ECN Electrical Forum - Discussion Forums for Electricians, Inspectors and Related Professionals
ECN Shout Chat
ShoutChat
Recent Posts
Increasing demand factors in residential
by gfretwell - 03/28/24 12:43 AM
Portable generator question
by Steve Miller - 03/19/24 08:50 PM
Do we need grounding?
by NORCAL - 03/19/24 05:11 PM
240V only in a home and NEC?
by dsk - 03/19/24 06:33 AM
Cordless Tools: The Obvious Question
by renosteinke - 03/14/24 08:05 PM
New in the Gallery:
This is a new one
This is a new one
by timmp, September 24
Few pics I found
Few pics I found
by timmp, August 15
Who's Online Now
1 members (Scott35), 266 guests, and 14 robots.
Key: Admin, Global Mod, Mod
Previous Thread
Next Thread
Print Thread
Rate Thread
Page 1 of 3 1 2 3
Joined: Feb 2013
Posts: 264
Potseal Offline OP
Member
I have an old 60's to 70's vintage SCR variable speed drive that I thought would make a good DC power supply for powering simple circuits on a breadboard. I had a simple circuit made up that included a relay taken from salvaged electronic equipment. Once power was applied to the circuit the relay began to chatter. I took the relay out of the circuit and tested it with a 9V battery. It "clicked" indicating proper operation. As well, I verified that the contacts opened and closed with a DMM.

I tried the relay with the SCR VSD again this time out of the circuit. Once the power was applied it chattered like before. I tried a different relay and it did the same. I tried varying the voltage ever so slightly +/- and it made no difference. I then tried using a different straight 12VDC supply. Both relays functioned properly. So why does the SCR VSD cause these relays to chatter? I assume it has to do with the design. The previous owner was a knife maker who used this SCR VSD for DC motor speed control.

Is the DC power not "clean" enough for my application? Would a bridge rectifier and capacitor make a difference? Any suggestions would be appreciated.


A malfunction at the junction
--------------------------------------
Dwayne
Stay up to Code with the Latest NEC:


>> 2023 NEC & Related Reference & Exam Prep
2023 NEC & Related Reference & Study Guides

Pass Your Exam the FIRST TIME with the Latest NEC & Exam Prep

>> 2020 NEC & Related Reference & Study Guides
 

Joined: Jul 2004
Posts: 9,923
Likes: 32
G
Member
The answer comes from a scope but I would start with a capacitor across the DC output. (hundreds if not thousands of MFD)


Greg Fretwell
Joined: Jun 2004
Posts: 1,273
T
Member
Scrap all of those ideas.

FREE switching power supplies are being thrown into the junk yard across the globe: ATX PSU from ancient PCs.

You can go to YouTube for any number of videos posted up by hobbyists.

All ATX systems are sophisticated enough to give you sweet DC voltages with trivial ripple.

+3.3 VDC
+5.0 VDC
+12. VDC

Even negative voltages:

-5.0 VDC
-12. VDC

And these boxes are being placed at the curb every day of the week -- everywhere -- FREE.

Just troll through CraigsList or call around for 'e-waste' or some such.

Because of the basic (DC) voltages provided, it's possible to straddle them to derive difference voltages:

+12. VDC minus +5.0 VDC = +7.0 VDC ===> and so forth.

The provided voltages are EXACTLY those required by hobbyists for all digital circuits.

Because of industry standards, these PSU are also surprisingly efficient -- WAY superior to ancient solid state junk. There's absolutely no comparison.

AND, these PSU are much safer to toy with. They're totally enclosed and have their own fan and fuses.

You can't beat them with a stick!


Tesla
Joined: Nov 2005
Posts: 827
Likes: 1
J
Member
I agree with Tesla on this one. I have one computer PS with 4 lighter sockets fitted for powering my RC chargers. My newer one was $4 at MPJA. I snipped the connectors off and lugged the conductors onto 5 way binding posts mounted to a piece of folded lexan. I also mounted a toggle switch, mini DVM, 2 LEDs, and a boost mode power supply. This gives me a 5VSB low current supply and an LED that are on as long as the PS is plugged in. Flipping the toggle lights up the second LED and turns on 3.3, 5, and 12 volt high current positive supplies, and low current -5, and -12 volt negative supplies. Connecting the boost regulator to the 3.3 volt output, and the mini DVM, gave me a variable 4-30 volt output with display. It gives me just about any DC supply that I need. I'll admit that the dual binding posts cost much more than the core PS cost, but I wanted the convenience.
Your SCR supply will definitely need tons of filter capacitance, to be of any practical use. The ripple amplitude will be higher at lower average voltages. There are a couple clever circuits out there that put diode/SCR bridges in front of linear 3 terminal regulators. They phase control the SCRs to provide a voltage to the regulator, just above the input/output differential requirement. That way, the regulator power dissipation is minimized for a given current.
Joe

Joined: Jul 2004
Posts: 9,923
Likes: 32
G
Member
I have one on my bench but one problem is on some of the supplies a fast change in the load like hooking up something big will trip them out.
If you restart it with the load connected it is OK again

These supplies do need a start wire grounded to get them going. On the old AT types (2 flat board connectors)it is hooking the red and black together in that 3 pin plug (the cable with a white too)
On the ATX with the rectangular 2 row board connector it is the green wire, usually #14, unless it is a Dell. They use a different pinout but it is still the green wire


Greg Fretwell
Joined: Jun 2004
Posts: 1,273
T
Member
Some brilliant ATX need a dummy load to stabilize.

