ARTA sweep of Dayton RS150 (just free air)

[contoursvt]

Hi everyone, not sure what i'm doing wrong - maybe everything, but I download ARTA and did a quick measurement of a pair of Dayton RS150 drivers which will end up as a mid for a 3 way system (200Hz-2Khz). I thought I'd look and see if I can see if the breakup of the metal cone is similar to what Dayton has on their graphs.

Following the below guide and using an ECM8000 microphone, I thought I'd be all set. Oh the audio interface was a Steinberg CI1 USB audio interface.
http://audio.claub.net/tutorials/FR%...ing%20ARTA.pdf


I put the driver on a tripod, facing forward at the same height as the microphone about 1 to 1.5 meters away. To make sure the driver didnt go anywhere, I used a thick rubber band that secured it around the magnet and held it to the tripod. Its in a fair sized room but lot of hard reflective surfaces such as wood floor, glass tables..etc, but thats why I thought the ARTA impulse measurement would be good. I'd basically ignore the reflections from what I understood.

So I'm confused as to why i'm not seeing the huge metal cone breakup and the graph looks too well behaved for me to believe seeing how its not in a cabinet and there is no crossover on it.






So to compare with something else, I took a spare 8" Camber driver which I used in my last DIY 8" two way project and a Vifa P17WJ00



Camber driver





Vifa P17WJ





So can I trust the measurement or could I be doing somthing wrong here? The fact that I have a dip in my Dayton RS150 results at 5K instead of a big peak is whats throwing me off...

Thanks in advance for any suggestions.

[critofur]

Peaks can be flattened if the mic is clipping.

Obviously, very little bass when the driver is not mounted in a baffle or box.

What does it look like if you measure it from about 6 inches away (be sure that the volume is set so that the mic is not clipping)?

For reference, do you have a tweeter you can measure (preferably mounted in a baffle).

[contoursvt]

Thanks critofur, actually I had tried a couple tweeters last night as well but I wasnt sure if it was relevant so I didnt post them. This is what I got for the tweeters - same 1 meter away but they were not on a baffle, they were just open.

Vifa D27TG45



Dayton RS28F



The volume levels and everything was exactly the same since it was the same session of trying to get measurements.

Peaks can be flattened if the mic is clipping.

Obviously, very little bass when the driver is not mounted in a baffle or box.

What does it look like if you measure it from about 6 inches away (be sure that the volume is set so that the mic is not clipping)?

For reference, do you have a tweeter you can measure (preferably mounted in a baffle).

[scottsehlin]

Peaks can be flattened if the mic is clipping.

Obviously, very little bass when the driver is not mounted in a baffle or box.

What does it look like if you measure it from about 6 inches away (be sure that the volume is set so that the mic is not clipping)?

For reference, do you have a tweeter you can measure (preferably mounted in a baffle).

I suspect your issue has been found. All of your plots top out at the same relative dB level (approximately +5). This indicates some clipping is likely taking place.

[contoursvt]

Awesome, I'll reduce the volume a bit (I had it set to be about 80 to 85db) and I'll try again. I never thought it might be clipping because it didnt look harsh. Thanks guys and I'll let you know what happens

[cjd]

Awesome, I'll reduce the volume a bit (I had it set to be about 80 to 85db) and I'll try again. I never thought it might be clipping because it didnt look harsh. Thanks guys and I'll let you know what happens

How are you measuring 80-85dB? I recommend putting an 8ohm resistor on the amplifier and measuring for 2.83v on the output side. Also check input amplification - it's entirely possible your amp is clipping there, not the mic. At any rate, 80-85dB at the mic should be well within its comfortable operating window.

C

[contoursvt]

I have a radioshack spl meter roughly where the mic is so when tge sweep happens, it registers a peak of around 85db or so. I never thought about the amp clipping but it might be possible now that you mention it since its a tripath 2020 based amp ... 6-7 watts rms per channel. Only other amp i have handy is a 200wpc soundcraftsmen so it may be overkill. Im reducing the volume a bit and trying again to compare results for now

How are you measuring 80-85dB? I recommend putting an 8ohm resistor on the amplifier and measuring for 2.83v on the output side. Also check input amplification - it's entirely possible your amp is clipping there, not the mic. At any rate, 80-85dB at the mic should be well within its comfortable operating window.

C

[cjd]

i meant the mic amp

also last i looked tripath amps werent usable in this setup.

[contoursvt]

Ah i see. In this case the mic amp is the steinberg ci1 usb soundcard. I did back off on its gains and then things improved. So tripath amps not good for testing with?

[Pete Schumacher ®]

Ah i see. In this case the mic amp is the steinberg ci1 usb soundcard. I did back off on its gains and then things improved. So tripath amps not good for testing with?

The tripath and some other switching amps don't necessarily provide the best load regulation, especially at higher audio frequencies, due to the output filter not being inside the feedback loop.

You would be better off with another small wattage class A/B. You won't need more than a 10W amp for sweeps, unless you're intent on doing thermal compression tests.

[contoursvt]

In that case I will just use the soundcraftsmen that I have. A bit big for the job but its a spare so no sense in buying another amp for now

The tripath and some other switching amps don't necessarily provide the best load regulation, especially at higher audio frequencies, due to the output filter not being inside the feedback loop.

You would be better off with another small wattage class A/B. You won't need more than a 10W amp for sweeps, unless you're intent on doing thermal compression tests.