Yamaha A-1000 Repair and Info Thread

[zaibatsu]

There should be no performance difference in the transistor, bias comp transistors have low demands (and I matched all numbers closely) - but the problems with the stock transistor are its flimsy mounting as you found out, and also the fact that it's mounted via an adapter with very low thermal mass. The latter is no big deal stock, but I usually have a slow fan pointed at these amps in Class-A mode and the temp tracing is more accurate if the transistor is directly on the heatsink - otherwise any airflow cools the little adapter which causes the bias to rise a bit more than ideal.

I experimented with location on the A-700 already so just picked the A-1000 mounting in one go, it works well (driver side of heatsink gets quite hot on the 1000). I've stability tested for 6 hours now, all is running smooth.

For fun, some photos I took with the FLIR thermal cam of one of the 700s warming up while playing with heatsinks:


Peak output transistor temp was 90.1°C.

 

[tmsears]

Update: lest anyone thinks I am slacking, I am in the process of getting a big order ready from mouser, of which, my needed transistors for this project are a part of. I got a little extra money from an odd job to help fund this order and I have just barely enough to get the transistors and resistors for the A-1000 and caps to restore my M-4 and C-6 with. This saves me a lot in shipping costs, as I am on a shoe string budget.

I have gotten over trying to kill my A-1000 via carelessness and am looking forward to finishing it up so that I can here it sing!

 

[zaibatsu]

It's probably worth grabbing a pair of these for 50c too:
http://mouser.com/ProductDetail/Nexperia/NZX9V1E133

There's a mistake on all A-1000s in the +/-22V supply on phono board (feeds preamp/flatamp board also):
D605, D606 should both be HZ-9C3 but they aren't (NZX9V1E133 is sub, one is wrong from factory)
D601-D604 should all be HZ-6C1L (NZX6V2B133 is sub, I changed them anyway but they were correct)

The amp still works without fixing it, but may as well keep handy.

 

[tmsears]

It's probably worth grabbing a pair of these for 50c too:
http://mouser.com/ProductDetail/Nexperia/NZX9V1E133

There's a mistake on all A-1000s in the +/-22V supply on phono board (feeds preamp/flatamp board also):
D605, D606 should both be HZ-9C3 but they aren't (NZX9V1E133 is sub, one is wrong from factory)
D601-D604 should all be HZ-6CIL (NZX6V2B133 is sub, I changed them anyway but they were correct)

The amp still works without fixing it, but may as well keep handy.

I put them in my order list.
Forgive my ignorance, but where did you come across this tidbit of info? Is is a typo in the SM or something?

 

[zaibatsu]

Just a build error, it should be a symmetrical arrangement to create the +/-22V rails but one is wrong (see D605,D606).
Both the parts list and schematic agree and appear to be correct.

Not critical but may as well have them handy.

 

[tmsears]

Just a build error, it should be a symmetrical arrangement to create the +/-22V rails but one is wrong (see D605,D606).
Both the parts list and schematic agree and appear to be correct.

Not critical but may as well have them handy.

Ah, so they put the wrong parts in at the factory, and you caught it during a repair?

 

[zaibatsu]

Yeah, I usually swap those six out because they are under high heat whenever the amp is on (being under the hot cement resistors, same as small resistors on BBQ board), although probably not necessary.

Found mistake while pulling factory ones, but then after searching I saw others had found the same. You'll see one says C-3 and one says C-1 when they should both be the C-3 (9V) ones.
Wouldn't bother messing with that (until amp working), just thought it best to mention while you were ordering.

 

[tmsears]

UPDATE:

After several financial delays and nagging illness, I was finally able to place my big order to Mouser yesterday. I got all the parts needed (that are known) to repair the A-1000, including those diodes @zaibatsu mentioned above. The order was big and took a while since I decided to get all of the above and the caps to recap my M4 and C6 with. Hopefully that will increase their value and my experience a little..

But first, the repair of the A-1000!

I am looking forward to seeing it live once again after having murdered it.

 

[tmsears]

Well, the parts are in and so far I have replaced the pre drivers with the new ones.

