5 Volt Power Supply Tripping

[murphy625]

The supply is a Condor...
Model HC5-6/OVP-A+
Input 120/240
0.9/0.5 amps
Output 5VDC @ 6amps

Here is my problem.
My 5volt dc supply is tripping out. Cycling the power to it restores it working immediatly.
It is now plugged into a battery backup unit with the computer. (its a small batter backup but it works).

The 5 volt supply is still tripping. It only seems to do it when the 480v 3ph motor shuts off after hitting a 24volt limit switch. (powered by the 24vdc Condor supply that works just fine)

The 5vdc supply will be on for HOURS, doing nothing really.. If I go around the machine and flip limit switches by hand, the OPTO relays turn on and off just fine. (they are powered by the 5volt supply)
But when the 480volt motor is operating, and gravity is working against it, the 5 volt supply will trip the second the mechanical movement hits the upper limit switch.
It does not trip when it hits the lower switch.
The mechanical action hits the upper switch, the motor shuts off (as the computer tells it to) and my 5volt supply trips out all at the same time.

The only thing the 5volt supply is hooked up to is the Opto22 relay boards and my ICM2900. There are 2 boards.
A G4PB32DEC and a PB16T and of course it powers the ICM2900.

When the supply is running, it is cool to the touch.. when it trips, it gets warm and continues to get warmer the longer I let it go before cycling the power on and off. Once I cycle the power, it turns back on immediatly and begins to cool off.
A multimeter says its pulling less than 200 milliamps when it blows.

Any ideas?
Thanks,
Murphy

[murphy625]

UPDATE,

I changed the power source from the 5volt supply. Instead of running off the same 220v 1ph circuit that the rotary phase converter is running off, I used an extension cord and tapped into a completely different power source. When I say different, I mean the power source (220vac 1ph) is coming from a different pole outside the building. Different transformer on the poll feeding a different home.
I dont think spikes from the rotary phase converter are causing this anymore.

Any ideas?

[CJL5585]

Model HC5-6/OVP-A+

The OVP in the part # indicates that the supply has OVERVOLTAGE PROTECTION.
My guess is that when the limit switch is made that the 5V supply is seeing a voltage spike causing the supply to shut down, or the supply is going into OVERCURRENT Limit.

Verify that the limit switch is not somehow shorting the supply causing a DEAD short. I believe your problem is related to OVERCURRENT. This is due to the fact that the supply heats up. Any type of heating would be related to a Short or very low resistance voltage path.
Jerry

[Al_The_Man]

I was about to mention that it appears you need a line suppressor as made by Aerovox these have a LC network to trap some of the ac line spikes. It is a good idea to use one of these regardless on any CNC control. your power supply shows OVP 'over-voltage protection' due to a spike and I would guess that this is kicking in and by cycling the power you are re-setting it.
Al

[murphy625]

The limit switch is powered by the 24volt dc supply.. The 24vdc is almost identical to the 5volt. Same manufacturer and it looks almost the same but larger.
The 24vdc has never given me a single glitch.

What side of the power supply are you suggesting the spike is coming from? The 110vac in of the 5vdc supply has a transformer on it. (spike goes threw transformer?)
And further, the supply is plugged into a different house. (neighbor who is on different transformer). Could the spike be coming from the 5volt (output) side?

From testing, I have gathered that it trips out and it doesn't matter if I plug into the same supply as the rotary phase converter, or a battery backup unit, or into the neighbors house.
Could a spike be coming from the other side? An "over voltage" on the 5volt side of the supply?

Here is something strange,
I have a G4PB32DEC card that holds 32 Opto22 G4 optically isolated relays. This board has a cable that runs from it to the back of my computer into a Contec Aux. IO board. On the bottom of this G4PB32DEC, there is a socket to test the G4 fuses. I have a fuse plugged into it and the little LED lights up indicating the fuse is good.
Now the strange part.
When the 5 volt supply is OFF, the fuse LED light is on, but very dim. When I turn the 5volt supply on, the LED light gets bright.
Is this normal?

Thanks again..

[murphy625]

Ok.. Im digging in again and found some weird stuff.

Using a multimeter and doing some checks here is what I get.

5 volt supply positive to earth ground = 5.3 vdc.
5 volt supply positive to 5 volt supply negative = 5.3 vdc.
5 volt supply negative to earth ground = 0.01 vdc (nothing)


24 volt supply positive to earth ground = ~2 vdc. and dropping.
24 volt supply negative to earth ground = 0.6vdc and dropping
24 volt supply positive to 24volt supply negative = 24.2 vdc. and steady.

I am getting the feeling I have something hooked up wrong, or should have run a line to ground when I didnt.

Also, with the 480 volt power off, my ICM2900 shows 0.0 volts across the gnd and McmdX terminals. When I turn the 480 volts on, it shows 2.49 volts.

