ECN Forum Forums Technical Discussion Electrical Theory and Applications SCR VSD causing relay to chatter

The following shows how I put the whole thing together. Please note that I used a 2.5 Kohm pot instead of the recommended 2 Kohm pot in the original variable voltage circuit (http://www.electronics-tutorials.ws/blog/variable-voltage-power-supply.html). Otherwise I followed it to the letter. I showed all the essential parts and labelled the conductors from the Molex connector that were pertinent to this diagram. There is one other yellow conductor that I used from a different connector to feed the relay coil.

http://www.electronics-tutorials.ws/blog/variable-voltage-power-supply.html

The following shows how I put the whole thing together. Please note that I used a 2.5 Kohm pot instead of the recommended 2 Kohm pot in the original variable voltage circuit (http://www.electronics-tutorials.ws/blog/variable-voltage-power-supply.html). Otherwise I followed it to the letter. I showed all the essential parts and labelled the conductors from the Molex connector that were pertinent to this diagram. There is one other yellow conductor that I used from a different connector to feed the relay coil.

OK, my thoughts:
1.) You already have jacks for +12V and 1.25-9V. Why give up one to have the other? If you are 5A capable, most bread boarding doesn't require that much current from all of the supplies combined. So you could run your +12 to the jack and an LM350T circuit, and just watch your combined output. Just change your variable output fuse to a 5A too.
2.) You didn't label your current setting resistor. If it's 240 ohms, you would be wasting most of your pot. A 390 ohm resistor would cause the supply to max out at around 9.25V.
3.)Keep the bodies of the resistors as close to the adjustment pin as possible since that is a very high impedance input. Twist leads going to the pot. If you add the 10uF cap for better ripple rejection, also add the extra diode across the 390 ohm resistor, cathode up. 1N4001, 1N914, 1N4148A would all be fine.
4.) The best place to learn how a chip works is usually straight from the mfg.
http://www.onsemi.com/pub_link/Collateral/LM350-D.PDF
OnSemi seems to have about the most complete datasheet for the LM350 these days. Page 2 is telling us that the reference voltage is typically 1.25V, adj. current 50uA, that I need to have around a 3.5mA load to regulate, and that I might be able to get 4.5A out of it. Page 5 shows how you can get more than 3A out of it(Fig.8), and how much more input voltage you need for different currents.(Fig.10) Figures 13 and 14 on Page 6, show how the 10uF helps. Pages 7 and 8 show useful applications.
Joe

Today I had a few minutes and tried a little test by placing a 4 ohm resistor across the output for the variable voltage. Measured the variable voltage setting at 5 volts with a DMM before connecting it to the resistor. I expected to see the analog ammeter read about 1 amp but it didn't move. Then I measured the voltage drop across the resistor and it was less than half a volt.

Looks like I have a problem

I liken this to a weak battery that without load reads a proper voltage but as soon as a load is applied the voltage drops to almost nothing.

I placed a 10 ohm resistor across the outputs of the + 12V output (I haven't made all the changes that Joe suggested, yet) and sure enough the ammeter showed over 1 amp of DC currrent.

Therefore something isn't right with my variable voltage circuit. Capacitors? Voltage regulator? Any help would be appreciated.

What is the input voltage to the 317 with it loaded?

I did an update of the circuit since I forgot to include the 1 Kohm resistor in series with the LED. Plus I tried to label every component properly:

Time permitting I will verify that after work today.

Have you checked your relay contact? What kind of heat sink do you have the LM317T mounted to? The chip has internal thermal limiting and will dial back rather than self destruct.
Joe

Well, I finally got it working.

Originally it was a minor mix-up that caused the problem. A somewhat embarrassing mistake.

Since I had to make some changes I went all out and revamped the entire circuit. I took Joe's suggestions and changed some of the components. As well, I simplified and streamlined the circuit board wiring and component placement and cut the space needed in half. Lastly, I made a separate compartment out of a repurposed AC Adaptor case to hold the output fuses on the outside of the device for easier access if they blow.

All in all my dumb mistake ended up being the motivation for an improved design.

Thanks to all who helped with good advice. I appreciate it.

The next upgrade will hopefully include a digital DC voltmeter

That's great! MPJA has a really nice red LED snap-in DPM for $3 in their Email special.
http://www.mpja.com/specialsview.asp?r=ref&s=2
More cheap fun in a boost converter for $5:
http://www.mpja.com/XL6009-Adj-5-35V-DC_DC-Boost-Converter/productinfo/31397%20PS/
Boost w/meter for $8:
http://www.mpja.com/Boost-Converter-Adj-4-35V-with-Meter/productinfo/32051%20PS/
Boost w/meter and heat sink for $13:
http://www.mpja.com/Adj-DC_DC-Boost-Converter-5-30VDC-50W/productinfo/31936%20PS/
Joe