The legs are bigger as noted by @zaibatsu, so I decided to tackle the problem by enlarging the through holes to 1mm from .08mm using a hobby drill from eBay:




As you can see, there is plenty of pad left to solder to.

I decided to go ahead and replace the pre drivers on both sides since the transistors were cheap and I had the drill out and because that's how I roll. I plan on replacing the transistors on both channels for the rest as well but only after verifying the repairs to the damaged channel first, so that I don't muck up the troubleshooting process:




Next up are the resistors, after which I will need help with putting the "dummy transistor" resistors in place and the testing of said setup.

 

[zaibatsu]

Nice work, that looks really neat!

Be careful with the orientation of the bias transistors when you put those in. For the resistors, just put one resistor across the B and E pads of where each output would go (so four in total). You can check for a relay click first on the DBT and then off the DBT as well if you like, before swapping the outputs in for the resistors and doing it again. To be honest this whole step is redundant with the DBT, but maybe good to be doubly safe.

I would personally relegate your replacement outputs to APS duty and keep the genuine ones as actual outputs (the "inner" two main transistors), though it's up to you.

 

[tmsears]

Yeah, I double-triple checked their orientation to the point of paranoia, as I was worried about the same thing.
When I put in the resistors for testing, can I leave the heatsink removed, or will the driver transistors overheat?

Also, you said four, are you including the outputs for the APS as well? I think you are, I just want to be sure.

 

[tmsears]

For fun, some photos I took with the FLIR thermal cam of one of the 700s warming up while playing with heatsinks:
View attachment 972874 View attachment 972875 View attachment 972877 View attachment 972878 View attachment 972879

Peak output transistor temp was 90.1°C.

Did you...did you get heatsinks for your heatsinks?

 

[zaibatsu]

Yeah, I double-triple checked their orientation to the point of paranoia, as I was worried about the same thing.
When I put in the resistors for testing, can I leave the heatsink removed, or will the driver transistors overheat?
Also, you said four, are you including the outputs for the APS as well? I think you are, I just want to be sure.

No definitely not for APS, I meant the four outputs. Don't put resistors on the APS transistors.

If you are confident in the other channel you can leave that side intact and only resistor the good channel (only two resistors used).

 

[tmsears]

No definitely not for APS, I meant the four outputs. Don't put resistors on the APS transistors.

If you are confident in the other channel you can leave that side intact and only resistor the good channel (only two resistors used).

Good to know, and glad I waited to make sure!

I will probably go ahead and use resistors on all four outputs..I got to thinking about it, and although I am confident the other channel wasn't affected, with my luck these days, better to be safe than sorry.
Thanks for the heads up!

Is there any voltages that I should be checking for or looking at when I fire the unit up using the DBT? What about after?

 

[zaibatsu]

Is there any voltages that I should be checking for or looking at when I fire the unit up using the DBT? What about after?

After you get a relay click with the resistors, you're good to swap the outputs in.

Then with all transistors back in (and a click), keeping it on the DBT (100W), check that you get an idle reading in AB. It might be lower than spec but you'll get some reading (probably 5-8mV), just make sure that matches on both channels and twiddle the pots if needed. Then you can take it off the DBT.

We're being overly careful here, basically if you've got a click you're good to roll.

 

[tmsears]

After you get a relay click with the resistors, you're good to swap the outputs in.

Then with all transistors back in (and a click), keeping it on the DBT (100W), check that you get an idle reading in AB. It might be lower than spec but you'll get some reading (probably 5-8mV), just make sure that matches on both channels and twiddle the pots if needed. Then you can take it off the DBT.

We're being overly careful here, basically if you've got a click you're good to roll.

Will do, nothing wrong with being too careful, especially when you are on a budget like me .

 

[avionic]

twiddle

We will have no twiddling here.

 

[avionic]

This is a respectful establishment......

 

[zaibatsu]

Don't listen to avionic he is new here, give them a little twiddle if needed.

 

[tmsears]

@avionic @zaibatsu .. Trying to do legitimate...serious...work here you guys are killing me .