Yamaha M-70 Thread

[Bratwurst7s]

So judging by the disturbed solder connections someone hit a lot of points on the ends of resistors on the 2 psu control boards. R606, 607, 621, 623, 629, 632 as well as C601, 605 and 611. But what they missed is a ring crack on the emitter of TR602. It's completely loose, I can see it move with a wiggle test. If I'm reading the schematic correctly the emitters of TR 602 and 603 are feeding voltage to the collector of TR601 through R606 and 607. So if TR602 isn't connected I would think that would play hell on the output of TR601. Maybe I just found something useful here.



Cheers,
James

 

[Bratwurst7s]

In spite of my intention to not replace any parts that aren't broken after thinking about what kind of stress that TR 602 might have seen I decided that anything that I pulled to check on this board would be replaced with a new part. In the end every single part did in fact check good but now the board is rebuilt and I feel good about it. That bad connection on TR602 feels like a smoking gun, it gives me hope that I might have found the original problem.

A couple of notes.
In spite of doing an exhaustive search and list of resistors to buy somehow I got stupid and didn't buy any 39k, 1.5k, 56k and 56 ohm resistors. Which are of course all on this board. It looks like I did get all of the others. So those ones were only removed, tested and re-installed. All were within 1% of nominal.
R226 and 627 are both 180k ohm and in series. I couldn't find any Vishay CMF or CCF in that exact value. But I did find 178k and 182k. And sorting through them I was able to make a pair that was very nearly perfect nominal for the combined value, at least a little better than the carbon film that they replaced.
With all of the resistors that I did replace I sorted through what I have and selected the ones that were closest to nominal, all inside of 1%.

TR602 tested at 266 hFe, TR603 at 322. I have a lot of KSA992FTA's and after sorting through them I was able to find 2 with an hFe of 333 and 334. TR604 tested at 182, I found a KSC1815Y with 178. TR605 measured 334, was replaced with a KSC1845 with 390 hFe. TR601 measured 115 and was replaced with a KSA1220YTA that has 185 hFe.

C601 measured 5.2µF/2.8ohms, was replaced with a Wima MKS2 4.7µF/50v, measured 4.86µF/0.07ohms.
C602 measured 95.2µF/0.59ohms, has replaced with a Panasonic FM 100µF/25v (measured 101.9µF/0.12ohms).
C603 (ceramic 0.001) measured 0.000942µF and was re-installed.
C607 measured 0.500µF/1.1ohms, the replacement Panasonic ECQ-U3A474MG X1 300v measured 0.432µF/0.66ohms.
C608, 609 and 610 measured 8940pF, 9395pF and 8937pF. I bought 10 of the replacement ECQ-U2A682KL 275v and sorted through them to find 6670pF, 6664pF and 6686pF.

I figured that the zeners D601 and 602 might be passing some heat so I mounted them a bit above the board.

VR602 measured 2.632k ohms, I pre-adjusted the 3339H-1-502LF trim pot to as close to that as I could before installing it.

A lot of the solder connections looked cold, and the ones that had been re-soldered looked really dirty so every one of them was de-soldered, cleaned and re-soldered. This being the ones where a component wasn't actually replaced.

I know that I'm going to need a 2nd multimeter to do the calibration of the PSU so I've been thinking that I might splurge and buy a Fluke 87 or 175, based on what I've been reading in some recent threads.



Cheers,
James

 

[Bratwurst7s]

 

[OMGCat!]

Very nice work so far! I bet you are right about the missed solder joint being a major culprit.
Which Bourns pots are those? I'm eventually going to go through my B2x which has pots in the same orientation and I've been wondering what I was going to use when the time came.

 

[HypnoToad]

I had an M70 and it would trip the AFCI circuit breakers, but was fine on all the others. Someone said it was due the the X power supply, never solved it, sold it to someone with an older house, they never had a problem. It had that suspect capacitor replaced with the correct X2 type and everything was aligned and running to spec. Great sounding amp with tons of power.

 

[Bratwurst7s]

Very nice work so far! I bet you are right about the missed solder joint being a major culprit.
Which Bourns pots are those? I'm eventually going to go through my B2x which has pots in the same orientation and I've been wondering what I was going to use when the time came.

Thanks OMG! I'm moving slow but am finally able to spend a few hours a day at the workbench.

