Hi all,
I'm trying to put together a recording system where I can simultaneously record a mic input and the signal generated by a vibration transducer Knowles BU-27135-141, which is attached to the speaker's neck. This has been done in the voice research field to estimate the aerodynamic pressure close to the vocal folds. Anyway, I've been researching for the simplest way to implement this and converging on using a Scarlett Solo as the A/D solution. However, I'm not 100% sure if it's as simple as I'm making it and wondering if anybody here could comment on the validity before I spend money on it.
From the linked datasheet above, the signal range is +/-1.5 V with the spectral response:
The vibration is lowpass and is expected to roll off steadily from the first harmonic of the voice. So the >3 kHz amplification shouldn't hurt.
The circuit shown on the datasheet is pretty darn simple, the sensor is driven by one AA battery, and the vibration (with dc bias) is measured across the sensor output.
Because the Scarlett Solo's input appears to be AC coupled (freq range 20-20kHz), can I simply hook up these output terminal directly to the Scarlett Solo's instrument input? I think I want to use the instrument mode to keep the battery drainage minimal. But, I really don't know a whole lot about for anything analog circuit or audio recording.
I'd greatly appreciate any inputs. Thanks! -Kesh
I'm trying to put together a recording system where I can simultaneously record a mic input and the signal generated by a vibration transducer Knowles BU-27135-141, which is attached to the speaker's neck. This has been done in the voice research field to estimate the aerodynamic pressure close to the vocal folds. Anyway, I've been researching for the simplest way to implement this and converging on using a Scarlett Solo as the A/D solution. However, I'm not 100% sure if it's as simple as I'm making it and wondering if anybody here could comment on the validity before I spend money on it.
From the linked datasheet above, the signal range is +/-1.5 V with the spectral response:
The vibration is lowpass and is expected to roll off steadily from the first harmonic of the voice. So the >3 kHz amplification shouldn't hurt.
The circuit shown on the datasheet is pretty darn simple, the sensor is driven by one AA battery, and the vibration (with dc bias) is measured across the sensor output.
Because the Scarlett Solo's input appears to be AC coupled (freq range 20-20kHz), can I simply hook up these output terminal directly to the Scarlett Solo's instrument input? I think I want to use the instrument mode to keep the battery drainage minimal. But, I really don't know a whole lot about for anything analog circuit or audio recording.
I'd greatly appreciate any inputs. Thanks! -Kesh
Here is a schematic of the Solo. It already has blocking capacitors at both +/- LINE input nodes.
Do NOT use the MIC input, which has 48VDC bias present.
Do NOT use the MIC input, which has 48VDC bias present.
The impedance levels are moderate, so avoid noise pickup as much as possible. Don't forget an on/off switch to save the battery.
Keep the circuit compact, and twist leads from the transducer to the jack. Same for the switch, battery, and resistor leads to the jack.
You may want a high pass filter to reduce low frequency noise pickup, which would be a series bipolar/nonpolar capacitor.
If you have excessive high frequency noise pickup, a small capacitor across the transducer could help.
Keep the circuit compact, and twist leads from the transducer to the jack. Same for the switch, battery, and resistor leads to the jack.
You may want a high pass filter to reduce low frequency noise pickup, which would be a series bipolar/nonpolar capacitor.
If you have excessive high frequency noise pickup, a small capacitor across the transducer could help.
This is a good idea. I wasn't thinking too much about it as the frequency range of my interest is up to 2 kHz, and the recorded signal is not meant to be listened to (just to be analyzed numerically). I'll keep my eyes on the need for a more elaborate hpf when I build the "1.5V phantom power" box for the transducer.
The switch is not necessary because the transducer will be connected to the box with a lead cable (~3ft from waist to neck). So, simply unplugging the transducer from the box will cut off the circuit to stop battery drain. My plan for this connectivity is to use a 1/8" TS jack on the box and use a run-of-a-mill 1/8" mono cable. But I didn't think about the nature of these cables. Do you think they are sufficiently shielded?
@kesh Great find. I’ve been thinking about how to measure speaker cabinet resonances, do you think this would work for that?
Also I’m confused about the data sheet schematic, it seems to be showing how they measured the device, but is that how you are supposed to use the device in practice? Confusing because there is an output pin on the device that looks to be unused.
Also I’m confused about the data sheet schematic, it seems to be showing how they measured the device, but is that how you are supposed to use the device in practice? Confusing because there is an output pin on the device that looks to be unused.
Whether the current is off depends on the wiring, but if the transducer is unplugged from the rest of the circuit, that will be ok.
Give the mono cable a try, but if there is excessive noise pickup, use a shielded twisted pair.
@maFrodite - I'd say yeah, why not. That would be a whole lot easier task to attach it to a face of a speaker cabinet than a neck of a human lol.
Voice researchers have been using this type of setup for at least a decade to measure vocal tract pressure, and the signals that they include in the papers look respectably clean (hopefully they aren't just showing the cream o the crop). I'm not savvy enough in speaker measurements or vibrations/acoustics to give you a definitive answer though. That being said, if you already have an A/D, this shouldn't cost more than US$100. So, it may be worth a try.
re: schematic, you and me both lol. I was very underwhelmed by the datasheet, but that's all there is to it as far as I could find. Others using this sensor indeed just leave the middle terminal unconnected.
Voice researchers have been using this type of setup for at least a decade to measure vocal tract pressure, and the signals that they include in the papers look respectably clean (hopefully they aren't just showing the cream o the crop). I'm not savvy enough in speaker measurements or vibrations/acoustics to give you a definitive answer though. That being said, if you already have an A/D, this shouldn't cost more than US$100. So, it may be worth a try.
re: schematic, you and me both lol. I was very underwhelmed by the datasheet, but that's all there is to it as far as I could find. Others using this sensor indeed just leave the middle terminal unconnected.
Would be worth testing both their test circuit, and a straight 1.5v DC power connection. Measure vibrations from the middle output pin referenced to the negative pin and see if it’s any different.
Or I was wondering if the middle pin could be the ground and + and - are balanced output. I'll find out.
Re: speaker resonance. The frequency range may be an issue for you. The transducer has flat response only up to 3 kHz, so likely it won't be able to provide much info for the audible high frequency range. Donno how critical it is for studying the speaker resonance.
Re: speaker resonance. The frequency range may be an issue for you. The transducer has flat response only up to 3 kHz, so likely it won't be able to provide much info for the audible high frequency range. Donno how critical it is for studying the speaker resonance.
@rayma - would you consider Mogami W2552 mic cable a better option to wire transducer to the phantom power box?
That cable is ok as long as you leave the transducer's center terminal floating. The conductors are not twisted, though.
Then you can float the shield on the transducer end of the cable, and ground the shield on the Solo end of the cable.
Then you can float the shield on the transducer end of the cable, and ground the shield on the Solo end of the cable.
That makes sense about connecting the shield only on one end. To be clear, this cable only goes to the battery (3" away), and a 1/4" TS patch cable connects the battery unit to Solo.
BTW, how did you figure out it's not twisted? I was hoping it is because it's made for mic but 🤷♂️
BTW, how did you figure out it's not twisted? I was hoping it is because it's made for mic but 🤷♂️