STK4191V and other STKs... and counterfeits?

[eccerr0r]

I've heard that STK4191 modules are no longer produced by (Sanyo) and a lot of clones and counterfeits are around. Have people been getting these clones and how bad do they sound? I also wonder ... haven't been able to see a site that details differences from the phonys.

I am wondering if I ended up with one of these counterfeit Sanyo STK4191Vs or not. Then again I really can't hear any bad artifacts that I can tell. I also am not sure how much I can pump up the volume before it blows my speakers or my ears (which is probably the limiter for my testing capability.)

A while back I had swapped my STK with a newer one which didn't perfectly match because the original one was dead. The replacement still works. I wonder if I got lucky with a good one, or is it impending doom from a counterfeit?

[rievax_60]

Re: STK4191V and other STKs... and counterfeits?

The counterfeits typically have smaller output transistors and are likely to fail. When the old and a failed counterfeit packages are opened, the different sized transistor die can be easily seen.

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

Is there a way to tell before they fail? Can they be tested without being in circuit?
Or are they visually impossible to tell until they fail or require surgery to tell?

[rievax_60]

Re: STK4191V and other STKs... and counterfeits?

No certain way to know as far as I know.

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

How many watts for a STK4191V do you think I need to drive out to test? Also duration - how long should it be able to sustain the load?

I suspect I will need to use resistors to emulate load, though I only have one 10 ohm 50W resistor to work with...

[budm]

Re: STK4191V and other STKs... and counterfeits?

STK4191xx?

http://www.datasheetspdf.com/datasheet/STK4191.html
http://www.datasheetspdf.com/PDF/STK4191II/528258/1

http://www.datasheetspdf.com/PDF/STK4191V/477351/1

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

I'm not sure how I can translate that to a test in a real system that may have different components than the test fixture

This is what the STK4191V looks like:

https://www.badcaps.net/forum/attach...1&d=1470855450

https://www.badcaps.net/forum/attach...1&d=1470855450

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

Ouch.

I was looking the amp that I have the STK installed in, it has a bargraph display (instead of dB number or a dial) to indicate relative output volume level:

0 LEDS - 0 watts
1 LED - 0.01 watts
2 LEDS - 0.05 watts
3 LEDS - 2 watts <- seems like a big discontinuity jump here for whatever reason...
4 LEDS - 10 watts
5 LEDS - 60 watts

Logarithmic, I guess it does make sense. Anyway the two speakers I'm using for a test are two junk speakers I had (open frame, no box), one is rated 28W 4 ohm and the other 30W 8 ohm...

I got the volume up to light up 3 LEDS... aaauuugh! TOO LOUD... then again I'm like 3 ft away.

No distortion that I can hear, just way too loud.

Switching over to 10 ohm resistors, now I can save my ears and still get it up to 10 watts. The resistors get very toasty...yow! Ran it for a minute or so, then switched back to the speakers. Still sounds ok so the STK didn't blow up yet.

I guess I can run it like this for a while and see how well it holds up, but not sure what other load I have that can handle more wattage without blowing up my ears to really test the STK module.

[Andrew F. Ali]

Re: STK4191V and other STKs... and counterfeits?

I have used the counterfeit STK ICs and have had no problems. I would sometimes, if space permits, install an fan.

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

Audio equipment should sparingly have fans! =o

How many watts was the STK module rated for and how many watts did you try to pull?

[stj]

Re: STK4191V and other STKs... and counterfeits?

a better question, how many W was it rated for, and how big was the heatsink!

most of the old systems i used to see with STK modules had pretty undersize heatsinks if you wanted to blast it for any real period of time.

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

Well, we have to assume that the original heatsink was sufficient for the original STK...

Both the original and this new STK should be 50 watts per channel. I don't know how much the module is supposed to be able to dissipate, but yes it should be fairly large...

Definitely see your point here but I think the reason why my original STK failed was because of one of the pins breaking, probably due to pre-assembly stress (versus transistor failure). Probably current flowing through that stressed pin, which initially worked, eventually failed. Then again there was way too much heat sink compound on the original STK (it was spread all around even on the plastic portions of the module!)...

The heatsink is fairly large but it's internal (no fins outside the unit). I think I put on a proper amount of heatsink compound now, at least it covers the metal plate and doesn't ooze all over the sides.

After the cleaning...

https://www.badcaps.net/forum/attach...1&d=1471537710

Again this was a long time ago when I repaired this unit...

