• 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

Hi Mike and John!


OPA1656 looks very promising, but after reading its datasheet I still have few questions, any chance you could shed some light for me?


My planned application is combined I/V stage, LPF and headphone output (not my design, just looking to improve/upgrade an existing one), but I also looking for good candidates for Headphone output stage on and by itself.



1. I note that applications section of datasheet does not mention filters or I/V stage explicitly, are there any reasons for OPA1656 to not to be used for these applications? (looks perfectly suitable based on specs)


2. I note that only typical capacitive load is stated (100pF) but not maximum, unlike in OPA1622 datasheet (> 600 pF) making it a bit hard to compare between the two. Have you tested higher capacitive loads by any chance? I would love to avoid series resistors if possible as I need low-impedance output


3. The output impedance is stated for 1MHz open loop, OPA1622 seems to have Zout around 5Ohm in the audio range, how about OPA1656?


4. Can OPA1656 be utilised as a buffer, similar to OPA1622 (which works really well in this application)?


5. Obviously power on/off pops are important for headphone amp (or rather lack of those), any suggestions how to reduce them provided they are present of course?
 
JohnC can fully answer, but it seems to me that OPA1656 isn't a replacement for OPA1622.

OPA1656 is a good all-rounder for audio purposes. OPA1622 is designed for heavy loads and probably has better distortion into low impedances. The package alone is a clue, having no exposed pad probably means it's not expected to be doing really heavy lifting.
 
Hi Mike and John!


OPA1656 looks very promising, but after reading its datasheet I still have few questions, any chance you could shed some light for me?


My planned application is combined I/V stage, LPF and headphone output (not my design, just looking to improve/upgrade an existing one), but I also looking for good candidates for Headphone output stage on and by itself.



1. I note that applications section of datasheet does not mention filters or I/V stage explicitly, are there any reasons for OPA1656 to not to be used for these applications? (looks perfectly suitable based on specs)


2. I note that only typical capacitive load is stated (100pF) but not maximum, unlike in OPA1622 datasheet (> 600 pF) making it a bit hard to compare between the two. Have you tested higher capacitive loads by any chance? I would love to avoid series resistors if possible as I need low-impedance output


3. The output impedance is stated for 1MHz open loop, OPA1622 seems to have Zout around 5Ohm in the audio range, how about OPA1656?


4. Can OPA1656 be utilised as a buffer, similar to OPA1622 (which works really well in this application)?


5. Obviously power on/off pops are important for headphone amp (or rather lack of those), any suggestions how to reduce them provided they are present of course?

To start, OPA1656 is definitely not a replacement for OPA1622, the two parts were designed for with very different applications in mind. OPA1622 was targeted specifically for headphone applications, while OPA1656 is a more general purpose device meant to address analog signal processing in line level audio circuits.

1. Active filters and DAC I/V circuits are perfect applications of the OPA1656.

2. OPA1622 will be significantly more tolerant of capacitive loads due to the nature of it's output stage design (extremely low open loop output impedance). I would not suggest using an OPA1656 in a headphone application without a series output resistor. If you still want to maintain low closed-loop output impedance, there are a number of circuits that close the loop around the output resistor at audio frequencies.

3. OPA1656 will not have the flat, 5 ohm, output impedance of the OPA1622. Combining a rail-to-rail output stage with a 3 gain stage architecture in CMOS normally results in regions where the output impedance is inductive. There will be a graph in the final datasheet.

4. Sure, there is no reason why OPA1656 couldn't be used as a buffer.

5. The best advice is to try to have both supplies ramp up simultaneously, and if there is still audible pops with that situation then you will need to add a muting circuit on the output. The OPA1688 CMOY thread on here has some very good examples.
 
If offset is important then OPA2156 would be a better choice. That device was targeted at precision applications, so the input offset voltage is trimmed. We do not trim the input offset voltage of the OPA1656, but there will be an offset voltage distribution in the production datasheet when it releases soon. The max offset spec of OPA1656 will be +/- 1mV.
 
If offset is important then OPA2156 would be a better choice. That device was targeted at precision applications, so the input offset voltage is trimmed. We do not trim the input offset voltage of the OPA1656, but there will be an offset voltage distribution in the production datasheet when it releases soon. The max offset spec of OPA1656 will be +/- 1mV.

Ideal offset for me would be around +-200uv but unless I hit jackpot it shouldn't be far off.

Last question here.
I'm looking at the ordering page, it shows the type is prototype. Will there be any changes?
 
