BF862 based SE Class A Headamp without the HEAT

Founder of XSA-Labs
Joined 2012
Paid Member
Using a Behringer UCA202 (self H2 and H3 of 15dB above niisenfoor of -110dB) I get this. Subtracting 15dB from this measurement gives a pretty close agreement between simulation and measurement.

UCA202 loopback:
591916d1484326615-mosfet-source-follower-headamp-uca-202-fft.png


BF862-ZVN4306 pocket amp:
591917d1484326615-mosfet-source-follower-headamp-bf862-vzn4306-fft.png
 
As for output cap: Low ESR means less potential effect from it being not a prefectly static/constant resistance (it's a chemical "resistor", mind you). Though it's more important to have no AC across the cap in the passband, that's why e-caps in a high-pass are usually selected to have 10x lower corner freq than actually needed.
Nice little project!
 
.....................Though it's more important to have no AC across the cap in the passband, that's why e-caps in a high-pass are usually selected to have 10x lower corner freq than actually needed.
Nice little project!
This is the message that D.Self started way back in the 1990s using different words.
It has still not got through to the general Membership.
 
If there is going to be any audible difference between AC coupling output capacitors, then the biggest difference will be between an electrolytic type and a plastic film MKT type.

The difference between MKT/S and MKP will be miniscule in comparison.
The added distortion when using an MKT as a filter is tiny and so low it is difficult to measure. But in this signal coupling duty the plastic film cap is not supposed to be acting as a filter. The MKT is supposed to be passing the audio signal.
 
I actually have Vishay MKP yellow one
:eek: I've never seen 470nF polypropylene that compact before. The highest capacity I've found in a 5mm pitch is 330nF, and those suckers are 8.5mm wide.

The difference between MKT/S and MKP will be miniscule in comparison.
Good point. And as far as the value goes? Is 470nF good, or should we consider something else?
 
if you want to pass 100Hz then I suggest you choose the F-3dB of the high pass filter to be one decade below that, i.e. 10Hz high pass filter will pass virtually all of the audio band from 100Hz upwards.

If you have F-3dB = 10Hz and the capacitor is 470nF, then the load (the R part of the CR filter) will be
R = 1 / {2 Pi F C} = 338r6
i.e. a 340ohms headphone will receive 100Hz and upwards when a 470nF coupling capacitor is used.
Frequencies below 100Hz will be progressively rolled off.
If one changes the headphones to 70ohms then the filter becomes 48.4Hz and frequencies above 484Hz pass through to the headphones.

It is simply manipulating the standard formula for a passive RC filter.
F-3dB = 1 / {2 Pi R C}
 
Founder of XSA-Labs
Joined 2012
Paid Member
I just redid the measurements with a fresh set of batteries and it looks really good now.

592184d1484436971-sound-two-transistors-clapping-pocket-class-good-fft.png


592185d1484436971-sound-two-transistors-clapping-pocket-class-good-fr.png


Note that the frequency response doesn't show any sign of falling off at the lower frequencies indicating that the 47uF cap is well suited for the 270R load.

This amp is the real deal - no kiddin' around here - the above plots are seriously excellent in my opinion.
 
Last edited:
:eek: I've never seen 470nF polypropylene that compact before. The highest capacity I've found in a 5mm pitch is 330nF, and those suckers are 8.5mm wide.

Those are MKT caps with 5mm LS, not MKPs.
As you stated, MKP would be much larger since that value would have a LS of at least 7.5mm.
MKTs are still quite good for audio coupling.

Baby, are you going to do a board with through-hole resistors?
Personally, I don't care whether it fits into a can or not.
 
Last edited: