Hi everyone,
I am currently looking for a fairly peculiar power amp for an acoustic laboratory:
- as many channels (i.e., >= 16) with as little form factor as possible
- relatively low power output/channel (i.e., < 10W/channel @ 4-8 Ohm)
- repeatable / time-invariant output (as this amp is supposed to drive small piezos or low-power loudspeakers for an acoustic laboratory, it is crucial that the gain is constant in time and preferably also across channels (definitely necessary)
- ideally with a frequency response as flat as possible from ~500Hz - 10 kHz (not strictly necessary)
- ideally with a variable gain (not strictly necessary)
- Edit: a very low cross-talk between channels is also very important
Does such a system already exist? If so, can someone point me in the right direction, please?
If it does not exist: Is it possible to construct such an amplifier in a reasonable time-frame (e.g., within a few weeks)?
Thanks a bunch,
Teddy
I am currently looking for a fairly peculiar power amp for an acoustic laboratory:
- as many channels (i.e., >= 16) with as little form factor as possible
- relatively low power output/channel (i.e., < 10W/channel @ 4-8 Ohm)
- repeatable / time-invariant output (as this amp is supposed to drive small piezos or low-power loudspeakers for an acoustic laboratory, it is crucial that the gain is constant in time and preferably also across channels (definitely necessary)
- ideally with a frequency response as flat as possible from ~500Hz - 10 kHz (not strictly necessary)
- ideally with a variable gain (not strictly necessary)
- Edit: a very low cross-talk between channels is also very important
Does such a system already exist? If so, can someone point me in the right direction, please?
If it does not exist: Is it possible to construct such an amplifier in a reasonable time-frame (e.g., within a few weeks)?
Thanks a bunch,
Teddy
Last edited:
you might manage a single channel in that time frame. You might even have time to test it properly after you have eliminated all the mistakes in the wiring.
I think this satisfies your requirements. Not sure about the delivery time though.
Free Shipping!!! mk153 LM1875 dual channel amplifier board / 2.0 Stereo / 30W * 2 beyond TDA2030A-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
This is a Solved Problem. Back in the 1920s the telephone company was stringing wires coast to coast and had hundreds of repeater amps to male-up line losses. If amps were 1dB out of gain spec the line would have + or - 100dB of gain error-- unusable. Of course trims were done (even automatic compensation for average temperature over the line segment) but it was also essential to be able to swap-out amps to restore a failed line without careful calibration.
Negative FeedBack moves 99+% of the gain accuracy from dubious tubes (transistors) onto (generally) the ratio of a pair of resistors which can be as perfect as desired. While formerly we might select 5% tolerance resistors for <0.5dB accuracy, today 1% resistors are dirt cheap, and some have better tempco than your lab technicians. Several dozen pair of excellent resistors may cost you 20 bucks.
There are several amp-chips in that power range, and most can be gain-set with two resistors.
How long? I been building a long time, and a 1-channel amp project was on my bench over a year. Lot of tedious wires (and lots of more pressing projects).
Well, do you want FIXED gain or variable gain? If there is a knob, it will be bumped. If your (perhaps digital) source can be level-set with a file, that's probably more "constant" than any knob.
Something like an LM1875 or LM3886 running on a ±15 V supply would be able to provide 10-15 W.
Your main challenge will be driving the piezoelectric drivers as they tend to be rather capacitive.
For someone with significant circuit design experience, a few weeks of design time is a bit aggressive but not impossible. You'll need to add a few weeks for circuit board manufacturing and assembly. Then add a chassis.
Tom
Your main challenge will be driving the piezoelectric drivers as they tend to be rather capacitive.
For someone with significant circuit design experience, a few weeks of design time is a bit aggressive but not impossible. You'll need to add a few weeks for circuit board manufacturing and assembly. Then add a chassis.
Tom
Here ya go.. The input(s) could be all common to each other to reduce possible input ground loop issues. Analog to reduce possible RF interference problems produced by digital based chip amps. Parallel as many as needed..
TDA2030A Module Single Power Supply Audio Amplifier Board Module-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
TDA2030A Module Single Power Supply Audio Amplifier Board Module-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
😕 😕 😕
I didn't mention the '7294. Or low gains. And the '7294 is a 100W chip, far past the "<10W" that the OP seems to be wanting.
Yes, all pre-baked amps have a minimum gain; and above the laptop field, often 20 or so.
The sheet-spec for '7294 is 24dB (about 16).
I'd suggest LM675 if low gains are what you're after (its a lower gain capable version of LM1875 if I recall)
I once looked at an ideal like this a 10+ channel amplifier per stereo channel in order to power a line array with multiple channels of DSP time alignment - as apposed to the simplified line arrays often seen here.
I once looked at an ideal like this a 10+ channel amplifier per stereo channel in order to power a line array with multiple channels of DSP time alignment - as apposed to the simplified line arrays often seen here.
- Status
- Not open for further replies.