• 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

OK, so what's with STK amplifier modules?

BrocLuno

AK Subscriber
Subscriber
I have a Mistubishi da-R20 with an STK 1060 power amp module and it generally sounds OK driving my EPI 100's.

I have a Yamaha R700 with an STK 3056 power amp that sounds really good driving my modified EPI 70's (B&G drivers). It kicks some serious controlled bass for a small amp and small'ish speakers (that now weigh around 30 lbs each!). And it has the "Spatial Expander" that is really fun to play with if you want to move the sound stage around a bit.

Both my classic MCS receivers use STK modules. They are more eye candy than real audio (knobs & switches galore), but they do an OK job for background or old vinyl input. They just hate over processed CDs and high input voltages.

So I guess my question is what is wrong with STK modules in general? Can they be influenced by the power stage caps? If trimmed for minimum offset, do they respond well? :scratch2:
 
Register to hide this ad
I can't help you with information about the STK Modules but I do have a related question to add. Is there a difference from one manufacturer to another for the same part number. For instance, I recently acquired a Sansui Classique A-701 system that has a bad channel. My question is not about the system, it is about the "STK2250" integrated output IC. When searching for a replacement I found prices ranging from $15.99 all the way up to $38.00 (and probably beyond). Since the A-701 appears to me to be a low to mid-fi pseudo-component system, is it likely that there would be an audible difference from the cheapest IC to the more expensive IC?

Originally posted by BrocLuno;
So I guess my question is what is wrong with STK modules in general? Can they be influenced by the power stage caps? If trimmed for minimum offset, do they respond well?
I am interested in exactly the same answer as well.

No intent to hijack the thread, just a related question that I suspect those in the know can answer in the same breath.
 
I just looked up the STK for my Yamaha and found a question and answer thread that says all STK amplifier modules are the product of Sanyo Corp? If true, I would email Sanyo asking for any differences in quality, etc. My suspicion is that the module is the same if it does not have suffix characters to indicate special run? If that's so, buy the cheapest and see if it works?
 
STK modules are made by Sanyo. There may be some knock - offs out there, but I don't know, and haven't heard of any counterfeits. I always buy the cheapest STK's I can find and they work well. Electronix is a good place to try. Also MCM Electronics.
The STK module is a case of cramming 10 lbs of @#$%^ in a 5 lb box. It seems to be that if you try to put an entire output stage, or even just the output transistors in a small enclosure, with a small metal plate on the back to transfer heat, you aren't going to get very good heat transfer away from the parts inside. The heat will build up and lead to premature failure. The fact that I have replaced so many STK modules seems to bear this out. As far as the sound goes, I haven't ever heard a receiver or amp that had these things sound as good as those with discrete outputs. However, since the STK's are a cost cutting measure, it's likely that corners were cut elsewhere in those units, and that may be what causes the mediocre sound, not the STK's. Now someone is going to bring up the Kenwood KR9600 receiver, which has output modules, but they are (were) Kenwood's proprietary design, not STK's. This receiver does sound wonderful when it works, but the modules do fail, proving my point that it's not a robust system.
 
Thank you both for the great info. I really do appreciate it. The info is pretty much what I expected but since I don't have the experience I didn't know. I will let you know in this thread how the replacement goes and what the results are.

By the way, I did order the IC and a bunch of other stuff from Electronix as a result of a Google search for the IC part number. It was my first order from them and I am very happy with them. Got some handy stuff and some pretty decent cables cheap.

Thanks again guys!:thmbsp:
 
Last edited:
I noticed that the Yamaha R700 has a large array of finned aluminum heat sinks the the modules attached. Looks like their engineers were taking the heat issue seriously?

So what's the best pratice to keep them alive? Stay below 1/2 throttle? Run the loads at 8 ohms only? Add resistance (non inductive) to keep the power-out down a ways - like add 2 ohm resistors to the speaker feed lines for 4 ohm speakers? Cooling fans make noise, but I could add a large'ish slow one to the top of the cabinet and drive off the switched outlet - it won't be visable in my rig?
 
1/2 volume on the amp / receiver does not mean 1/2 output. With a high enough input level, 1/2 volume level can mean full output. If you then boost the bass or use the loudness control, you are trying to further amplify the bass beyond what the amp is capable of, and the output will clip. This will destroy your outputs and your speakers. So to preserve the STK's, keep the tone controls at flat, the loudness off, and the volume to well below where you hear distortion. Also stick with 8 ohm speakers. Don't add resistance in series with the output, it will lower the damping factor, the control the amp can exert on the speakers.
 
