• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

STK Not Good?

Theophilus Punoval

The Queen says it's OK...
A technician told me that a Yamaha R300 receiver with an STK chip is, quote, "by definition not high end" but it "sounded great". I don't particularly care if it's high end because it's going to a college kid and, most likely, will blow out after getting beer (or worse) poured into it sooner rather than later. But I don't want to give away junk either.

My question is why, if an STK chip is in a unit that sounds really good, it isn't high end. Isn't sounding good the whole point of high end? I'm not saying the R 300 IS high end or arguing his point.

He hinted the STK chip was a shortcut of some sort. Yamaha used to have a good reputation and I'm surprised to think that they would sell poorly built products. I have a Nikko 320 with an (a?) STK chip as well for the kid's sister and I wouldn't want to give them second rate equipment, just things that will get them through school without being embarrassed.

BTW, I'm sending re-foamed Mini Advents with the crossover caps replaced to go with these sleds. Are there any opinions about if they fit well with either or both receivers? I have Infinity RS 1000s and small ARs to send instead.
 
Last edited:
Register to hide this ad
You're going to get a lot of replies on this one. :blah: :biggrin:

I'll just say that in general, I don't pretty much care for the STK and other amplifier modules as they contain the pre driver, driver, and output transistors plus tone controls (in some) in a single package. Kinda sucks when only one transistor in the pak shorts and you gotta replace the whole damn thing instead of a single discreet part.

High end? I would say yes & no as there were 1000's created for all different kinds of uses especially some which were used as video amplifiers in projection televisions, and from black plastic crap to bigger amps and receivers.

However, I will say that I really like my Fisher CA-880 which uses one STK-0080 per channel and is rated at 80 watts per channel. It really has the power to drive most any speakers you throw at it.
 
Oh ****... Not this again....

Cademan is right tho... Some are awesome... Some are turds..... You'll get the buggers on here gripe about them (and they think transformer or capactior coupling is good)

One universal truth tho, it that they're one day gonna be unobtanium.

The DPP series are good as they're not overly complex, and the set designer has more influence.

I have personally seen alot of Hybrid IC amps outgun discreet amps. And see it the other way aswell... But you'd be hard pressed to rival the Big discreets (Sanken MT200 devices) in the top liners (Onkyo M510, Rotel RHB 10, Technics SE-A3MK2, Yamaha MX 10000)

I shall leave a wheel of cheese here to compliment the whine that will invariably chime in
 
I thought DiscReet was a Frank Zappa record label, and also meant circumspect. "Discrete" means separate. Forgive me, but my mother was a teacher and always corrected my usage and I picked up her habit. This has led to...well, let's say, consequences. Not now, I hope.
 
High end has nothing to do with performance, and everything to do with price. There is no correlation. Cheap amps regularly outperform expensive junk. As Douglas Self, expert amp designer claims: it shouldn't be called high end - it should be called expensive end.
 
Discrete component snobs will poo-poo modules. However, it is possible to create and use a seriously good module. The op-amp is a good example. The real problem with modules is obsolescence. You can't repair what you can't get parts for.

You can do things with a module that you could not possibly do with discretes, such as precise trimming and temperature matching/compensation, etc. For example, I would like to see the discrete component snobs have a crack at an Athlon processor in discretes. That would be something to see.

Enjoy,
Rich P
 
If a STK voltage amp is in a BPC Sansui B1000, which can be had for around 60 euro's over here, including tuner and preamp, this STK voltage amp is a sign elcheapo design.

If this very same STK is in an Accuphase E203, it suddenly does not matter anymore......:biggrin:
 
Simply put, some IC output blocks are unobtanium. Serviceable and that you can substitute around beats the alternatives. Some of the modules sound good, but I will still prefer discrete components as they are more serviceable.
 
The STK modules are what they are. The performance is built in- some are pretty good, some not so good, and there's little you can do about it. My main complaint is that you usually can't get originals for repair. The copies or clones aren't as reliable, and may not perform the same way in the circuit. I've gotten some that tend to oscillate on signal peaks, where the originals were perfectly stable. No doubt millions of them have worked well, and continue to work, but I'd never pay much for an STK based amp or receiver.
 
I'll agree with what everyone has said and add, specific to the R300, that it is a pretty nice receiver.
I picked one up for $15 a while ago and after cleaning it up, let a friend borrow it. The only thing I did to it besides blowing the dust out was to renew the heat transfer goop on the output STK and add a stick on heat sink to the voltage amp STK. It gets plenty hot and seemed like a little cooling help would go toward keeping it alive longer.
Make sure to keep the selector knob tight as the seem to loosen and get lost on quite a few of them.
 
I got my exact STK module new for $20, which was more than I paid for the Nikko which it went into (or "into which it went" for grammar snobs). The R 300 seems nice. It was a $1 pickup plus $25 for pots and knobs because they had been broken off.
 
STK's ST-inK..

Joke....

But I usually try to get by amps that carry them.

Still, my LXI AM2600 carries lots of punch.
 
The Ford JBL amps used in a lot of 80s and early 90s Ford/Lincoln/Mercury products rocked a pair of STK chips. They don't age overly well, and may are dying now. Original was a 2240, but a 2250 drops in. They work and and sound nice when operational, but its the whole dealing with obsolete nonsense that makes them suck.
 
The issue for me is they, if not unubtainable now, will be at some point in the future. So when I look at putting $ into rebuilding a vintage piece this is a huge consideration .....
.......I dont want to put good money into something that will become unfixable.
 
".....I dont want to put good money into something that will become unfixable."

Doesn't most "vintage" equipment have unobtainable parts?
 
a lot of new stuff does....the firmware inside certain parts is unobtainable for the common man, many times the part itself too, and it may be produced during a few years (or months) only.
 
Not long ago I dealt with an unobtainium STK. It can be done. This was a lower powered one so the ubiquitous LM3886 was just fine. I thought of just putting in pre out jacks but nope, the tone controls were in the global feedback loop.

Bottom line is that STKs are not a performance issue at all. Look at the internal diagram of them and see that this is how you would design it anyway. Seriously. And you have the bias regulator really really thermally coupled. Someone really should make modern replacements but alas, there is simply not that much demand.

There are three major types. One is the Darlington power pack which only has the outputs, drivers and bias regulator. The transistors in those are just as good as anything out there, they do not have more nonlinearity which would cause distortion, and other important amp specs are determined largely by the design, not the STK. The 0029 up to the 0100 series are like this. The performance is up to the designer for the most part, as stated - a shortcut.

Then there are the really integrated ones that are the whole power amp section on a chip. Engineers don't just save money using them, they save time and PC board space. And holes. Back in 1990s if you called a circuit board manufacturer for a price, before he even wanted to know anything like the size or whatever, he asked "How many holes ?". I know why. And that same reason is why SMDs have profilerated so well. That stuff is hard to drill and mill. Actually for holes they punch it. You'll notice in some things they just punch a row of holes and break it, it is actually cheaper and faster to do that than to cut or drill it.

That is one of the major reasons that ICs, including the STKs became so popular.

And then came the SVI series. Mainly Panasonic/Technics, These were a cut above, some new class A. Expensive as hell too, I got a Technics amp in the boneyard, the ?I?C for it is A HUNDRED BUCKS. That is the B&D price and I really do not want to get it anywhere else, not THAT part.

And then we get to the Sonys for example that use the STK module to drive the outputs. Now that sucks. They get to say it is a discrete output stage, but got out of designing is by using an IC. I don't want to get into trouble here because I know the owner of this site likes Sony and has some killer Sony stuff. but the fact is, their mid line **** is **** IMO. The ESS line is totally different. But the BPC is as bad as the Pioneer BPC. And all of them are as bad as it gets as far as I am concerned.

Thing is, it is not so much that they made their own ICs, they didn't. Sanken or whoever made them and all they do is set up the right transistors, and two resistors to set the gain and the engineering is done. Anything having to do with sound quality was not done in house. You did not buy a Sony, you bought a brand X with a Sony output and transformer. This is my problem with it.

And then, some will say that ICs, not just STKs but any ICs in the audio path simply have too damn many transistors. I do not disagree. I do not totally agree but the got a point. Every damn thing that audio goes through affects it and there is no argument about that. but then there is the pretty much highly regarded Yamaha "Wedge". I refuse to work on them. A circuit like that SHOULD be in an IC.

Engineering wise, I am a faithful follower of the KISS church. Keep It Simple Stupid ! Consider audio like water. A straight pipe flows best. When you convolute it all over the place you lose that straightness, the immediacy, the something. How do you spell that French phrase Je Ne Say Qua ? It means that which cannot be described.

But sensitive testing devices CAN describe it. Bob Carver proved that a long time ago.

There is yet another speaker wire thread, I am not getting into it even though the point has been proven to me. My boss many years ago said OK, on the test bench setup which was not the best of speakers but good enough to hear ****, take this brand new package of elcheapo speaker wire and put it in line with your bench speakers. There was no question, it sounded ALOT different. I was convinced right then and there.

Anyway, using STK Darlington power packs you still get YOUR sound in an amp. But when you take to using the driver ICs, you just took yourself out of the loop. You are no longer an audio designer, your job is to make the bluetooth n **** work. The wifi. The surround sound, not the sound. And they have all done it. All the mid fi manufacturers seem to have gone the same way. If you want something new and really good, you know that boat in the backyard ? Sell it.

Try to imagine now, in fact some of you are qualified to do so and if so, do comment.

What would it cost to build this stuff today ? Start with something relatively easy, say a Piojneer SX-850. Think you could do it for a grand ? And the numbers do not warrant it. Let's take it a step further, a Marantz 4270. You could buy a Corvette for that kind of money.

Anyway, the main problem with STKs is not being able to get them. Sonically they are no different than any other bunch of transistors, and in fact are probably matched better. But now people like me have to figure out how to build a replacement for them because it is so damn hard to get a new one. I will take a 2SC3281 and 2SA1302 and screw them in the mounting holes along with the drivers right on top of them. But then I have to deal with the bias. It is not adjustable, it has to be right the first time. I can do it, but not for the $14 the STK used to cost twenty years ago. (and we bitched about that)

I can outdo almost all STKs with more modern components, however the class A ones not really. The SVI ones, understand if I refit an amp it will no longer be "new class A". I am simply not willing to do it. I can get it running, but it will not be class A, sorry. You want that you need to send B&D a hundred bucks for one of the few last ICs you can get for it.

But for most of them I can find what is needed to make it work. And it might not be pretty but it will work reliably and sound good.

Nother thing I have noticed, remember how the FTC rules only apply to the front two channels of a surround amp ? Well I have noticed that alot of them have discrete components on those channel;ls but use a chip for the rest, center, surround etc.

Why are they doing that ? Why ? Because that is where their ratings are. Even a Carver, has a custom IC for voltage amp and discrete outputs on the main fronts, and a big STK for all of the other channels. (HR895 IIRC)

Actually, I can't really bitch about STKs because eventually you might just have to pay me to get past the problem of not being able to get one. Only thing is, if it comes to a Technics new class A amp or receiver, I am removing the brand name and model number because it would be unethical to sell one that is not class A without disclosure.

OK this is plenty long and you are probably tired of it. Get some sleep. (I am on a FUBARed schedule myself, there is no hope for me)
 
Back
Top Bottom