Phonograph pick-up in current mode

[JohnRoberts]

[quote author="mediatechnology"]I measured 757 ohms DCR on my AT MM cartridge. Still no reason to rule out bipolar inputs.

Sure it has rising OP impedance with frequency. So not exactly low Z like a microphone but not a guitar pickup either.[/quote]

Sorry, more opinion than arguable fact on my part. Being terminated with a few hundred pF probably helps drop that Z too. I don't know if you can trick the IC to lower it's input stage current density (sinking or sourcing some dc current at both emitters) for the higher source Z than microphones. Again probably more tweaking than is justifiable, an old habit of mine when messing with phono preamps. If it's worth doing, it's worth doing differently. :grin:

JR

 

[pstamler]

In practice 5532/4 chips are plenty quiet enough on the input for MM cartridges. The rule of thumb: If you have to go stick your ear against the speaker to hear the noise (at operating volume), it's quiet enough. Heck, even a TL07x is quieter than any LP I own, although you can hear it without putting your ear to the speaker. (I needed to use TL071s once for a simple preamp in a high-RFI environment.)

Jung's first preamp for the PAT5 mod used inverting gain on the second amp. Later he modified it with some kind of trick input, don't remember the details. A couple of issues later he further modified it with passive EQ.

Peace,
Paul

 

[bcarso]

Typically the important impedance in the MM cartridge comes from the substantial inductance, which makes the current noise of the preamp and the ubiquitous 47k terminating R's noise more important. It still may well be fine with a ~5532 but the resistance per se is not the main consideration to decide this.

Of course the equalization is helping a lot, but IIRC not enough to invalidate the above. The discussion in the Nat Semi Audio Handbook and then the revision in a later App Notes publication is apt.

 

[JohnRoberts]

[quote author="bcarso"]Typically the important impedance in the MM cartridge comes from the substantial inductance, which makes the current noise of the preamp and the ubiquitous 47k terminating R's noise more important. It still may well be fine with a ~5532 but the resistance per se is not the main consideration to decide this.

Of course the equalization is helping a lot, but IIRC not enough to invalidate the above. The discussion in the Nat Semi Audio Handbook and then the revision in a later App Notes publication is apt.[/quote]

I typically used discrete JFET front ends in my later ('80s) phono preamps but I did one disco mixer in the '70s that used a bipolar national chip (LM 387) that was dialed in for phono pre use. The old National app notes were really quite good.

In keeping with my philosophy of often doing things a little different , I found a way to mitigate what IMO was a shortcoming with that part/topology. The National part was decompensated requiring a significant minimum gain for stability. Putting a zero in the feedback network for this min gain was detrimental to RIAA accuracy, admittedly mostly above 20 kHz. I kept the opamp happy by putting the minimum gain resistor on top of the complete RIAA EQ network. As far as the opamp was concerned it had the higher minimum gain, but I grabbed my signal feed from the top of the RIAA network so it's response kept dropping even below that minimum gain and was only limited by the NI topology to unity. This feed was a high impedance internal point so it's loading was minimal and well behaved.

My later designs were either inverting or combinations of passive and inverting for RIAA accuracy.

JR