• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

xrk971 Pocket Class A Headamp GB

That looks familiar - the 8mA bias on the BF862 may be a tad high but probably fine. It will depend on large part, the actual BF862 variations. I have tested hundreds of them with an in situ bias current test rig and then try to match left and right channels for bias current at the same exact conditions. Interesting mosfet - never heard of it but if you say transconductance is low, should be good.
gm is actually much higher (nt lower) than ZVN4306
 
Founder of XSA-Labs
Joined 2012
Paid Member
Model has been used and verified by many folks over in Pass and Headphone Amps forum.
You did not look very hard - but that's ok.

Look at top line in my LTSpice schematic for the model that I got from one of those threads.

588533d1483117660-mosfet-source-follower-headamp-juma-aksa-bf862-lu1014d-schematic-v1.png


Predicted FFT:
588535d1483117660-mosfet-source-follower-headamp-juma-aksa-bf862-lu1014d-amp-fft.png


And I built the above amp, sounds very nice:
588699d1483173489-mosfet-source-follower-headamp-aksa-xrk971-jfet-ha-02.jpg


Matched JFETs here:
LU1014D JFET's

Alas, I stray OT...
 
Last edited:
thanks, I was looking for the spice file on the internet . Could not find it as I just had to look for it on the diyAudio site:D.
Generally speaking, I have issues with using such spice "file" as it lacks many infos. Only the very minimum is there so that the simulation can run.
Another point is that I run my simulaiton with the official BF862 spice model. Yours differs significantly from the one I use.
 
Founder of XSA-Labs
Joined 2012
Paid Member
My BF862 model also comes from DIYA members - ElFishi's model:
LTSPICE model for your BF862 JFETs

But has been verified by myself as fairly accurate in terms of getting the DC set points in the circuit to match an actual build, and FFT to match measurement.

There are a couple of threads on BF862 models, and a more complex one involves using the measured Idss to set the interpolation between the bounding book-end cases. I use the simpler model and it gets close enough.

Mark Johnson's model:
LTSPICE models of worst-case BF862 JFETs

I did not know there was an official BF862 model from the manufacturer. Can you share the model?
 
Last edited:
sure :
* BF862 SPICE MODEL MARCH 2007 NXP SEMICONDUCTORS
* ENVELOPE SOT23
* JBF862: 1, Drain, 2,Gate, 3,Source
.SUBCKT BF862 1 2 3
Ld 1 4 1.1n
Ls 3 6 1.25n
Lg 2 5 0.78n
Rg 5 7 0.535
Cds 1 3 0.0001p
Cgs 2 3 1.05p
Cgd 1 2 0.201p
Co 4 6 0.35092p
J1 4 7 6 JBF862
.model JBF862 NJF(Beta=47.800E-3 Betatce=-.5 Rd=.8 Rs=7.5000 Lambda=37.300E-3 Vto=-.57093
+ Vtotc=-2.0000E-3 Is=424.60E-12 Isr=2.995p N=1 Nr=2 Xti=3 Alpha=-1.0000E-3
+ Vk=59.97 Cgd=7.4002E-12 M=.6015 Pb=.5 Fc=.5 Cgs=8.2890E-12 Kf=87.5E-18
+ Af=1)

.ENDS BF862
 
10k! Wow - planning on a nice eBay business in the future... Remaining stock on Mouser is 23k and guys like you will deplete the reserves very quickly. :p

At least in a full reel of 3k units they are $0.16ea.

You can still order tons of reels from NXP., mouser or digikey if you want. Last production run is foreseen in one year from now. I'll buy those for the lab, not for me personnally.
I design multichannel preamps. In my last batch I used 1000 units..
 
xrk,
the more i investigate your design , the more qyestions i have for you.
Why did you chosse to reduce the jfet gain to about 3 ?
increasing the gain to 5 to 6 significantly improve the output voltage swing ( clipping on the negative rail)
there must be a reason .. maybe TH2 improves a low gain ?
Not really sure how i could simulate the TH2/TH3 actually...
 
I ran a simulation with gain 5 and another MOSFET : a dual device from ST : STL40DN3LLH5
measured TH2 seems ok.
The simulation looks promising. Unfortunately, the spice does not include all features allowing to extract the total noise.
However, driving that little beast at 150 mA , VDS = 10V should not be an issue, even without a heatsink.
I simulate the circuit with output impedance as low as 8 ohms.
 
Founder of XSA-Labs
Joined 2012
Paid Member
You need something to draw the voltage on the cap down on initial power up to reduce turn on pop. But I find that having a bit extra load on 250ohm headphones to being overall impedance down to circa 120ohms sounds better. It's also convenient when measuring the amp I don't need to add dummy load. Probably 1k dummy is fine to drain the cap if you want to use small caps and don't plan on driving low impedance headphones.
 
Founder of XSA-Labs
Joined 2012
Paid Member
Sodacose's Estudiante Amp

Well this is my first tube amp, ever. Thanks to Sodacose for providing me the tube and schematics and info (available here in his blog).

Schematic:
An externally hosted image should be here but it was not working when we last tested it.


I had some issues getting it to run initially because I did not have the correct "HV" version of the LM317. So I am using a simple 270R 5W resistor as the MOSFET current sink. So very similar to the PCA in topology. It's running off of 45.5v rails at present (3.5v hit due to MOSFET cap Mx). Gives me about 89mA bias current which is plenty for the OB-1 or DT880-250 cans.

I did modify the circuit a bit - I added a 1uF 63v Wima MKS input coupling cap plus a 680pF 100v film cap to ground on the input to filter out RF pickup and added extra 1000uF for 1470uF total output cap (just cheap no-name Chinese caps - but measured relatively low ESR). Right now the LM317HV is simply replaced with 270R wirewound cement 5W resistor. I am using IRF610's as I have those on hand. Works fine with the IRF610's and some local heatsinks. When I do get the LM317HV's the four TO220's will provide a 4-point under-hung bolt pattern to the bottom panel of an aluminum case for a heatsink.

It's a neat little circuit and sounds wonderful. Just test listening right now. No measurements yet but I can tell it has the classic SE Class A harmonic profile that we all have come to love from the PCA. This is a stereo tube hybrid amp that fits on a 6cm x 8cm veroboard. The cap Mx is probably not needed if you have a clean linear source of 48v. I am using a DC-DC step up SMPS so, amd trying to filter out as much as possible - plus I wanted a soft start mechanism to gradually apply power upon turn on to avoid in-rush current that would trip the SMPS into shutdown.

I'll probably start a new thread to show details of this build at a later point.

attachment.php
 

Attachments

  • Estudiante-amp-P2P-01.jpg
    Estudiante-amp-P2P-01.jpg
    371.2 KB · Views: 467
Last edited: