LMV721.pdf (application/pdf Object)

I’m actually stepping back in time. I needed to set up a turntable for my dad’s unit finally wore out. So… it looks like its time to build a RIAA preamp.

I took a look through my parts inventory, and the only thing audio op amp related was the NE5532. It was a good part 15 years ago. Pretty low noise, but today, the LMV722 really shines. 9nv/sqrt(Hz). Its also low powered, and can run at low voltages. Make that only run at low voltages. I blew one out, by running it at 12V by not measuring my lab supply after moving it. I must have hit the knob, and sure enough, the mysterious smoke left the LMV722 non-functional. As such, be SURE to stay under 5.5V…..

National makes 2 versions, the single 721, and dual 722. However, if you want something thats easy to work with, go for the 722. The LMV721 comes in a SOT-23-5 and a SC70-5. So… I tried it out. Its a royal pain to solder by hand. The SC-70 sure is silicon dust, even when working under high megnification. As such, I prefer the 722, even when I only need one opamp. Its easy to solder, even without magnification.

Now some interesting stuff.

The PSRR varies whether on the + or – side. Interesting, still its pretty decent. Greater than 70db at 100Hz, which probably means that at 60Hz, one is still good to go. Otoh, if you are using a switcher, make sure the ground is quiet. Otherwise, it will go right into your output signal.

Common mode is pretty cool at 60Hz as well. Its too bad the market for guitar effects pedals is pretty dead. One could have done some really cool products with this part. Then again, for the input stage and micamps, it may well be an ideal part. The best data on a discrete pro input stage was given by the audio seminar by Analog Devices years ago. Unfortunately, when Walt retired, apparently much of the analog audio world retired as well. They have only a fraction of the cool stuff they used to have.

Gain and phase plots are pretty decent. One should be able to really crank these up with lots of gain before getting in trouble. Thats a major improvement from days of old.

These parts are also rail to rail output, and can swing a little below ground if you are careful. Be sure to add a clamp for protection if you anticipate getting anywhere near the -0.3V destrutive limit.

The part is good to go at 4700pF output loading, and can even run at unity gain, but the phase margin does take a real header if you do so. If you are concerned, you can use a feedforward method to allow more drive… but you will sacrifice the slew rate.

As always, good quality caps are recommended. Much improvement has occured in SMD films caps over the years. Still be sure to watch your profile. Overtemp will kill them, or worse, make them really flakey. Been there done that in volume production. Its not fun.

[Slashdot] [Digg] [Reddit] [del.icio.us] [Facebook] [Technorati] [Google] [StumbleUpon]