If there's no fire potential, there's no fun. 🙂
Red LED district (where's Jacco when you need him?!) has a definitely different ring, though. 😀
Could it have been OPA211 John? http://www.ti.com/lit/ds/symlink/opa211.pdf It has rail-to-rail output.
Or the opa1602, which is the audio-spec equivalent. Either way, it's TI/BB's low-noise/high GBW BJT part.
Gahhhhhhh, I've LOOOOONNNG since been scooped, then. I'm sure it's a thing of humor for many of us (I've been making that joke for a number of years).
Likely the OPA1611 or OPA1612 which are based on the OPA211 core.
OPA1602 has slightly lower bandwidth and is actually based on the OPA209 core with some changes. Much to my dismay they did not use the super-beta input devices from the OPA209 🙁 but I wasn't the product definer at the time, OPA1622 was my first.
These are both lousy op amps for audio, not only because they are made out of dirty sand but, to add insult to injury, the silicon is heavily contaminated with germanium.
Or maybe it's the other way around, since germanium is know as sounding so good, these are amazing op amps for audio.
I don't know, somebody has to listen while peeking or peek while listening to figure it out.
We have the nerve to let all sorts of dopants contaminate our once pure silicon! And don't get me started on the blasphemous unholy matrimony in a SiGe process! 🙂
Like all op amps, some percentage of the population hates these, others love'em, and we keep making new ones to try to make everybody happy.
You'll not make Mr. Curl happy by cancelling the 49990 altogether, think of the cost of having to scrape off the AD logo.
Johnc124, I think you are right about the IC that did not work as well was the OPA1611. However, I truly believe in IC's and I do use the LME49990, and OPA 134 from TI in the JC-3. In fact, what is the situation with the LME4990? My boss, just today told me that it is being phased out. Is this true, or is it just a different part #? Please don't make me go back to Analog Devices! '-)
Find thee an AD part 😉
It's being phased out in the (eventual) wash-out of TI's purchase of National.
It is indeed going away. Right now its on lifetime buy but I'm not sure on the exact date when production officially stops. Better get an order into your distributor. 🙁
Initially a much larger group of the LME devices were going EOL, but thanks to much screaming some of them seem to have been spared (I believe the LME49600 is off the chopping block for now). But LME49990 wasn't one of the ones brought back from the dead.
Initially a much larger group of the LME devices were going EOL, but thanks to much screaming some of them seem to have been spared (I believe the LME49600 is off the chopping block for now). But LME49990 wasn't one of the ones brought back from the dead.
You can get the LME49600 and stuff it. It's a 36V part that's not doing anything that (a perhaps sorted subset of) the classic BUF634 can't already do.
Saving the 44V LME49860 buffer could do some good for audio, but then the entire former National high voltage process(es) for audio was trashed without any chance to appeal.
Which leaves the OPA604 the only TI high voltage (48V) op amp that could be used for audio, since the dual OPA2604 spec was downgraded to +/-20V and it is unlikely it will be revisited. Kill the OPA604 (I think it's in an old dielectric isolated process) and it's all done, no more parts for high dynamic range audio.
Don't count out the OPA2604 yet, I was actually in a meeting this morning on the ongoing work on that part to restore the spec. It's definitely not on a DI process though (never was), just on Burr-Brown's old bipolar process with an extension for 48V operation.
Audio Opamps Linearity & Stability
Now about linearity or linearity "threshold"... In discrete amplifiers I pay great attention to transistors linearity, because I 'think' I can perceive its effect. But it is so 'fuzzy' that may be impossible to ABX. But I believe that our brain/feeling can process what our ears cannot(!)...
Now with opamps, how can we 'measure' this metric, or know how or when an opamp is operating at its most linear region?
Now about stability... I believe that stability plays an important rule in amplifiers sound quality, especially when perceived in prolonged listening. I think, for audio (and only for audio opamp), the opamp should be produced to avoid unnecessary instability. Optimum bandwidth or slew rate as to avoid stability compensation networks externally or internally...
LM4562 interests me. I found that it competes very well against other opamps when the gain is high. I believe this is because of improved stability, so I have had a plan to compare 3 schemes (by ears):
(1) Non-inverting buffer (LF353) --> Inverting signal gain 40x-100x (LM4562) --> Inverting signal gain 0.4x-0.01x (JRC2068)
(2) Non-inverting buffer (LF353) --> Inverting noise gain 40x-100x (LM4562) --> Inverting buffer (JRC2068)
(3) Non-inverting buffer (LF353) --> Inverting signal gain 40x-100x (LM4562) --> Resistor voltage divider ---> Inverting buffer (JRC2068)
What do you think the result will be?
Now about linearity or linearity "threshold"... In discrete amplifiers I pay great attention to transistors linearity, because I 'think' I can perceive its effect. But it is so 'fuzzy' that may be impossible to ABX. But I believe that our brain/feeling can process what our ears cannot(!)...
Now with opamps, how can we 'measure' this metric, or know how or when an opamp is operating at its most linear region?
Now about stability... I believe that stability plays an important rule in amplifiers sound quality, especially when perceived in prolonged listening. I think, for audio (and only for audio opamp), the opamp should be produced to avoid unnecessary instability. Optimum bandwidth or slew rate as to avoid stability compensation networks externally or internally...
LM4562 interests me. I found that it competes very well against other opamps when the gain is high. I believe this is because of improved stability, so I have had a plan to compare 3 schemes (by ears):
(1) Non-inverting buffer (LF353) --> Inverting signal gain 40x-100x (LM4562) --> Inverting signal gain 0.4x-0.01x (JRC2068)
(2) Non-inverting buffer (LF353) --> Inverting noise gain 40x-100x (LM4562) --> Inverting buffer (JRC2068)
(3) Non-inverting buffer (LF353) --> Inverting signal gain 40x-100x (LM4562) --> Resistor voltage divider ---> Inverting buffer (JRC2068)
What do you think the result will be?
The LME49990 is a nice chip - I used it one of my pre's inside the f/back loop of a LME49710.
Another alternative is the TPA6120. Its not an open loop buffer like the LME49990 but it is very wide bandwidth (re-purposed ADSL/Video driver IIRC), very low distortion etc. This chip easily drives a pair of 32 ohm headphones. Other alternative is to go discrete - then you can also go class A if that appeals. I did this on my last preamp and got good results.
- Status
- Not open for further replies.