Building Dual Power Supply In LTSpice, having troubles

Thread Starter

Evan Moore

Joined Dec 18, 2021
8
Hello!
I recently posted a thread about how I am building a dual power supply and was given a circuit, however there seems to be some issues with the design. First of all the LM317 is outputting a negative voltage, which I do not think it should do, as it is a positive voltage regulator. also it seems that while the LM317 output varies from the range of almost 0V to -30V, the LM337 output varies from +27.8V to +39V. this makes even less sense. Can someone please look at my circuit and tell me what I am doing wrong?
 

Attachments

Papabravo

Joined Feb 24, 2006
18,073
Your schematic is faulty for a couple of reasons
  1. You did not include the symbols and the subcircuits for the LM317 or the LM337.
  2. You do not have a coupling coefficient statement for the transformer. EDIT: Found it
  3. The default diode model may not produce expected results. Check the default model parameters, or use an actual diode.
  4. 8H and 1H are ridiculous values for transformer windings.
  5. The specification you want for North American line power is SINE(0 169.7 60). In LTspice we specify the peak value of an AC source, not the RMS value.
  6. You should avoid using the global ground symbol on both sides of the transformer.
This is what I see when I open your file:
1640192317436.png
My system cannot find a symbol named "LM317", and the LM337 symbol my system found is not compatible with your drawing.

The proper procedure for getting community help is to put EVERYTHING needed to run the simulation in a ".zip" file and post that file.

One more thing. Not having the component value next to the component is quite annoying. Do you really do schematics this way?
 
Last edited:

Papabravo

Joined Feb 24, 2006
18,073
Your asc file seems corrupted. Component values are missing. Component designators have become comments.
I know it looks like that, but what the TS did is move all of the component values away from the component to make a pseudo BOM list. It threw me when I first looked at too. Check your LTspice setting to make comment text a distinctly different color. See my screen clip in post #2. Component values and reference designators are BLACK. Comment Text is BLUE.
 
Last edited:

dl324

Joined Mar 30, 2015
14,306
Can someone please look at my circuit and tell me what I am doing wrong?
You shouldn't move the component values away from the components. LTspice schematics are usually difficult to read (usually not the program's fault), but moving the values as you've done makes it even more difficult.

My usual complaint with LTspice schematics is that there's too much "whitespace". In your case, there isn't enough.

As I said in your other thread, you'll learn more if you don't use a simulator until you understand the basics. LTspice doesn't always give correct results and will let you operate components at conditions that would destroy them in a real circuit. You need to know enough to recognize invalid results.
 
Last edited:

Papabravo

Joined Feb 24, 2006
18,073
One thing I do know is that you need to take extreme care with a center tapped transformer and a bridge regulator. It is quite easy to create a circuit that WILL let the magic smoke out with potentially lethal consequences. This is especially true if you have large electrolytic capacitors in metal cans.
 

Papabravo

Joined Feb 24, 2006
18,073
He wants to be able to adjust both supplies to 0V. I gave him the relevant circuitry in his other thread.
Yeah. OK, I didn't know there was another thread. I thought he gave up after your comments in post #5. I'll take a look.
When I opened the original schematic I couldn't find his symbols for the LM317 and LM337. My schematic might have the same problem since my symbols came from the library by @Bordodynov. I keep forgetting that minor detail.

On the plus side it shows how to parameterize a power transformer with an arbitrary turns ratio. If anybody is interested I can gather up the pieces parts to create a project folder with everything required to run the simulation without having the requisite libraries.
 
Last edited:
Top