Getting 31v on speaker outputs...

[diy1995]

Hello,
I have finaly build amplifier with STK442-090.
I build it on prefboard but I did my best to make it as best as possible.
The problem is that im getting -31vdc out of both speaker outputs...
I have checked everything. But I can't find any error.
I'm getting +-31v out of psu to the board, both rails have power.
What should I check, measure?

Sub gnd should be connected to gnd rail right?

Help me please

 

[JonSnell Electronic]

The circuit you show is the test example from Sanyo.
Check your work, make sure there is zero DC on the input lines and the feedback line GND is ground.

 

[scottjoplin]

Hi, have you any more information on the chip than this? HTTP 301 This page has been moved
How have you mounted it, is the mounting bracket metal and is it isolated from the Power rails?

 

[diy1995]

Hi, have you any more information on the chip than this? HTTP 301 This page has been moved
How have you mounted it, is the mounting bracket metal and is it isolated from the Power rails?

Hello, the link that you posted has all the informations that I could find.
I havent mounted it yet.

 

[diy1995]

The circuit you show is the test example from Sanyo.
Check your work, make sure there is zero DC on the input lines and the feedback line GND is ground.

Hello, I will measure input lines. Which line is feedback line?

 

[JonSnell Electronic]

The bottom of the 1k8 resistors should go to ground through the capacitors giving AC gain to the amplifier. Gain = 56k + 1k8 / 1k8 = 57.
pins 4, 5, 6, 7, 10, 11, 13 and 14 should be close to Zero from GND.
pins 8 -ve and 9 +ve rails.

Chip datasheet; STK442-090 SANYO, STK442-090 Datasheet

 

[diy1995]

On both input lines i get 90mV dc when I connect dmm. If falls down to 10mV in about 15seconds. Than it raises back to 30mV and it floats around 30mV.
EDIT: measured across input gnd.

 

[Mark Whitney]

It looks like the input GND should connect to the output GND for it to work correctly.

 

[diy1995]

Pin 4: -31v
Pin 5: -31v
Pin 6: -31v
Pin 7: -31v
Pin 8: -31v
Pin 9: +31v
Pin 10: -31v
Pin 11: -31v
Pin 13: -31v
Pin 14: -31v

These are the readings

 

[diy1995]

There is -26v on input lines if i measure across gnd instead of input gnd...

 

[diy1995]

With speaker and phone connected there is buzz on the speaker. When I disconnect mains I can hear few, realy distorted kicks (from music) until capacitors discharges... if that tells anything...

 

[scottjoplin]

Don't run it for any length of time without a heatsink. Have you connected input and output grounds as Mark says?

 

[diy1995]

Don't run it for any length of time without a heatsink. Have you connected input and output grounds as Mark says?

I run it without heatsink just for test. I have to drill holes to the heatsink and I cant do it today. I am monotoring chips temperature all the time.
Should I connect output and input gnds together? If i measure input lines voltage across input gnd I get only few 10 of mV. But when I measure it across gnd/output gnd i get -31v so should I connect them together?

 

[scottjoplin]

Don't worry about the voltage you are measuring there, the input ground must be connected to 0V

 

[diy1995]

Don't worry about the voltage you are measuring there, the input ground must be connected to 0V

Ok. I will connect them together. Cant do it right now. Im not home. Will report asap when i get home.
Thank you

 

[diy1995]

Don't worry about the voltage you are measuring there, the input ground must be connected to 0V

Hello,
I connected input gnd to gnd and it works After having another look to the schematic it`s logical that it should be connected as it is writen GND next to the connection, but it wasn`t clear enought to me.

Thank you very much.

 

[scottjoplin]

Good. Well, Mark spotted it first, easy to overlook. Can you see why it had the effect that it did? It provides a reference for the input and controls the feedback, without it the output floated towards the power rail

 

[Chris Daly]

It would seem to infer your initial power supply is not correct.The data sheet STK442-090 SANYO, STK442-090 Datasheet suggests to use a centretap transformer and to derive a positive and negative voltage. It also indicates 500 ohm resistors on each rail to ground. My guess is you have unequal voltages at the power supply hence the output contains an offset now indicating that error.

Check to see that you have positive voltage at Pin 1, and negative voltage at Pin 2

 

[diy1995]

Good. Well, Mark spotted it first, easy to overlook. Can you see why it had the effect that it did? It provides a reference for the input and controls the feedback, without it the output floated towards the power rail

Yes, I understand a little now. But I don`t have much knowledge about details about amplifiers. This is my second big build (i dont count easy ones with only positive voltage and very few components ). First was single tube EL84 amplifier.

 

[diy1995]

It would seem to infer your initial power supply is not correct.The data sheet STK442-090 SANYO, STK442-090 Datasheet suggests to use a centretap transformer and to derive a positive and negative voltage. It also indicates 500 ohm resistors on each rail to ground. My guess is you have unequal voltages at the power supply hence the output contains an offset now indicating that error.

Check to see that you have positive voltage at Pin 1, and negative voltage at Pin 2

Hello, I already fixed problem by connecting input gnd to gnd.
I have center tapped transformer, same power supply that is on the schematic, except for 2 6800uF instead of 10000uF capacitors (had the capacitors laying around and I don`t know if is necessary to use 10000uF. Sanyo stated the same psu for all stks from 15w to 200w rms).