Not so bad

I honestly believe that STK output stages can produce great sound quality, even if they are a cost-cutting measure. Then again, they are not as robust as discrete when driving lower-impedance speaker loads.

I use my Sanyo dca-611 amp with STK-0059s in the output stage all the time in my garage, and it gets damn hot in there in the summer. It has been extremely reliable. Maybe this amp was designed correctly, since the STK's each have their own massive heatsink (massive for 60wpc, that is), and the service manual indicates a sophisticated thermal protection scheme. That's probably the reason why both STK's are OEM Sanyo parts after some 30 years.

FWIW, I see no real sound quality difference in an STK design vs a discrete , class AB design.
 
STK modules are made by Sanyo. There may be some knock - offs out there, but I don't know, and haven't heard of any counterfeits. I always buy the cheapest STK's I can find and they work well. Electronix is a good place to try. Also MCM Electronics.
The STK module is a case of cramming 10 lbs of @#$%^ in a 5 lb box. It seems to be that if you try to put an entire output stage, or even just the output transistors in a small enclosure, with a small metal plate on the back to transfer heat, you aren't going to get very good heat transfer away from the parts inside. The heat will build up and lead to premature failure. The fact that I have replaced so many STK modules seems to bear this out. As far as the sound goes, I haven't ever heard a receiver or amp that had these things sound as good as those with discrete outputs. However, since the STK's are a cost cutting measure, it's likely that corners were cut elsewhere in those units, and that may be what causes the mediocre sound, not the STK's. Now someone is going to bring up the Kenwood KR9600 receiver, which has output modules, but they are (were) Kenwood's proprietary design, not STK's. This receiver does sound wonderful when it works, but the modules do fail, proving my point that it's not a robust system.


Great observations, thanks Warren. I think it is important to realize what was going on in electronics manufacturing back in the day when these units were new. As home electronics moved from discrete to thick film and I.C. devices, sometimes the quality suffered and the MTBF often took a hit. I believe there is a direct correlation between cost cutting and the amount of abuse a given system was designed to tolerate, so I choose my vintage gear accordingly
 
Over the time I had more then a few units which used STK's. Most of the units had faulty ST's when I bought'em. However although I listened a lot of STK units I think that all sounded different.

As an example, I can say that a Pioneer silver REciever with STK's sound a lot different then a Sony HMK 80 .

Last words, by a very big margin the best sounding STK based unit I heard was theAurex 12 series power amplifier with it's matching pre. I gave that as a present to some friend. It sounded so good and powerfull.

Mertol
 
In my opinion, the STK was a sign that the manufacturer was not being creative, and was using a cookie cutter amp stage. If you look at Harman Kardon, for example, their amplifiers were discrete, and utilized cascode loading it the driver stage.
Early failures in STK were common. An effect called secondary breakdown commonly results in shorted outputs. STKs probably had secondary breakdown safe operating areas that were not the best available.
 
AFAIK it iis all in the surrounding design, frex, I really could not identify in a double blind test whether it was a Yamaha CR-820 or a CR-840 when both are properly adjusted.
 
Good explanation, Nakdoc. To elaborate, every power transistor has a safe operating area, described by a graph of voltage vs. current and temperature. There is a shaded in area ofthe graph that shows the conditions the device will function in. If you go outside the area, you risk destruction of the device. So, for example, a transistor might be ok at 10A, but only with 20V across it. Or 200V but only at 1A. And these ratings degrade as the temperature increases.
 
On Yamaha R-900's, for example, if the STK devices are replaced (driver and output) with new devices, the amplifier will oscillate in the right channel. Replacing compensation caps (of which there are a dozen or so) will not snub the oscillation. Yamaha has a "mod" for the feedback loop, evidently intended to suppress oscillations, but this is not effective, and appears to have been workable only on R-900's that developed oscillations while the units were still relatively new (mid-80's).

My conclusion has been that new-production STK devices are often slightly different in internal compensation than old-production STK devices. This may make them "better" when used in new designs, but it makes them slightly incompatible with older circuit topologies.

I've had THREE R-900's that could not get fixed with new STK's and the officially-approved mod.

Sigh.

Fred Longworth
 
Fred,

Did you try decoupling caps on the power supply pins of the STK? Are the outputs integrated into the pack also? If discrete outputs try a 100pf cap between base and collector.

I've had THREE R-900's that could not get fixed with new STK's and the officially-approved mod.

Sigh.

Fred Longworth
 
Only the R-1000 and R-2000 had discrete outputs, I have one of each looking for the STK voltage amps. The R-700 and R-900 had bothe STK voltage amps and STK power amps - still obtainable bot as noted maybe not electrically exact as the originals.
 