The side adjust pot is a 5k 4 turn Bourns, 3339W-1-502LF,
https://www.mouser.de/ProductDetail...=sGAEpiMZZMvygUB3GLcD7rufck0TIBKywTr7dQ1ppt0=

For the other 5k top adjust I used 3339H-1-502LF,
https://www.mouser.de/ProductDetail...=sGAEpiMZZMvygUB3GLcD7rufck0TIBKyhmwWeQKT7IU=

For other values of the same type it looks like the "502" part of the part number is the key. For 1k use "102", for 500ohms use "501", 100 ohms "101" etc and plug that into the string and do a Mouser search. So the 1st & 2nd numbers are the base ohms and the 3rd number adds a zero position(s).

The 4 turn should be a bit less sensitive to adjust than the originals. I would have liked to go with a 20 turn but would have had to do a fair amount of lead bending to make them fit and settled for the 4 turn.

I got started on the Control-2 board yesterday, we'll see if I get it finished today.

Cheers,
James

 

[Bratwurst7s]

For 25 turn trimmers there are these:

Side adj:
https://www.mouser.de/ProductDetail...=sGAEpiMZZMvygUB3GLcD7mDahPR1oQ7G/Ye13IhhaC8=
Top adj:
https://www.mouser.de/ProductDetail...EpiMZZMvygUB3GLcD7oqoMetRlnqQNo%2b%2bGRO5/to=

Both of these types (3339 and 3299) only have 2.5mm offset spacing on the single leg and the position needs 5mm (side adj) and 10mm (top adj). We know that both can be made to work with some careful lead bending but the 3339 leads seem to be a little longer (about 8mm vs 5mm) which makes them a bit easier to fit.

The drawing for the 3339W shows "min. 4.75mm" lead length and the 3299P shows "6.22 +/- 1.44" but the 3339's that I have here measure closer to 8mm and the 3299's closer to 5mm. Go figure.

Cheers,
James

 

[amr2]

 

[HypnoToad]

I wish I had the confidence to strip something down totally and rebuild it, my hat is off to you.

 

[Bratwurst7s]

I wish I had the confidence to strip something down totally and rebuild it, my hat is off to you.

Thanks very much, but in reality the complete vote's not in yet. As to weather I'm being confident, bold, or just plain dumb.

It's a bit of a dellima at the moment.
1) I don't really know just exactly what the original problem is, and I can only hope that I've found all of the secondary damage caused by the previous person(s?).
2) I do know that just turning it on to explore can be potentially bad.
3) And using a DBT to explore with apparently is bad because of the complex triac PSU.
4) I haven't decided on the €150+ cost of a variac yet.
5) Every indication so far points to previous exploration concerning a power supply and protection problem.
6) Then there is the ugly DSBG problem, and leaky caps, and bad/old solder connections worry.

So I am very cautious about turning it on before at least going completely through the power supply, which I am only very dimly beginning to understand.
On the other hand there is the painfully learned principle of not doing wholescale restoration etc before first finding and correcting the original problem. See points #1 and 5.
On the other other hand every time that I remove one of these boards or re-solder a connection I risk unintended damage, so I'd like very much to only do each board once.
Plus, once one has de-soldered a semi-conductor and has new replacments on hand is it better to replace it or re-install it. Is it a potential sacrificial lamb?
On the other hand did I do my due diligence correctly and are all of the replacement subs correct?

So, at the moment my confidence is based largely on hope and beer.

Still, I've mostly gone through Control-2 now. Yesterday I spent a good bit of time with google learning how to test a triac with a multimeter. From what I've learned SCR602 & 603 check good. My 328 tester also shows SCR601 thermister as good. I replaced them all anyway once they were removed. I removed enough parts to isolate the IC601 IG04080 control chip and test it with my little 328 tester. And while it obviously can't completely show the chip as functioning correctly at least it showed it as not kaput. Right now I'm just waiting on a new Mouser order to arrive (hopefully tomorrow) with more resistors and film caps. Every resistor, diode and cap checked good (with the exception that I don't know how to function test a zener as anything but a normal diode) but are being replaced with a metal film resistor and poly film cap and hopefully quality diode, triac & thermister.

The photo shows the results of checking the IG04080.

Time will tell as to if I'm being confident, bold, or just dumb.



Cheers,
James

edit: One note. A lot of the old solder connections looked cold and many of them spat out ugly old flux(?) when I de-soldered them. Everything is getting de-soldered, cleaned with acetone and re-soldered.

