Re: Power supply transformer

You are doing very well. A couple of minor comments are:

It is OK to use a transformer on lower input voltage as you are doing but not on higher line voltage.

You will lose a couple of volts across your bridge rectifier.

What you then have calculated, is the the no-load voltage output with the assumption that the RMS input voltage to the bridge is
correct.

When there is a significant load on the transformer, the voltage will drop because of the winding losses and it will be less
than the no-load voltage. Further, between the charging peaks, the voltage will drop even more and you will get "ripple" which
is the difference between the peak charging voltage and the lowest voltage during the off cycle. The bigger the capacitor, the
less the ripple.

As you see, there are now three voltages to think about:
1. No load voltage
2. Peak voltage under max load
3. Min "valley" voltage under load

Sometimes an average voltage of the peak and valley voltages are used.

The no-load voltage at high line voltage (105%) determines the capacitor and semicondutor ratings with proper safety
multipliers.

If you are designing a regulated supply, the minimum voltage is critical since it determines when the regulator are going out of
regulation.

The minimum valley voltage at low line voltage (85%) determines the lowest output voltage available under load.

One last comment: A transformer listed as, for example, 120V - 24V normally is designed to provide an output voltage of 24 V at
rated load. Therefor the turn ratio will be adjust for that and its no-load voltage will be a little higher.

The drop in voltage with load is called regulation and varies from a few percent for very large transformers to 10 to 20 % for
small cheap ones.

Bertho Boman
Vinland Corporation
================================================


Ray Henry wrote:

> From: Ray Henry <rehenry@...>
>
> A lot of good comments have come out on power supplies so I'm trying to put
> one together.
>
> It occurred to me to question whether you really need a 110 or 220 input
> and 24 volt out. Seems to me that what I should be looking for is the
> ratio of input voltage to output voltage. As long as the high side of the
> transformer is rated higher than your AC voltage you should be able to feed
> it. What you get back should be a voltage equal to the ratio of the coils.
>
> So I went out to the shop and dug up some old machine tool control
> transformers that I liberated years ago from stuff headed for scrap. I
> found one with dual winding on the primary 220/440 with + or - 15% taps and
> with 110 on the secondary. It's rated for 650va. Correct my numbers here
> when I screw up.
>
> 1. Volts
>
> My power line is 122 most of the time. (yes my filament life is short)
> Feed line to the 440 side of this transformer and get 30.5 back. Peak
> voltage should yield 43 volts DC. (rms/.707) Add the 15% up and I should
> have a 49 volt supply.
>
> The surplus steppers and centent drivers that I liberated from some x ray
> equipment ran on a 48 volt supply.
>
> 2. Amps
>
> I get real iffy here! 650 va divided by 122 tells me that I should limit
> the AC side to 5.33 amps so I'll sugggest a 5 amp common fuse. (Someone
> mentioned that a transformer is rated for continuous duty so it should do
> more.) On the secondary side I should be able to draw four times that or
> 20 amps. (Isn't this limited by the wire size in the coil?)
>
> 3. Load
>
> Centent tells me that their driver uses 1/3 of motor rated amps for series
> connected motors and 2/3 for a parallel connected motor. Since each motor
> is rated for 4.5 amps I should be able to run at least six of these 150
> in/oz motors (18.09 amps) from this transformer. Wow six axis! I could
> run the table top nist hexapod.
>
> I plan to pile all this stuff on the basement table (I don't have a
> hexapod) and try it with my EMC software in the next couple of days so
> correct me where I'm wrong or the significant other will rush down the
> basement stair yelling, "I SMELL SMOKE!"
>
> Ray