So I've cobbled up 2 automobile lamps -- soldered -- to a female Molex connector.

(12V and 5V)

That's enough to get the smartest PSUs to fire up all rails.

RAIL = DC VOLTAGE VALUE

Red = + 5.0 VDC
Orange = + 3.3 VDC
Yellow = + 12. VDC

Black = 0 VDC -- the return conductor

The other whips constitute either legacy negative rails -- not used on any daughter card in the last quarter-century -- or are essential for smart switching of the PSU by the MSFT Windows program.


Purple = + 5 VDC -- hot at all times limited amps: Standby
Green = + 5 VDC (fed by purple rail) during PC - 'off'

It's fed across a (stiff) resistor... and swings HIGH when its path to ground is broken -- the 'off' state for the PC.

Should you physically short this conductor to ground (any black) its new Boolean value ("0") triggers the PSU to spool up to full output -- specifically the +12 VDC and + 3.3 VDC rails. { with the smarter PSUs insisting on a load from the + 12 VDC rail -- ie CD-ROM drive/ HDD/ FDD. (If all three are missing then the shorted signal is a false one, and the PSU STILL won't fire up. This feature is not a universal one.)

Brown + 3.3 VDC feedback detection circuit. This puppy is typically snuck into the Molex connector. It's a totally 'unloaded' run of +3.3 VDC power that permits the latest brainy motherboards to evaluate and adjust the voltage drop hitting the memory and CPU rails. Beyond a certain threshold the voltage gap -- at the motherboard end between this conductor and the orange rail (the power run at +3.3 VDC ) will cause the CPU to instantly shut down -- without any notice at all. This is to preserve the 'machine state' -- data quality -- so that the PC does not run-on with flakey memory values. The brown conductor will appear to be seriously undersized. It's carrying virtually no actual current. Think of it as a perpetual DMM running in the backround.

Gray = Actually an analog-Boolean time-delay value. This, like brown, 'talks' to the CPU and memories. It's a delay line in their Boolean start up logic. It stops them from even beginning to load digital information until the +3.3 VDC 'rail' is steady and sweet. The designers could've simply put a time-delay sequence on the motherboard. Instead they've elected to force the PSU to tell the motherboard that enough time has elapsed -- AND -- verify that the + 3.3 VDC is sweet and stable; all of the capacitors are saturated with a DC potential.

Yet later designs have dropped much of the above -- and deliver +12 VDC only. This drastically simplifies the PSU -- as the CPU voltage is provided by a DC to DC power supply right next to the CPU on the board.

It's this element that has all of the big caps soldered in parallel from the Pentium 4 and onwards... which is another story.


Tesla
Joined: Jul 2004
Posts: 9,923
Likes: 32
G
Member
The -12 is used by the serial port. We don't hook modems to them very much these days but they use a serial port for a UPS.
The -5 was used for memory in some old XT era systems.


Greg Fretwell
Joined: Feb 2013
Posts: 264
Potseal Offline OP
Member
Thank-you for the replies.

Basically you guys are suggesting I move on from the old SCR - alright.



A malfunction at the junction
--------------------------------------
Dwayne
Joined: Feb 2013
Posts: 264
Potseal Offline OP
Member
Well, I did as Tesla suggested and grabbed a few PC PSU's.

Then I started looking at some projects online and found someone's design where they made a variable DC voltage power supply using a PSU. I took that initial circuit and added a couple modifications and ended up with a Frankenstein's Monster version of a +12 VDC and +0-9VDC power supply. Behold...

[Linked Image from i61.tinypic.com]

[Linked Image from i62.tinypic.com]


A malfunction at the junction
--------------------------------------
Dwayne
Joined: Feb 2013
Posts: 264
Potseal Offline OP
Member
Incase you were wondering that PSU is bolted directly to a gutted RCA ohms/volt analog meter turned on it's side. I replaced the original gauge with a 1-5 DC amp gauge. Probably fairly useless but it fit the existing hole.

The lay-out for the switches, ind. lights and outputs was mostly determined by existing holes. I tried to build this device as economically as possible and the looks speak for itself. As far as performance, it does exactly what it is was intended to do.

Last edited by Potseal; 04/03/15 06:25 AM.

A malfunction at the junction
--------------------------------------
Dwayne
Page 1 of 3 1 2 3

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5