from another friend (beacuse of the discussion about the quality and reliability of the calibration file). I know it is far from the perfect test, but at least I found it interesting. I just swapped mic during the test (nothing change at all, not even wiring) and captured two plots.




To me, they look VERY similar. Strangely enough the level seems to be a bit different, but the graph is very similar for sure.

[weltersys]

Originally Posted by barco

I had the opportunity to borrow a second ECM8000 from another friend (beacuse of the discussion about the quality and reliability of the calibration file). I know it is far from the perfect test, but at least I found it interesting. I just swapped mic during the test (nothing change at all, not even wiring) and captured two plots.

To me, they look VERY similar. Strangely enough the level seems to be a bit different, but the graph is very similar for sure.

Test results look like you both got some good units.

[barco]

The Kraken's are a bit more integrated in my environment now



Added a front for the openings to make it more "stealth".



I have the sofsci measurement stuff now (+-2db 10-20 Hz, +-1db 20-20kHz) so I will be posting a more "correct" freq response shortly.

These horns are amazing. That the deep base could rumble a lot of things I understood, but the quick, clean and tight base for music surprised me.