• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

OPA1656: High-Performance CMOS Audio Op Amp

OPA1656 is less noisy than the OPA827, 2.9nV/rtHz vs. 4nV/rtHz both flat.

See attached, Jan is using the opamp for his project MRK 3 input box (input attenuation & gain) prior an ADC for measurement purposes :redhot:

Currently it is not clear, how the measurement (customer) setup was really done... balanced input & output of his upcoming gear

While did many noise simulation with LTSpice for a SVF notch (state variable filter) and the winner was always the OPA827 even in LTSpice FFT graphs.

Hp
 

Attachments

  • Noise OPA827 vs OPA1656.png
    Noise OPA827 vs OPA1656.png
    19.5 KB · Views: 932
The input stage on the OPA2156 is rail-to-rail, and uses slew boosting to improve the settling time (notice the higher slew rate). Slew boosting is something I don't think most customers are aware of, and we never really do it in audio power op amps.

PCM1704 datasheet
Figure 5 shows single channel circuit connections for a typical PCM1704 application. It shows the PCM1704 interface to the digital filter, the I/V converter, and the DAC post filter. Selection of an appropriate op amp for the I/V converter is critical for obtaining optimum dynamic performance from the PCM1704. The OPA627 is recommended for this application. Op amps with similar characteristics and faster settling times may also be used.

ts Settling time

OPA627 0.01% 550 ns

OPA1656 0.01%, 800 ns
OPA2156 0.01%, 600 ns

Why it happens? Or is i / u not an audio?
 
Last edited:
What should I look at? OPA1656 is clearly lower noise.

If you believe some simulation results over the datasheets that's your choice (and problem, I would say).

After time, here to drawing's to compare:

. OPA1556 for audio

. OPA827 for instruments


while Jan's auto ranger is for measurement and sholul be ways better than the DUT.

IMHO, or 2 cents
 
Hi, Would like to thanx Johnc124 for all the good advices , and all members to this thread.

just replaced my already very good opa2134 on my dac analog pre-output with the opa1656, and not only it keep warmness of the opa2134 but feeds more details to my Naim amp hifi system , which is amazing !

so one more time thx

G.
 
I have a couple of those sitting here on my desk that I have had about a year. When it came out it had the lowest noise numbers there was if I am not mistaken. I wrote them and got some samples sent over. Wish I could get my hands on a D50 schematic. I'd like to swap it into those. Take that $250 DAC up another level.
 
Yes, there are definitely plans, the team is just spread really thin at the moment. But stay tuned, the single and quad channel options are on the way.


Cut a metal mask option of the 1656 & disconnect the other channel to make a single 1655, like in the old days?



Seems like mostly a matter of product & test engineering resources, which are probably already stretched thin.
 
Hi, Would like to thanx Johnc124 for all the good advices , and all members to this thread.

just replaced my already very good opa2134 on my dac analog pre-output with the opa1656, and not only it keep warmness of the opa2134 but feeds more details to my Naim amp hifi system , which is amazing !

so one more time thx

G.


I had almost the same experience after replacing the OPA2134 that followed by DAC (CS4398) on a Popping D30 DAC I picked up from Craiglist. The Popping performed OK with the OPA2134 but not much better than my other budget DAC that used a PCM1793 DAC followed by an OPA2134. But after swapping in the OPA1656 into the Popping DAC, there was an significant improvement in Detail. I'm really loving being able to so easily hear the improvement in the snare drum and cymbals, something that usually sound somewhat muted or not distinct at all. And a very noticeable job on violins, something that seems to be a real problem to reproduce once the sound is digitized. Best DAC improvement so far. Tried also in my phono preamp, with improvement in detail, but may not be a good idea there.
 
If the circuit has low impedance (1k ohm) use a BJT input opamp. If the impedance is high (10k ohm or higher), then a FET input op amp may be better.

The effective input noise depends on both the input current noise & input voltage noise. With a FET input op amp input current noise is lower, but the input voltage noise is higher.

Generally a FET input has a higher 1/f noise corner than a BJT input.

Also typically a MOSFET input has a higher higher 1/f noise corner than a JFET input. The input bias current of a MOSFET input op amp is mostly due to the input ESD protection diode junction leakages.

For a BJT input op amp the voltage noise is lower but the current noise is higher.

So the choice of op amp depends on the circuit around it and the application. Low impedance inputs and feedback resistors favor a BJT opamp.

High impedance circuits (10k ohms or higher) favor a FET input opamp for low noise applications.
 

Attachments

  • Compare-OPA 156 and 1642 input voltage noise 1 over f corner.JPG
    Compare-OPA 156 and 1642 input voltage noise 1 over f corner.JPG
    100.1 KB · Views: 548
There's a discussion going on over at this thread:
https://www.diyaudio.com/forums/sol...ounding-op-amps-lowest-thd-8.html#post6555025
that is saying that the OPA1656 is quite a bit noisier than the OPA1642 AT LOW FREQUENCIES (100 Hz). This appears to be confirmed by the datasheets and is not a trivial difference---about 6 db; although at 1K Hz, the 1656 is lower by about 1.5 db. Hmmmmm............

Yep, the JFETs used in the OPA1642 input stage still have an edge over CMOS (1656) for low frequency noise. However, I should point out that broadband noise in most audio applications is way more important. Although we're used to seeing frequency on a log scale, it tends to distort our perception of how much the low frequency noise contributes to the total integrated noise over the audio bandwidth. Just consider, there is 19x more bandwidth between 1kHz and 20kHz than there is between DC and 1kHz.

The exception is phono preamps. Because the RIAA eq curve has highest gain at low frequencies, 1/f noise matters. And on MM phono preamps, the cartridge inductance is high enough that amplifier current noise may also matter. I like phono preamps because you can't ignore any noise contributors like you can in other circuits!
 
  • Like
Reactions: 1 user
Hi Johnc124,


I would like to ask if TI now have any opamp to replace lme49990 at the voltage follower buffer in 3-opamp balance input? OPA1612 is very low noise but have a rise in distortion when driving 600 ohm as voltage follower. LM4562 can drive 600 ohm or lower very well but are not particularly low noise device.


Thanks in advance