Yamaha YST-SW030 full recap

[goodpsusearch]

I bought this subwoofer for 60€ as used but in perfectly working condition.

I opened it up to check the insides and found several capacitors that had already started going bad: ESR was high and capacitance has dropped bellow 20% tolerance. I am mainly talking about the ones being near the 2 transistors in the photos.

The 2 complementary NPN and PNP transistors are slightly discolored from heat and the area on the PCB indicates that they get quite warm. But they tested good and the sub is working, so I didn't replace them. I tried to bend them to make sure they are not cooking the new caps.

The transistors are D2531 and B1642. The sub woofer is based on the STK404-120 hybrid power amplifier and C4570HA operation amplifier. It relies on the TA7317PG for its protection and the bridge rectifier is RBV-402.

The old caps were Jamicon SK 85C. So this is how these Yamaha subs are so cheap for their specs and sound quality. They save on the caps by not using Japanese capacitors and even worse using 85C caps even in places where they are expected to get a lot of heat and not enough airflow.

I decided to do a full recap. I mainly used Rubycon YXF, Nichicon HE and Panasonic FR. Unfortunately I was out of stock for 47uF value and had to use 85C caps there. But that area of the PCB is not near parts generating heat.

I also plan to replace the 2 6800uF primary caps with 2x 10.000uF ones hoping this will enhance the dynamic response of the sub woofer.

Some pics attached before the recap.

[goodpsusearch]

And here are some "after" pics.

When recapping, I stayed with the original values because of my lack of knowledge on the amplifiers, I didn't want to risk breaking something by going up in uF.

[goodpsusearch]

Originally Posted by goodpsusearch

I also plan to replace the 2 6800uF primary caps with 2x 10.000uF ones hoping this will enhance the dynamic response of the sub woofer.

I revisited this project and replaced the caps with 2x 10.000uF 50V Nichicon GU 105C

[momaka]

Originally Posted by goodpsusearch

The 2 complementary NPN and PNP transistors are slightly discolored from heat and the area on the PCB indicates that they get quite warm. But they tested good and the sub is working, so I didn't replace them. I tried to bend them to make sure they are not cooking the new caps.

Most likely these regulate the +/-12V (or +/-15V... or whatever typical supplies most op-amps use) for the pre-amp section.
I almost always see these running hot in newer audio stuff, simply because they drop the main +/- rails down to 12-15V, meaning they often drop around 20V or more. So even with just 50-100 mA of power draw, the P_d will easily jump to 1 Watt or more... hence the heat output.

Since these are used for linear regulation, you could technically attach them to a piece of aluminum (they have conveniently isolated tabs too ) and relocate the whole assembly a little further - preferably held down with screws onto something, so the vibrations don't knock things out.

Originally Posted by goodpsusearch

The sub woofer is based on the STK404-120 hybrid power amplifier

Ah, STK modules. Not a big fan of them (in terms of servicing / finding replacements), but they do simplify the PCB layout and design.

Originally Posted by goodpsusearch

The old caps were Jamicon SK 85C. So this is how these Yamaha subs are so cheap for their specs and sound quality.

I think it's just a 90's (and onwards) thing.
Most audio gear from the early-mid 80's usually only came with Japanese caps. Then somewhere between then and the early 90's, they started switching to cheaper brands to lower costs... and here we are today.

Originally Posted by goodpsusearch

They save on the caps by not using Japanese capacitors and even worse using 85C caps even in places where they are expected to get a lot of heat and not enough airflow.

Well, the good thing with most audio equipment (at least stuff that doesn't use an SMPS) is that none of the caps have to deal with high frequencies. So self-heating from high-frequency ripple current is virtually non-existent... and therefore even 85C caps will still have very good life. The problem with heat is just poor design (dropping large voltages in a linear fashion instead of having a separate tap on the transformer for lower voltages for the op-amps), poor PCB layout (caps placed next to hot stuff, as you noticed), and not heatsinking components that should be heatsunk.

That's not to say this Yamaha subwoofer is bad in terms of design and build quality. Compared to other subwoofers, it's actually good. But compared to old, properly-designed amps (like your Pioneer A66X or other 70's and 80's amps), it's not as good. And compared to modern wireless blutooth "soundbars"... no ... those are just modern throwaway junk.

I like that you got the Cherry veneer style, BTW.

[goodpsusearch]

Perfect advise, thanks momaka. I was thinking to add some heatsink to them but then I was worried that this would dissipate even more heat to the nearby caps, so I didn't do it.

But your advise is great, this would definitely keep them cool and also eliminate heat sources from this sub, as it barely gets warm during use and the STK is attached to the classic subwoofer amp external heatsink that conducts heat outside the sub.

Now that I Know what these transistors do it might also help to replace their output caps with something bigger in capacitance.


Will keep in mind next time that I need to revisit the internals of the amp, Hope it's not soon though.

[momaka]

Originally Posted by goodpsusearch

Perfect advise, thanks momaka. I was thinking to add some heatsink to them but then I was worried that this would dissipate even more heat to the nearby caps, so I didn't do it.

Well no, the heat dissipation will be exactly the same as before. If the regulator is outputting 1 Watt of heat, it won't matter if you put a heatsink on it or not - it will still output 1 Watt of heat either way. The major difference is how nearby components will get affected.

Without the heatsink, the regulator's surface area for heat dissipation is much smaller. Therefore it will run hotter, and any nearby components (especially those connected with traces/tracks directly to the regulator) will heat up more or nearly to the same temperature as the regulator (not good for nearby electrolytic caps.) But further away components may not get heated as much.

In contrast, if you add a heatsink, the heat will be more spread out across a bigger area, so more components will get heated up, but to a lower and more consistent temperature. In other words, the components near the regulator won't get heated to such high temperatures as the regulator, but further components may get heated up a little more.

All in all, though, the temperature inside the enclosed box will still be the same, regardless if you add the heatsinks or not. So again, the only difference the heatsinks will make is how hot components will get near the regulators... and also that the regulator will run cooler overall, which may extend its useful life.