AmpCasq: an OPA1622 integrated headamp project

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey Chris...

the amp is done...blue led is pretty!! LOL..

Just wanted another board to be able to build another and possibly one for a friend.

If you want to keep them thats ok for someone that really wants to build one thats fine otherwise I would indeed take one for sure. No hurry etc.

I am on vacation for a few days myself.

Thanks
Alex
 
Just got the 12vdc regulated linear adapter from Jameco...

I am totally amazed how good it sounds compared to the older IBM laptop 16vdc supply....

There is an audible difference to me with this adapter! I never tried just batteries, but the amp is working very well compared to before and I am very skeptical on this stuff.

Its like the sound or music is very much more "clear" ..IMO. Could be just me but its definitely not hurting anything.

With the T1's even on the lower 3db gain I can drive the T1's very well and the vol is around 1pm position.

Glad I got this adapter...now to get Sergeys Croc working...its giving me the frits!! LOL

Alex
 
Member
Joined 2015
Paid Member
Just got the 12vdc regulated linear adapter from Jameco...

I am totally amazed how good it sounds compared to the older IBM laptop 16vdc supply....

There is an audible difference to me with this adapter! I never tried just batteries, but the amp is working very well compared to before and I am very skeptical on this stuff.

Its like the sound or music is very much more "clear" ..IMO. Could be just me but its definitely not hurting anything.

With the T1's even on the lower 3db gain I can drive the T1's very well and the vol is around 1pm position.

Glad I got this adapter...now to get Sergeys Croc working...its giving me the frits!! LOL

Alex

Good news, yes it is amazing, you should confirm with a battery or another supply

Chris
 
Final Pix:
 

Attachments

  • WP_20170612_11_53_04_Rich.jpg
    WP_20170612_11_53_04_Rich.jpg
    701.9 KB · Views: 667
  • WP_20170612_11_53_18_Rich.jpg
    WP_20170612_11_53_18_Rich.jpg
    594.8 KB · Views: 636
Member
Joined 2015
Paid Member
Hey Chris...

the amp is done...blue led is pretty!! LOL..

Just wanted another board to be able to build another and possibly one for a friend.

If you want to keep them thats ok for someone that really wants to build one thats fine otherwise I would indeed take one for sure. No hurry etc.

I am on vacation for a few days myself.

Thanks
Alex

Board shipped.
 
I finally managed to do some measurements of the amplifier, lent by AIM65.
Measurements were done at an output between 0.3Vrms and 0.4Vrms, where the ADC of the soundcard gives the best results.
FFT averaged over 30 acquisitions to reveal spurious noise.

1. Measurement system loop-back - 1kHz FFT
2. Measurement system loop-back - Distortion vs frequency

3. AmpCasq - 1kHz FFT - amplifier set at unity gain
4. AmpCasq - 1kHz FFT - amplifier set at unity gain - with Sennheiser HD-600 load (300Ohms)
5. AmpCasq - 1kHz FFT - amplifier set at full gain (pot at max position, input lowered to keep output at 0.4Vrms
6. AmpCasq - Distortion vs frequency

All in all, except for the little mains induced noise, excellent performance all around, I hardly measure anything other than the limit of the ADC.
 

Attachments

  • THD vs Freq_-10dB_AmpCasq.png
    THD vs Freq_-10dB_AmpCasq.png
    33.3 KB · Views: 163
  • AmpCasq HP.png
    AmpCasq HP.png
    41.1 KB · Views: 171
  • AmpCasq HP _HD600.png
    AmpCasq HP _HD600.png
    76.5 KB · Views: 184
  • AmpCasq HP _1k.png
    AmpCasq HP _1k.png
    76.1 KB · Views: 579
  • THD vs Freq_-10dB_loop-back .png
    THD vs Freq_-10dB_loop-back .png
    62.5 KB · Views: 592
  • loopback _1k.png
    loopback _1k.png
    73.6 KB · Views: 607
Last edited:
Member
Joined 2015
Paid Member
I finally managed to do some measurements of the amplifier, lent by AIM65.
Measurements were done at an output between 0.3Vrms and 0.4Vrms, where the ADC of the soundcard gives the best results.
FFT averaged over 30 acquisitions to reveal spurious noise.

1. Measurement system loop-back - 1kHz FFT
2. Measurement system loop-back - Distortion vs frequency

3. AmpCasq - 1kHz FFT - amplifier set at unity gain
4. AmpCasq - 1kHz FFT - amplifier set at unity gain - with Sennheiser HD-600 load (300Ohms)
5. AmpCasq - 1kHz FFT - amplifier set at full gain (pot at max position, input lowered to keep output at 0.4Vrms
6. AmpCasq - Distortion vs frequency

All in all, except for the little mains induced noise, excellent performance all around, I hardly measure anything other than the limit of the ADC.

