Hey Oleg, these are also really nice: Locking 1/4'' Chassis Jacks - Neutrik (prices at Mouser).
Those Neutrik chassis jacks will easily fit a 10 mm front panel, can be fastened with M3 screws, are extremely rugged, and look the part as well.
I have finally assembled the BOM for the OPA1622 boards with and without the PSU.
Now I am thinking which gain should I choose? It seems that unity gain should be sufficient for my needs but the gain of 2 would be more practical according to what is typically recommended. So I think I'll go with the gain of two using a pair of 768 Ohm resistors as in the datasheet.
By the way, is there any downside to using 402 Ohm resistors instead? It would give 201 Ohm resistance seen by the inverting input and it would match the input resistance of 200 Ohm if I use 100 Ohm resistors at the output of the preceding unity gain buffer (which I plan) and at the non-inverting input of the OPA1622.
Also what capacitor value to choose for the feedback/input filter? 470p value would give ~1.7Mhz cut-off frequency with 200 Ohm total input resistance. It seem to be a bit too high a frequency for an input filter. Should I go for 1n or 1n5 capacitors instead assuming 200 Ohm input resistance is fixed or it is better to rise the value of the input resistor?
Regards,
Oleg
Now I am thinking which gain should I choose? It seems that unity gain should be sufficient for my needs but the gain of 2 would be more practical according to what is typically recommended. So I think I'll go with the gain of two using a pair of 768 Ohm resistors as in the datasheet.
By the way, is there any downside to using 402 Ohm resistors instead? It would give 201 Ohm resistance seen by the inverting input and it would match the input resistance of 200 Ohm if I use 100 Ohm resistors at the output of the preceding unity gain buffer (which I plan) and at the non-inverting input of the OPA1622.
Also what capacitor value to choose for the feedback/input filter? 470p value would give ~1.7Mhz cut-off frequency with 200 Ohm total input resistance. It seem to be a bit too high a frequency for an input filter. Should I go for 1n or 1n5 capacitors instead assuming 200 Ohm input resistance is fixed or it is better to rise the value of the input resistor?
Regards,
Oleg
Glad to see you're making progress!
For my ODAC (2.1 Vrms output voltage) feeding an O2 amplifier I'm using unity gain with the HD650 headphones. That allows me to use about 75% of the pot range. I guess though that we would need to change just one BOM entry to adjust the gain, right?
For those who want to use an input filter a lower cut-off frequency would probably be better.
Would it be okay to omit the input filter if I put a THAT1646 differential line driver directly between a DAC and your OPA1622 amp? Plus, if the DAC only provides a single-ended output, would it be preferable to feed a single-ended signal into the OPA1622 (given that your design allows for that approach as well), or should I use a separate line driver like the THAT IC?
For my ODAC (2.1 Vrms output voltage) feeding an O2 amplifier I'm using unity gain with the HD650 headphones. That allows me to use about 75% of the pot range. I guess though that we would need to change just one BOM entry to adjust the gain, right?
For those who want to use an input filter a lower cut-off frequency would probably be better.
Would it be okay to omit the input filter if I put a THAT1646 differential line driver directly between a DAC and your OPA1622 amp? Plus, if the DAC only provides a single-ended output, would it be preferable to feed a single-ended signal into the OPA1622 (given that your design allows for that approach as well), or should I use a separate line driver like the THAT IC?
Yeah, the progress is slow due to other things stepping in the way.
As for the gain selection, it is really easy to adjust changing one BOM entry. But keep in mind the impedance imbalance seen by the inputs as discussed at the beginning of this thread (within the first 20 or so posts). Also if I understand the designer's intent correctly the differential amplifier straight after the voltage output DAC is the main application for OPA1622 (as shown in the datasheet). So using THAT1646 in front of the OPA1622 should be fine, but keep in mind the distortion figure of THAT1646 driver compared to the OPA1622. THAT1646 has significantly higher THD+N than OPA1622. May be johnc124 or Sergey888 could answer your question better if they still follow the thread.
As for the gain selection, it is really easy to adjust changing one BOM entry. But keep in mind the impedance imbalance seen by the inputs as discussed at the beginning of this thread (within the first 20 or so posts). Also if I understand the designer's intent correctly the differential amplifier straight after the voltage output DAC is the main application for OPA1622 (as shown in the datasheet). So using THAT1646 in front of the OPA1622 should be fine, but keep in mind the distortion figure of THAT1646 driver compared to the OPA1622. THAT1646 has significantly higher THD+N than OPA1622. May be johnc124 or Sergey888 could answer your question better if they still follow the thread.
Hi ammel68,
Sorry for silence regarding your suggestion. After your post I actually went reading on the subject... and returned only now I followed your suggestion and ordered 1k resistors and 470pF caps. I also ordered some values in between. I'm still unsure if adding series resistance is the way to go in my case. I understand that compared to the common mode input impedance of the op-amp 1k is negligible but I planned to use 10k volume pot before the OPA1622 and this can interfere with the filter if its corner frequency is already low. I probably have to add the buffer after the pot before the OPA1622 as suggested earlier in this thread. I'm still undecided how to proceed. I'll come back to it when I get the parts and start putting things together. The parts from mouser should be in my hands within a couple of hours, the modushop small galaxy case probably tomorrow. I already have most of the other things. So stay tuned, more questions will follow as I progress with the build
Regards,
Oleg
Sorry for silence regarding your suggestion. After your post I actually went reading on the subject... and returned only now
I followed your suggestion and ordered 1k resistors and 470pF caps. I also ordered some values in between. I'm still unsure if adding series resistance is the way to go in my case. I understand that compared to the common mode input impedance of the op-amp 1k is negligible but I planned to use 10k volume pot before the OPA1622 and this can interfere with the filter if its corner frequency is already low. I probably have to add the buffer after the pot before the OPA1622 as suggested earlier in this thread. I'm still undecided how to proceed. I'll come back to it when I get the parts and start putting things together. The parts from mouser should be in my hands within a couple of hours, the modushop small galaxy case probably tomorrow. I already have most of the other things. So stay tuned, more questions will follow as I progress with the build Regards,
Oleg
Finally I came to the assembly point, but I have a couple of questions. I plan to go with the gain of 2 in the non-inverting configuration and just take the values from the datasheet (example head amp design but in single ended configuration). So the max output would be around 2 Vrms. Is this output level OK for the nowadays common 32 Ohm headphones? Another question is which supply voltage to select? It seems that taking +-6V should be sufficient or I should go higher to let say +-9V?
Regards,
Oleg
Regards,
Oleg
If +/-6V gives you the output voltage range desired then I wouldn't go any higher. Unlike older op amps, whose open loop gain would increase somewhat with higher supply voltages (as well as some other parameters, ultimately giving lower distortion), there is no benefit to using excessively high supply voltages with the OPA1622, quite the opposite in fact. Higher supply voltages increase the power dissipated in the amplifier, increasing the die temperature and potentially limiting the maximum power able to be delivered to low impedance headphones.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.