Hello, yet again. I'm trying to hook a piezo disc, as a contact mic, up to the 2.5w mono amp and a 3w 4 ohm speaker. It makes sound, but it's pretty low. Whole setup works just fine and is decently loud enough w/ audio from ipod/computer thru a regular 1/8 in input. Wondering how I'd go about increasing the piezo mic volume here. Considered possibly wiring 2 2.5w amps together somehow, and adding another 4ohm 3w speaker - but not sure how I'd do this or if it'd work.
Thoughts? Thanks, YET AGAIN, for your help!
2.5 watt mono amp - increasing DBs.
Moderators: adafruit_support_bill, adafruit
Please be positive and constructive with your questions and comments.
- adafruit2
- Posts: 22194
- Joined: Fri Mar 11, 2005 7:36 pm
Re: 2.5 watt mono amp - increasing DBs.
you will need to pre-amp the contact mic to 'line level' voltages, e.g. 0.7vpp or so
check out online for piezo pre-amp circuits for some ideas!
check out online for piezo pre-amp circuits for some ideas!
- 203
- Posts: 11
- Joined: Mon Jan 11, 2016 7:06 pm
Re: 2.5 watt mono amp - increasing DBs.
Actually tried that right from the beginning, using this circuit - http://i2.wp.com/cdn.makezine.com/uploa ... =600%2C403, with the piezo's input ground wired to amp's A-, the + of the piezo (coming from the preamp's + out) to the A+ of the amp, and a 7805 to drop the 9 volts down to 5 for the amp, but the amp/speaker didn't like that very much - lots of popping in addition to a little sound. Using the preamp circuit above tho, I can plug right into a stereo or mixer and get quite decent boost and DBs. I wonder now, maybe the 2.5 watts just isn't enough for something like a piezo contact mic? But that seems odd too. It DID make sound through a 4ohm 3 watt, just ever so little...
- 203
- Posts: 11
- Joined: Mon Jan 11, 2016 7:06 pm
Re: 2.5 watt mono amp - increasing DBs.
OH, actually, let me clarify - piezo input ground & amp ground all wired to 9v ground. piezo+, after the preamp, wired to A+ on amp, and then amp vIN to batt+ and amp ground to batt ground.
- adafruit_support_mike
- Posts: 67485
- Joined: Thu Feb 11, 2010 2:51 pm
Re: 2.5 watt mono amp - increasing DBs.
Hmm.. one of Collin's old circuits. ;-)
JFET preamps are great for working with high-impedance sources, but they usually want about 10v of supply voltage. For 5v and below, it's easier to work with a mosfet-input op amp:
For the values given, it will have gain of about 20. To adjust the gain, replace the 22k resistor with a 10k and a 10k pot whose wiper is connected to the MCP6001's output.
The MCP6001 is a jellybean op amp that happens to have an input leakage current of about 3pA, which is comparable to a JFET.
The voltage dividers at the input will center the op amp's input and output around VCC/2, giving you the widest possible swing.
JFET preamps are great for working with high-impedance sources, but they usually want about 10v of supply voltage. For 5v and below, it's easier to work with a mosfet-input op amp:
For the values given, it will have gain of about 20. To adjust the gain, replace the 22k resistor with a 10k and a 10k pot whose wiper is connected to the MCP6001's output.
The MCP6001 is a jellybean op amp that happens to have an input leakage current of about 3pA, which is comparable to a JFET.
The voltage dividers at the input will center the op amp's input and output around VCC/2, giving you the widest possible swing.
- 203
- Posts: 11
- Joined: Mon Jan 11, 2016 7:06 pm
Re: 2.5 watt mono amp - increasing DBs.
AHA. - Thanks for the incredibly thorough response -! Very very curious to see how this works.
Just to clarify, and out of novice curiosity - are you saying that the LM7805 somehow didn't play nice with the 2.5w? That the 9v powering Collin's circuit and the drop to 5 required for the amp just wouldn't work well?
Just to clarify, and out of novice curiosity - are you saying that the LM7805 somehow didn't play nice with the 2.5w? That the 9v powering Collin's circuit and the drop to 5 required for the amp just wouldn't work well?
