Using chokes

[2muchstuff]

With everybody worried about filter caps what about the use of chokes/ inductors. Am I correct in thinking that a filter choke stores up current in a magnetic field and will release it when there is a need for it as a load is applied.

Since I have yet to find a big enough torroid choke, is it possible to use a steel core from a variac as a choke form.

BTW- The power supply that I'm building is 48 voltDC @ 20 amps feeding Gecko's.

[Al_The_Man]

There was a time when power supplies were designed with series chokes, connected with two filter capacitors in the form of a low pass or Pi filter, you have one capacitor immediatly after the bridge followed by a series choke with the second filter capacitor connected, the result is very low ripple without limiting the current, the choke offers a high impedance to the ripple content but will pass the DC without limiting. The lower value of capacitance required over just a large electrolytic filter alone, places less demands on the transformer. The choke core generally has to be quite large due to the DC current present.
Al.

[2muchstuff]

Where can I find a formula to calculate the inductance using core and wire size along with number of wire turns.

[Al_The_Man]

If you are looking to implement a Pi filter for your DC supply, it might end up being an academic exercise, as to design it is no mean feat and may not only not be necessary in your case but impractical due to the core size of choke required with the effect of the high DC current affecting the impedance.
The practice was one commonly used in high- voltage low-current supplies, where the physical choke size was not an issue, but has mostly fallen out of favour due to this reason.
Where it was desirable to minimize the ripple content was in high powered Audio amps where the ripple would cause audible hum etc.
It is used in switching supplies, but with high frequencies and toroid construction, makes it more practical.
For the typical supplies used in servo applications the common capacitor smoothing alone is usually sufficient.
If you wish to pursue this further, you could search for Pi filter, LC low pass filter or tuned circuit designs.
Al.

[gar]

050625-1112 EST USA

2muchstuff:

In your other thread I referenced Signal Transformer.

In an equivalent circuit for a transformer one would reflect everything to the primary or secondary depending upon your interest. If I neglect resistance and capacitance, then the circuit looks like a Pi network of three inductors. One inductor is the shunt inductance across the input, the second is a series inductor between input and output (known as the leakage inductance), and the third is the shunt inductance across the output.

On my test transformer, the Signal A41-175-24, I ran two measurements. The first was with the secondary open circuit (low voltage side open), then the second was with the secondary shorted. The two secondaries were in series for the 24 V output selection. The primary was connected for 120. I measured the inductance and Q using a General Radio 1650-A bridge at 1 kHz.

Viewed from the primary the open circuit measurement was 0.642 Henrys and Q=2 (high loss). The closed circuit test resulted in 0.017 Henry and Q=11.

The turns ratio of this transformer is about 4.41 to 1. The square of the turns ratio is 19.47. If we reflect the leakage inductance to the secondary. In otherwords view the leakage inductance from the secondary. We get 0.017/19.47 or 0.87 millHenrys. This is an E I type of lamination. A toroid of equal VA rating would have a lower leakage inductance.

In the old radio days of vacuum tubes power supplies were built with full wave center tapped circuits and choke input filters. Sometimes the choke was the coil that supplied the speaker magnetic field. It was not practical to build bridge rectifiers. Thus the choke had to be in the DC path.

With bridge rectifiers you can put a choke on either the AC side or the DC side of the bridge rectifier. On the AC side you eliminate DC core saturation. The leakage inductance provides series inductance on the AC side of the bridge rectifier.

The series inductance in the circuit along with series resistance controls the peak transformer, diode, and capacitor charging current. If you want to use large output capacitors, and want to limit peak steady state current, then add inductance on the AC side of the bridge rectifier. Also do not bother with a toroid transformer. Keep in mind that the 175 VA Signal transformer I tested was about 95% efficient.

To experiment you need some way to measure RMS current, and a scope would be useful. A Simpson 260/270, or a Fluke 27 does not measure AC RMS values. A Fluke 87III measures RMS if the peaking is not too great. A hot-wire ammeter would work. You can also mathematically process the scope wave form to get the RMS value.

You do not need an inductor to create a continuous current on the AC side, but one sufficient to avoid having to oversize the transformer. This gets back to using the criteria in the Signal catalog and no external inductor.

.