Phono pre-amp function and other questions.

[jheatac]

My brother gave me a cheap phono preamp by Pyle

The only active device is a F4558 dual op amp. Aside from that, a couple voltage dividers and a few filtering capacitors and an "on" LED. The unit operates off of a 9v battery or a 12v (marked) wall wort that puts out about 19v at open circuit. 18.8v at input with amp on. Then it runs through a resistor and a diode to provide 12v to the device.

He says it doesn't work, but it appears to work to me, other than I think it is just a garbage device.

I have 3 pictures of the scope attached. All 3 are 2mv +/-. They are 100hz, 1khz, 10khz.

I can't understand how you can have an audio amplifier and not have a negative rail voltage - but that may just because I don't know what I am doing.

The 100hz seems like an acceptable output - 2mv in, 750mv out. Wave is slightly offset of center
The 1khz, the gain shrinks considerably and the output is completely positive - 2mv in, 200mv out. Wave is completely above zero
The 10khz I had to change the scale to 50mv/div, but it is has a considerable ?dc offset? - 2mv in, 50mv out, 80mv offset.

10khz is high for audio, but I am just trying to understand more than anything.

While technically the same tone from the speaker, this output would only push the speaker and not pull it - correct?

OP amp is 12v+ and ground for input. Audio goes in the non-inverting input.

I realized that I had the input probe set at 10x so that has the input wave scaled wrong, but the signal generator was 2mv

How did I get more than a 2mv negative wave on the output without a negative rail?

[BrianHG]

Do you have a 50kohm load to GND on the output?
The output should have a series DC filter cap.  Maybe said cap is old and leaking, so your 10megaohm scope probe might not be able to load the DC filter output cap properly.

Also, phono pre-amps are not flat.  They boost the low frequencies by an order of magnitude compared to the >10khz.  Note that this intentional filter has a different gain and slope characteristic depending if the pre-amp is designed for a MM (Moving Magnet) cartridge, or a MC (Moving Coil) cartridge.


That 'DC' shift may be happening somewhere else in the circuit, like at the low frequency compensation filter.  In this case, it may be dying caps, or a fault in the design.

[TimFox]

Search for “RIAA equalization” to see the frequency response curve for a magnetic phono cartridge.
“Moving magnet” cartridges usually are designed for 47 k\$\Omega\$ input impedance.
“Moving coil” cartridges have much lower impedance and output voltage, but require the same equalization response.

[BrianHG]

[jheatac]

Do you have a 50kohm load to GND on the output?

I do not. With my minimal knowledge I was assuming the output would essentially hit the input of an op-amp at the line-in input of whatever it is attached to. I will add a 50k resistor to the output tomorrow and re-check.

I didn't realized they were not an equal amplification across the board.

I may try to draw a schematic and send a couple pictures tomorrow as well. There is really nothing to this thing.

Thank you for taking the time to respond.

[BrianHG]

RIAA gain curve:

[jheatac]

That's very interesting. It makes sense to me now, but it isn't something I had ever given thought to.

I don't think this device is complex enough to do that intentionally but maybe it is. I'll learn a little more tomorrow.

Thanks again.

[edpalmer42]

I couldn't stand to watch that video.  I found it long and tedious.  So, this info might be in there.

My understanding of the purpose of the RIAA equalization is that it has nothing to do with your hearing.  Its purpose is to compensate for characteristics of the vinyl.

The recording side cuts the low frequencies to reduce the amount of room that they take on the record.  This allows more music to be placed on the record.  The playback side boosts the bass to return it to the proper level.

Meanwhile, the recording side boosts the high frequencies because vinyl tends to have high frequency hiss when played back.  The receive side cuts the high frequencies to return the levels to normal and, in the process, attenuates the hiss which improves the overall SNR of the medium.

Since the circuit has full control of the impedances within the unit, it's trivial to implement these filters with simple RC stages that are carefully designed to have their effect at the appropriate frequencies.  In a high quality preamp, you might find 1% tolerance components for these filters.

You shouldn't have any trouble finding schematics online that show real-life implementations of these filters and associated amps.  They're kind of a jelly-bean circuit.

Ed

[BrianHG]

Its in the recording side where they cut the bass frequencies.  They had to do this since a very strong cut in the vinyl for a loud bass drum would physically knock the needle out of the track causing a skip.  This is why the pre-amp needs to boost the gain on the low frequencies by an additional 20db.

[jheatac]

Here is a schematic of this little preamp. I also included a board picture with my color coded traces.

I don't have values for the ceramic SMD capacitors

The op amp is a SOP dual (F4558) which I could not find in LTSpice.  In looking at how basic the schematic is - is the skew to the amplification built into the op amp?

[tooki]

I can't understand how you can have an audio amplifier and not have a negative rail voltage - but that may just because I don't know what I am doing.

Indeed. Single-supply amps are commonplace. You basically bias your signal somewhere between the Gnd and supply voltage (typically at 1/2) and capacitively couple your signal.

While technically the same tone from the speaker, this output would only push the speaker and not pull it - correct?

Well, you wouldn’t run a speaker directly off the preamp, and one would expect/hope that the power amplifier would have its inputs AC-coupled (capacitively coupled) to ensure no DC bias gets in.

But in theory, if your entire signal chain was DC-coupled and had a significant DC bias, then you’d be correct that you’d only be pushing the speaker cone.

How did I get more than a 2mv negative wave on the output without a negative rail?

Again, AC coupling using capacitors. They strip away the DC bias, recentering the wave at 0V.

[tooki]

I don't think this device is complex enough to do that intentionally but maybe it is. I'll learn a little more tomorrow.

Sure it is. I suspect that many cheap devices implement a simplified, practically straight (i.e. no humps) curve. Remember, the RIAA curve was conceived in the 1940s*, so had to be capable of inexpensive implementation using 1930s-1940s technology. That meant very simple filter circuits using tubes. Compared to those, an op-amp based filter is luxury.

*The record companies fought back and forth for years about which curve to use, so it wasn’t until the 1950s that they finally agreed on this specific curve. But equalization curves had been in use since the 1930s, with each record company using its own.

[tooki]

Here is a schematic of this little preamp. I also included a board picture with my color coded traces.

I don't have values for the ceramic SMD capacitors

The op amp is a SOP dual (F4558) which I could not find in LTSpice.  In looking at how basic the schematic is - is the skew to the amplification built into the op amp?

If it has any 4558 (or you find the spice model to download elsewhere), use that. And if not, choose something like a 5532. It’s not identical but will perform similarly enough for this purpose.

[wasedadoc]

C7, C8,  R9, R10 are the RIAA shaping feedback network for one of the channels.

[TimFox]

A magnetic cartridge (as opposed to piezoelectric) gives an AC voltage proportional to the needle velocity.
That velocity is proportional to the product of amplitude times frequency.
Since the amplitude must be less than the radial spacing between grooves, the minimum adjustment to the frequency response would give constant amplitude with respect to frequency and velocity proportional to frequency.
The “equalization” is a further adjustment to improve the overall signal-to-noise, consistent with the power spectrum of the music to be recorded and played back.  Without equalization, the playback response (for magnetic cartridge) would be just 1/f, or -6 dB/octave.
Analog magnetic tape has similar requirements, since the recorded flux density is limited by the saturation of the magnetic material, but the standard equalization functions (e.g. NAB) are somewhat different than the RIAA curve.