Tom Christiansen Audio HPA-1 Headphone Amp Review

[trl]

Speaking of competence... I just noticed that I have a bug in the block diagram. The output stage is actually a composite amp with a high-speed, high-current output. I'll fix it and update it in the post above (and on my website).

And me thinking that there are paralleled OPA1656 under the heatsink.

Given that you already designed and implemented probably the best measuring "gainclone" from Earth, I would say that there might be a couple of LM3886TF under the heatsink. A similar approach was done by Lake People inside G-103 where they used LM1876TF, so at least my guessing seems plausible.

 

[trl]

@ tomchr

Are you aware that a web search of HPA-1 results in mostly links to the Pass Labs HPA-1? There is also a LD Systems HPA 1

Let's not forget about @March Audio's HPA1 too.
But sometimes a bit of confusion might actually help, so this is a highlight.

 

[ElNino]

This is precisely why the HPA-1 has differential signalling all the way to the volume pot.

I hope this isn't off-topic, but since we're discussing volume pot configurations, I was wondering if you could comment on where you'd recommend adding a volume pot to a preamp based on your upcoming universal buffer board.

Would you recommend getting two universal buffer boards, configuring one for balanced->SE, inserting a volume pot there, and then configuring the second board for SE->balanced again?

 

[tomchr]

And me thinking that there are paralleled OPA1656 under the heatsink.

Given that you already designed and implemented probably the best measuring "gainclone" from Earth, I would say that there might be a couple of LM3886TF under the heatsink. A similar approach was done by Lake People inside G-103 where they used LM1876TF, so at least my guessing seems plausible.

As I described previously in this thread, the output stage in the HPA-1 is a composite amp with a 300 MHz, high-current, current-feedback opamp as the output stage.
The OPA1656 is a 53 MHz voltage-feedback opamp. The LM3886 is an 8 MHz voltage-feedback audio power amp. The LM1876 is a 7.5 MHz voltage-feedback audio power amp. All these specs are readily available on TI's website. I'm curious how you manage to conclude any of those would fit the "300 MHz, high-current, current-feedback opamp" description.

Let's not forget about @March Audio's HPA1 too.
But sometimes a bit of confusion might actually help, so this is a highlight.

Yep. The HPA-1 is a common name for a headphone amp. I'm not concerned.

I hope this isn't off-topic, but since we're discussing volume pot configurations, I was wondering if you could comment on where you'd recommend adding a volume pot to a preamp based on your upcoming universal buffer board.

Sounds like a great question for an upcoming Universal Buffer thread on DIY Audio.

Tom

 

[Doodski]

As I described previously in this thread, the output stage in the HPA-1 is a composite amp with a 300 MHz, high-current, current-feedback opamp as the output stage.
The OPA1656 is a 53 MHz voltage-feedback opamp. The LM3886 is an 8 MHz voltage-feedback audio power amp. The LM1876 is a 7.5 MHz voltage-feedback audio power amp. All these specs are readily available on TI's website. I'm curious how you manage to conclude any of those would fit the "300 MHz, high-current, current-feedback opamp" description.


Yep. The HPA-1 is a common name for a headphone amp. I'm not concerned.


Sounds like a great question for an upcoming Universal Buffer thread on DIY Audio.

Tom

Off topic but on topic. Can we have a slew rate specification test at Audio Science Review and for your TCA HPA-1?

 

[tomchr]

I hope this isn't off-topic, but since we're discussing volume pot configurations, I was wondering if you could comment on where you'd recommend adding a volume pot to a preamp based on your upcoming universal buffer board.

Just launched: DIY Audio: Universal Buffer achieving -140 dBc (0.00001 %) THD

Tom

 

[BYRTT]

Off topic but on topic. Can we have a slew rate specification test at Audio Science Review and for your TCA HPA-1?

If you look for spurs of "300 MHz, High-current, current-feedback" remember that fine spec is selective for output opamp, in a normal DUT setup circuits upstream of output opamp incuding RFI filters is included.