[stj]

Re: STK4191V and other STKs... and counterfeits?

i'm pretty sure that heatsink would need free airflow at a minimum.
did it hang out of the back of the case?

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

No, this was completely internal, but there are slots/perforations in the case above and below the heatsink, so convection definitely was enabled.

[budm]

Re: STK4191V and other STKs... and counterfeits?

BTW, what the power supply +/- Voltage being provided to the STK?

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

I don't recall, it was something less than ±49V and above ±40V...

Edit:
From the STK4191V Datasheet: "Built-in muting circuit to cut off various kinds of pop noise."
Does this mean the STK will cut off Taylor Swift?

[momaka]

Re: STK4191V and other STKs... and counterfeits?

I got the volume up to light up 3 LEDS... aaauuugh! TOO LOUD... then again I'm like 3 ft away.

Put the speakers with the cones facing down on your floor. Even better if you have carpet. That will muffle some of the annoying mid and high frequencies and improve bass. Don't expect miracles here, but when you have no box, this is better than nothing.

Oh, and as an added bonus: when the speaker is put face down on the ground (or in a box), its coil becomes loaded better, since the cone can't push/pull air as easily. Thus, the speaker will draw more power. Same happens when you put it in a box.

So that setup should be able to load your amp and speakers more.

Both the original and this new STK should be 50 watts per channel. I don't know how much the module is supposed to be able to dissipate, but yes it should be fairly large...

Well, if you want to do 50 W/ch and those modules are class AB...

Max efficiency for class AB is around 60-65%. So with 50 W/ch output and worst case scenario efficiency of 60%, the amp module will be drawing 83.333 Watts total. Since 50 Watts goes to the output, the heat being dissipated is = 83.333 - 50 = 33.333 Watts max. If the module has two channels (stereo), the heat output will be twice that.. i.e. 66 Watts.

Just about any modern CPU heatsink can easily handle that and keep things cool, if you got a fan on it. But without a fan, that heatsink will need to grow in size A LOT if all you're relying on is convection cooling. I'd guesstimate at least 3 to 4x bigger.

That said, I seriously doubt you will be running your amp at max output constantly. Thus, you could probably get away with something the size of a CPU heatsink without a fan.

[eccerr0r]

Re: STK4191V and other STKs... and counterfeits?

Hopefully Sanyo was smart and put the output transistors far away from each other on its heat spreader, which would keep localized temperatures lower. It looks like the official Sanyo part says it can get up to 125°C which means it's okay for a smaller heatsink than CPUs which typically are rated a bit lower... then again CPUs have thermal throttling these days.

Looks like:
Thermal resistance: 1.8°C/W
Max Junction temp: 150°C
Max substrate temp: 125°C

I suspect that music at max volume and something needs to be playing at the time to make it dissipate the most amount of power. (IIRC the amp does have a large capacitor on the output so no DC dissipation...)

[momaka]

Re: STK4191V and other STKs... and counterfeits?

Hopefully Sanyo was smart and put the output transistors far away from each other on its heat spreader, which would keep localized temperatures lower. It looks like the official Sanyo part says it can get up to 125°C which means it's okay for a smaller heatsink than CPUs which typically are rated a bit lower... then again CPUs have thermal throttling these days.

Looks like:
Thermal resistance: 1.8°C/W
Max Junction temp: 150°C
Max substrate temp: 125°C

But as you know, heat is the enemy of all electronics, so you shouldn't let it get that hot . IMO, 60°C should be the absolute max to push things.

I suspect that music at max volume and something needs to be playing at the time to make it dissipate the most amount of power.

In almost all cases, music will NOT be able to make it dissipate that kind of power. Only a pure Sine wave at maximum volume can. That said, I typically use Audacity's tone generator to make a 50-60 Hz sine tone and play it at max volume. This allows me to push a really high amount of power through my speakers without going deaf or burning out my speakers (lots of cone movement is good for cooling the voicecoil... though on that note, if you really want to push A LOT of power through a speaker, go for 35-45 Hz). As an added bonus, you can use any cheap multimeter on AC voltage to measure the output RMS voltage swing on the speaker terminals if you are playing a 50-60 Hz sine tone .

(IIRC the amp does have a large capacitor on the output so no DC dissipation...)

Are you sure about that? I haven't seen many amps rated for more than a few Watts use series capacitors on the outputs.

[budm]

Re: STK4191V and other STKs... and counterfeits?

It use +/- power supplies so no need to have the output coupling cap but it should have speaker relay protection in case of large DC offset.
The spec sheet also shows the PD vs PO graph.