The Muffsy MH-1 Prototype 2 has been soldered, and it's time for some testing.

• Both channels (yes, it's dual mono) use LT1115, with LT1097 and OP177 as DC servo, testing is done with 750/47 ohms gain resistors and 100 ohms input impedance.
• Input signal for testing: 0.5 mV

Here are all the tests:

• DC offset: < 0.1 mV. Similar performance with OP177:

• Circuit does not oscillate

• Frequency Response (graph shows the -3 dB point at 0 dBr):
  • 20 - 20.000 Hz: +0.0 / -0.1 dB
  • 1 - 45.000 Hz: +0 / -3 dB (irrelevant because of the QA400's frequency response, see below)

The frequency response drops off to -3 dB at ~45 kHz. This seems to be a limitation of the QA400 which has a loopback frequency response of, you guessed it, 45 kHz. The -3 dB point as defined by the low pass filter in the gain network should be around 300 kHz.

• IMD (ITU-R): 0.001 % (difference between 19/20 kHz pair and 1 kHz mixing product: -97.9 dB)

The value of the Intermodulation Distortion is of somewhat academic interest. In reality, the 1 kHz result will be way below the noise floor of the Muffsy Head Amp; I had to increase the input signal by almost 40 dB to even find it.

• Measured noise profile:

• Overall noise is -105.6 dB (20-20.000 Hz)
• Peak noise value: -133.03 dBV
• Output with 0.5 mV input: 8.5 mV / -41.44 dB

Hmmm... What's with these measurements?

The SNR, according to these readings, is 64.2 dB. But theoretically, it should be over 70 dB...

I get the exact same measurements with nothing connected to the QA400, and the specifications say that its input noise performance (20-20.000 Hz) is 102/104 dBV (low/high impedance). Add 3 dB for A-weighting. I've hit the limits of the QA400... This tells me that the SNR must be greater than, or equal to these measurements.

How much better, I can't exactly tell. I do know that the LM4562 in Prototype 1 measured just shy of the 60 dB it was supposed to according to the calculations, which also is on the limit of the QA400's abilities, so I can only guess that the Prototype 2 performs according to the calculated value, which is 72 dB. It should at least be greater than 70 dB.

I have redone the noise measurements, using the Muffsy MC Head Amp itself to lower the noise floor according to this article: http://www.quantasylum.com/content/Home/tabid/40/Post/205/Extending-the-QA400-Noise-Floor-with-a-Pre-amp

I can now tell exactly what the signal-to-noise ratio is for the Muffsy MC Head Amp is:

113.2 dBV (total noise) - 41.4 dB (signal strength) = 71.8 dB

• THD / THD+N

With an SNR of at least 70 dB, the noise floor of the Muffsy MH-1 is at the very least 5.8 dB below that of the QA400. So I'm increasing the input signal by 5.8 dB to take that into account.

• THD: < 0.1 %
• THD+N: < 0.6%

• Connecting the Muffsy MH-1 to the stereo

This is quite the test. The first prototype made some noises, so fingers crossed.

I connected a 0.5 mV signal to both channels to see if it did what it was supposed to do, disconnected the signal and hooked up the MH-1 to my stereo on the input with the Muffsy Phono Preamp (with the volume off, of course).

I turned up the volume very slowly, no sound (yay!), then further and further until I had reached max volume.

There was a very tiny, minuscule white noise sound that could only be heard by pressing the ears into the loudspeaker. I think that could be called a successful test. :)

The first test was with 100 ohms input impedance, I tested again with 50 ohms and got the exact same great result. The test was done with LT1097 as DC servo, I'll redo the test with OP177 as well.

But... Here's the thing... I don't own an MC cartridge... I should be able to solve that though, stay tuned.

Well, I got an MC cartridge and the Muffsy MC Head Amp sings. And it's every bit as good as I hoped for, and then some!