Fixing a Yamaha M-65 For A Friend

acetate

Active Member
I have a buddy who loves Yamaha and he brought over a unit that was in protection. It was pretty obvious where at least one issue was because a transistor (TR144) and five resistors were completely burnt up - the traces separated from the foil side too. I read on here that when these units fail they tend to take down a whole bunch of components. So I started pulling transistors in the area. I found one more that was dead (TR140). I then pulled all the transistors on the rail and found three more TR126, TR132, and TR134. I have pulled all of those out of the unit but I am trying to figure out if there are more.

I still seem to be getting -65V at the relay and the unit still shows as in protection. So I assume I should keep pulling and testing?
 
Found five more transistors TR118, TR124, TR112, TR110, and TR108. Also a diode D114. And a bunch more toasty resistors.

I keep pulling them but the unit stays in protection. The relay now shows one pin at -45V, the rest are 0V - 0.1V.

At this point I have tested every transistor and every diode on the bad channel (the same one with -45V at the relay).

Any pointers are appreciated.
 
Have you checked IC102,TR104 and 106 ? Check your cement block .22 ohm emitter resistors. as well.
 
Last edited:
I will check those out tonight. How do I check the IC?

Its a dual JFET.(2SK389) Check it like two transistors. pin #4 is substrate and not used.

I use a PEAK DCA to check all the active parts in the M-series amplifiers. If your using a DMM You can easily check for shorts between the D-S-G pins for the two FETS.
 
Last edited:
Out of the last 3 M-65's I have looked at . I've had to find two replacement 2SK389's. If R103/104 are cooked ? The JFET is toast as well most likely.

R103/104 220Ω
1/4 watt film resistors
 
Last edited:
Thanks Avionic.

You were dead on about the JFET. It is toast. I pulled TR104 to test and it was bad after all (a very weird test though), and both emitters (.22 ohm) were dead. I had already pulled R104 as it was toasty.

At this point I have every transistor, outside of the protection circuit, pulled and just resting in socket if good or pulled out if it is bad. I am now getting -14V at the relay. It keeps dropping so I can see the progress. I assume now that I should re-focus on diodes?
 
On these units. If I was doing the repair. I would replace all the diodes(including zeners) and transistors on the dead channel. Due to the fact that they were all stressed during the failure event.
Diodes are all available from Mouser or Digikey.
Transistors are all available at B & D Enterprise.
http://www.littlediode.com/product.php?productid=390692
A new protect relay would be a good Idea as well as long as your in there.

A drop in replacement.
http://www.mouser.com/Search/Produc...virtualkey65300000virtualkey653-G4W2212PUSTV5
New bias trimmers...
http://www.mouser.com/Search/Produc...virtualkey65210000virtualkey652-3386G-1-102LF
 
Last edited:
I think I see why. This is maddening. I will just do that.

Is the best replacement for the 2SK389 two 2SK100s?
 
One more question: I am going to also recap it while I am in it. Options are limited for the main caps because of size. I was thinking of using the Nichicon LNR/LNT line. Any input there would also be appreciated. Thanks for all the help.
 
For the two 1000µf/80vdc I would opt for Nichicon KS or KG, these are snap in style mounting with the correct lead spacing I believe.
 
Good to know for the 1000uf. I hadn't gotten there yet. My main concern is the four 33,000uf. The LNR/LNT are about $25/ea and look to be decent but never used them.

Yikes $33/piece for the 2SK389BL!
 
Yours got all the old crusty glue corroding the copper buss bars, jumpers wires between the buss bars and everything else it contacts ?
 
Yes I got a nasty case of the glue. I will scrape it out as much as possible. I just did a Sansui AU-717 that had the same wonderful stuff.

I think I have everything figured out for parts but a few questions.

1) The 2SC2240s that didn't explode are marked as GRs. I just want to ensure that they all are GRs. I see the HFE is double between the GR and BL and just want to be sure. I verified against the good channel but just want to double check.
2) I see that I have 2SA1106 and 2SA1106B (same with 2SC2581/B) but I can't tell the spec difference and BDEnt only seems to stock the non-B version. Are they inter-changeable?
3) For the diodes I have as follows:
Orig Replacement
HZ24-2 TZX24B-TR (24V Zener Vf 1.5V .5W)
HZ6C1L 1N5233B-TAP (6V Zener Vf 1.1V .5W)
HZ16-3L NZX16C (16.7V Zener Vf 1.5V .5W)
HZ12B2 NZX13B (12.85V Zener 1.5V .5W)
RD3.OEB2 BZX55B3V3-TR (3.3V Zener 1.5V .5W)
1s1885 MUR160RLG (1A Rectifier)
30DF1 UF5402-E3/73 (3A Rectifier)
D5FB20 GBJ1502-F (Bridge 200V Vf 1.05V)

The LNR/LNT fit by diameter and lead spacing so I will use those.
 
For the outputs use these SANKEN's from BDEent
2SA1492 and 2SC3856
They are beefier than the 2SA1106 and the 2SC2581.
D5FB20 GBJ1502-F (Bridge 200V Vf 1.05V)
Unless its bad- replacement not required-
I would just stick to the diodes in the amplifier circuit that burnt up, not the entire M65.
 
The 2SC2240s that didn't explode are marked as GRs. I just want to ensure that they all are GRs. I see the HFE is double between the GR and BL and just want to be sure
Either ranking will due just fine in this app."GR's" are probably cheaper and easier to get.
 
Thanks Dave. I will leave off the bridge. I just hate having to place second orders mostly because of time and having to pay shipping again. Getting the order in today so I can rebuild next weekend.

For the traces that are pulled away should use some adhesive to hold them back in place? Or just try to stiffen them up with extra solder?

This photo is pre-part pulling. That whole charred part is now pulled away from the board but still intact.

attachment.php
 

Attachments

  • WP_20150302_17_57_21_Pro.jpg
    WP_20150302_17_57_21_Pro.jpg
    102.5 KB · Views: 152
Thanks Dave. I will leave off the bridge. I just hate having to place second orders mostly because of time and having to pay shipping again. Getting the order in today so I can rebuild next weekend.

For the traces that are pulled away should use some adhesive to hold them back in place? Or just try to stiffen them up with extra solder?

This photo is pre-part pulling. That whole charred part is now pulled away from the board but still intact.

Maybe both..:scratch2: Need to clean that area up and see whats under that "carboned up" area.Might have to save some of your capacitor lead clippings. Excellent source for jumper wires going point to point on the same trace.Can be used as trace stiffeners in a pinch.
 
Today I finally got some time to work on this. I changed out every diode and transistor on the main board. I pulled every component, observing silk screen for any errors. I still had a negative voltage at the relay. I decided to go ahead and replace all of the parts anyways. When I bring it up on variac at 20VAC I get -1.8V at the relay still. I pulled the main caps to get at the diodes under them and have not put them back in. But I still get the same reading.

I tried comparing good channel to bad channel but nearly every reading is off between the two. I don't get it. Hopefully someone can point me in the right direction.
 
Back
Top Bottom