« Duty Cycle and DC Gain | Main | Power MOSFET Gate Drive »
August 22, 2001
Start-up Bias Power
Question: What can be done about poorly regulated bias power?
Original Question: I am designing a two-switch forward converter rated for 500 Watts, which has a PFC front end. I am facing a problem with the Vcc derived from the PWM transformer. Its regulation is bad. The supply voltage to the PWM IC varies too much with load on the converter output. If the turns are kept low, then the converter won't start at low loads. If the turns are sufficient, then at high load Vcc goes very high, damaging the IC or zener. Can you suggest some solution for this problem?
Answer: I'm not sure I fully understand the problem so let's talk about it.
First, I don't see how the power factor correction (PFC) circuit gets into the act. The forward converter should just see a dc voltage across its input capacitor that varies with line voltage as modified by the PFC. So let's eliminate it from the discussion.
Next, a forward, or any buck-derived converter, usually has fairly good load regulation even when open loop. The DC transfer function to the first order is simple Vout = n*D*Vin, where D is the duty ratio of the switch and n is the transformer turns-ratio and load does not enter into it.
Now we are to the heart of a very common problem. We can get a semi-regulated voltage for Vcc just by adding a winding, rectifier, filter to the main power transformer (or even an auxiliary winding on the inductor). This works fine after the converter is up and running, but where does the power come during start-up? Usually from a special startup circuit. If you just brute-force it from the voltage across the input capacitor, you run into the problem you describe. If you design it to work at low-line voltage you get problems at the high-line voltage. Either the high-line voltage is high enough to damage parts or the power loss either damages parts or raises havoc with efficiency.
The solution usually results in a non-trivial startup circuit that gets power from the input during the first several milliseconds of startup and then gets power from the semi-regulated winding discussed earlier when converter operation starts and voltages come up. When Vcc starts getting its semi-regulated power within acceptable limits, the original starting circuit is turned-off or disabled so it no longer draws power. There are dozens of ways to do this and you can examine a few application notes or reference designs to get some ideas. It does not have to be a forward converter. All converters have this problem and need a solution, either in terms of parts selection or circuitry.
Posted by Jerrold Foutz at August 22, 2001 03:06 PM