simple_psu.jpg
Here are the answers and further discussion.
1. Why does the transformer RMS secondary voltage need to equal the DC output voltage?

The voltage across C1 is (approx) 1.414 * the AC secondary voltage. This is higher than the output voltage by enough to overcome the regulator's Vin-Vout limitation, for most instances. Your low-line performance may not be what you expect. Making C1 larger can help. The important thing is to ensure that when the circuit is driving its rated load, the lowest voltge across C1 is still above the Vin-Vout minimum. You need to look at the voltage with a scope to see the waveform.

2. Nothing here, but do take note, because the transformer current varies with the rectifier configuration.
3. If you raise the secondary voltage, what happens to the regulator?

If you raise the secondary voltage, there is more voltage across C1, but also more voltage across the regulator. The regulator has no choice but to turn it into heat, which you have to get rid of via a heat sink. If the regulator dissipation is low enough, then you get to omit the heat sink.

4. What are the consequences of low line voltage?

See #1.

5. Why is the connection to the negative side of C1 drawn the way it is?

The object here is to keep the ripple current from the transformer/rectifier/filter system confined to just that. Ripple currents can be high enough that the voltage drop caused by conductor resistance can affect things downstream.

6. nothing here
7. What is the consequence of making C1 obscenely large?

You may need to create some sort of soft start circuit like a resistor between BR1 and C1. This limits the charging current and forces C1 to charge more slowly. A relay or FET in parallel with the resistor shunts the resistor after an appropriate time period (a few seconds).

8. Nothing here.
9. You should put the finished supply on a Variac and vary the input voltage through the regulator's dropout point. Monitor the output voltage using a scope that is AC coupled so you can see any ripple getting thru. Put the supply on some sort of load. When the regulator operates below the Vin-Vout minimum, it goes in and out of regulation with every half-cycle of the AC line. The cleanliness of the regulator's output under these conditions is unspecified.
10. Why do you need to pay attention to the regulator's Vin-Vout minimum voltage?

This is very important, and is the reason for making the transformer's secondary voltage AT LEAST equal to the desired DC output voltage. See #8.

11. Why the local bypass capacitor at the regulator input?

The local bypass should be a capacitor with good HF characteristics, like a monolithic ceramic. 100nF is sufficient. It needs to be right at the regulator.

12. What is the exact formula for the regulator's output voltage?

Refer to the regulator's data sheet for more detailed information.

13. What about the value of R1?

Refer to the regulator's data sheet for more detailed information.

14. What about the minimum output current?

Refer to the regulator's data sheet for more detailed information.

15. What about the power dissipation in R3?

P = E^2/R. The resistor should be rated at least 2x this amount.

16. Nothing here
17. Do I really need D1, D2 and D3?

Diodes are cheap insurance. Regulators less so. Don't Muntz these out.

18. What about C2?

Refer to the regulator's data sheet for more detailed information.

19. What about C3?

Refer to the data sheet.

20. What's all this heatsink business anyway?

If in doubt: make the heat sink bigger. You can use your handy bio-digital thermometer (your thumb) to check the case temperature. If your thumb really doesn't want to be there, then you need more heat sink.

Back 

Last modified 07/27/2018.