Hello,
Thanks in advance! I Have a very basic LM317 adjustable power supply. Getting some odd readings.
Here is what I have:

28V Transformer. Altronics T2885 ---> 10A Bridge rectifier. --->DIY LM317. (27.3V reading Max.)

My DIY supply is built to the T.I. schematic. w/o capacitors. All is working what however next I decided to add a capacitor. For whatever reason I chose To add it to the input of the supply, coming off the bridge rectifier. Cap is: 3300uF 50V. Huge cap for this position in the circuit. Voltage out went from 27 as the high to 43! everything seems to act ok with the extra voltage. I just want to understand how that happened.

 

In the diagram below, the dotted waveform is what you would be getting without the capacitor. Your meter would read the average voltage of this waveform. When you add the capacitor, you can see that the output voltage is held at the peak voltage by the capacitor so now your meter will be reading the peak voltage.

Note that the maximum voltage between input and output of the LM317 is just 37V. It will not be happy with 43V across it. That means that your minimum output voltage is 6V and you must be very careful not to short circuit the output.

 

I get it! Thank you so much. Is the resultant output waveform what we call half wave?

 

Well it was my impression that the voltage coming out of the rectifier was flat?

Why would you think that? It's rectified AC and there's always going to be ripple.

You told us about the filter cap, but nothing about the load. A schematic is a better way to convey circuit details...

I was already aware that im over on the voltage. However it is not showing any sign of being unhappy.

A good design will derate components from their maximums; and would *never* exceed them.

Also one of the first things I did was short it. To see how it would handle it. 10 seconds or so. maybe 4 times.

Fortunately for you, the regulator has safe area protection built-in... When it becomes "unhappy", it will cease to regulate and the load may also become "unhappy".

 

I get it! Thank you so much. Is the resultant output waveform what we call half wave?

Nope, that's a full wave. A half wave would have every other peak removed, since only the positive-going peaks make it through a half-wave rectifier. The full-wave shown passes the positive peaks and 'inverts' the negative peaks so that they come through also.

One thing to note regarding your confusion is that your meter was reporting an average 27V voltage to you, and nothing about the waveform. If you had viewed it on an oscilloscope, you would have seen that your steady 27V was in fact a series of waves going from 0V to the peak voltage, ~43V. Adding the big capacitor locks in the peak voltage and with no load there will be no ripple. Increasing the load increases the drop from the peak during the discharge period and gives a ripple. So your meter will show a decreasing voltage as the load increases. Under very heavy load, the cap discharges completely and the ripple is once again full-wave. In that case the capacitor is just adding to the load and not really smoothing anything.

 

Without the cap to hold it up the voltage coming out of the FW rectfier is a series of half sine wave pulses as the figure above shows. The cap helps to smooth these out. Helps because the cap is only charged at the very peak of the sine wave, and will discharge until the next peak.

The LM317 is limited to 40 volts between input and output, which is why it is working with over 40 volts input. BUT... when you short the output the full input voltage is across the device. Now to meet that 40 volt spec the device can handle more than 40 volts, but how much more is a mystery only found by destructive testing.

So stop shorting the output until you reduce the input voltage.

If you can prevent shorts or overloads then this device will work as a prototype, but may soon fail otherwise. Be prepared for that.

 

I did read all the comments. Points taken. Much to consider. Thank you all so much. Good help.