Capacitors: discharge tool design

[slowbeard]

Hi,

Newbie here - I'm beginning to learn electronics with the aid of an arduino, breadboard and YouTube, hope this is the right place for my question.

Projects so far have all made sense and pretty easy, and now I want to venture out into creating some things.

I'm planning an infrared sensitive, motion activated wildlife camera using an old optio point and shoot digital camera.

After accidentally shocking myself on the capacitor (which I was actually trying to be very careful not to touch haha) I'm trying to first build a discharge tool!


The capacitor I'm dealing with is 100mf 330v, and I was going to use a 20k 5w resistor soldered to heavy duty wire, insulate everything well and use the wire ends to discharge.

The other design I've seen uses a bank of 7 1n4001 diodes in a circuit.

Questions:

1) Is there any main reason to prefer one design over the other?

2) for the first design how do I calculate the appropriate resistor values and discharge times to ensure I'm doing this safely?

3) could someone explain how the diode based discharge circuit mentioned in the below link operates, I don't think I understand the principle?

http://repairfaq.cis.upenn.edu/sam/strbfaq.htm#strbsww


Thanks,

Slowbeard

[Simon]

Don't use the diodes. A resistor will suffice, The discharge time to fully discharge is 5xRxC you may like to put some crocodile clips on your resistor.

[Yansi]

The link below doesn't work.

[slowbeard]

The link below doesn't work.

Sorry, link should be fixed now!
Thanks for pointing it out

[timb]

The link below doesn't work.

Sorry, link should be fixed now!
Thanks for pointing it out

He actually tells you what the diodes do:

"The two sets of 4 diodes will maintain a nearly constant voltage drop of about 2.8-3 V across the LED+resistor as long as the input is greater than around 20 V. Note: this means that the brightness of the LED is NOT an indication of the value of the voltage on the capacitor until it drops below about 20 volts. The brightness will then decrease until it cuts off totally at around 3 volts."

The 25W resistor is what actually discharges the capacitor. The diodes are just there to make the LEDs a crude voltage indicator.

One piece of advice, large caps can have a memory effect, whereby a voltage can reappear on them after being discharged. It's called dielectric absorption. In some cases, deadly voltages can reappear minutes after discharge. So be careful!


Sent from my Tablet

[slowbeard]

Don't use the diodes. A resistor will suffice, The discharge time to fully discharge is 5xRxC you may like to put some crocodile clips on your resistor.

Thanks for this.

So the time to total discharge would be 22 seconds?

20,000 (ohms) x 0.0001 (farads) x 5 (time units) x 2.2 (time units to seconds conversion)?

Where does the watt rating of the resistor come into it?

[slowbeard]

The link below doesn't work.

Sorry, link should be fixed now!
Thanks for pointing it out

He actually tells you what the diodes do:

"The two sets of 4 diodes will maintain a nearly constant voltage drop of about 2.8-3 V across the LED+resistor as long as the input is greater than around 20 V. Note: this means that the brightness of the LED is NOT an indication of the value of the voltage on the capacitor until it drops below about 20 volts. The brightness will then decrease until it cuts off totally at around 3 volts."

The 25W resistor is what actually discharges the capacitor. The diodes are just there to make the LEDs a crude voltage indicator.

One piece of advice, large caps can have a memory effect, whereby a voltage can reappear on them after being discharged. It's called dielectric absorption. In some cases, deadly voltages can reappear minutes after discharge. So be careful!


Sent from my Tablet

Ahhhh. I did read that but it didn't sink in at first, so the diodes just stop the led from blowing by keeping the volts steady instead of allowing the charge to all be dumped into it at once.

Rightyo.

And yep, definitely being on the side of caution with capacitors - one frying was enough!

[Simon]

What is the units to seconds conversion for ? A capacitor discharges by 2/3 in RC seconds, it fully discharges in 5RC seconds, same applies to charging. You could al add the LED, maybe have a resistor in series with an LED and a zener across the LED to protect it.

[slowbeard]

What is the units to seconds conversion for ? A capacitor discharges by 2/3 in RC seconds, it fully discharges in 5RC seconds, same applies to charging. You could al add the LED, maybe have a resistor in series with an LED and a zener across the LED to protect it.

Got a bit confused when i was reading another page on dischargong. not sure where I got that conversion from!

 so the result is simply 10 seconds

20000 * 0.0001 * 5 = 10 seconds.

[Electric flower]

Am I only one here who uses 230V incandescent light bulb to discharge capacitors?

[AF6LJ]

Am I only one here who uses 230V incandescent light bulb to discharge capacitors?

I guess I am the only one here who uses a "Jesus Stick".

[dom0]

Larger caps don't like the dead-short method much.

As for deadlyness, it's mostly about the energy stored above a certain voltage threshold. Now we know that the energy stored in the electric field of a capacitor is E=¹/₂ * C * U². This means that after tau (R*C) the stored energy decayed not to e^-1 (~0.37) but e^-2 (~0.14) the initial energy. Already after only two tau we're down to ~2 % of the initial energy, which won't be lethal with usual cap sizes (at most a couple hundred µF at 300 V) anymore.

[AF6LJ]

A crappy screwdriver and a baaaam works all the time, from 10v 100000uf to 630v 330uf. Higher energy calls for more respecting, but as long as you have a pair of glass, it won't do much harm.

What about your pants?

That is a Jesus Stick.
When it goes BAM you go JESUS CHRIST!

[Ian.M]

Many ceramic cased wirewound resistors are rated for a 5 to 10 times pulsed overload. i.e, for a one second pulse, a 7W resistor may well be able to stand 70W.  Convert the overload rating and the pulse duration into an energy in joules and make sure you don't use it to discharge any caps with more energy than that.  ((1/2)*C*(V^2))

However beware of the resistor's max voltage rating - you may need several in series. 

