I need to design an audio amplifier as shown in the attachment. Since, I'm hearing impaired, I don't know anything about audio system design. I did do what research I could on each of three sections for this question.

PREAMPLIFIER:
-placed close to the to the sensor to reduce noise and interference
-it is used to boost the signal strength to drive the cable to the main instrument without degrading signal-to-noise ratio
-noise performance is critical
-provides voltage gain, but no current gain
-has volume control
-three types: current sensitive
parasitic capacitance
charge sensitive

MAGNETIC CARTRIDGE
-a mechanical device that converts vibrational energy from a stylus riding in a record groove into an electrical signal

MOVING COIL CARTRIDGE
-an electromagnetic generator with the coils attached to the stylus
-output signal is a few hundred microvolts and easily swamped by noise

RIAA COMPENSATION
-for the playback of phonograph records
-for permitting greater recording times
-pre-emphasis on recording and de-emphasis on playback
-not a simple low-pass filter. Transition points in 3 places: 75us, 318us, and 3180us, which corresponds to 2122Hz, 500Hz, and 50Hz

POWER AMPLIFIER
-drives the speakers
-amplifies low power signals at frequencies between 20Hz and 20kHz
-input is a few hundred microwatts
-keys: frequency response
gain
noise
distortion

EQUALIZER
-adjusts balance between frequencies
-adjust signal strength of specific frequencies

I found a sample circuit for RIAA compensation and I've uploaded that picture. I just don't know what's in the box "RIAA_INV". I'm not clear on how this part is supposed to work.

I'll need to find simple circuits for each of these sections preferably using MOSFET's because that's what I'm most familiar with. I'm guessing that, for the preamplifier, I'll need a differential amplifier along with some other sections. That sounds like I could use a basic op-amp using discrete components and give it enough of a gain to get up to the 10V asked for. I've attached a simple op-amp design.

I don't have circuit ideas for the other sections yet. Can anybody give me some ideas for a circuit?

Thanks for your help,
Mike

 

http://sound.westhost.com/dwopa.htm

 

That is one hell of a homework problem. It would take an experienced team of designers at Kenwood a few months to complete this work. Not counting the parts that are impossible or undefined, like an RIAA compensating preamp out to 100 kHz.

To answer your question, inverse-RIAA is a test device, not a part of thee stereo amp system. Once you have designed and built a preamp with RIAA compensation, there are two ways to test it to make sure the compensation is accurate. One is to calculate the exact gain at several frequencies through the audio band such as octaves or half-octaves, connect the input to a sinewave oscillator at one frequency, measure the output, compare it to the calculated output, then move on to the next frequency and repeat. When you have all the data, plot the results of your preamp on top of the calculated RIAA curve and see how accurately they match. That takes almost forever.

Another way is to design and build a specialized filter, a frequency shaping network that is the inverse of the RIAA compensation in the preamp. The net effect of a signal going through the inverse network and then through the preamp should be that the signal does not change at all. Now you can just adjust the sinewave oscillator to any frequency and see if the output equals the input, a much easier thing to see or measure without any calculations.

Elliot Sound Products (westhost.com) has many pages of great design guidance for audio products.

ak

 

I'm not clear on what RIAA compensation does. What I found suggests it's part of a phonograph record head, the part that reads the record. Is it supposed to boost those 3 frequencies while leaving the rest alone? Also, where should the circuit for that go?

Thanks,
Mike

 

A phono cartridge, either moving magnet or moving coil, is an inductor, and its output voltage is proportional to frequency. The same is true for the cutting head that makes the master disc from which phonograph records are pressed. The audio bandwidth of 20 Hz to 20 kHz is 1000:1 frequency span. For a single pole element, that is a 60 dB change in output voltage from one end to the other, assuming constant energy in. That means that if you put flat audio into the cutting head, its range of motion has to be greater than 1 million to 1. That's a lot to ask of a mechanical system designed in the 30's. So the frequency response of the incoming audio is adjusted to reduce the range of motion required of the cutting head. The curve boosts the high frequencies relative to the low frequencies. To restore the audio band to its original frequency characteristics, a compensating frequency response curve is applied in the photo preamp, attenuating the highs so they are back to their original relative amplitudes with respect to the lows. The curve is not a simple single-pole highpass or lowpass filter because again electronics in the 30's had a tough time with 60 dB amplitude ranges. So the RIAA equalization applied in a playback preamp looks like a single-pole rolloff with a flat part in thee middle. The longest time constant starts the rolloff at 50 Hz. The middle time constant at 500 Hz cancels out the first one and so the response curve goes flat as frequency continues to increase. The third corner freq at 2122 Hz re-establishes the -6 dB/octave rolloff out to 20 kHz. Because of the step in the middle, the overall gain difference from 20 Hz to 20 kHz is 40 dB, not 60 dB.

