diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   STK4050V and it's use in Definitive Tech PF15TL subwoofer (http://www.diyaudio.com/forums/chip-amps/97929-stk4050v-its-use-definitive-tech-pf15tl-subwoofer.html)

rkc7 12th March 2007 02:37 AM

STK4050V and it's use in Definitive Tech PF15TL subwoofer
 
I just bought a broken Definitive Technology PF15TL subwoofer from a guy for cheap... nice big 15"er with what is claimed to be a 325w amplifier. But, looking inside is a single STK4050V chip??

The guy said it just stopped working one day, but the power LED still comes on. Which I confirmed, light comes on, it makes a little noise shortly after powering, but no output from signal.

So, probably means the chip is blown? Not a big deal, can be replaced and all the resistors look OK, so probably not a hard fix?

But, this chip is only rated for 200wRMS at 8ohms, yet they claim 325w, plus the woofer measures 3.9ohms. I'm guessing the amp may put out around 300 some watts at 4 ohms, but is that a safe load for it? Also, it appears to use a Silpad but there's also some thermal grease used between both the chip and pad and pad and heatsink... I thought you weren't supposed to use grease with those pads?

Did this company do something special with this chip to get 325w and be able to drive a 4 ohm woofer? Could that or something else affected the longetivity and cause this thing to blow? It's about 5-6 years old, but for an originally $700 speaker, that's not too good.

I don't have a schematic let alone any specs on the subwoofer, couldn't really find anything online.

rkc7 12th March 2007 08:01 PM

2 Attachment(s)
I did some testing and I'm not sure the chip is blown. No pins were shorted at all and the "diode test" results seem to be consistent with the equivalent circuit in the datasheet. I did find a blown 1/2 watt metal film resistor which I think is 19.1k ohms that connected to pin 4 and then to the collector of an MPSA92.

Someone at Definitive has said they can give me some information about the amp, which I'm hoping includes a schematic, so that should help.

I've attached a picture with the pin voltage readings I measured with a copy of the equivalent circuit pasted in as well, if anyone is interested at looking at it. To clarify, the two lines at the bottom with '0' meant no reading, all pin combinations not shown had same reading. All readings taken from chip out of circuit.

Stuart Easson 12th March 2007 08:37 PM

pin 4
 
Looks to me like pin4 is usd to bias the front end of the amp. If the transistor that 'supplies' this pin is blown, I'd expect the amp to be silent...

Without more info it's difficult to diagnose with any accuracy, but fixing the transistor and it's associated components may take care of the problem...

regarding the power rating...they are perhaps using some short term power delivery specification, not continuous RMS watts. The size of the transformer and heatsink will give a better indication of it's long term capabilities...

HTH

Stuart

FastEddy 12th March 2007 09:02 PM

Slightly off topic
 
I have Q & A for the sub woofer amp builders:

Situation: a stereo pre-amp, how do I connect these outputs and "combine" them for the third single channel to drive an active sub ??

(I have a bottlehead tube pre-amp kit coming and I want to output to the two main amps (stereo) and also to my sub woofer (a power MOSFET single channel amp.) :confused:

BWRX 12th March 2007 09:26 PM

A simple resistive summing network will sum the left and right channel signals. It will also attenuate them, so some extra gain will be required in the sub amp to normalize the levels.

Make a Y with three resistors. Use 3 10kohms for example. Ground the lower terminal of the Y and connect each source channel to an upper terminal of the Y. The summed signal is at the center of the Y, and in the case of 3 10kohm resistors, each channel's voltage level is reduced by a factor of 3.

rkc7 12th March 2007 09:35 PM

Re: pin 4
 
Quote:

Originally posted by Stuart Easson
Looks to me like pin4 is usd to bias the front end of the amp. If the transistor that 'supplies' this pin is blown, I'd expect the amp to be silent...

Without more info it's difficult to diagnose with any accuracy, but fixing the transistor and it's associated components may take care of the problem...

regarding the power rating...they are perhaps using some short term power delivery specification, not continuous RMS watts. The size of the transformer and heatsink will give a better indication of it's long term capabilities...

HTH

Stuart

The resistor was the only defective component I found. Still not sure if the chip itself is bad, but it seems it may be okay. I'm hopeful that replacing the resistor would fix the amp, but I still have more components to test and I'm waiting on a schematic before I attempt to fix and test.

