• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Yamaha B2 Lateral MOSFET

H

hazze

New Member
About two years ago, I bought a B2 at an auction without knowing if it worked or not. The seller made no promises, so I took a chance and got it for 4k SEK.

Upon getting it home, I never plugged it in; just one look under the hood showed that things were not as it should be. I removed all of the VFETs and measured them as described in other threads here on AK, and all except two of them were bad.

Well, it was not working and was not in the best shape otherwise, so what to do? Getting four pairs of matched VFETs seemed impossible. Anyway, I started restoring it; I took it apart, cleaned everything, went through all the PCBs checking every component, and followed all the advice in the different restoration threads here on AK. A big thank you for making all this information available!

At the same time, I started doing simulations in LTspice with the driver card and Exicon ECW20N20/ECW20P20 double-die laterals for outputs. I was trying to find a solution that had minimal impact on the driver cards, making it fully reversible if I ever got my hands on working VFETs. It looked promising, and I made some PCBs for the laterals that would use the same connectors as the original outputs.

Long story short, it took me about six months of working on and off with this project, sorting out all the issues—where oscillation was the biggest problem. But now, I have had it working for approximately 18 months.

It measures as it should and sounds exceptional with my NS-1000.

More posts will follow, describing what has been done to get it working, along with schematics and some measurements.

I am in the process of selling our house and moving, so it will take some time to gather my notes and do the writeup. Finally I must add that I am not a professional, just a hobbyist enjoying restoring classic vintage hifi and good sound.


P1010301.JPG
P1010304.JPG
P1010311.JPG
 
Register to hide this ad
In short what has been done.
D 112, 113, 116, 117 removed - protection for the VFETs, but will do quite the opposite for the Laterals.
TR 120, 121 removed - auto bias for the VFETs, but will also here work the wrong way for the Laterals.
R 153 jumpered.
VR 102 100 Ohm, 470 Ohm resistor in parallell (in case VR 102 goes open).
R 151, 152 changed to 680 Ohm.
C 107 changed to 4,7 pF.
C 109 changed to 15 pF-6,8kOhm+0,5 nF-68 pF (TMC type compensation). Unsure of the values here have two notes, has to open the B2 to check for sure.

Quite some laboration with the compensation and the gate stoppers to make the amplifier stable without killing the speed of the Lateras.
225 mA bias for the Laterals, giving it aprox 100 mA for each die (being dual-die). Heatsinks at aprox 40 degrees C at 20 degrees room temperature, no prolem with termal runaway, the Laterals haing negative temperature coefficient.
 
Maybe you want to clamp the input for the driver transistors with a say 10 V zener to the power rails to protect the gates of the laterals. For Exicon devices it is only +/-14 V.
 
hazze said:
Leaved that part out, so the letters on the schematic would still be readable. No changes there from the original.
Click to expand...
You could also plonk the .asc file in a zip file and post it. The pic is hard to read and harder to replicate. If there are models for the Exicon devices (I used the Hitachi 2SJ162/K1058 ones for my Sony TA-4650 mod somewhere on this forum) add them too. I have no B-2 but if I come across one I would like to study your work first.
 
I bet this amp sounds swell. I read an article by a guru of amplifier design in a technical audio magazine here that, the sound signature of an amplifier comes primarily from the magma flow in its transformer(s). As such, this modified B2 sound character must be close or very similar to that of the B2 itself. Would be cool to compare.

Great work. :thumbsup:
 
Last edited:
Sparkplug said:
Would be cool to compare
Click to expand...
I did just such a comparison between two TA-4650 amps, one with original SITs and one with laterals (Hitachi). I used a relay box with blind switching on the outputs and a good CD-player as input, properly leveled to within less than one dB. But no matter what I played, classic, jazz, house, heavy metal, vocal, I could not hear any difference let alone detect the Hitachi laterals. Maybe my ears are just not good enough...
 
PE9ZZ said:
You could also plonk the .asc file in a zip file and post it. The pic is hard to read and harder to replicate. If there are models for the Exicon devices (I used the Hitachi 2SJ162/K1058 ones for my Sony TA-4650 mod somewhere on this forum) add them too. I have no B-2 but if I come across one I would like to study your work first.
Click to expand...
Attached you will find the .asc and a .txt file with all models in, put them in the same directory and it should work out of the box. For the input I used LSK170A, as models for the originals does not exist and I don't have the skills to make them.

The sim fairly well reflect how it works in reality, but voltages and currents differ quite some bit.
 

Attachments

Sparkplug said:
I bet this amp sounds swell. I read an article by a guru of amplifier design in a technical audio magazine here that, the sound signature of an amplifier comes primarily from the magma flow in its transformer(s). As such, this modified B2 sound character must be close or very similar to that of the B2 itself. Would be cool to compare.

Great work. :thumbsup:
Click to expand...
Thank you! The first time I plugged it in it was clear that it was different from all other amps I have had, rich full bass, smooth midrange and treble, the NS-1000 never sounded better.
 
PE9ZZ said:
vocal, I could not hear any difference let alone detect the Hitachi laterals.
Click to expand...
Sometimes it takes a few days or even weeks to discern sound characteristics, it does for me anyway. That’s why after making changes, I spend days listening to fully absorb what really sounds different.
 
Last edited:
I will post some measurments soon, but have a garden that needs attention today, which will take a while :)
 
Sparkplug said:
Sometimes it takes a few days or even weeks to discern sound characteristics, it does for me anyway. That’s why after making changes, I spend days listening to fully absorb what really sounds different.
Click to expand...
Makes sense. Only this was a straight A/B comparison, not a before/after evaluation. Those never work. Hearing has no memory.
 
PE9ZZ said:
Makes sense. Only this was a straight A/B comparison, not a before/after evaluation. Those never work. Hearing has no memory.
Click to expand...
See your point. Though you could consider it a before/after on the same model and not necessarily the same unit, assuming all the same models sound basically the same out of the factory. I really wonder now whether the transformers are a lot more important players than they’re usually given credit for … :biggrin:
 
You must log in or register to reply here.
Back
Top Bottom