Yamaha CA-2010 possible restoration

[nj pheonix]

Its 40 yrs old. If you can hear the difference (and like it) run class A. They were designed to be run that way.

 

[Oldsansui441]

Its 40 yrs old

Quite true, it has lasted this long .

 

[mbz]

OK, still early days, but what are listening impressions. In particular phono, compared to the CA-1000 if possible.

Following up on Nichi KG's. I did my "training" on a NAD3150. Did almost everything to it, still sounded ordinary.
Swapped out the KG's for KW's and a veil lifted. The amp still sounded below par, preamp is ordinary despite
opamp rolling (LM4562 suited best).

The KG's look great. The problem is what are the alternatives in that size?

 

[Oldsansui441]

OK, still early days, but what are listening impressions. In particular phono, compared to the CA-1000 if possible

Hi, l have had it running for a few hours today but only on Aux, l am in the middle of repairing the only turntable l have at the moment, so can't give you an impression of that yet sorry.
I have never heard or owned a CA-1000.

Following up on Nichi KG's. I did my "training" on a NAD3150. Did almost everything to it, still sounded ordinary.
Swapped out the KG's for KW's and a veil lifted

Are you saying you used KW's in filter cap positions?

 

[mbz]

Are you saying you used KW's in filter cap positions?

Yes, in the NAD3150, they have 2 for each rail, 4700/63V. Used KW 6800uf/63V. Need to keep an eye on the ripple,
they handle about the same maybe 10% lower than the KG's. Would not feel comfortable with KW 4700/63 in that scenario.

 

[Oldsansui441]

Yes, in the NAD3150, they have 2 for each rail, 4700/63V. Used KW 6800uf/63V

Ahh l see. Forgot to say the CA-2010 sounds really nice, definately Yamaha neutral.

 

[Bassblaster]

Here are the numbers

2SA673A
2SA777
2SA872
2SB537
2SB566ANS
2SC1213A
2SC1509
2SC1775
2SC1918
2SD381
2SD476ANS
2SD526

Thats very close to any pioneer receiver from the 70's...my SX-636 has almost all of those in it.

 

[Oldsansui441]

Thanks Bassblaster, I have worked on quiet a few 70's Pioneers and in those cases l replace the 2SA725/726's, 2SC1312/1313's and 2SC1451's pre-emtively. That is interesting your SX-636 uses most of these same transistors as the CA-2010. My SX-939 only uses the 2SB536 (should be same as 2SB537) and the 2SD381. I really like the x3x series receivers.

 

[mbz]

Picking your brains, what did you use for C403 (10uf/25V) and C406 (100/16), the low impedance types
on the psu.

 

[Oldsansui441]

All power supply/protection got UPW (apart from the 4 bipolars) which covers the low impedence requirements of those two positions as well.

 

[Oldsansui441]

Sorry it has taken me so long to update this thread, here are some photos of what l actually replaced. So in all l replaced:

-all electrolytic (apart from main filters as these tested fine) and tantalum capacitors
-the three heat sinked power supply transistors
-necessary fusible resistors
-both relays (soft start and protection)
-driver transistors and output transistors
-trim pots
-l went over every soldered joint and reflowed many
-upgraded the meter lighting to LED's as the original lighting was fairly poor. (this can easily be put back to original if required)

 

[Oldsansui441]

Here are some more

 

[Oldsansui441]

And some more

 

[zaibatsu]

Looks beautiful, nice work!
Very even spread on that lighting too, worth the fabrication efforts.

How does she sound?

 

[Oldsansui441]

Looks beautiful, nice work!
Very even spread on that lighting too, worth the fabrication efforts.

How does she sound?

Thank you, she sounds very nice, nice quiet background and very neutral sounding in that true Yamaha way. I do however like to run it with the bass notched up a little .
The lighting isn't quiet as bright as it looks in the first two pics l think in person it is just right, not too out there.

 

[mbz]

Looks beautiful, nice work!

Second that. Nice clean work.

 

[Oldsansui441]

Second that. Nice clean work

Thank you.

For those wondering, l replaced the tantalums on the function board with Nichicon UKL's. There is not enough room on the topside of the board to fit all of those e caps so l fitted two underneath board, there is plenty of clearance between board and lower chassis for this.

