By using an isolation transformer-variac-full wave bridge rectifier and a capacitor, I have built a AC to DC power supply. Now, I would like to regulate the output. My desire is to get a 150V DC constant output. For that I decided to use this LM317HV and want to apply the configuration given in the link. http://www.ti.com/lit/an/snva583/snva583.pdf

Can I directly connect this to the output of an Isolation Transformer-Variac-Bridge Rectifier-Capacitor group as shown in the image? Or, do I need something connected in between?

Also, should I use the negative output of the rectifier as a ground connections of the regulator? Or, by leaving the negative output unconnected, should I use the earth ground as shown?

Thanks in advance!

enter image description here

  • \$\begingroup\$ NSD134 is a 10W transistor. LM317 can dissipate more heat itself. \$\endgroup\$ Jan 12, 2017 at 4:58
  • \$\begingroup\$ If you want constant 150 V DC, you don't need a transformer with variac. A single secondary winding without variation will do, but you need some reserves for an undervoltage of the 120 V AC. The combination of a 1:1 transformer and a variac as shown in your schematic would not work with 10 % undervoltage, it might work for 5 %, but you should calculate the voltage loss at the bridge and the minimum input voltage necessary for the regulator to produce a stable 150 V DC output. \$\endgroup\$
    – Uwe
    Jan 12, 2017 at 10:36

1 Answer 1


There are a few things that would be good to add.

On the left, add some parts for safety. You need a fuse. Learn about type X and Y capacitors, and a common mode choke. These are sold all packaged together as a 'line entry filter'. These parts protect the circuit from high-voltage spikes that are common in ordinary household power. Transients of 10kV are common and 30kV is ordinary. You don't notice them, because your other devices have the necessary protection circuitry. These circuits also help prevent radio interference.

A snubber circuit on the secondary of the transformer can help. This is covered nicely in The Art of Electronics Third Edition, and less nicely in the Wikipedia article.

The 20K pot is going to use over 1 Watt. It needs to be full-size.

The 25mA current is all that the circuit can support, but the current limiting of the LM117H is much higher than that. Not sure what you want to have happen if there is too much current. Perhaps the thermal shutdown of the LM117H will protect the transistors? I don't know much about the NSD134, and you will probably need to substitute something that is more available. Perhaps R3 will act as a fuse.

I would add a 1uF Mylar 630V capacitor in parallel with the 1000uF electrolytic. This will help prevent RFI from the diode bridge getting into the regulator. Something like the Kemet R463N410000N2M is sufficient.

Here is my boilerplate advice for working with high voltage: High voltages can be lethal. Take all safety precautions. Wear non-conductive shoes. Wear safety glasses. Probe with your left hand in your back pocket. Don't work alone. Make sure someone with you knows CPR. Know where the nearest fire extinguisher is.

  • \$\begingroup\$ Thanks Tom. I wired the regulator as shown above and I get stable DC voltages depending on the 20k resistor's value. I will definitely add the safety equipment that you have recommended in your first paragraph, I believe that you are referring to the primary side of the isolation transformer. Ok, I will read about X and Y capacitors, and a common mode choke. For the snubber section, I have the book, so I will be reading those. Also, thank you for the other comments and precious warnings about the safety!!! \$\endgroup\$
    – CMarine
    Jan 12, 2017 at 6:38
  • \$\begingroup\$ I have seen a fuse before and one after the transformer at power supplies. Two fuses are not necessary, but they protect against failures in the transformer as well as failures in the volatge regulator or the load. \$\endgroup\$
    – Uwe
    Jan 12, 2017 at 10:22
  • \$\begingroup\$ Tom, actually I think, I have kind of tested the thermal shut down case. If I connect a resistor that wants to draw a current higher than it can supply, the output voltage swings around 31-32 Volts. However, if I use a resistor with higher resistance. I get a nice stable high voltage output. 140 volts in my case. Is there any reason it gives lower voltages? Is that a thermal shut down thing? \$\endgroup\$
    – CMarine
    Jan 13, 2017 at 6:48
  • \$\begingroup\$ In order to figure out what is going on, I would use LTspice to simulate the circuit, and see if I could get the measured voltages in the circuit to match the simulation. This applies to both the over-current and the normal setup. I would use a DC voltage source and only simulate the circuitry to the right of the 1000uF capacitor. The Youtube "Tom Anderson Simulation Series" parts 1, 2, 3, 9, and 10 cover a lot of what you need. \$\endgroup\$ Jan 14, 2017 at 18:59
  • \$\begingroup\$ Also, yes to the line entry filter going on the left side of the transformer. \$\endgroup\$ Jan 14, 2017 at 19:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.