Dragster Amplifier bad SMPS

[ben12345]

Hi,
I got this amp from a friend, and since I dabble with some electronics, I thought i'd have a go.
Its a Dragster DB 505 5 channel amp.

http://www.flickr.com/photos/55331647@N00/1571391294/

I replaced all obviously burned components, ie. all of the output devices and corresponding power resistors, the power supply transistors, input capacitors and one or two other misc components. The big resistor by the terminal block is replacing a smt resistor that was open circuit incase you were wondering.

I disconnected the rectifier diodes to aid troubleshooting. I'm only getting less than half the output from the toroid that I was expecting. I'm getting about 9vac per secondary!

I know the basic idea of SMPS, but I must be missing something obvious.

The four power supply transistors all get warm, but the two in the far left get hotter. The voltages at the legs of the devices all compare the same (ac and dc), and i'm getting about 4vac on each primary of the toroid.

Hope someone can give me some advice on what to check next.

Cheers.

[Perry Babin]

Measure the DC voltage on the power supply transistors. There are two banks of transistors. Post the three voltages on each bank with the amp powered up. All of the transistors in each bank should have the same voltage. For this, touch the black lead to the ground terminal of the amplifier.

AC voltage with a multimeter can mean many things because they don't all respond the same way to high frequencies. DC is somewhat more reliable. Nothing beats a scope but I realize not everyone has one.

[ben12345]

Managed to fire it up again today;

Bank 1 both mosfets measure the same
G 6v
D 14.4v
S ~0v

Bank 2 both mosfets measure the same
G 5.9v
D 14.4v
S ~0v

I do have a digital storage 'scope, but i'm waiting for some new probes to arrive, so i'm limited to the DMM at the moment.

[Perry Babin]

Those voltages are what I'd expect.

Were the replacement FETs the same as the originals?

[ben12345]

Ok,
Progress! I have a 10v p-p square wave on the gate pin of both banks of transistors.
25v p-p square on the drain pin.

I remembered I still had a crappy probe set put away in the loft.

Looks like the SMPS is working?

[Perry Babin]

The FETs shouldn't be getting hot.

Were the replacement FETs the same exact part as the originals?

Is the gate drive waveform square? The right, falling edge is most critical.

[ben12345]

And yes, they were exactly the same type.

[ben12345]

Sorry if it appears that my posts are coming through at odd intervals, i'm still being moderated.
Here is a pic of the wave on the gate pin:

[Perry Babin]

Calculate operating frequency by using the timebase setting and the number of divisions per cycle. It should be near 30kHz. If it's not, that could be the reason the FETs are heating up.

[ben12345]

Period is 25us so my my reckoning thats 40KHz.
Should the FETS remain cool? I dont have them attatched to a heatsink at the moment.