 

[Doodski]

If you look for spurs of "300 MHz, High-current, current-feedback" remember that fine spec is selective for output opamp, in a normal DUT setup circuits upstream of output opamp incuding RFI filters is included.

That's why I requested amirm do slew rate test. To get a overall view of the circuit rise time and fall time relative to frequency with reactance shown at instantaneous values. Otherwise I could use the specifications <.pdf> for reference but that would not be real and would not indicate a power supply's effectiveness.

 

[trl]

300 MHz, high-current, current-feedback opamp

Well, it's quite easy not reading the datasheet while being in bed and reading & typing is done from the cellphone. Thank you for the reminding me the main specs about the composite amplifiers.

I think that THS3061 THS3062 could fit the initial specs, although I'm not sure it worth what's inside as long as the headamp measures so very well.

 

[JohnYang1997]

Well, it's quite easy not reading the datasheet while being in bed and reading & typing is done from the cellphone. Thank you for the reminding me the main specs about the composite amplifiers.

I think that THS3061 THS3062 could fit the initial specs, although I'm not sure it worth what's inside as long as the headamp measures so very well.

Quote from last page "I just noticed that I have a bug in the block diagram. The output stage is actually a composite amp with a high-speed, high-current output". I don't think more hint is needed.
Let's focus on the implementation or the performance instead of what chip is being used. I can guarantee most would have the same exact circuit schematic but with 3 times worse performance. Yet with another person there might be even better performance.

 

[tomchr]

Let's focus on the implementation or the performance instead of what chip is being used. I can guarantee most would have the same exact circuit schematic but with 3 times worse performance. Yet with another person there might be even better performance.

I agree. The schematic is an abstraction of the circuit. The PCB layout participates as part of the circuit as well. This is why I pay attention to the layout. It is also why I characterize the crap out of my circuits. Where else do you find a set of measurements or specifications as comprehensive as mine? Many manufacturers won't even give you fundamental specs.

Tom

 

[dragonspit4]

I agree. The schematic is an abstraction of the circuit. The PCB layout participates as part of the circuit as well. This is why I pay attention to the layout. It is also why I characterize the crap out of my circuits. Where else do you find a set of measurements or specifications as comprehensive as mine? Many manufacturers won't even give you fundamental specs.

Tom

oh, you should ask amirm to update the picture on this forum (first page) to the final production pictures. The current pictures makes the amp looks a bit "cheap" (if you get what i mean).
Just a suggestion.

 

[Music1969]

Hi @tomchr

Any issues with this amp driving planar magnetics like the Mr Speakers Aeon Flow Closed:

Sensitivity = 92 dB / mW

Impedance = 12 ohm flat

Cheers

 

[tomchr]

12 Ω is rather low, but should be fine. The distortion will be slightly higher than the 32 Ω data indicate, but will still be way below audible.

Tom

 

[solderdude]

Tom will also be integrating an LDR volume control as that sounds better and the measured results will be even better because he will finally be rid of the slide contact in a potmeter now that he knows potentiometers are not suited for audio.

 

[Music1969]

Tom will also be integrating an LDR volume control as that sounds better and the measured results will be even better because he will finally be rid of the slide contact in a potmeter now that he knows potentiometers are not suited for audio.

Is this true @tomchr ?

Or were you just making a friendly request @solderdude ?

 

[solderdude]

Just teasing Tom

LDR volume control embodies everything Tom is NOT looking for.

 

[restorer-john]

LDR volume control embodies everything Tom is NOT looking for.

Lirpa Labs.

Maybe Tom could team up with them and get a product to market by, let's say, early April 2020?

 

[Music1969]

Just teasing Tom

LDR volume control embodies everything Tom is NOT looking for.

Hehe

You're a fan of the Alps RK271-series ("Blue Velvet") pot to be used in the HPA-1?

 

[solderdude]

I have very positive experiences with that pot. The ones I have used all are about 30 years old (no recent ones) and have worked flawlessly till today.
No channel imbalance, no scratching, still the original 'feel'.