F5 Headamp ?

[xrk971]

EUVL,
I am grateful for the layouts and schematics you have provided for the F5HA. It's one of the best I have heard and just want to thank you for making it available. We realize you have a business and understand if certain designs are not "free". What does a HA protection PCB cost to buy from Xen Audio?
Cheers,
X

 

[RollE2k]

I bought three pieces of those boards, payed 80HK$ each, including shipping.

 

[e_fortier]

Not bad at $10 USD / PCB including S+H.

 

[EUVL]

Just to demonstrate the performance level of the F5X-HA when properly built :

For 1Vrms no load,
H2 -112dBV
H3 -104dBV
H4 <115dB (limit of the measurement equipment)
H5 -112dBV


Patrick

.

 

[EUVL]

For 1Vrms into 75R
H2 -118dBV
H3 -99dBV
H4 <115dB (limit of the measurement equipment)
H5 -118dBV

Patrick

.

 

[EUVL]

For 1Vrms into 30R
H2 -97dB
H3 -85dB
H4 <115dB (limit of the measurement equipment)
H5 -113dB


Patrick

.

 

[EUVL]

At 2Vrms into 300R, relative to the 6 dBV fundamental :
H2 -109dB
H3 -103dB
H4 <115dB (limit of the measurement equipment)
H5 <115dB (limit of the measurement equipment)

As a benchmark, the Pass HPA1 is quoted to have a THD <-86dBV at 1Vrm, 30R.
At 2Vrms into 300R,
H2 -83dB
H3 -112dB
H5 -112dB

https://passlabs.com/amplifier/hpa-1
Pass Labs HPA-1 headphone amplifier Measurements | Stereophile.com


Patrick

.

 

[RollE2k]

At what bias is this? And when do you expect the boards to be for sale?

 

[EUVL]

With inputs shorted to Gnd, output noie referenced to Gnd is < 25uVrms,
measured uing a Wurcer 1000x LNA.

http://www.diyaudio.com/forums/soli...1000-low-noise-measurement-amp-ikoflexer.html


Patrick

 

[EUVL]

Properly built means :
-- curve-trace matched FETs at working temperature from a large pool
-- selected and if necessarily matched passive components
-- appropriate low noise power supply
-- proper PCB layout
-- above all a thorough understanding of how the circuit works


Patrick

 

[EUVL]

Also, to demonstrate the performance difference of normal single-point Idss and Vgs match,
For 1Vrms no load,
H2 -90dBV ( 12.6x worse than curve-tracer match )
H3 -102dBV
H4 <115dB (limit of the measurement equipment)
H5 -112dBV


Patrick

.

 

[jean-paul]

At what bias is this? And when do you expect the boards to be for sale?

I asked the same question but find no answer while boards apparently have been for sale. The fact that many don't know says the "GB campaign" was not successful.

Patrick please give a link where we can find the original boards with prices etc.

 

[EUVL]

Bias was 150mA.
Increasing bias to 250mA will lower distortion into 30R load even further.
But we need a better heatink enclosure first.

F5X-HA will not be published.
As already mentioned it will only be available if there are at least 10 subcribers.
We reserve the right to whom we supply or not (only for serious DIYer's who know what they are doing and appreciate what we do).


Cheers,
Patrick

 

[sikahr]

In that case it's time to publish my design files, maybe someone not serious enough want to do something with it.

 

[e_fortier]

I love the KISS principle si I guess I'm not serious..thank you for the file, I'll look at it later.

Could someone find the bias value so that THD is mainly H2 and H4 with very low H3 and H5 (the worst). My headphone ar 56 ohm.

Thanks,
Eric

 

[sikahr]

Gerber printscreen. Looks OK to me.

 

[xrk971]

Sikahr,
I can't tell if above is double sided - maybe not? Would it be too much to ask for a single-sided home-etching friendly version and pdf's?
Thank you.

 

[AndrewT]

It is double sided. The left side is top.

 

[mlackey]

LTSpice models for FQP3P20/FQP3N30

Does anybody have working LTSpice models for FQP3P20 and FQP3N30?

 

[EUVL]

The V2 GB PCB sets have been shipped by registered surface mail today.
You will be informed separately per email for tracking info, etc.

Thank you for your patience,
Patrick