When I check the 0 volts terminal on the servo drive to ground it is 0 volts with or without the 480 turned on.

I have a ground loop here????? Something is a fishy...

[Al_The_Man]

OK, To start with I make a point of bringing any AC supply ground to a common ground point, to this I also ground (make common) any power supply common, whether it be PC power supply, separate DC supplies, i.e. 5v, 24vdc etc. plus all shielded cable shields. Your opto board, wether it be 5v logic or 24vdc should also have the common grounded, I think the problem you are having due to glowing LED is due to floating common.
Going back to your OVP power supply, most of these type of supplies have a 'crowbar over-voltage' sensor, this detects a overvoltage and immediatly shunts the DC which causes the DC to collapse by imposing high load on the DC which can remain untill reset. ( some are auto re-settable). This would explain your temperature climb on the supply.
BTW transformers do have an inherent suppression to spikes, but there are always ones of a low frequency nature will get through.
Al

[murphy625]

Al,
That's it...

Are you suggesting that I send the negative side of the 5volt supply and the 24 volt supply to ground (brown wire)??? I thought they were supposed to be isolated.. Am I to understand that I am WRONG AGAIN ??? (boy if I had a penny for every time) LOL

All grounds be it AC, DC, or Filtered power supplies should go to the same common EARTH GROUND ???

I feel like an idiot. I've been wiring machine tools for 15 years but this is the first time I have used DC supplies and filtered AC Transformer supplies for computer controlled equipment. Whenever I wired a control transformer, they would always be 480/220 primary (wired for 480) and have 110vac secondarys which I would run one side of the secondary straight to earth ground and fuse the otherside for the control voltage.

Thanks for all your help Al, (and everyone else too!!)

[CJL5585]

The dim LED indicates a ground loop. Current is flowing in the circuit with the power off on the supply.

As I said in a previous post, you have a probable short circuit or a low resistance from +5V to the 5V common (Gnd) when the one limit switch makes. The problem might be on the relay board, the wiring to the relay, or the wiring on the switch itself.

My suggestion: Unplug any relay which is energized by the limit switch and see if you have the same problem with the supply going offline. If not, the problem is in the relay board wiring. If the problem is still there, check limit switch configuration and wiring.

Jerry

[Evodyne]

Murphy,

Al knows what he is talking about. You may have other problems, but a poorly connected system is an invitation for headaches. If, as Al had suggested, you tied your commons/grounds together, then your entire system has a single, common point of reference. As it stands now, yes, you measure 5v and 24 across you respective supplys-but try measuring from the 24 to the 5. Not what you expected, right?

Without a common ground or earth the different supplies can "float" at unknown voltages relative to each other. And, like another member meantioned, this difference in potential can cause a current to flow. A current of this type flowing through a ground lead between devices is known as a ground loop.

The best practice is to use correctly sized conductors, good connections, and to bring all the separate grounds to a single point. Google ground loops and you'll find plenty of information on how to protect yourself.

Good luck!

[Al_The_Man]

Originally Posted by murphy625 Are you suggesting that I send the negative side of the 5volt supply and the 24 volt supply to ground (brown wire)??? All grounds be it AC, DC, or Filtered power supplies should go to the same common EARTH GROUND ???

Yes, even if you have a 120vac supply transformer for control supply etc, the 'neutral' should be at the same ground point.
I did not come up with this scenario, This has come about over time when initially 120vac relay control was mated with low voltage logic, as in early CNC using noisy SCR drives and AC relays, when this noisy environment was wedded with 5vdc logic.
The practice to bond all commons, shields and grounds has evolved due to the neccessity to eliminate the noise problem.
I have seen a few systems where all the supply commons have been kept separate and personnally have seen the problems that can occur.
Just to qualify things, you have to establish that the supply common is capable of being grounded ( most are). If you have a home built supply where the ac secondary of the transformer is grounded then you cannot ground the DC supply derived from it.
Al

[murphy625]

EVODYNE,
Al obviously knows what he is talking about. Seems to me he is the CNC god here...
I no way shape or form was I questioning his advice.. Just making sure I understood it correcty!!!

Yes.. I am now almost 100% sure he is correct. I am re-wiring grounds, and sheilds to meet at the same copper ground bar. It just seems "off" to me that I am running white, brown and green wires to the same place. I guess it will seeem normal once I get more experience in mixing AC and DC stuff in the same box.


Al (or should I call you CNC_GOD from now on?? )
You said something about a home built supply and not being able to run dc grounds to it. What do you mean by that? and Why?
None of my supplies are home built,,,, just home installed..

I hooked up all the grounds to a common point with jumpers. The 2.47 volts that I was seeing is now gone. All earth grounds, transformer neutrals, and DC Grounds are now tied to the same copper bar.

The little LED light on the G4PB32DEC card is still doing the same thing. Dim when supply is off, bright when its on.

Could that little light be pulling power from the computer board?

Thanks to everyone...