LME49713 preamp

[klewis]

I would like to make a preamp around the LME49713. From what I've read, there are some issues with relatively low impedance that might affect the volume control scheme. I've posted my schematic and would like help and critique.

Regards,

Ken

[Ludwik]

Constant input impedance at the second stage? Very interesting. You must add a series resistor, maybe 1 kohm, on the output of the first stage to prevent a shortcircuit.

[klewis]

Added resistor before the pot. Other thoughts? what if the the second op amp was a buffer - Lme49610?

[kaos]

One of the things to watch for in low noise bipolar amps is the input noise current density. That can easily swamp input noise voltage in moderate to high impedance circuitry. The LME49713 data sheet lists a whopping 16 pA per root Hz at 1KHz. The 49713 also shows a typical input bias current of 1.8 uA, which means that the input bias resistor can create a significant IR drop at some settings, the effect of which will cause a varying offset ranging from a few millivolts to around 50 mV (which will be amplified by the DC gain of your circuit) as the volume is changed. The choice of 50K for the “volume” pot may not be the best choice given those facts. Something closer to 5K may yield somewhat better results. I don’t know how much of an improvement the LME49610 would be in this regard, given similar input bias currents and a mislabeled input noise current graph in the data sheet. The choices are yours to make, but you may want to rethink a few aspects of the project.

[klewis]

will this work?

[klewis]

Originally posted by kaos
One of the things to watch for in low noise bipolar amps is the input noise current density. That can easily swamp input noise voltage in moderate to high impedance circuitry. The LME49713 data sheet lists a whopping 16 pA per root Hz at 1KHz. The 49713 also shows a typical input bias current of 1.8 uA, which means that the input bias resistor can create a significant IR drop at some settings, the effect of which will cause a varying offset ranging from a few millivolts to around 50 mV (which will be amplified by the DC gain of your circuit) as the volume is changed. The choice of 50K for the “volume” pot may not be the best choice given those facts. Something closer to 5K may yield somewhat better results. I don’t know how much of an improvement the LME49610 would be in this regard, given similar input bias currents and a mislabeled input noise current graph in the data sheet. The choices are yours to make, but you may want to rethink a few aspects of the project.

kaos,

thanks for the suggestions. I'm ordering parts to build one channel with the 5k pot you suggest. I would like to use a realitively "realistic" load for testing, what should I use? I've thought about a 1k resistor with a small capacitor in parallel, say 0.1uf. Does this make sense. I could then look for dc offset with my scope. yes?

thanks

Ken

[kaos]

A .1 uF load is huge for a pre-amp to drive. I’d just leave it off and look for an offset with a meter, and signs of oscillation with the scope. If you have a signal generator check the squarewave response at several frequencies for excessive overshoot and ringing, which can be indicators that the layout may have problems.

[Tolu]

Hi klewis

did you finish the preamp?