Discussion in 'Yamaha' started by Bratwurst7s, Jan 17, 2018.
Your very welcome, yes they would be a great choice or Nichicon UPW-UPM.
Bridge Rectifier subs and links to where I found the recommendations.
1D2C1 = STTH1002CT or 2x SRP300K-E3/54. Neither are available.Still looking for something else.
S2VB20 = GBPC1206W
4D4B41 = GBPC604-E4/51
S5188 = Still looking. Haven't found a datasheet yet, only a couple of statements that it is 200v 11A. I did buy a GBPC2504-E4/51 but am not sure that it's actually suitable. Probably not.
S1WB = Mouser appears to have these but if I need to replace it I'll just make one out of 4x SBYV27-200 diodes like I did for my DCD-2700.
edit: A few trusted members name came up quite often as I searched for answers.
1S1555 = 1N4148
1S2076A = 1N4148
MV12 = 2x 1N4148 in series.
1S1885 = 1N4004, UF4004 or MUR160RLG. UF4004 not available, bought UF4005 and have 1N4004 on hand.
1SS82 = BAV-21. I have lots of BAV21 on hand.
10D4 = 1N5404.
30DF1 = UF5402-E3/73
WO3B = 1N4003, 1N4004.
WO6C = 1N4004.
You are a bit ahead of me Karl, I'm just getting started collecting datasheets. But the 2SA1146/2SC2706 are supposed to be O or Y class. The 2SA1186/2SC2837 that Reichelt and DigiKey are listing appear to be the base class and they are showing an hFE of only 50. How much of a performance hit would I be taking there compared to the mhz and cob differences?
I found a 2SC3263 (60 ft, slightly slower) that l thought may be ok for outputs, however it's complement 2SA1294 only has an ft of 35. Also the ones that are available are the standard class so hFE is only 50 as well.
I forgot to mention yesterday that I made a large Mouser order that's already on it's way to me. All the caps for the entire amp + extras, zener diodes, most of the other diodes, a large potfull of different transistors and some of the MOX resistors as well as the protection relays. Largest order that I've made with them so far, almost €300. But some of it is extras for my benchstock and almost all of the items I ordered 10+ or 100 for the quantity discounts. It's almost perverse that ordering 100 resistors or diodes is in some cases almost cheaper than ordering 10.
I was really planning on waiting until I had all of the subs nailed down before making an order but Mouser only had 4 of the relays in stock and I didn't want to have a situation where I had to wait 16+ weeks for a backorder. I will have to make another order in the next week for resistors and a few other things.
They didn't have any of the big NJW transistors so I'll have to check Digikey or wait. The other thing that I haven't done is check with BDent to see if they have any of the original outputs. That might be the best thing to do.
I'll continue with listing the various sub parts later today. I've now found almost all of the datasheets and will start printing them out and making comparisons.
Sounds like you have been very busy, yes it would be a pity to have to wait on the relays for 16 weeks preventing your build. Looks like you are going to be doing just about everything in this very lucky amplifier.
Are you going to be replacing the outputs just as a preventative measure or are you going to use the originals, l think you stated in an earlier post that the originals were ok?
I have a lot of time on my hands at the moment. So I am using the chance to dive into this.
I still have to sit and critically compare the installed outputs with what the SM shows using datasheets. They do appear to be usable so I think that I'll at least use them to do initial testing and make a final decision later.
My plan at the moment is to go step by step through the entire power supply and check everything. I won't replace anything except the caps (*!) but will check/test everything that I can and do a complete re-flow/resolder of the entire amp. In the background I'll continue to research the parts and substitutes for the entire amp and do a series of purchases to have parts on hand for anything that may be needed. All of the leftover/unneeded parts will become a welcome addition to my general benchstock.
For the first power up I'll leave the outputs out. I ordered a bunch of 470 ohm 3w resistors for that. Once I get the amp to power up and come out of protection I'll install the outputs and see what happens. That's hoping of course that I'm able to get the amp to power up and come out of protection.
After I get the amp working I can decide on any PM parts replacements/upgrades. I've made the mistake in the past of not getting the problems fixed before doing a general PM. I'm trying to not repeat that mistake but it's a little hard because I don't really know exactly what the problem(s) is(are).
(*!): Any parts that look/measure questionable during testing will be replaced.
edit: I will replace the protection relay and go through that section also before the first power up.
Btw, I hope that you stick around during this adventure. Having someone here that has had their hands on an M-70 is a big help. Plus I just like your and Karl & Zaibatsu's posts, as well as the other members that have stopped by with moral support.