Hi CaféNoir,

Thanks for those measurements, great results !

Any idea for possible cause of the small 10K to 20K spikes ?, If I may, a measure with amp at full volume and input shorted may give us some insights.

I definitely should buy such device, currently the Focusrite 2i2 seem to have great value for price but I'm also interested by Frex projects (on DIYA) associated with the Jans Didden autoranger....
 
Pushing the limits!

I did some new measurements, trying to extract the best I could from the measurement system.
All I can say is the opa1622 is very challenging to measure, and the AmpCasq is doing a good job at taking advantage of its performance.

First is the ESS DAC - ADC loopback, with the source set at 192kHz 32bit, and ADC at 44kHz (its performance degrades at higher sampling rate).

Second is exactly the same, with the AmpCasq inserted between the DAC and ADC, at unity gain.

Compared to the loop-back alone, We can see the amp is adding minor PSU noise, and some low peaks above 10kHz, but the overall distortion is extremely low (all harmonics below -120dB from fundamental).

Compared to previous measurements, a peak at 50Hz is reduced (I noticed the power cord of my laptop created a ground loop), and some hf peaks due to ADC are reduced as well.
 

Attachments

  • loopback _1k_new_192kHz.png
    loopback _1k_new_192kHz.png
    38.6 KB · Views: 496
  • AmpCasq _1k_new_192kHz.png
    AmpCasq _1k_new_192kHz.png
    42.4 KB · Views: 495
Last edited:
Member
Joined 2015
Paid Member
I did some new measurements, trying to extract the best I could from the measurement system.
All I can say is the opa1622 is very challenging to measure, and the AmpCasq is doing a good job at taking advantage of its performance.

First is the ESS DAC - ADC loopback, with the source set at 192kHz 32bit, and ADC at 44kHz (its performance degrades at higher sampling rate).

Second is exactly the same, with the AmpCasq inserted between the DAC and ADC, at unity gain.

Compared to the loop-back alone, We can see the amp is adding minor PSU noise, and some low peaks above 10kHz, but the overall distortion is extremely low (all harmonics below -120dB from fundamental).

Compared to previous measurements, a peak at 50Hz is reduced (I noticed the power cord of my laptop created a ground loop), and some hf peaks due to ADC are reduced as well.

Hello CaféNoir,

Many thanks for those measurements !

Chris
 
Hey Chris!

5 weeks + here after surgery and recovering nicely...been a long road here!! Some minor issues but very hopeful!!

I havent had a chance to listen much lately, but re-organized my listening systems and am listening to your amp via a new all in one pc via foobar to a new Schitt Audio Bimby multibit dac via their "USB gen 5" circuitry..

The dac is really awesome and its making your amp sound VERY good with my T1's (600 ohms) even with the low gain setting....actually as good as most amps I have heard.

How is your dac project coming along?

Alex
:>)
 
Member
Joined 2015
Paid Member
Hi Alex,
Happy to get good news from you.
Your new DAC looks like a pretty nice piece of hardware, it’s build around an R-2R DAC chip from Analog Device; do you have others DAC to compare to?

Blueprint of my DAC is 70% ready and drawn into Kicad. I plan to have various output stages, including a discrete one. This is why, before moving to the DAC execution, I’m currently building a bench supply (regulator board tested and debugged today) which will be followed by a signal generator, not high end but good enough to do some serious work… This take time, I hope AKM won’t release the successor for the AK4497 before I finish my project!

Chris
 
Compared to previous measurements, a peak at 50Hz is reduced (I noticed the power cord of my laptop created a ground loop), and some hf peaks due to ADC are reduced as well.

What source are you using? I always have this source noise added on my amplifiers.:(

I’m currently building a bench supply (regulator board tested and debugged today)

Can you share the schematic of the PSU?
I always have this source noise added on my amplifiers (Exactly the same as the pictures above). I can not identify the cause exactly. :(
Being that using any internal sound card in the computer we do not have that type of noise, and use PS switched.
 
Member
Joined 2015
Paid Member
It’s a bit early to release the diagrams, testing is not finished. It’s a 723 based supply with LDO style implementation.
To which source noise do you refer to ? 50Hz and 100Hz peaking ?. Because 50 is the main frequency which radiate everywhere and 100Hz is the result of full wave rectification made inside the external power adapter of AmpCasq. But having -125dB at 50Hz is not that bad, -110dB for 100Hz to…
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.