Paralleling Power Supply Units Made Easy ???
I am considering paralleling a couple of 20a variable adjustable psu's for 40a charging. According to what I have read on the world wide inter-web, it can be done quite easily, provided the following ...
Forum Discussion
Salvo wrote:
“That's not quite right. The supply that sees higher voltage than it's setpoint will just shut off. The pwm duty cycle will go to zero percent. This is a very safe condition for the supply.”
Correct, this is one of the condition that I think could happen.
“When the first supply reaches its setpoint it goes into constant voltage mode. When the other supply charges the battery voltage above that set point the first supply's control loop tries to lower the voltage to its setpoint, it can not do that. It will keep trying and will saturate. At that point the first supply is operating open loop. Depending on the supply design it may just shut the output off, a good thing.”
Condition 2:
“It is possible that the error amp, being over driven, will saturate and reverse its output if it is not designed well. This would turn on the output to full voltage, a bad thing!”
Salvo wrote:
“ You're not going to overstress the error amp either. The amp is probably powered by 12V. The reference voltage to the error amp is usually around 5V. If the first supply is at 15V then the second supply needs to go up to 15V * 12/5 = 36V. That's highly unlikely that the two setpoints are that far apart.”
It has been many years and my memory is not what it use to be! My terminology may be wrong when I said error amp is over driven. What I am trying to say is that at the point that the control loop, while trying to lower the output voltage, will at some point saturate. At this point I agree with what you say. “The pwm duty cycle will go to zero percent.” At this point the control loop is now open loop. When it is open loop there is no feedback to drive the + and – inputs to the error amp to the same voltage. A long time ago operational amplifiers had a nasty habit of reversing or latching their outputs in the opposite direction than it should when this happened. This would turn the converter output full on, a bad thing. Maybe this condition can no longer happen, I don't know. With parts being made in China I would not bet on anything.
From MEXICOWANDERER's last post it seems as if something bad happens when paralleled! Maybe what I just said?
“That's not quite right. The supply that sees higher voltage than it's setpoint will just shut off. The pwm duty cycle will go to zero percent. This is a very safe condition for the supply.”
Correct, this is one of the condition that I think could happen.
“When the first supply reaches its setpoint it goes into constant voltage mode. When the other supply charges the battery voltage above that set point the first supply's control loop tries to lower the voltage to its setpoint, it can not do that. It will keep trying and will saturate. At that point the first supply is operating open loop. Depending on the supply design it may just shut the output off, a good thing.”
Condition 2:
“It is possible that the error amp, being over driven, will saturate and reverse its output if it is not designed well. This would turn on the output to full voltage, a bad thing!”
Salvo wrote:
“ You're not going to overstress the error amp either. The amp is probably powered by 12V. The reference voltage to the error amp is usually around 5V. If the first supply is at 15V then the second supply needs to go up to 15V * 12/5 = 36V. That's highly unlikely that the two setpoints are that far apart.”
It has been many years and my memory is not what it use to be! My terminology may be wrong when I said error amp is over driven. What I am trying to say is that at the point that the control loop, while trying to lower the output voltage, will at some point saturate. At this point I agree with what you say. “The pwm duty cycle will go to zero percent.” At this point the control loop is now open loop. When it is open loop there is no feedback to drive the + and – inputs to the error amp to the same voltage. A long time ago operational amplifiers had a nasty habit of reversing or latching their outputs in the opposite direction than it should when this happened. This would turn the converter output full on, a bad thing. Maybe this condition can no longer happen, I don't know. With parts being made in China I would not bet on anything.
From MEXICOWANDERER's last post it seems as if something bad happens when paralleled! Maybe what I just said?
