Help Troubleshooting a blown amp (SAP16)

[twistedsymphony]

I've never built an amp... though I might consider it now that my receiver has died.

I have a Marantz SR6200... it worked beautifully for years but then a couple of weeks ago when I powered it on, it sparked, shot flames and smoke and then put itself back into standby.

Opening it up it looks like a Sanken Electric SAP16 (1413) Transistor for the Center Channel exploded.

I can't find any sign of damage to any other components other than this one part.

It looks like the circuit is setup with 2 SAP16s per channel, a 1413 and a 1417 only one of them popped though... the 1413

There are no signs of burnt traces and I've tested all of the resistors and capacitors on the board and not found any additional problems. I'm not entirely sure how to test the remaining components on the board. FWIW the leg all the way to the right on the SAP16 burned through completely, separating it from the chip right at the base of the chip.

Here are some pictures:




Here are my questions:
1. What could cause something like this to happen? I'd replace the SAP16 with a new part... but if it's just going to pop again I'd rather replace the ROOT problem first

2. Where can I get a replacement part? I found some places that carry the SAP16 but different model numbers of it... if these can be used as direct replacement that'd be fine but ideally I'd like to replace it with the same exact part....

Worst case scenario this is a 6.1 channel amp and I only use 5.1 so I could steal parts from the rear center channel to rebuild the front center... but again I don't want to risk doing that until I know

Any help would be GREATLY appreciated... thanks

[jaycee]

The 1413/1417 will be batch numbers. You will find that each device is marked either "SAP16N" or "SAP16P". They are NPN/PNP Darlingtons with built-in bias circuitry and emitter resistors.

At a guess, it is a SAP16N that has blown, and it is the Emitter lead that has melted. Is it possible a speaker lead has shorted intermittently and caused this?

[twistedsymphony]

the connection on the back of the speaker is clean but I suppose it's not out of the realm of possibilities that there was a short where the speakers connected at the back of the unit... I was using bare wire (I've already ordered some banana plugs to remedy this)... and I did move the unit out a bit and back in place earlier that day to connect an S-Video wire...

so a short on the speaker wire could cause this to pop? That's a welcome relief if it's the case...

Would anything bad happen if I removed both SAP16s from the rear center and just left the circuit empty and used those for the front center?

[ceharden]

Before putting changing the output devices it's worthwhile trying to check for other faults first.

A very useful tool for this kind of repair is the series connected lightbulb. Wire a 60-100W mains filament lamp in series with the mains inlet. I have one on my bench with a bypass switch. With a bulb in series it normally allows enough current to flow for an amplifier with no faults to power up, allbeit probably slower than usual. If there is a fault however (eg a both NPN and PNP output deviced turned on together) the lightbulb heats up and limits the current through the circuit, hopefully stopping any more devices getting fried.

If all seems well then you can use the switch to bypass the bulb and run the amp as normal.

Unfortunately as the bias circuitry is half inside the output devices themselves it's difficult to check that without replacing them. Might still be a good idea to have a probe round and check for anything unusual, you've got other channels to compare to at least.

[twistedsymphony]

It sounds like you're basically using the lightbulb as a fuse... that's rather ingenious...

with the amp board removed I have probed around with a multi-meter and compared the front center channel to the rear center channel... I even removed the blown component as well as the matching part on the rear center channel so that both circuits are identical...

all of the resistors and capacitors checkout ok... the only thing I haven't tested are the smaller transistors.

Then there is of course the amp board plugs back into the mainboard and due to the way the board is installed I'm not entirely sure what's happening between the amp board connectors and the speaker connector... From metering around I can tell it's not a straight through connection... thinking about it now I should be able to at least test for shorts. What would go between the amp board and the speaker outputs? would there be one last RC circuit or something to help cleanse the output? or something more substantial?

[jaycee]

The lightbulb trick works because the filament's resistance changes with the current drawn through it. At idle current, it lets enough to run the amp pass, but if there is a short that causes the load to draw too much power, the bulb lights up and the power is used in the bulb instead of in the amp.

It's also called a "poor man's Variac"

Without a schematic to see how these devices are used, it's tricky to say what to check. If there is a trimmer resistor in the channel, check it's not fried. Sometimes, the transistors driving the SAP16 devices fry as well.

The speaker connection is probably via a Thiele network and with a zobel (R-C) on the output. You might see a coil with a 2W resistor near it of low value. It's also possible it goes through some protection relay.

[KUL FIR CHICK N]

Rather than repairing the left front (which was blown), I swapped the signal and output wires in the connectors between the amp and the receiver from the rear center channel. Then, I cut the traces that run from the power supply to the blown section. Everything now works like a charm, except now I don't have 6.1 which I don't use anyway.

I found that there are relays between the amp and the speaker terminals which cut off the rears when in stereo mode.

[jackinnj]

Guess why Sanken obsoleted the SAP16's??

-- they've been replaced by the STD03N and 'P -- looks like the same device w one pin sheared off -- the problem was a funky emitter resistor in situ.

You're gonna have to find a distributor for the SAP16's (Allegro?) and replace them all with STD03's and an external 0.33 or 0.22 ohm resistor.

Sanken hinted at this problem in their product file -- again, ALL the P and N devices should be replaced with the STD devices and a emitter resistor rated per the wattage it has to dissipate.

[twistedsymphony]

Originally posted by KUL FIR CHICK N
Rather than repairing the left front (which was blown), I swapped the signal and output wires in the connectors between the amp and the receiver from the rear center channel. Then, I cut the traces that run from the power supply to the blown section. Everything now works like a charm, except now I don't have 6.1 which I don't use anyway.

I found that there are relays between the amp and the speaker terminals which cut off the rears when in stereo mode.

I ended up doing pretty much the same thing... I just re-pinned the input and output connectors to the amp board to swap the front center with the rear center, it's only 5.1 now but it's running fine in that configuration.

swapping out the SAPs for the STDs sounds like the "best" fix for this but it' would probably cost me more than it's worth. I ended up buying a new receiver (a Denon 989) and I'll probably just throw the Marantz up on CL or eBay or something.

thanks for all the help here guys...