Power supply capacitor

[itsme]

Hi there,

Well, I wired up my power supply components today (mounted in a PC case) and amazingly it all worked!!! Hooray!!!

The transformer says 25VAC on the box, but I'm getting about 28V. My original '35VDC' supply is now about 38V. The max my board can handle is 44V, so I think it should be fine...

The reason why I am posting here, is to find out what the deal is with the capacitor. I plug the supply in and it all works nicely, but once switched off off, the capacitor is obviously still holding a charge (don't ask how I found that out...). The instructions for my Xylotex board say DO NOT wire it up with the power on. Now, technically, the power is not on, but the cap is still holding a charge, so when I wire it up it could possibly damage the board. This may not be true, but I don't want to take any chances... Any suggestions? Should I discharge the capacitor with a resistor before attempting to connect the supply to my board??? The cap is 63V 22000uF and the only resistors that I have on hand are a 10k 2W resistor and a 470R 2W resistor.

Thanks
Warren

[zoltan]

For your benefit discharge the capacitor. You do not need a resistor, but you can use 470 ohms resistor. Without resistor it will be a small spark..nothing major.

Zoltan

[itsme]

Hi,

Thanks for the reply. I'm willing not to use a resistor - as long as it definitely won't hurt the supply (or anything else around it [like me...]).

When it suddenly occurred to me that it was storing charge (a blonde moment...) I accidentally touched the wires. There was a loudish 'electrical' pop and a spark. On closer inspection, I noticed that the copper wire strands had melted slightly...

I want to get these motors running now, but definitely mustn't rush. Besides, I've got 3 months of summer holiday to waste now, so there'll be lots of chip cutting and lots of my mom complaining about aluminium getting stuck in the carpets . hehe

Regards
Warren

[itsme]

Woohoo!!!

I just noticed that I'm an apprentice!!! Yippee!!!

[gar]

050628-1615 EST USA

itsme:

The time constant of a simple RC circuit is R times C. One time constant is the time for a 63% change in capacitor voltage. Take 0.37 to the 4 th power and you get .02 or 2%. Thus to reduce the the voltage on the capacitor from 38 V to 0.02 * 38 = 0.76 V will take 4 time constants.

Your 10 K resistor at 2 W is fine for power dissipation ( 38 squared/10,000 = 10 mw ). The time constant in seconds is 0.022 * 10,000 = 114 sec, multiply by 4 and we have 456 sec or 7.6 minutes to reduce the voltage to 2% of its initial value.

Your 470 at 2 watts dissipates 1444/470 = 3.07 watts. You would not want this permanently attached. A higher wattage resistor would be required. For example 5 or 10 watts. Here the time constant is 470/10,000 = 0.047 of the previous time constant or 0.36 min to discharge to the 2% level. This is a more reasonable fixed bleeder to leave on the power supply.

In all situations safety is the number one item to consider. Do not use a dead short to discharge your capacitors with this much stored energy.

For a faster discharge you might use 47 ohms and maybe a 5 watt resistor. Only connected for discharge because of the high power consumption and overloading of the resistor. Ohmite or similar wire-wound resistors are fairly tolerant of short time overload. In this case the nominal value is probably 50 ohms. Use a switch to contain the arc on closure, or use an SCR as the switch for no arc and trigger with a low current.

It is a good idea to have your power supply output capacitor discharged before any other circuits are connected. Also any time you turn power off for safety reasons.

.

[Imagineering]

As I have a Heart problem, I'm very considerate of ANY stray Voltage in ANY equipment I want to work on, so I've also considered this problem with my Power Supply.
I have a 16Volt 10Amp Output with 20,000uf of Capacitance also. What I have done is to connect a 7805 Voltage Regulator across the Capacitor which then feeds my 12V PSU Cooling Fan. When the power is switched off, the Fan winds down over a minute or so, thus discharging the Capacitor. Slight additional cooling is an added benefit.

(gar - I've chickened out of the 35Volt PSU and built a 16Volt 10Amp one instead as I already have most of the parts. Three 12Volt Transformers feeding Three Bridge Rectifiers and then common the DC lines to 20,000uf as above.)

[itsme]

Thanks for the replies.

I managed to discharge the capacitor using a resistor and all worked well.

I also plugged in a stepper motor last night for the first time and it worked!!! I was quite impressed with myself, because I assembled the control board and the power supply (first board I've ever assembled).


Today, I can hopefully get my machine running on 2 axes. That should be fun.

Regards
Warren

[Imagineering]

OOOPS . . .

Quote myself {I have a 16Volt 10Amp Output with 20,000uf of Capacitance also. What I have done is to connect a 7805 Voltage Regulator across the Capacitor which then feeds my 12V PSU Cooling Fan. When the power is switched off, the Fan winds down over a minute or so, thus discharging the Capacitor. Slight additional cooling is an added benefit.} /Quote

I should have written 7812 V Regulator NOT 7805 - Sorry . . .