Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › TLC2272 Op Amp in a CMoy
New Posts  All Forums:Forum Nav:

TLC2272 Op Amp in a CMoy

post #1 of 10
Thread Starter 
I replaced the OPA2132 in a nothing-special CMoy with a TLC2272 (page, datasheet). I happened upon it while looking through TI's cross reference stuff (for TS922AIN). It's advertised as rail-to-rail, runs at 2.2v - 8v per rail, 50mA output current. I barely can grasp those numbers, but I figured it'd be worth a try for low impedance canalphones or K81DJs.

It dropped in with no issues/heat at gain=5, and it sounds better--maybe the sound is fuller, maybe more midrange, seems like less distortion with a single 9v. I can hear more good stuff in the SomaFM IPR stream, but it's milder overall. Definitely more separation between channels. ~1.5mV offset. In short, Pretty good in the KSC75s.

Since TI doesn't advertise it as "unity gain stable," am I to assume it's not? I would like to lower the gain of this CMoy before moving on to bigger betters. Does an OpAmp oscillate when the gain is too low for "stable?"
post #2 of 10
Low slew rate, relatively high output impedance, and it's not designed to drive low impedance loads. What you're probably hearing is the sound of higher distortion, not lower. Distortion performance even with a 10K ohm load isn't all that hot.

If it isn't listed as unity gain stable, then you shouldn't run in unity gain. However, several of the specs seem to be tested at Gain 1, which seems to indicate that it is unity gain stable. I just did a quick scan of the datasheet, though.
post #3 of 10
Thread Starter 
smooth, creamy distortion

I guess what I hear less of is clipping due to the single 9v NiMH batttery I'm using. I wonder how the TS922AIN compares--I think its THD is like .005% vs the TLC2272's .0013% but some of the fiine print numbers are conspicuously different. This is clearly one of those couple-times-a-year times when I need to have learned more math
post #4 of 10
I think "50mA output current" is the damage threshold for external current forced into the output pin

the graphs for +/-5 V operation such as "MAXIMUM POSITIVE PEAK OUTPUT VOLTAGE†
vs
OUTPUT CURRENT"
fig 17-21 are more relevant/indicative, with only single digit mA output capability this op amp is really not suited to drive any dynamic headphone

of similar TI parts the TLC072 is far more suitable for headamp application
post #5 of 10
Thread Starter 
Well I would like to thank you all for shooting down my favorite Op amp Seriously though, I appreciate the help understanding the critical data. I can now see why it's unwise to trust the cross reference charts to have headphone-amp parameters in mind.

And for the record, I do not recommend this Op amp to anybody as fit for any head-fi related purpose
post #6 of 10
Quote:
Originally Posted by jcx View Post
[...] of similar TI parts the TLC072 is far more suitable for headamp application
The TLC071/072/074 are CMOS single rail supply opamps, mostly meant for amplifying logic levels at high speeds. It does not take a negative supply - so it is definitely not suitable for anything which has to do with audio. If you give it a negative supply it will more or less explode.
Other than that the TL071 and family is a dirt cheap but average performance opamp with a highish offset voltage, but is ok for a BJT base input opamp. Don't use it for high amplifications without making compensation for it.

For ULV amplifications i'd really recommend the OPA350, which is also very suitable for audio, if you don't need higher than about 5V.
post #7 of 10
Quote:
Originally Posted by Daroid View Post
The TLC071/072/074 are CMOS single rail supply opamps, mostly meant for amplifying logic levels at high speeds. It does not take a negative supply - so it is definitely not suitable for anything which has to do with audio. If you give it a negative supply it will more or less explode.
The opamp has no connection to ground, so it doesn't know the difference between "single supply" and "split supply" except for the difference between its + and - supply pins. There are some opamps that are "optimized for single supply use", and that usually means the input stage within the opamp can handle signals that swing below negative supply voltage without undue damage. That is all. Nothing would explode unless you exceed the supply voltage limits.

Quote:
Other than that the TL071 and family is a dirt cheap but average performance opamp with a highish offset voltage, but is ok for a BJT base input opamp. Don't use it for high amplifications without making compensation for it.
TL071 is JFET-input with low offset. You were probably thinking of something else. It is cheap and not a good performer, though.
post #8 of 10
Quote:
Originally Posted by heretical View Post
Lowish slew rate and bandwidth are good for headphone amps.
How do you figure?
post #9 of 10
Quote:
Originally Posted by heretical View Post
Keep in mind that we were talking of unbuffered headphone amps. Would you deny that generally, the slower the opamp, the better its capacitive load drive? The AD8532 as opposed to the AD8397 is a perfect example. As a hyperdetail fan(atic) (I think of the AD8058), perhaps you'll disagreee.
Well, stability with a capacitive load relies on a number of factors. Don't go picking out slow op-amps just because you think you'll have stability. Linear has those C-Load op-amps which are both very fast and, at least allegedly, stable with "unlimited" capacitive loading. At the very least, I've seen some high speed op-amps that claim stability with several thousand (10000, for example) pF of capacitance. Output impedance, phase margin, open loop gain, your noise gain vs. signal gain, and a variety of other things are going to factor into stability. The LM6171 is blisteringly fast and has better CL drive than op-amps a few hundred times slower. Generally, though, yes it does help. C-Load mostly seems to work by pulling back bandwidth and slew. It isn't the only road to stability, though.

The AD8532 is considerably inferior to the AD8397 as a headphone driver, IMO. I wouldn't even use it over the AD8616 or AD8656. Capacitive load drive capability isn't the only thing to consider when trying to drive headphones. Slew rate is going to take a dive when you hook up that low impedance reactive load, and you're pretty much asking for distortion with this recipe of low slew, olb, and high olg.
post #10 of 10
Quote:
Originally Posted by heretical View Post
Btw, my concern wasn't just stability but lack of overshoot/ringing effects that would compromise the quality of sound (AD8397 seems to suffer from this problem in that it tends to have a slightly coarse mid-treble; AD8532 seems not).
If the AD8397 overshoots and rings, then it's due to improper application and design. I'll let you have a look at the "Specifications" section of the Mini³ website, which contains extensive tests of the amp, one of the two flavors of the amp is based on the AD8397. In particular, look at the square wave response oscillograms. Overshoot? Ringing? Nada!
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › TLC2272 Op Amp in a CMoy