- adafruit_support_mike
- Posts: 67485
- Joined: Thu Feb 11, 2010 2:51 pm
Re: 2.5 watt mono amp - increasing DBs.
JFETs have wonderful qualities.. among the highest input impedance, lowest noise, and fastest response of any active device. You can find other devices that match them on one quality, or maybe two, but no other device hits all three. They don't do low voltage without a fight though.
JFETs are normally-open devices, and for N-channel JFETs you pull the gate voltage lower than the source voltage to shut them off. For most JFETs, the gate has to be 5v to 15v lower than the source to shut current off completely.
Any JFET will have a current level that puts the source voltage near the gate voltage, but that brings us to the second weakness of JFETs: their tolerances suck. If you take two adjacent JFETs from the same reel (usually meaning they were fabricated side by side on the same wafer), one might zero the gate-source voltage at 3.5mA and the other might do it at 6.5mA.
The processes for making JFETs are actually very good, but they're trying to measure one dopant atom per 1e15 atoms of silicon, which is tough. Worse yet, the physical locations of the dopant atoms in the silicon matrix have an effect on a JFET's performance. We won't have close-tolerance JFETs until we learn how to build them one atom at a time.
You can ignore the tolerance issues if your supply rails are far enough apart. In the days of vacuum tubes, you'd lose 20v across a diode, so 5v was practically precision equipment. For the voltage levels we use these days, about half of any circuit that uses JFETs is devoted to keeping them in the expected working range.
A typical design looks like this one by analog guru Jim Williams:
The NPN transistor controls the current through the JFET. The two 10M-100nF filters track the average DC voltages of the input and output with a time constant of about 1 second. The op amp watches the two filters and adjusts the JFET current to keep the average values the same. The buffer provides power to drive the output, and the diode protects the JFET from reverse current.
That one is just a JFET buffer, so it doesn't have any gain.
JFET designs like Collin's above have to be lovingly hand-crafted if you want to make them work between rails much closer than 10v apart. You have to tweak the values of the resistors above and below the JFET so the average output voltage is near the center of its working range, and then tweak the ratio for the right amount of gain. You also have to leave room for the gain to change as the JFET's performance drifts, which it will, in response to both current and temperature.
JFETs are normally-open devices, and for N-channel JFETs you pull the gate voltage lower than the source voltage to shut them off. For most JFETs, the gate has to be 5v to 15v lower than the source to shut current off completely.
Any JFET will have a current level that puts the source voltage near the gate voltage, but that brings us to the second weakness of JFETs: their tolerances suck. If you take two adjacent JFETs from the same reel (usually meaning they were fabricated side by side on the same wafer), one might zero the gate-source voltage at 3.5mA and the other might do it at 6.5mA.
The processes for making JFETs are actually very good, but they're trying to measure one dopant atom per 1e15 atoms of silicon, which is tough. Worse yet, the physical locations of the dopant atoms in the silicon matrix have an effect on a JFET's performance. We won't have close-tolerance JFETs until we learn how to build them one atom at a time.
You can ignore the tolerance issues if your supply rails are far enough apart. In the days of vacuum tubes, you'd lose 20v across a diode, so 5v was practically precision equipment. For the voltage levels we use these days, about half of any circuit that uses JFETs is devoted to keeping them in the expected working range.
A typical design looks like this one by analog guru Jim Williams:
The NPN transistor controls the current through the JFET. The two 10M-100nF filters track the average DC voltages of the input and output with a time constant of about 1 second. The op amp watches the two filters and adjusts the JFET current to keep the average values the same. The buffer provides power to drive the output, and the diode protects the JFET from reverse current.
That one is just a JFET buffer, so it doesn't have any gain.
JFET designs like Collin's above have to be lovingly hand-crafted if you want to make them work between rails much closer than 10v apart. You have to tweak the values of the resistors above and below the JFET so the average output voltage is near the center of its working range, and then tweak the ratio for the right amount of gain. You also have to leave room for the gain to change as the JFET's performance drifts, which it will, in response to both current and temperature.
Please be positive and constructive with your questions and comments.