High Current Regulated Power Supply For Gainclone: Design Thread

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

After looking at a few options for the BPA300, i have decided on using a regulated supply. I have been searching for ways to increase the output current of the LM317, i found various ways of doing this, one with the TIP3055.

The idea is to use the LM317 as a reference for the 3055's, to put out 30V.

The peak current required to be around 10A at full load, which a pair of 3055's should be able to handle with a lot of heatsinking.

I would like to be able to regulate the + and also - supplies, which i belive can be done with this regulator.

Can anyone see any errors in the schematic, or any suggestions?

IMGDEAD]
 
Hi,
you don't need 22//22//22mF of smoothing before the regulator.

The peak current should only be a short term transient. It should not significantly heat the output/pass transistors. It's average current that creates most of the heat.

Use a dual secondary transformer with a bridge rectifier on each secondary.
Then cascade your two regulators to create a dual polarity supply. Connect X1-1 to X1-2 to form the zero volts connection.

I think C4 could usefully be made a much higher value.

R7 is going to run pretty warm.

The ratio between R4 & R7 does not seem right for 30V output.

Don't build the ground line like your schematic shows.
R5 should connect to X1-1, not to R2
 
AndrewT said:
Hi,
you don't need 22//22//22mF of smoothing before the regulator.

The peak current should only be a short term transient. It should not significantly heat the output/pass transistors. It's average current that creates most of the heat.

Use a dual secondary transformer with a bridge rectifier on each secondary.
Then cascade your two regulators to create a dual polarity supply. Connect X1-1 to X1-2 to form the zero volts connection.

I think C4 could usefully be made a much higher value.

R7 is going to run pretty warm.

The ratio between R4 & R7 does not seem right for 30V output.

Andrew,

Yes, i was going to use a bridge rectifier on each secondary, and simply reverse the outputs from the regulator for a negative supply.

I will bump up C4, and set the correct ratio between R4 and R7 you are correct it is incorrect, i was thinking about using a trimpot to ajust the voltage.

Can you see any modifications that would be usefull? How about adding another 3055?

I have fixed the ground on my local schematic.
 
for a 7Arms output the resistors will share the dissipation ~4.9W for 50% duty cycle.

7Wrms ~ 300W into 5r4

It would appear that during testing to full power into 5r4 each resistor is required to dissipate ~1.3W. That indicates that 2W would be sufficient.

I suspect that during heavy domestic use, 600mW will only get warm.
 
rhysh said:
R4 and R7. 215R and 5K should be okay?
Maybe.
try breadboarding it.
It appears that T1 is adding extra fairly fast gain to give an adjustable reference voltage to the adj pin.
Reg out is 1.25V above this ref V.
It may need a trimmer for either R7 or R8 to set the output voltage and the drop across the 317. R7 must be protected from going too low, otherwise T1 gets an enormous Vbe.
Delete R3 (=6r0), surely this is a dummy load for the simulator?
 
the base of T1 is fixed by the resistor ladder.
The emitter of T1 must be that voltage less ~600mV.
The current through R7 must give that emitter voltage for T1 to work.

Now find what currents are flowing above T1 to find what resistors need to be altered to give your regulated 30V output. Look at adj pin current, R4 current and R5 current. You don't need a simulator to do that.
You'll learn more by breadboarding it and measuring voltages and inserting them with the appropriate currents on your schematic.
 
Hi,
What is the peak current demand of an 8ohm speaker?

200W into 8r0 approximates to 7.1Apk for sinewave power.
An 8ohm speaker can require three times this, i.e. 20Apk on fast transients.

Where can that current peak come from?
Can some of this demand be met by the decoupling close to the output stage?

What is the peak transient current that is permitted in a warm or hot 3055?
How many 3055 are required to meet the remaining peak transient current demand of that 8ohm speaker.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.