Aksa Lender HPA

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Founder of XSA-Labs
Joined 2012
Paid Member
I wanted to bring this preamplifier to the attention of folks in this forum to show how well it can work as a headphone amp, when modified to have higher bias current and lower voltage rail. Originally designed to drive high impedance loads up to 40Vpp, a few resistor tweaks and swapping output device from KSA1381 to 2SA1837 (n.b., pinoutss are flipped), increasing bias from 17mA to 130mA, and dropping voltage from 48V to 24V. But all in same basic package. This is a SE Class A amp with a modified LTP input stage. No fancy components, no matching needed, just nice performance and great sound. Very easy to build. Thanks to Aksa (Hugh Dean) for circuit design, and JPs64 for superb professional layout work.

Original thread here:
AKSA's Lender Preamp with 40Vpp Output

Headphone Amp build here:
AKSA's Lender Preamp with 40Vpp Ouput GB

Amp looks like this:
657048d1516059914-aksas-lender-preamp-40vpp-ouput-gb-aksa-lender-hpa-amp-main-jpg


The amp core modules fit on low cost swappable daughterboards so changes to gain, harmonic profile etc can be adjusted or swapped in/out:
657047d1516059914-aksas-lender-preamp-40vpp-ouput-gb-aksa-lender-hpa-amp-cores-2-jpg


For those inclined, an SMT version (both MELF or 0805/1206 available) can be done with lower bias current of 80mA for higher impedance cans:
654036d1514801644-aksas-lender-preamp-40vpp-ouput-gb-smt-melf-build-jpg


Here is schematic of amp core daughterboard, no input or output caps here as they are on the main board. I will post changes made to this schematic to enable use as HPA below:
649512d1512425408-aksas-lender-preamp-40vpp-ouput-gb-jps64-aksa-lender-preamp-th-daughterboard-schematic-png


Here is measurement for 2.0Vpp into 50ohms (title is incorrect with regards to voltage):
657054d1516060213-aksas-lender-preamp-40vpp-ouput-gb-aksa-lender-hpa-130ma-1-0vpp-50hms-png


Here is same 2Vpp into 260ohms:
657111d1516086583-aksas-lender-preamp-40vpp-ouput-gb-aksa-lender-hpa-110ma-bias-2-0vpp-260ohms-hp2760-jpg


It goes without saying that it sounds great, as all of Hugh’s SE Class A creations usually are.

For above schematic, change R15 to 11ohms (two 22R 1% metal thin film 1/4w in parallel), R10 to 6k8 carbon film or 4k7 if you want lower 12dB gain, R14 2.2ohm 1% metal thin film 1/4 watt, Q4 becomes Toshiba 2SA1837. And drop Vcc to 24v at the amp (circa 28v upstream of cap Mx).
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
I get a lot of questions on those caps! LOL

they were only about $1.60ea and yes, from Aliexpress. I literally used them to give a nice racy look to a bland looking green PCB! :D

They sound fine and as good as an equivalent part from a European brand. All my other caps are real Nichicons, Pansonic, Wima (except for the 100uF 63v bias smoother on the daughterboard - those are from Ali but measure exceptionally well on LCR meter and I have a bag of 100 of them).

My listening tests with critical high resolution test tracks with 192kHz FLAC show that this amp with all of its coupling caps, is just as resolving, if not even more so than a DC coupled Schiit M3, but with superior soundstage and transparency.

So I say, for $1.60, it's harmless and go ahead and buy them and try them. Another good source of nice MKP tube shaped audio caps is Dayton speaker crossover caps. About $3 for the same thing but in 2.2uF and 250v and 1%. I'm sure they are from China but don't say so on the cap.

Dayton Audio PMPC-2.2 2.2uF 250V Precision Audio Capacitor
 
Last edited:
I get a lot of questions on those caps! LOL

they were only about $1.60ea and yes, from Aliexpress. I literally used them to give a nice racy look to a bland looking green PCB! :D

They sound fine and as good as an equivalent part from a European brand. All my other caps are real Nichicons, Pansonic, Wima (except for the 100uF 63v bias smoother on the daughterboard - those are from Ali but measure exceptionally well on LCR meter and I have a bag of 100 of them).

My listening tests with critical high resolution test tracks with 192kHz FLAC show that this amp with all of its coupling caps, is just as resolving, if not even more so than a DC coupled Schiit M3, but with superior soundstage and transparency.

So I say, for $1.60, it's harmless and go ahead and buy them and try them. Another good source of nice MKP tube shaped audio caps is Dayton speaker crossover caps. About $3 for the same thing but in 2.2uF and 250v and 1%. I'm sure they are from China but don't say so on the cap.