Always have a clip on lead for the earthy end of the discharger because yoou *really* don't want to be poking around with two hands round charged high voltage caps.  That's a good way to get a pine box if you are unlucky.

[vk6zgo]

A crappy screwdriver and a baaaam works all the time, from 10v 100000uf to 630v 330uf. Higher energy calls for more respecting, but as long as you have a pair of glass, it won't do much harm.

What about your pants?

That is a Jesus Stick.
When it goes BAM you go JESUS CHRIST!

That's exactly what I've always used---never had any complaints from capacitors!

We used to do  this to short out the big oil-filled caps in Broadcast Transmitters.

After a while,the Boss got a bit fed up with all the big bites out of the really large screwdrivers we used,so he had us make up proper shorting sticks,with a long insulating handle,an earth strap & a "shepherd's crook " hook on the business end.-----Still no series resistors!

I sometimes worry about people making up mechanically complex "discharging devices".
One faulty joint,& you will not discharge anything.

[AF6LJ]

A crappy screwdriver and a baaaam works all the time, from 10v 100000uf to 630v 330uf. Higher energy calls for more respecting, but as long as you have a pair of glass, it won't do much harm.

What about your pants?

That is a Jesus Stick.
When it goes BAM you go JESUS CHRIST!

That's exactly what I've always used---never had any complaints from capacitors!

We used to do  this to short out the big oil-filled caps in Broadcast Transmitters.

After a while,the Boss got a bit fed up with all the big bites out of the really large screwdrivers we used,so he had us make up proper shorting sticks,with a long insulating handle,an earth strap & a "shepherd's crook " hook on the business end.-----Still no series resistors!

I sometimes worry about people making up mechanically complex "discharging devices".
One faulty joint,& you will not discharge anything.

I went through a short stage of wanting to make up a slow discharge device and decided that I couldn't make it as reliable as a cast off screwdriver.

[Ian.M]

The problem with shorting the cap without a resistor is the extremely high transient current pulse.  If you are even slightly careless where you connect the ground lead, the resulting ground bounce can blow the s--t out of delicate chips elsewhere on the board.

[AF6LJ]

The problem with shorting the cap without a resistor is the extremely high transient current pulse.  If you are even slightly careless where you connect the ground lead, the resulting ground bounce can blow the s--t out of delicate chips elsewhere on the board.

That is why you use a heavy ground clip, and a reasonably heavy ground lead. (don't get crazy with wire size).
Like others here I have never seen any adverse effects from just shorting them out.
The potential for death increases with voltage, as a member of the "I came in contact with 1600V at a reasonably current because of my carelessness and survived to talk about it club." I can say that when it comes to discharging caps in a power supply; I want to make sure they are discharged.

We each have to decide what is most important.

[Ian.M]

I speak from experience with discharging the EHT on large screen CRT TVs.  If you clipped the Jesus stick's ground lead anywhere other than the CRT's aquadag grounding spring, there was a high risk of killing a signal processing chip or the syscon.   Due to the very high voltages, it wasn't easy to obtain rugged enough bleeder resistors, so the Jesus sticks tended to be the direct shorting type.

When you are only dealing with a few hundred volts, there is no excuse for trying your hand at spot-welding when all you want to do is make the caps safe to work on.    Depending on the circuit configuration, especially if there are more than one energy storage cap,  the transients can do a lot of damage.   A robust bleeder resistor lead is easy enough  to make, and if you are worried about it failing, test it after every use.

IF you are working with HV cap banks storing potentially lethal amounts of energy, after safely discharging it, TREAT THE CAP BANK AS LIVE UNTIL YOU HAVE A SHORTING LEAD CLIPPED ACROSS IT.

[Cubdriver]

That is a Jesus Stick.
When it goes BAM you go JESUS CHRIST!

I always thought it was the ground rod for shorting out and holding the HV parts at ground potential so you could work on them.  If you neglected to use it, you might get to go meet Jesus!

-Pat

[all_repair]

I have made the LED type and have been using it for a few years.  It is well worth the effort.   Be sure is much better than be sorry.  When you are working on something, you tend to rush and can't wait, you need the LED as traffic light to pace you.

[tautech]

One piece of advice, large caps can have a memory effect, whereby a voltage can reappear on them after being discharged. It's called dielectric absorption. In some cases, deadly voltages can reappear minutes after discharge. So be careful!

I heard that that was true for oil filled large capacitors. Not so common on PCB's

Can someone elaborate please?

timb is quite correct, however the phenomenon is only of risk to person with high voltage caps.
To see it, charge a suitably rated cap to say 200 VDC, short it out and with nothing across the terminals connect a DMM. Magic before your eyes. 

Practically when discharging a HV cap, after the initial arc and a short pause, more discharge can sometimes be had from it.

Just so you know.

[AF6LJ]

That is a Jesus Stick.
When it goes BAM you go JESUS CHRIST!

I always thought it was the ground rod for shorting out and holding the HV parts at ground potential so you could work on them.  If you neglected to use it, you might get to go meet Jesus!

-Pat

Actually that is true.
But I had to be a bit colorful with the post.

[kripton2035]

bought this tool from dx some years ago - still working fine, well built, no sparks when touching capacitor
http://www.dx.com/p/repair-tools-sparkpen-capacitor-discharge-pen-for-camera-mobile-phone-home-appliances-129179

[kripton2035]

yes one led on each side, the led that light depends on the capacitor polarity...
but as there are a red and a black probe, you always put the good probe and see the right led
there is another model on dx web site with the 2 leds on the same side and also with sound but at $37 it's a bit too much
better open this box and move one led on the other side if really needed.