 

[zaibatsu]

My philosophy on my CA-800/-1000 is to use class A all the time, except on background "noise" duties. The ripple will be tough on those Nichi KG type 1's(?) and you may need to replace them a little earlier, but that's a small price. To be honest I don't really like the KG's, they were upstaged by KW's in smaller gear. I'm trying United Chemicon and boutique (Mundorf) in my next projects.

I've heard you say this a few times now so I'll bite - did you give them plenty of time to settle in?
Were the comparisons done with nothing else changed?

I can't really imagine how they could have much impact on sound when on PSU duty unless the ripple current rating is insufficient - especially in Class-A where current draw should be relatively constant. I've had good results with the KG-TypeIII on rail filtering / decoupling duties but not sure if the difference from the Type I you used is just in terminal material or if there are internal differences too (Nichicon's specsheet grammar is a bit flaky).

Anyway the reason I ask is because I've considered the 50mm diameter KGs as a "very distant future" PSU cap due to direct factory fit in a number of my amps, but your posts make me wonder. I've heard the European items are generally preferred for PSU duties (eg EPCOS, KEMET, etc - high ripple ratings) but are a bit tough to find a matching fit in my application. Will be interested to hear what you find after you try a few more, since I probably won't need to worry about choosing for a few more years.

 

[mbz]

I've heard you say this a few times now so I'll bite - did you give them plenty of time to settle in?
Were the comparisons done with nothing else changed?

The NAD 3150 was probably my first refurb, recapped with cerafines and KG and FW for preamp psu.
After replacing 4 cerafines in speech path with Silmic II's I noticed an improvement so replaced all cerafines with Silmic II's, no further significant improvement.
Rolled the opamp a few times NE5542->OPA,,, ,,,->LM4562, sounded better but not great. Changed KG for KW-> improvement. And a few other mods along
the way including Cree rectifier diodes, metal film resistors,,, All done over maybe 2 years. The KG's would have had 100+ hours on them before being swapped.
I understand the placebo effect, but over the 2 years I change so many things that made no difference, I was surprised that the KW's made a difference.

The KG's are a great looking cap, I was tempted to try the type III's but a bit costly.

I compared the datasheets for the KG's and KW's and noted that the KW's had a lower ESR/Tan delta

I tried United Chemicon (SMH) and Mundorf Mylitics in a Yamaha CA-1000. Sounded about the same but need to give them time.

I can't really imagine how they could have much impact on sound

I tend to think that the amp is powered by the main caps and that the AC and rectification just charges up the caps. So it's the
energy in the caps that used to amplify the audio signal from mV level to a "few" volts. OK this is just a basic understanding...

It's been said a million times,cap selection is very subjective. Use what you are happy with, but every now and then try something different.

 

[zaibatsu]

That's not really the same application as here though is it (largest PSU filter caps in a Class A amplifier)?
Or did you try the KGs along with the Mundorf and SMH in your CA-1000 (as the main PSU caps), with all else equal?
If so, fair enough. I need 22,000uF 80V in 50mm diameter so KW is out.
United Chemicon SME and Nichicon LMR are options - SME has higher ripple, but LMR has lower tan delta.

I tend to think that the amp is powered by the main caps and that the AC and rectification just charges up the caps. So it's the
energy in the caps that used to amplify the audio signal from mV level to a "few" volts. OK this is just a basic understanding...

I think Yamaha amps, particularly the higher end ones use many stages of local capacitance to feed transistors, and as a result, in amps like the one in this thread I would think that the local capacitance makes a far bigger difference. On lower end amps the main filter capacitance is much lower + without running in Class A there is a far more variable load on what is potentially insufficient capacitance - which is why a lot of those low-to-midrange amps have audible benefits with main filter cap upgrades. No doubt for the same reason they are more sensitive to their performance too.

Anyway, I'm open to correction - but I do feel it's important to test under identical conditions as the application does make a difference. For what it's worth, I've found the actual capacitance varies a lot when measured too (between brands and models), so it's nice to check and eliminate that as a point of variation.

Edit - apologies to @Oldsansui441 for mild thread hijack.