Dayton Audio PMPC-2.2 2.2uF 250V Precision Audio Capacitor

Will try them, thanks :)
 
Founder of XSA-Labs
Joined 2012
Paid Member
Hi X,

Amazing to see how versatile this circuit/board can be! Which version shall I build now :confused:???
I also love your 'hydraulic' pot knob :D !
Cheers,

Jacques

Well, the nice thing is with this flexible design is you don't need to make a whole new board for each amp. Just make a second set of daughterboards with the revised parts and have on hand for a headphone amp. You could also, just use the HPA as a preamp if you don't need to large voltage swing. The daughterboards and the components are very inexpensive to make.

Another variant is to run Vcc=44V and 71mA bias and that would work as both a high swinging preamp and a headphone amp suitable for 120ohms or greater (or high sensitivity IEMs) headphone amp. Jack of all trades. Dissipation on outputs would be 1.56w/ea the same as 130mA at 24V.
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Hi X,
Thanks for the additional information on the headphone amp.

You might want to mention in this thread that the pin configuration for the 2SA1837 is different to the KSA1381 transistor.

Good point, but I figured folks would realize that when they seen that the 2SA1837 mounts on the backside of the board - which is a good thing too for two reasons: there is not enough room on one side for two heatsinks, and secondly, the mass distribution is balanced so the card doesn't lean :)

I had a few more hours of critical listening last night and played with the input caps. I removed the 4.7uF Wima and left the ChiFi 1.0uF MKP. The bass extension was reduced by -0.5dB at 20Hz and 1.3dB at 10Hz, as predicted. But this is a fine value given that there may be improved claroity and resolution from using just one input cap.

657277d1516169951-aksas-lender-preamp-40vpp-output-aksa-lender-hpa-110ma-bias-input-cap-study-fr-jpg
 
I get a lot of questions on those caps! LOL
they were only about $1.60ea and yes, from Aliexpress. I literally used them to give a nice racy look to a bland looking green PCB! :D
Humble Homemade Hifi - Cap Test
They have a slightly different character to a standard MKP like the Intertechnik Audyn Cap QS4, just to name one of many. The "Audiophiler" has an emphasis on the lower treble range making them a bit upfront or foward sounding. This nature does make them subjecively more dynamic which can work well in certain systems or with certain types of music such as solo piano or string quartets. Not completely neutral but they can be usefull in certain situations. They remind me a bit of the Solen PPE-MKP-FC in their presentation. Play vocals and they can get a bit shouty sometimes. Compared to a Bennic FPP the "Audiophiler" are slightly rougher, less smooth. All in all not bad but also nothing special. Like the Solen they are a good step up from standard MKT capacitors often found in factory built speakers in being a fraction more spatial and open.

Dan.
 
Founder of XSA-Labs
Joined 2012
Paid Member
All cased up

A very utilitarian aluminum Bud box...
The DC-DC step up is one of the flat metal core ones, and is mounted underneath the main PCB attached directly to the bottom wall with a layer of Kapton in between. 5.5mm barrel jack for power in from either a SMPS laptop supply or Class 2 wall transformer (12v 1A works well). RCA input jacks on the back. Power indicator light on rocker switch and separate LED shows voltage on the PCB after the cap Mx charges up.

attachment.php


Closeup showing DC booster mounted underneath main PCB:
attachment.php


attachment.php
 

Attachments

  • Aksa-Lender-HPA-Amp-Case-01.jpg
    Aksa-Lender-HPA-Amp-Case-01.jpg
    333.8 KB · Views: 2,727
  • Aksa-Lender-HPA-Amp-Case-02.jpg
    Aksa-Lender-HPA-Amp-Case-02.jpg
    317.6 KB · Views: 1,967
  • Aksa-Lender-HPA-Amp-Case-03.jpg
    Aksa-Lender-HPA-Amp-Case-03.jpg
    299.6 KB · Views: 2,513
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Hi X,
are you running the risk of adding noise/ distortion to the main board by have the DC-DC boost board underneath it?

The proximity of the DC-DC step up doesn't seem to affect the noise pickup on the board at all. These are some of the lowest noise background measurements I have seen. Here is 2.0Vpp into 42ohm load.

attachment.php
 

Attachments

  • Aksa-Lender-HPA-130mA-2.0Vpp-43hms.png
    Aksa-Lender-HPA-130mA-2.0Vpp-43hms.png
    76 KB · Views: 1,042
I let the amp warm up for a bit and put the cover on and the THD dropped down to 0.01% from earlier value of 0.023% for 43ohms and 2.0Vpp:
Is this THD reduction due to temp, or due to the cover ?.
I have noted change in amplifier sound according to no cover, steel cover, aluminium cover or timber cover....same amp, covers belonging to other amps in the shop at the time.

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