The input signal is feeding through to the output for the first 4.5us. Then, the PNP output transistor starts conducting (producing the ramp).
Ed
Ed
Not for a MC phono preamp though - AD797 knocks it out of the water for voltage noise. And the NE5534A beats it for MM on noise (I've a preamp design that uses it, the output offset is of no concern due to the rumble filter that follows). But the OPAx192 is a great performer, if a little expensive though. "Best" is only meaningful against a set of requirements really.
Most audio applications have no need for such low offset voltage, though its great to have if its free, but its not, the trimming steps directly add to the chip's cost. I can see the value for the typical applications quoted in the datasheet (current sensing being a great example).
I've a stash of NE5532's that were about 10% the price of the dual OPA2192 (no, they were not from eBay!). They'll do for many things, and I guess when a JFET input is needed the AD2192 could be a good choice, but not as cheap as the OPA1652 which I've used.
The LM358 has no bandwidth to speak of, its not suitable for audio signals, its commonly used to set DC control signals in equipment as its very cheap, low power, and can handle DC tolerably well, basically protecting the output pin of a DAC... Personally I avoid it because it its easy to be bitten by its limitations (you forget how slow it really is!) and there are literally 100's of much much better rail-to-rail opamps these days.
The full swing bandwidth is only 5kHz. Normally devices for audio tend to quote a full-swing BW of 100kHz -- several MHz, because to keep distortion under control it helps to have plenty of headroom for gain and for slew-rate. The 741 manages twice the full-swing bandwidth of the LM358 despite having the same slew-rate of 0.5V/us.
Only because the LM358 datasheet authors have chosen to plot full swing bandwidth with single ended supplies, while the uA741 datasheet plots full swing bandwidth with twice-as-big supplies, namely plus AND MINUS 15 volts. Chips work the same, chip marketers spin them differently.
Too bad the LM358 can't reproduce a 5 kilohertz sinewave when operating at a gain of -1, despite having a bandwidth of 500 kHz at a noisegain of 2 (which is an inverting gain of -1)
Too bad the LM358 can't reproduce a 5 kilohertz sinewave when operating at a gain of -1, despite having a bandwidth of 500 kHz at a noisegain of 2 (which is an inverting gain of -1)
According to the TI datasheet, the slew rate of the LM358 is typically 0.3 V/us. The LM358B has a typical slew rate of 0.5 V/us.
According to the ST datasheet, the LM358 has a typical slew rate of 0.6 V/us and a minimum slew rate of 0.3 V/us.
None of them are a FET op-amp, so they are all off topic.
According to the ST datasheet, the LM358 has a typical slew rate of 0.6 V/us and a minimum slew rate of 0.3 V/us.
None of them are a FET op-amp, so they are all off topic.
Last edited:
Besides poor slew rate LM358 and LM324 have a ~class-C output which is very poor choice for audio. They were intended for applications like automotive and battery chargers and have a couple important features being low power and "single supply" ie the input operates down to the negative rail. That was a long time ago and I'm surprised anyone would consider them for audio.
In those days, a decent bipolar op-amp was the dual 4558 and the quad 4136, although the 4136 had a non-standard pin-out. There are several variations on that circuit from different vendors. Of course, the best bipolar were the 5532 and 5534. The LM318 was fast in those days and BGW used it as an IPS.
In those days, a decent bipolar op-amp was the dual 4558 and the quad 4136, although the 4136 had a non-standard pin-out. There are several variations on that circuit from different vendors. Of course, the best bipolar were the 5532 and 5534. The LM318 was fast in those days and BGW used it as an IPS.
LM324 is still doing good business in audio. Everytime i visit nearby parts store for general purpose components like 1N4007, electrolytics often i see someone buying LM324 based tone control, low pass filter or something like those & it's not surprising! But sometimes i feel irritating 
Actually still a large number of people don't care about 'sound qualities' & most of them never listened to hifi/hi-end audio. Cheap audio systems for truck is an another example.
Anyway back to topic. Is there any good quad low noise jfet opamp?
Actually still a large number of people don't care about 'sound qualities' & most of them never listened to hifi/hi-end audio. Cheap audio systems for truck is an another example.Anyway back to topic. Is there any good quad low noise jfet opamp?
Last edited:
I was looking for a budget JFET op-amp recently and found out something that may be of interest:
TI (quietly?) changed the spec of the TL07XX op-amps.
While the cover page still reads:
Texas Instruments TL071 datasheet
I find the THD / price ratio pretty good.
TI (quietly?) changed the spec of the TL07XX op-amps.
While the cover page still reads:
There is something interesting in the change log - Changes from Revision T (December 2021) to Revision U (December 2022):
Texas Instruments TL071 datasheet
I find the THD / price ratio pretty good.
The bandwidth is a bit higher at 5.25MHz rather than 3MHz. Hope its at least as stable as the TL07x is known for needing little decoupling.
Beware the TL07xH's voltage noise, very high flicker knee.
Beware the TL07xH's voltage noise, very high flicker knee.
Beware that all TL07 from TI except the PS / NS package and the M devices have new specs.
Means they are somewhat similar to the new H devices, but still different.
Some specs improved, some got worse.
I just ordered some, but have no idea whether I'm going to receive the legacy devices (old stock) or the new ones.
Means they are somewhat similar to the new H devices, but still different.
Some specs improved, some got worse.
I just ordered some, but have no idea whether I'm going to receive the legacy devices (old stock) or the new ones.
I was disgusted when I saw the updated TL07X datasheet.
The equivalent circuit and all graphs in the datasheet have been replaced with those of the TL07xH. This confuses the viewer. In the first place, it seems unreasonable to integrate the conventional TL07x with bipolar process J-FET input and the C-MOS TL07xH into the same datasheet.
The equivalent circuit and all graphs in the datasheet have been replaced with those of the TL07xH. This confuses the viewer. In the first place, it seems unreasonable to integrate the conventional TL07x with bipolar process J-FET input and the C-MOS TL07xH into the same datasheet.