As for the heatsink and transformer, the chip is mounted to a 4mm thick, 406mm x 178mm plate with a 122x102mm heatsink with 3mm base, and 13 17mm high fins mounted on the outside of the plate behind the chip. The transformer is 105X115X92mm and pretty heavy.

FastEddy 12th March 2007 09:51 PM

" ... Make a Y with three resistors. Use 3 10kohms for example. Ground the lower terminal of the Y and connect each source channel to an upper terminal of the Y. The summed signal is at the center of the Y, and in the case of 3 10kohm resistors, each channel's voltage level is reduced by a factor of 3. ..."

I knew about this, but wanted something that more closely match common practice in the tube world = hoping to reduce cross talk between the stereo channels as much as possible. I would have used 100K rather than 10K and then paralleled 50K trim pots to get gain and balance under control.

Is there an interesting op-amp solution, like a summing op-amp with input impedence >> 1 meg? (ala my avatar image the AD825) ... that way I could use a low pass active filter / crossover as a part of the circuit. I suppose I should just dust off the ol' op-amp cookbooks and roll my own. I was just hoping for the latest DIY wisdom here.

Stuart Easson 12th March 2007 10:34 PM

sounds big enough to me...
 
Definitive rate all their current subwoofers in RMS watts, so I'd assume this one is rated the same way...and they have increased the rating on the PT15TL+ model to 500w RMS...

The transformer and heatsink sound pretty beefy, not sure if they can do 350w RMS continuously or not, but given the very +ve reaction people seem have to Definitives subs it sounds more than adequate...

Anyway the rails used to drive this amp have to be at least +/-72v to get 325w RMS, probably with 20% more to allow for component tolerances, transformer regulation, mains dips etc. Please be careful, there's more than enough energy to hurt you.

Here's a WAG: if muting is implemented by a 19.1k resistor used from pin4 to ground (or worse +ve), with the mpsa92 stealing current to -ve, the 1/2w resistor is close to it's rating with ~85v rails (to +ve would exceed it's power rating). If it is also an overload or protection mechanism, the resistor could be subject to high temps if the mute/protect is on too long...did the previous owner like to play it loud?

HTH

Stuart

rkc7 12th March 2007 11:41 PM

Re: sounds big enough to me...
 
Quote:

Originally posted by Stuart Easson
Definitive rate all their current subwoofers in RMS watts, so I'd assume this one is rated the same way...and they have increased the rating on the PT15TL+ model to 500w RMS...

The transformer and heatsink sound pretty beefy, not sure if they can do 350w RMS continuously or not, but given the very +ve reaction people seem have to Definitives subs it sounds more than adequate...

Anyway the rails used to drive this amp have to be at least +/-72v to get 325w RMS, probably with 20% more to allow for component tolerances, transformer regulation, mains dips etc. Please be careful, there's more than enough energy to hurt you.

Here's a WAG: if muting is implemented by a 19.1k resistor used from pin4 to ground (or worse +ve), with the mpsa92 stealing current to -ve, the 1/2w resistor is close to it's rating with ~85v rails (to +ve would exceed it's power rating). If it is also an overload or protection mechanism, the resistor could be subject to high temps if the mute/protect is on too long...did the previous owner like to play it loud?

HTH

Stuart

Would a higher rated resistor be better? Not sure if that exact of a value is available in higher wattages, though

I didn't ask the owner how hard he liked to play it, but it's probably a good bet he had it loud when he used it. Could the resistor acted in a protective manner by burning out due to high clipping and left the amp chip intact along with everything else?

Hope to have a schematic soon, which I'm sure would be more helpful in diagnosing the problem. I'll see if I can trace down the circuit a bit more, it's just difficult since it's a double-sided board.

Stuart Easson 12th March 2007 11:50 PM

bigger resistor
 
Bearing in mind I am guessing about whats going on here:

The use of a bigger resistor would be an excellent idea, I'd use something more like 2W or 5W if the connection point is to +ve instead of ground.

The app note for the chip shows a 10k resistor to ground, perhaps Definitive doubled the value to make the +ve connection safer...anyway you can always use a pair of 9 or 10k resistors to get more power handling at the value you need, I'm pretty certain this value isn't critical.

Let us know what they send you, I'd be very interested in how they hook it all together...

Stuart


All times are GMT. The time now is 12:33 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2