 

[Bratwurst7s]

All of the Control-2 resistors are now 1% metal film, sorted and tested to get as close to nominal as possible, most are within 0.5%.
VR601 tested at 2.57k ohms and I pre-set the Bourn to as close as that as I could.

C604 was replaced with a Wima FKP2, 2.5% 100v.
C605 was replaced with a Wima MKS2 5% 63v.
C606 was replaced with a Wima MKP4 5% 250v.
C611 was replaced with an Epcos/TDK B3268 series poly 1250v.

The S1WB20 bridge rectifier was replaced with 4x Vishay SBYV27-200 ultra fast-soft recovery diodes.

The zener diodes were placed a little high to dissipate heat.
The 2x 10D4 diodes were checked and reinstalled, mainly because I missed the fact that the 1N5404 diodes that I bought have a 1.2mm lead and I would have had to drill the board. But the solder pads are so small that there wouldn't have been much left to solder on.

The only component that wasn't checked was the opto-coupler. Because I don't know how and didn't want to risk removing it. It did get carefully re-flowed, as did the IG04080 chip.

Everything went back together nicely and I removed the old bits of cut off mains wires, so there is room on the posts now to do a wrap when I solder the mains back on. But I'm leaving the cord off right now because it will make removing the back panel easier.



Cheers,
James

edit: I forgot to mention, the big blue Epcos/TDK cap lead spacing is a bit less than what it replaced. But Yamaha was nice enough to provide alternate mounting holes in the board so in the end it was a drop-in fit using 1 of the alternates.

edit 2: WARINING! The subs that I chose for SCR601 & SCR602 work but their pin-out is different than the original parts. Legs 2 & 3 must be position swapped and the parts installed 180° from the originals. See post #157!!!

 

[Bratwurst7s]

 

[Bratwurst7s]

 

[Bratwurst7s]

 

[Bratwurst7s]

I used a sil-pad under SCR603.

The control assy is now re-installed and I've started accessing the 12v stabilized PSU and capacitor boards up front.



Cheers,
James

 

[Bratwurst7s]

Removing the Electrolytic Cap board was actually more work than reworking it. But it's out without causing any unintended problems. That I know of.

The 2x 10000µF caps measured 8620µF and 8589µF, the 18000µF measured a surprising 19830µF and 19600µF.
The replacement KMH measured 10930µF, 11010µF, 20390µF and 20460µF. All disappointingly low but still at least above the OEM nominal and with that 105°c rating.



Cheers,
James

 

[Bratwurst7s]

To install the 12000µF all I had to do was snip off the 3rd dummy leg (the one near the minus pole) and bend the other pins slightly in towards center to get a 22mm distance and then it drops in. Very easy. For the 22000µF all one needs to do is snip off the tops of the dummy pins, leaving the wide portion of the leg as supports and then the cap fits snugly in the outer edges of the slots. Again, very easy.

My Elektromeister friend assured me that a single tinned 0.8mm wire would be enough to bridge the cut section of traces. Naturally I used 2 pieces per trace.



Cheers,
James

edit: Added note for the interest of others in the future. The cap board has some main cap mounting flexibility built in. The 10000µF can be mounted with 24mm, 20mm and 15mm lead spacing. The slots for the 18000µf caps allow roughly from 27mm to about 17mm (perhaps 15mm, I forgot to measure the slot lengths before soldering over them).

 

[Bratwurst7s]

The 22k and 12k 1w MOX resistors checked a bit high but good. (22.8k). But they look more like a 2w when compared to the new ones that I bought, which also checked a bit high. So I cleaned and re-used the old ones. The 1.5k and 820 ohm 1/4w resistors were replaved with Vishay RN metal film. The board is heat discolored under the 1/4's so I mounted the mounted them a bit high.

Concerning the bridge rectifiers. I didn't do my homework well enough and the 15A bridges that I bought have far too small of a lead spacing to fit. The old one checked good so I re-installed it. The 4D4B41 is getting replaced with a GBP604-E4/51 400v 6A bridge, although it isn't installed yet.



Cheers,
James

 

[Bratwurst7s]

Oh the Humanity! The horror of DSBG!

C702, the one that I thought looked like it was leaking? The minus lead is completely eaten through, it's just hanging in place with the plus lead.
R704, 705 and 706, D703 and 704. All badly corroded. So, some items to replace on this board.



Cheers,
James

 

[Bratwurst7s]