Yamaha B2 with bad VFETs....what are my options. Need advice.

[TommyW]

I just came across a Yamaha B2 for cheap. Paid $100 for it and a Yamaha C2 knowing that the left channel was not working on the B2. Previous owner said they were playing it, heard a pop and it went into protection. I did not know anything about this amplifier before I purchased. Long story short, I got it home pulled what I thought were the output transistors on the left side and 3 of the 4 VFETs on the left side tested bad with just a simple diode test on my DMM.

I am somewhat of a novice, especially when it comes to VFETs amplifiers. At this point, given the availability of replacement VFETs, I am unsure on what to do. This unit has the 2SK76A and 2SJ26A upgraded vFETs. I don't really want to part the unit out. I had a couple of questions for any of the B2 experts.

1.) Somewhere I read the unit can be ran on just one pair of VFETs per channel. Did I misunderstand that? I could not find any documentation around this and if this is possibly what's the downside?

2.) Has anyone had any luck sourcing replacements? I do see 2SK76(non-A) versions for sell on Ebay for $150 from China, but I am skeptical these aren't just fakes. Also, could I replace the 2SK76As with just the 2sk76s?

Obviously, if I did go the route of purchasing replacements, I would want to identify the root cause of what occurred on the left channel before installing new ones, but I will cross that bridge if this is what I decide.

 

[avionic]

I am somewhat of a novice, especially when it comes to VFETs amplifiers.

Maybe bad. Maybe not. VFETS don't DMM test like a normal BJT.


See post #25 in the following link.

Post in thread 'How expensive are replacement V-fet's?'

Sep 4, 2019

Yes thats true. I just assumed that they ARE dead, but thats hopefully not the case.....

To the OP - there is a sony service bulletin over at hifiengine.com that has a very basic procedure for testing if a VFET is OK, by simply using a DMM.

Complements of Glenn aka echowar
Bottom portion of this pdf is VFET testing with a multimeter.

• avionic

• 

? I do see 2SK76(non-A) versions for sell on Ebay for $150 from China,

Junk !!!! Any VFET's from China are junk.

 

[avionic]

@rottalpha ,@Vint Age

 

[avionic]

I am somewhat of a novice, especially when it comes to VFETs amplifiers.

B-2 is not a very good candidate to practice on. IMHO.

 

[TommyW]

B-2 is not a very good candidate to practice on. IMHO.

Understood. I'm restored about 100+ receivers and amplifiers, I'm just a novice to the VFET ones. I have enough knowledge to not cause any additional damage at least....hopefully. At this point if the vFETS are as unobtanium as they appear to be, I have a feeling it is going to be sitting on a shelf until my wife inherits it

 

[TommyW]

Maybe bad. Maybe not. VFETS don't DMM test like a normal BJT.


See post #25 in the following link.

Post in thread 'How expensive are replacement V-fet's?'

Sep 4, 2019

Yes thats true. I just assumed that they ARE dead, but thats hopefully not the case.....

To the OP - there is a sony service bulletin over at hifiengine.com that has a very basic procedure for testing if a VFET is OK, by simply using a DMM.

Complements of Glenn aka echowar
Bottom portion of this pdf is VFET testing with a multimeter.

• avionic

• 

Junk !!!! Any VFET's from China are junk.

Also, meant to say this is the method I used. Found it in a thread earlier. They are definitely bad.

 

[Vint Age]

1.) Somewhere I read the unit can be ran on just one pair of VFETs per channel. Did I misunderstand that? I could not find any documentation around this and if this is possibly what's the downside?

I've never tried it myself but yes, in theory you should be able to set the B-2 to run with just 1 pair of V-FET's per channel. You'd have to adjust the bias to half of what's normal. The down side is less power. That said, it will still sound really nice.

Note that the B-3 only uses a single pair per channel ...

2.) Has anyone had any luck sourcing replacements? I do see 2SK76(non-A) versions for sell on Ebay for $150 from China, but I am skeptical these aren't just fakes.

Like Dave says, stay away from V-FET's from China.

Also, could I replace the 2SK76As with just the 2sk76s?

Yes you can but all the V-FET's need to be very well matched within a channel making it even more difficult to find replacements.

Obviously, if I did go the route of purchasing replacements, I would want to identify the root cause of what occurred on the left channel before installing new ones, but I will cross that bridge if this is what I decide.

Yes indeed!

 

[PE9ZZ]

There are two options. Get NOS VFETs (actually they are calles SITs for Static Induction Transistor, VFET is a type of MOSFET), match them carefully and put them in. Option #2 is use laterals but you need to exchange the gate drives and lower the gate voltage. You could even try actual V-MOSFETs like IRFP(9)240 but they have a different package. Again the gate voltage is different as these are all enhancement FETs and the SITs are depletion FETs. Look it up...

But whatever you choose to do is make very sure the drive is correct and matches the gate voltage for your output FETs. You do this by putting a 1k resistor across the gate and source terminals of the top and bottom FETs. Once you are confident the drive is correct you can put the FETs in.

