Without a bulb yes, its possible to turn the bias up so high that the transistors pass enough current to overheat them. A bulb would just begin to light and the rails collapse preventing damage.
If you appear to have no bias current flowing then first make sure that the resistor I asked you to remove is put back in place.
If you appear to have no bias current flowing then first make sure that the resistor I asked you to remove is put back in place.
i did not remove the resistor finally but the TP1 was tried before found that the fault caused by the driver (as you advice).
So i connect in series on LIVE 240v cable a bulb of any wattage and the bulb should be completely off, in case that it start lid i should adjust the trim to turn it off. But what should be the correct voltage on the emitter resistor? As i understand less than normal may cause a problem also.
So i connect in series on LIVE 240v cable a bulb of any wattage and the bulb should be completely off, in case that it start lid i should adjust the trim to turn it off. But what should be the correct voltage on the emitter resistor? As i understand less than normal may cause a problem also.
The bulb goes in the live side as you say.
The bulb may light dimly depending how much current the amp normally draws... and with multiple pairs of output transistor it could well be enough to light the bulb.
TP1 should be initially set for minimum bias current and that setting is when the wiper of the preset is turned away from R25 and toward R27. In other words if the base of Q28 is connected (via the wiper) to R25 then you have it turned to the wrong end.
The circuit is a little unusual in having the wiper and one of the other preset pins connected and so I'm taking the diagram at face value and assuming its correct.
The bias voltage you measure across the emitter resistors will be very small, for example 50 milliamps in 0.1 ohm (check what value is fitted) would measure as just 5mv of DC voltage across the resistor.
The bulb may light dimly depending how much current the amp normally draws... and with multiple pairs of output transistor it could well be enough to light the bulb.
TP1 should be initially set for minimum bias current and that setting is when the wiper of the preset is turned away from R25 and toward R27. In other words if the base of Q28 is connected (via the wiper) to R25 then you have it turned to the wrong end.
The circuit is a little unusual in having the wiper and one of the other preset pins connected and so I'm taking the diagram at face value and assuming its correct.
The bias voltage you measure across the emitter resistors will be very small, for example 50 milliamps in 0.1 ohm (check what value is fitted) would measure as just 5mv of DC voltage across the resistor.
The bulb will not go completely dark, but a dim orange glow is what you want. Then adjust bias. Then play music. First with no speaker. You should be able to turn it up to clipping without the bulb getting very bright. If it passes that test, connect a speaker and play music at a low level. The amp should behave normally. The bulb will get brighter the louder you play, and return to dim when you turn it back down. Only then do you hook it up without the bulb, and re-adjust bias.
100-150 watt bulb is correct for PA amplifiers, for smaller amp go down to 60. I use old PAR38’s out of my disco light setup which were replaced with LEDs because buying a “light bulb” is next to impossible now. Higher wattage 56 or 64’s will work also, especially for bigger amps. Save em. Don’t throw them out when you go to LEDs in your rig. You’ll need a power line limiter eventually.
100-150 watt bulb is correct for PA amplifiers, for smaller amp go down to 60. I use old PAR38’s out of my disco light setup which were replaced with LEDs because buying a “light bulb” is next to impossible now. Higher wattage 56 or 64’s will work also, especially for bigger amps. Save em. Don’t throw them out when you go to LEDs in your rig. You’ll need a power line limiter eventually.
Guys thanks a lot both of you for your valuable info.
Wg_ski... when you said “only then do you hook it up without the bulb and readjust the bias” why do i have to continue adjusting the bias? I will bring it to 5mv and then discommect the bulb.
Also is it possible by playing at hi gain the bulb to light almost full?
Wg_ski... when you said “only then do you hook it up without the bulb and readjust the bias” why do i have to continue adjusting the bias? I will bring it to 5mv and then discommect the bulb.
Also is it possible by playing at hi gain the bulb to light almost full?
The power supply voltage will be a little low with the bulb in series. If you adjust the bias that way, it will be a little high when you operate directly off the line. Not dangerously high, but you may want to bring it back down a smidge. It is very circuit dependent - I’ve seen it go both up and down with increasing Vcc, as well as not changing at all. The usual is a few % increase due to the VAS stage running hotter (normal temperature) at full Vcc.
You can run it with the bulb, but it’s output power will limit to somewhere in the 20 watt peak range. The bulb will come up to about half to 3/4 brightness and then the amp just won’t put out any more power. I would suggest doing this as it is a good sanity check to be sure the amp runs properly as the supply voltage drops. If the amp latches to the rail or goes into oscillation during this test it could indicate unresolved problems or issues with the design. If it gracefully comes back after reducing the volume, everything is fine.
You can run it with the bulb, but it’s output power will limit to somewhere in the 20 watt peak range. The bulb will come up to about half to 3/4 brightness and then the amp just won’t put out any more power. I would suggest doing this as it is a good sanity check to be sure the amp runs properly as the supply voltage drops. If the amp latches to the rail or goes into oscillation during this test it could indicate unresolved problems or issues with the design. If it gracefully comes back after reducing the volume, everything is fine.
I kind of suspected the amp may draw more quiescent current than a bulb might be happy with (read post #24) due to the parallel pairs of outputs.
If the amp is working OK on full mains then the 'danger point' of something still being faulty (or the bias preset being drastically incorrectly set) and causing catastrophic failure is past. Just set the bias up carefully and your good to go
If the amp is working OK on full mains then the 'danger point' of something still being faulty (or the bias preset being drastically incorrectly set) and causing catastrophic failure is past. Just set the bias up carefully and your good to go
Trying to set the bias (without bulb), i found some strange behaviour. If i measure the good channel with gain at minimum i have only 0.8mV, when i increase the gain at half i have about 6.8mV and from half to full gain it drobs again at about 2mVolts. So the bias is changing with gain without any input or dpeakers connected. Is this normal?
I have try the repaired channel an it measures about the same with difference that at full gain it drops at 1.3mv if i set it to start at 0.8mv.
What are yours opinions guys?
Thanks
I have try the repaired channel an it measures about the same with difference that at full gain it drops at 1.3mv if i set it to start at 0.8mv.
What are yours opinions guys?
Thanks
Its not expected behaviour, that's for sure. I wonder if the stages before the power amp (before the volume control) are introducing some high frequency noise of some kind.
The only sure way to know what is happening would be to use an oscilloscope and look at the amplifier output under these conditions.
Although not expected, it may be 'normal' for this set up, particularly as both channels behave essentially the same.
The only sure way to know what is happening would be to use an oscilloscope and look at the amplifier output under these conditions.
Although not expected, it may be 'normal' for this set up, particularly as both channels behave essentially the same.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.