James, if you are looking for components take a look here too : https://www.taydaelectronics.com/ It is cheap , the wares are genuine ( I bought a lot of stuffs) and ship via Dhl.de Ciao
Having mbz, Oilmaster and Avionic look over my shoulder and give critical comment would also be a very welcome event. I'm still a novice at this stuff.
Thanks BG! I always enjoy your help and comments. I hope that you stick around during the thread also. I'll check them out today.
I am your faithful follower and pupil Ciao
Sounds like a good plan, test all components, get it up and running then decide on any restoration/upgrade work.
I will definitely be hanging around, very kind of you to say about our posts, l have found your posts to be of great interest. To be honest l didn't have to get in too deep with mine, l had to make a new power supply board and replace most of those components, reflow any bad solder joints, replace that cap for the relay timing, set power supply voltages, bias and offset and diagnose a fault in the LH meter circuit which ended up being a bad trace. I didn't do any restoration work on mine but l did check all of the large filter caps and miraculously they were all within about 5% of original values. I believe It had not been used since the 80's when it was put aside after the power supply "fire", which wouldn't be ideal for the caps but there you go.
So I just removed the left output bank and heatsink. And having taken a closer look at the solder points before removing them I've come around to believing that they are not original. The joints are just too shiny to be 35yo.
Next I took a little more critical look at the outputs and drivers. The labelling on the drivers is crooked and looking at the close-up photo they look re-labelled. And look at the outputs. None of the labels are exactly square and they are on a black patch marking. Some of the letters don't line up with the others.
I took some acetone and a rag and the marking on one of the drivers wiped right off with one light swipe. The one output that I checked didn't immediately wipe right off but started to with a little work. The edges of the cleaned driver are not sharp and clean like a moulded case, they look like they have been somewhat rounded off from sanding.
I'm coming around to thinking that the drivers and outputs are not only not original but also possibly fakes. What do you guys think?
Acetone will remove text from even genuine devices, personally I wouldn't use it as a "general solvent", rather best left as a specific choice for DSBG and a few other things (it is however safe for the printing on Yamaha PCBs, even though it's been debated here before, I think everyone who has used it extensively agrees). Isopropyl is a better choice for general use, and for dissolving that old white paste, I've found household olive oil can help break it down gently before using your solvent of choice. Anyway, I don't think losing the text to acetone classifies the devices as fake.
I think it's hard to comment on whether the devices are genuine just by looking at them though, I've definitely seen less-than-perfect printing on genuine devices before. I suppose if you are suspicious of fakes it might be safest to replace/sub all the drivers and outputs.
The other point I wanted to add is - some of the high speed output device recommendations in this thread are LAPT transistors (parallel devices within single device), I'm not sure if that's appropriate if the factory-spec devices weren't also LAPT. Non-composite ("single"?) transistors these days are usually slower than what was available back in the day, it's just a sacrifice you have to make if substituting - particularly for drivers and outputs. I think the 0281G / 0302G you suggested will work fine, and better than the currently installed subs anyway.
It's probably more important to make sure you sub the drivers appropriately, since the A913 / C1913 devices are 120MHz
The Toshiba A1837 / C4793 might be a decent sub @ 70 / 100MHz (slightly higher CoB too, but not bad)
Bdent still had them last I checked (OMGCat and I got some recently), but the pair are NLA at mouser. I'd pick those over the MJE150xx driver sub options, if you can find them.
Probably just because you meant 1SS82
Good info, thanks very much. I have to say that the drivers tested good, perhaps I'm just being paranoid. Of course just because I'm paranoid doesn't mean that they aren't out to get me.
I'll look around at Bdent this week.
Yes! That's what I meant, 1SS82. I must have had one eye closed while typing.
I just checked and they still have them:
If you do replace all the outputs then it's probably good practice to replace the drivers with them.
Maybe it can wait until after you're sure the rest of the amp is working though (PSU, rail voltages, etc).
Way cool Zaibatso, thanks for looking those up. The price isn't bad either. And they have the 2SA1186/2SC2837 also.The photos show Y and O class, but that might not mean anything.
I may just buy some of all of those. Can't be a bad thing to have around even I don't end up using them for this project.
I haven't been able to find original Sanken datasheets for the 2SA1106/2SC2581 but did find some from New Jersey Semi and Savantic.
They don't show Cob or ranking but do show that they are 10A/100w, have a base hFE of 30 and 50, and are only 20mhz. I don't think that I'm going to use these. I'm thinking that I'll be placing an order with Bdent and Digikey in the next week.
Separate names with a comma.