If Marantz made STK's, theyd be golden

I see the modules have a stigma attached to them depending on who makes them, (as usual) referring to Warren's 1st post.

Abuse a solid state amp by overheating it repeatedly, it will eventually fail, doesnt matter if it's STK's, Kenwood Trio's (Sanken), or Toshiba discrete outputs. Basically, all solid state complimentary Class AB amps sound the same, the differences are about 90 percent in the power supply, and the other 10 percent cannot be heard....only imagined. Those that say the metal plates on STK modules are too small are'nt getting out their rulers and making any measurements, by comparison most modules I've seen have more surface area than a combined set of discrete output pair.(Even TO-3 types) Digging a little more deeper into the thermal foray, You will find that the junction to case thermal coefficient of these devices are not as low (efficient)as a discrete power transistor in most cases, and this accounts for just about all of the "these things fail under heat" opinions. It is ESSENTIAL that these devices have as good of thermal coupling to their heatsinks as can be had, make sure thay are gooped up well, secured tightly, and you'll be fine. This will not make up for the above thermal coeffiecient, as that is a function of heat transfer within the device itself, BUT, The above mentioned thermal "weakness" is adequate however, to protect the units when operated within spec and sinked adequately.
There is absolutely NO reason you should "keep the volume down" or "leave Your tone controls flat" if You use a unit employing these modules than You would any other solid state devices. Driving them into clipping also is no more an issue for STK modules than it is for a discrete design. There is No reason to operate into strictly an 8 ohm load, most of the modules are safe into 4 ohms, and yep they get hot as hell just like discrete outputs under those conditions. I have repaired dozens of receivers and amps that had blown output stages, and there have been just as many with roached discrete output transistors as there have been STK modules.
Well designed amps have thermal protection, again, does not have to be (and is not in my experience) more elaborate protection because STK modules are employed, thats just good design regardless. Heat is the arch enemy of all solid state electronics, period.

With an adequate power supply behind them, as in the case of the Sanyo 2050 receiver for instance, they DO sound just as good as a discrete output design. Once again, the stigma determines the sound here, not the ears, as has pretty much been the case around here forever with anything labeled Sanyo. And as for advanced topology and circuit designs, it all depends on the STK, as some do have CCS's employed for their input stages, as well as other advancements that were being developed and employed at the time. Some have very advanced designs that sport specs just as perfect as their discrete amp counterparts. As we all know, ears dont hear numbers and if one skimps in other areas of the design (as Warren mentioned), any amp can sound crappy regardless of its schematic. If you were blind to what an amp employed (discrete vs. STK) and all you could view was the test equipment, You simply could not determine from the measurements which amp You were testing.

In the beginning of their use, STK modules were NOT a cost cutting measure from a parts only standpoint, in fact, they were more expensive than discrete output counterparts of the same power level, the savings came in labor, You just bolt them on with two screws to a flat, undrilled surface, no sockets, no mica insulators and extra wiring to the power transistors.

Why do KR9600's fail, has nothing to do with their modules, has everything to do with the fact that they get abused because of what they are, by mostly the non-audio loving groups. "Check this out man, this was my Dad's- it' got 200 WATT's dude, I can rattle my playstation controller right off the top of this iPod dock, watch man"......................

Not being able to replace the Yamaha's STK module without oscillations is clearly a result of poor design, something is wrong there, and it aint the STK's, I have replaced many of these on 25 year old receivers, and never had a problem with a "new" replacement STK module. A commercially produced amplifier should never oscillate.....ever. Someone dropped the ball on design there, or there was way more cost cutting going on at Yahama than even, oh my gosh, that crappy Sanyo stuff. Of course, the models being mentioned are not Vintage receivers to my knowledge, and basically anything produced after about 1980, had nowhere near the overbuild mentality incorporated into it, as units that fit into this time frame (before 80).....

SO here's my advice, if you use a receiver built after the Vintage era (the Real Vintage era), then I would suggest Warren's advice, keep the volume way low, turn off Your loudness and leave your tone controls to flat, regardless of whether Your unit employs STK's or has twenty four discrete devices in parallel, and use strictly 8 ohm loads, but for those of You that have SX-780's for instance, go ahead and push all You want and enjoy, Your no more likely to fail than anything else, and 100 times less likely than anything built later by Pioneer, like those with "V" in their model numbers, or ones that say "elite" on them..........
 
Just wanted to thank @sanyofreak and the others with experience for weighing in. I just picked up a Sansui 221 and it sounded considerably better than I expected it to. It runs an STK013.
 
Back
Top Bottom