The standard place for RIAA equalization is in an opamp stage at the phono input. The general idea is that after equalization, the phono signal is flat and at the same amplitude as the direct input signals from a tuner, CD player, or line input. That way, everything after the selector switch (tone controls, equalizers, power amplifiers, etc.) is the same for all inputs.

Or something like that; working from memory.

ak

 

Can I use the attached circuit for the preamplifier? The differential amplifier section helps suppress noise, but is it enough for audio? Also, the design calls for a distortion of less than 0.02%. What is distortion?

Thanks,
Mike

 

Can I use the attached circuit for the preamplifier?

You are not designing an Integrated Circuit, the innards of which is what you posted.

I suggest that you should be looking at applying off-the shelf standard I.C parts to your project; not designing a circuit that needs to be fabricated in an I.C. foundry...

 

Can I use the attached circuit for the preamplifier? The differential amplifier section helps suppress noise, but is it enough for audio? Also, the design calls for a distortion of less than 0.02%. What is distortion?

Thanks,
Mike
View attachment 96479

If you put a pure sinewave, at a single frequency, into an amplifier you should get a pure sinewave, at a single frequency, out. That would be 0% distortion.

Why are you using 5V for your Vcc?

 

Elliot Sound Projects has a lot of information concerning audio projects. I suggest you visit them.

The output of the magnetic cartridge can be the "inverse RIAA" curve. Your RIAA circuit will take that cartridge output and produce a flat response.

If you are looking to simulate the "inverse RIAA", there is a project at ESP which refers to this article as the source.

ESP project 91 describes all the 78 RPM and RIAA standards.

 

I can't use IC's for this assignment, but I can use the innards, which is why I posted that circuit. The 5 volts was for another project, not this one. I'll probably use something like 15 or 20 volts for this one because I need an output of 10V from the preamplifier. Thanks for the link's Joe.

 

You could probably just cut highs and boost lows with a simple op amp type EQ Circuit and get acceptable phonograph playback results unless RIAA specs are required. Would make the filters much simpler.

 

I can use IC's for the equalizer section only. What do you mean cut high's? How would I do that in an op-amp?

Mike

 

I can't use IC's for this assignment, but I can use the innards, which is why I posted that circuit. The 5 volts was for another project, not this one. I'll probably use something like 15 or 20 volts for this one because I need an output of 10V from the preamplifier. Thanks for the link's Joe.

You also can't use the innards of an IC to build a real circuit either. You actually need real parts to do that and you haven't indicated any real parts on your schematic so it's hard to figure out what you might be up to. Your active devices look like FETs but the non-standard symbols are all the same so I'm pretty sure if you make it out of real parts it won't work like you expect it to.

 

Alright, does anyone have a simple circuit design I can use for the preamplifier? I'm pretty sure the input will be around 300uV, which could be boosted with a voltage divider at the input. A differential amplifier should be used for the input to reduce noise. I don't know what other methods can be used to eliminate noise. What else would I need?

 

Alright, does anyone have a simple circuit design I can use for the preamplifier? I'm pretty sure the input will be around 300uV, which could be boosted with a voltage divider at the input. A differential amplifier should be used for the input to reduce noise. I don't know what other methods can be used to eliminate noise. What else would I need?

A voltage divider will not boost the input, they only attenuate things.

There are tons of preamplifier circuits on the web that you can understand, build and test.

Here is one using a common BJT, the 2N3904
http://www.rason.org/Projects/bipolamp/bipolamp.htm

Here is one using common JFETs, the MPF102
http://www.rason.org/Projects/jfetamp/jfetamp.htm

Here is one using a quality amplifier chip
http://sound.westhost.com/project88.htm

Google "audio preamp circuit design" and take your pick.