I built a Pearl 2 and it works but has too much gain. Even with R16 at 15K it gives distortion on loud passages, I'm using a moving magnet cartridge. Is there a way to reduce the gain of the first stage? I tried 27R for R21 - R24 but the top end disappears.
In the 40dB gain output stage you could reduce the 100k resistor R16, to around 30k instead.
But check for stability when doing so.
Don't mess with the first stage, it won't be doing the overloading.
But check for stability when doing so.
Don't mess with the first stage, it won't be doing the overloading.
Have you tried searching for an answer in the original Pearl Two thread? I seem to recall discussions from some with similar issues.
The first stage has a gain of 35dB ( x56 ), working on +24VDC.
Even if your MM cartridge puts out 100mV peak (unusually high), that's only 5.6V peak before the RIAA network.
Look to the second stage, after the RIAA, for the problem. To repeat, don't mess with the first stage.
No need for an inverse RIAA to troubleshoot this problem. You can just use a sine wave.
Set the source to 1V, with an L pad attenuator of series 10k and shunt 10R. This gives 1mV out.
Input a 1kHz sine and see if there is overloading before the series RIAA resistor. There won't be.
Then increase the input until you see overloading. Adjust the input attenuator as needed for the larger input.
Change the series resistor first to 1k, then to 100R. Probably it will not overload until about 200mV at the input,
after the L pad.
Then look at the main output under the same conditions. It will overload sooner, as it should.
Even if your MM cartridge puts out 100mV peak (unusually high), that's only 5.6V peak before the RIAA network.
Look to the second stage, after the RIAA, for the problem. To repeat, don't mess with the first stage.
No need for an inverse RIAA to troubleshoot this problem. You can just use a sine wave.
Set the source to 1V, with an L pad attenuator of series 10k and shunt 10R. This gives 1mV out.
Input a 1kHz sine and see if there is overloading before the series RIAA resistor. There won't be.
Then increase the input until you see overloading. Adjust the input attenuator as needed for the larger input.
Change the series resistor first to 1k, then to 100R. Probably it will not overload until about 200mV at the input,
after the L pad.
Then look at the main output under the same conditions. It will overload sooner, as it should.
Last edited:
You can play with the values of R28, but this will necessitate adjusting P1
Increasing the values of R21-R24 will slightly increase the noise of the first stage, but should not affect the frequency response. I would leave the first stage alone.
While an inverse-RIAA network is nice to have, just measure the response at 1kHz.
The main output LF overload point at <50Hz will be much lower ( ten times lower ) than at 1kHz,
due to the +20dB higher gain at low frequencies relative to 1kHz.
Opposite happens for HF >2kHz due to the lower gain (-20dB) relative to 1kHz.