Sony VFET Amplifier Part 2

[Zen Mod]

Happy New Year to all.

Just got back to using the VFET2

From my last post Changed C1,C2 to 220uF and C3,C4 to 50uF (2 100uF in series) as suggested and didn't hear a putt putt (1 beat per second low frequency}

usingTea Bag boards and Botte changes

Now I have a problem in one of my channels (call it Channel A) . I turn on the power and bias current is about 2A and DC offset is 600mV. I can lower bias and offset by trimming P1, P3 and P4 - P2 seemingly doesn't have any effect. After an hour or so of tweaking I have 1.25A of bias and 50mV offset. Left amp on for 3 hours - everything stable. Channel B is at 1.25A of bias and 15mv of offset. Turn off amp.

Next morning Channel B is1.25A of bias and 15mV upon turning on the power
Channel A has reverted to 2.25A bias and 550mV of offset - which I can adjust in about an hour to 1.25A and 50mV offset.

Turn off power - high current and offset return to Channel A

Thoughts anyone?

Best and thanks

Bob

I would start changing those TL431 critters and rebiasing

maybe putput situation drove semi-berserk one of them

 

[flocchini]

Will do - you know I have avoided that
Thanks

 

[flocchini]

Changed the critters

Now I start with bias of 3.5- 4A and 8V offset. If I lower the bias with P3 and P4 offset up - P1 and P2 don't seem to do much - I think I will put it away and start over tomorrow.

 

[Nelson Pass]

Could you please explain the purpose of resistors R41 and R42 in the “Sony VFET Amplifier Part 2” article. In “The DIY Sony VFET” article from 2016 the same resistors are designated R23 and R24.

These provide a DC connection from the output of the front end to the
input of the output stage. As you see, they are bypassed by capacitors
which hold the bias voltage (since we want low AC impedance for driving
the output stage). AT DC we don't need such a low impedance, and the
resistors allow the front end to adjust the output DC.

 

[gl]

Thank you Nelson.

Graeme

 

[gl]

These provide a DC connection from the output of the front end to the
input of the output stage. As you see, they are bypassed by capacitors
which hold the bias voltage (since we want low AC impedance for driving
the output stage). AT DC we don't need such a low impedance, and the
resistors allow the front end to adjust the output DC.

So if we used similar resistors across the corresponding capacitors in the F4, could we then eliminate the offset adjustment?

I am building a BA-3B and want to use the latest VFET front end, hence the questions.

Graeme

 

[flocchini]

I have changed the TL431s.

Then back to basics.

Lifted R11 and R12 on one side. Adjusted P3 and P4 to get 1.8V across R25 and R26 with approximately 10mV offset at R20. Turned off power and then checked next day. Everything returned to original measurements so I think this is stable.

However with R11 and R12 still lifted I used P1 and P2 to set voltage at R1 and R4 so that I know VFETs are shut down. I can get -14V at R1 but only can get a few mV at R4.

Next move ???

Best and thanks

Bob

 

[Zen Mod]

all that with disconnected outputs?

 

[flocchini]

Yes

 

[Zen Mod]

in this moment , all what's important of entire amp is just two voltage regs made of TL431 ........ so , it's simple - think of them as plain small regs and nothing more

check surrounding resistors , same as loading of regs

are R4 and R5 still OK ....
are C5, C6 OK .........
are R41 , R42 OK .........
are Z3, Z4 OK ..........

 

[Nelson Pass]

So if we used similar resistors across the corresponding capacitors in the F4, could we then eliminate the offset adjustment?

You would still need the resistors going to the supplies to bias up the
bias network. Also, you would then have to cap couple the front end
and provide an offset adjustment for the Jfets.

 

[flocchini]

Will do

 

[gl]

You would still need the resistors going to the supplies to bias up the
bias network. Also, you would then have to cap couple the front end
and provide an offset adjustment for the Jfets.

Understood. Thank you again.

Graeme

 

[flocchini]

from ZM
in this moment , all what's important of entire amp is just two voltage regs made of TL431 ........ so , it's simple - think of them as plain small regs and nothing more

check surrounding resistors , same as loading of regs

are R4 and R5 still OK ....
are C5, C6 OK .........
are R41 , R42 OK .........
are Z3, Z4 OK ..........

Visually nothing smoked

Measuring in circuit

C5,C6 222uF
R5,R6 8.5k
R41. R42 15k
Z3 18k, Z4 =15k not sure that's what I should be measuring but wanted to check that there wasn't an open
R4 220

Thanks ZM

 

[Dennis Hui]

Hi Bob,

Any chance that P1 could be broken/shorted out?

Dennis

 

[Zen Mod]

Bob , R4 was a typo , I meant R5 and R6

whatever ....... as I said - you're troubleshooting "just" two , in fact - just one reg circuit

change everything in that reg circuit if needed , don't fret about reason why something broke ..... so - lift resistor leg and measure it , check pot ( as D. said) , replace TL again .... replace all mentioned parts , but don't let funny reg to spoil your fun

 

[flocchini]

ZM, Dennis

Thank you !!!

I will proceed

 

[flocchini]

OK the latest

replaced P1 ( thanks Dennis ) it was flaky

pulled and checked R5,R6, R41, R42 they are OK

replaced c5,c6

replaced z3.z4

replaced the critters

now no negative on R1 and about .5V on R4

I did some stuff on bottom of board so I need to check everything - something could be backasswords or not soldered correctly - will let it rest for a day or two

Thanks guys

 

[kawihornet]

Any Idea if a new batch of these would ever be released again?
Thanks

You would still need the resistors going to the supplies to bias up the
bias network. Also, you would then have to cap couple the front end
and provide an offset adjustment for the Jfets.

 

[Dennis Hui]

Papa mentioned a new design is forthcoming:

https://www.diyaudio.com/forums/pass-labs/307071-2sj28s-35.html#post5638929