Speaker Testing - Amplifier needed

[BellMan]

I am currently setting up a test system for speakers and speaker enclosures.

I have been reading about the sound and vibration measurement suite, and I see that I have the option to output a swept sine wave.  I would like to use this to drive the speakers.  I beleive that I could use the analong outputs of a 4461, or a 5404 to output the sine wave, however I don't beleive that this signal will be of sufficient power to drive the speaker.  Therefore an amplifier would be needed.  What I am looking for is a precision amplifier that I can use to amplify this signal.  I am looking for something that I can control via LabVIEW, to set the output power to match the speakers rated power.

Does NI offer such a device?  If not, any thoughts on where I might find one?

[Mr_Hardware]

Would something like this work ? ...

  www.electronics-lab.com/projects/audio/044/index.html

This device has a volume adjust such that you could calibrate it.  I think you might have trouble finding a "precision" audio amplifier, at least economically.

Good luck.

[BellMan]

Certainly we could build something like this, for that matter we have an off the shelf stereo reciever/amplifier that we have used in the past, but it still requires manually adjusting a knob and measuring the output. 

I was asked if I could find an amplifier that we could set the output with labview and not have to maually set it every time.  Something that comes with a calbration certifiicate would be even better.

I hear what you are saying about what we might be able to find "economically", but I wasn't necessaryly asked to find something cheap.  I was asked if I could find something to do what we what to do here, and see what it would cost, so I would welcome any options at this point, even if they are expensive options.  I just need to be able to tell my boss what our options are and what they would cost.

Thanks again for your help.

[Seth_B.]

BellMan,

NI does not currently offer anything like that.  What you will basically need is an amplifier with a serial port (or GPIB if you would like to purchase an NI GPIB card).  You can check the NI Instrument Driver site to determine what amplifiers we already have drivers for.

However, I'm a bit confused as to why we will need to programmaticly change the gain on the amplifier.  If you need to alter the output, you could scale your sine wave on the DSA board and see the same effect.

[johnsold]

BellMan,

Back in LV 1.2 days we built a system with a nominal 120 dB dynamic range where we sent calibrated sound level pulses to the test subjects (rats).  We used an off the shelf consumer grade audio amplifier and speakers.  We used a high quality microphone and preamp to calibrate sound pressure levels at various frequencies (and across different amps and speakers).  Using a lookup table we set the levels by programming digitally controlled attenuators.  We did not actually achieve 120 dB dynamic range and could not have measured it if we did.  (We could not convince the user that the system did not need to be that good for what he wanted.)  The system was quite stable and reproducible over the top 60-70 dB and was probably working down to about -80 or -90 dB from the peak although noise limitations prevented good measurements at those levels.  The rats couldn't hear that anyway.

Your speaker measurements are probably ratios of sound pressure levels to power input at various frequencies.  If so, set the amplitudes close to what  you want and measure the power with sufficient accuracy.  This may be much easier than finding or building a precision amplifier.

Lynn 

[Now_With_Underscores]

I would recommend using any of our DSA cards and then piping the output into a 3rd party precision amplifier, such as the HeadRoom Balanced Desktop Amp

[BellMan]

At a minimum what I am looking for here is an amp that is linear I can set the gain to a known value, and preferable it is either known to be linear, or I know the characteristics.

The reason I would like ot be able to set the gain from the program is so I will know exactly what the gain is set to every time.

If the amp gain is set with volume control knob then every measurement has one more opportunity for human error.  If the gain were programable via RS232, GPIB, or other means, I would expect this to be more consistent that just turning a nob.

If the gain is not adjustable then we would have to know the full range we would want to test when we set up the test equipment.  Choose a range that is to large, and we give up resolution.  Choose a range that is too small and we might not be able to test higher powered units in the future.

From what I could tell most of the amps that were listed on the NI drivers site were designed to amplify low voltage senor signals to a higher voltage to improve the resolution of the measurement.  For speaker testing we would need a amp that can push some power.  Our product are using relatively small speakers, but we still get into 1W speakers quite often, and 5W or 10W speakers might not be out of the question in the future.

[AnalogKid2DigitalMan]

/* hack mode on */

Most older amps have the volume knob which is an analog variable resistor (aka potentiometer) be it a linear or audio response curve (nowdays I think the majority of them may use encoders). There are digital potentiometers that accept a digital code to set their resistance. With a little research one could replace the analog pot with an equivalent digital one and control it setting to at least 8 bits of resolution. You just have to figure out the communication protocol and interface, which is already documented by the manufacturer.

/* hack mode off */

Just a thought

-AK2DM