~700w ampifier

[megajocke]

You could try making a short connection from chassis to the ground terminal block on the board if there isn't already one. Is the input floating when doing this? Does a shorting plug in the input change it?

I'd clean up the wiring a bit too, maybe tie together power supply wires and twist speaker hot and return.

Also, are you sure the amp is oscillating? It could be picked up from a cheap and noisy switcher somewhere. Does it go away if the lid is put on the amp and/or the amp is turned off?

edit:
I just remembered what topology the amp is - noise on the mains is easily coupled into the output if the transformer doesn't have an electrostatic shield. Maybe this is what you are seeing.

 

[gootee]

Looks pretty good so far!

Does it oscillate with a top cover in place?

Some wire pairs should probably be twisted together.

What happens if you adjust the trimpot?

Have you tried adjusting the feedback capacitors? What values did you use?

 

[krazatchu]

Here is the output pic....

It seems to decrease as I bias into A... about half the amplitude...
I'm assuming it's A, the dim bulb gets brighter which would indicate more power consumption and hence class A, right?

It seems the oscillation is still there with the power off...

Thanks!!

 

[krazatchu]

Lid makes no differnce...

It's about half when it's turn off..

I will tidy up the wires and add an case gnd to the terminal block
Currently the only ground is the socket to the case.... So I assume its floating...

Main feedback is still 56p, forgot to get a high voltage 22pF when I was at the market...
Opamp feedback is 100pF...

Will also read up on electrostatic shielding...

Thanks!!

 

[megajocke]

Here is the output pic....

It seems to decrease as I bias into A... about half the amplitude...
I'm assuming it's A, the dim bulb gets brighter which would indicate more power consumption and hence class A, right?

It seems the oscillation is still there with the power off...

Thanks!!

Then it's just the lack of electrostatic shielding. Oscillation at 40kHz would be a bit unexpected. It's probably coming frome some switching power supply somewhere. It could be the computer.

I'd follow djk's advice on biasing:
"For your use I would turn down the bias to about 340mV across R7 and R10 (optimal class B bias)."

Gives class AB+B and low dissipation. Unless sound quality is important for ultrasonic welding.

 

[krazatchu]

After adding chassis gnd and tiding wires, no noticible change...

Shorting input cuts amplitude in half and causes a negative offset of about -15 mV..

As its occuring with the power is off it seems to be coming in thru the gnd...

I will play with it more on monday... Today is a national holiday in Korea, I didn't get the memo so I'm at the office alone... I'm hungry and no one is here to order lunch...

Thanks!!

 

[krazatchu]

Hmm... electro static sheilding....

Tinfoil??? Makes a great hat!!

hehe

 

[megajocke]

Good for protection from alien mind control beams too!

 

[bicikle]

Hi! I just seen this amplifier, but between what is speaker? Cause cases of output transistors are sirectly conected on GND, and out for speaker is on GND, there will be null potencial, so there will no any sound..... I am just beginer, sory if i ask something stupid...
I am confused cause in other amplifiers that i maded, usualy speaker is between emiter of tranzistors and ground, and cases are directly cnected on the potencial (+ and -), so they must be isolated. I cant stand this tupe of amps, so please some one help me about that...

 

[angeltech]

Hello to all,

A bit late for this thread, as I rescan this project, the problem of oscillation was not given answer, does anyone pursue this project and successfully eliminate the occurring problem?


Regards,

Angeltech