 

[TommyW]

There are two options. Get NOS VFETs (actually they are calles SITs for Static Induction Transistor, VFET is a type of MOSFET), match them carefully and put them in. Option #2 is use laterals but you need to exchange the gate drives and lower the gate voltage. You could even try actual V-MOSFETs like IRFP(9)240 but they have a different package. Again the gate voltage is different as these are all enhancement FETs and the SITs are depletion FETs. Look it up...

But whatever you choose to do is make very sure the drive is correct and matches the gate voltage for your output FETs. You do this by putting a 1k resistor across the gate and source terminals of the top and bottom FETs. Once you are confident the drive is correct you can put the FETs in.

Given finding just one 2SK76/2SJ26 or 76A/26A is hard, the odds of me finding all 4 matched seem non existent. Given I have $100 for the amp and C2 preamp which works perfectly, it's really no loss for me.

When I pulled the current vFETS, they all have writing on them underneath(e.g. 2269, 2270) , which I assumed was from a previous tech for matching purposes.

 

[Nashou66]

@Vint Age Is it possible to use Sony V-fets in a B-2 with some possible circuit changes ?

I thought i read some where that it has been done.

Athanasios

 

[PE9ZZ]

@Vint Age Is it possible to use Sony V-fets in a B-2 with some possible circuit changes?

You can us 2SJ18/K60 provided you have them reasonably matched for V(gs) and you carefully make sure drive is correct. You could even use a single pair which means you can use two transistors from the other channel provided you don't overload the amp. 2SJ18/K60 are 63 W devices, the J26/K76 can handle 100 W.

Edit: the Sony devices have much higher V(gs) requirements so this prolly is not a good idea.

 

[Vint Age]

Edit: the Sony devices have much higher V(gs) requirements so this prolly is not a good idea.

Yup, not a good idea ... The Sony J28/K82's might work but you'd need a bunch of each to get a possible matched set. In fact with those you could probably get away with a single pair per channel due to the dual die.

 

[Nashou66]

Oh so you would have to modify the power supply I guess as well as other things.

Athanasios

 

[PE9ZZ]

The Sony 2SJ28/K82 are rated for 95 W but may still have the larger V(gs) requirements than the 2SJ26/76 made by Yamaha.

 

[PE9ZZ]

You can even cheat and replace the whole shebang with regular bipolar transistors like MJ15024/5. You can toss a whole slew of components in the process. No one will hear the difference

 

[Sparkplug]

No one will hear the difference

LOL, I think anyone who can’t hear the difference has no business buying a B-2 to begin with. The first thing I noticed when I bought my B-2 was the difference in audio quality and then there was no going back.

 

[nakmandan]

I think you are already ahead just getting the C2 pre for $100. Sell off the B2 for parts or repair and you will have paid for the C2 and have some extra cash left over. Win win.

 

[Sparkplug]

… in theory you should be able to set the B-2 to run with just 1 pair of V-FET's per channel. You'd have to adjust the bias to half of what's normal. The down side is less power. That said, it will still sound really nice.

Absolutely, look no further than B-3, one of the best sounding amps Yamaha ever made. If this was my amp, I’d try this approach. Seems straight forward (in theory) enough to try. I’m sure though other adjustments may be required. After all, how often do you even listen to more than 10W?

 

[rottalpha]

I would want to identify the root cause of what occurred on the left channel before installing new ones,

I can help you with that: do not operate a 50 years old amp for daily use and expect any sense of sustained reliability

If you want something closer to the "why" of the actual big-bang: check the drivers and check the bulk caps on that channel. My money is on the driver section and that the amp must have been used without proper air circulation - read: inside a cabinet or with that C2 stacked on top of it. V-fets can take a lot of heat. Those original drivers were already used in the upper end of the SOA, and the probability of them shitting the bed, was always high. In the B3, Yamaha started using more robust drivers for half of the output vfets....Not sure why they did not have a recall.

Forget about simply repairing that channel...something else is likely to crap up in short order after. The amp need full restoration to be used with any sense of reliability.

in theory you should be able to set the B-2 to run with just 1 pair of V-FET's per channel. You'd have to adjust the bias to half of what's normal.

like he said^^^^ - that is your most feasible option.

actually they are calles SITs for Static Induction Transistor, VFET is a type of MOSFET

They are ALSO called SIT but also V-Fet, based on die geometry and a name used by Yamaha, Sony, NEC whom are the reason these devices were even made available for audio, not to be confused with V-MOSFET. V-Fet is closer to a J-fet as in they are both enhancement mode. Not much to do with the Metal Oxide part of the mosfet. As you called out, some MOSFETS can be enhancement mode bus still, there are other differences that need mitigation.

That said, lateral MOSFETs sound quite nice as well as you know from your Sony conversion

Edit: the Sony devices have much higher V(gs) requirements so this prolly is not a good idea.

True but it depends on rank. The lower rank K60/J18 could operate within the same same range Vgs as the Yamaha ~ 9-16Vgs @~ 60Vds, or the circuit can be slightly modified.
Getting matched pairs would actually be the more difficult part as the Sony rank span is 2.5V where Yamaha matching for same set of 4-vfets per channel is typically within less than 0.5V.

The Sony J28/K82's might work but you'd need a bunch of each to get a possible matched set

Yes, these would likely be the better candidates but rank needs to be checked for Vgs and for tight matching.

 

[avionic]

B-2 is not a very good candidate to practice on. IMHO.