Not specificly about the OPA1656 ....but :

Here are some questions for John124 . While looking for opamps for an HP amp , I come up with this problem : Nearly all audio opamps have protection diodes on their input . The worst are the bipolar NPN's inputs with anti-parrallel diodes between inverting and noninverting input ( like OPA 1611) and like all others have diodes to the supply rails , so max input voltage goes 0,5 to 0,7 V above the rails . Only the OPA627-37 go 2V over the rails . Why did T.I . abandon BB's DiFet type of opamps ? The OPA627 replacement : OPA 827 , has only 0,5 V . Not even the new CMOS amps go beyond the 0,7 V . Why not go beyond 2 or even 3 V over supply for ESD and overvoltage protection? Mosfets usually can have 20 V GS voltage. Why do I ask and or want this ? The HP amp will be connected to outputs like computer line out , which can give max 4Vpp AC and a CD player that can give even bigger peak voltages ( with bass drums banging against 0 dB), while the HP amp will not be powered up . But even powered up : I like DC coupling but that would get those peak voltages through the 10 mA max diodes (because the sources , CD , computer out ,... are low impedance ). Yes ofcourse a blocking cap on the input would block DC voltages , but what about low frequency bass peaks ? And yes a resistor on the input will get that current down to the 10 mA max, but do I really want 100-500 ohm extra source resistor , not profitting from the low impedance of the source outputs ? Inverting opamp is a solution for the input resistor , but then I need another inverter after that and datasheets show that inverting amps usually have worse specs than noninverting . Data sheets also show the lag between output and input ( propagation delay in digital cicuits). The lag can be 100 to 500 ns and more (with slow audio opamps) .... so during that time , current will flow through the anti-parallel diodes when the opamps is used as a noninverting buffer ?

Unlike most here on the forum with an engineering degree and distortion & spectrum analysers , I'm just a hobbyist , so go easy if there are stupid questions.
 
Not specificly about the OPA1656 ....but :

Here are some questions for John124 . While looking for opamps for an HP amp , I come up with this problem : Nearly all audio opamps have protection diodes on their input . The worst are the bipolar NPN's inputs with anti-parrallel diodes between inverting and noninverting input ( like OPA 1611) and like all others have diodes to the supply rails , so max input voltage goes 0,5 to 0,7 V above the rails . Only the OPA627-37 go 2V over the rails . Why did T.I . abandon BB's DiFet type of opamps ? The OPA627 replacement : OPA 827 , has only 0,5 V . Not even the new CMOS amps go beyond the 0,7 V . Why not go beyond 2 or even 3 V over supply for ESD and overvoltage protection? Mosfets usually can have 20 V GS voltage. Why do I ask and or want this ? The HP amp will be connected to outputs like computer line out , which can give max 4Vpp AC and a CD player that can give even bigger peak voltages ( with bass drums banging against 0 dB), while the HP amp will not be powered up . But even powered up : I like DC coupling but that would get those peak voltages through the 10 mA max diodes (because the sources , CD , computer out ,... are low impedance ). Yes ofcourse a blocking cap on the input would block DC voltages , but what about low frequency bass peaks ? And yes a resistor on the input will get that current down to the 10 mA max, but do I really want 100-500 ohm extra source resistor , not profitting from the low impedance of the source outputs ? Inverting opamp is a solution for the input resistor , but then I need another inverter after that and datasheets show that inverting amps usually have worse specs than noninverting . Data sheets also show the lag between output and input ( propagation delay in digital cicuits). The lag can be 100 to 500 ns and more (with slow audio opamps) .... so during that time , current will flow through the anti-parallel diodes when the opamps is used as a noninverting buffer ?

Unlike most here on the forum with an engineering degree and distortion & spectrum analysers , I'm just a hobbyist , so go easy if there are stupid questions.
One question, have you ever experienced unwanted behaviors using those opamps?
If the answer is no, why bother.
If the answer is yes and you want something suitable for such applications, try opa189.
 
Last edited:
One question, have you ever experienced unwanted behaviors using those opamps?
If the answer is no, why bother.
If the answer is yes and you want something suitable for such applications, try opa211 and opa189.

His questions are valid. OPA211 is similar to the 1611 he is complaining about...

The answer is probably that you really need to provide external protection yourself or actually disconnect (switch) the input if this is an expected condition.
 
One question, have you ever experienced unwanted behaviors using those opamps?
If the answer is no, why bother.
If the answer is yes and you want something suitable for such applications, try opa189.

All my other (audio) opamps sit behind relais so they never get a input voltage while being powered down . But now I'm looking for opamps for a HP amp that will be connected to an active source ( CD player or computer line out) . I'm just doing due dilligence . I can't be the only one who has questions about these input protection diodes ? The OPA211 is a very nice opamp but again with anti-parrallel diodes on inv & noninv NPN inputs . The OPA 189 seems to be a chopper opamps , not really for audio . Extremely good DC spec's though.