Discharging caps to ground?

[gdomeier]

When discharging filter caps to ground, where does the electricity go if the amp is unplugged?

All the how to's about discharging caps mention to unplug your amp, then discharge the caps by connecting one a bleeder setup to ground and the other to various points inside the amp.

I don't understand how the electricity can really go out of the circuit if the amp isn't grounded. Sitting the chassis on a wooden table or such doesn't seem to me to leave anywhere for the the electrons to go.

[antik]

Think of a cap as 2 steel or aluminum plates that are next to each other separated by air. When the capacitor has a charge it means there are more electrons on one plate than the other, ie there is a net difference in the number of electrons on each plate. This also means there is a voltage between the two plates. The voltage is between the two plates of the capacitor. It has nothing to do with earth ground. It is like a battery. The battery has a voltage that is really not related to earth ground. (You can kill a battery without every connecting it to earth ground. Discharging a charged cap is very similar)

Whenever there is a difference in the number of electrons in one place and another, energy is stored. Given half a chance, electrons like to spread out evenly and not be bunched up in one place when there is somewhere else they can go that is less crowded. When there is a path for those electrons to go from one plate to the other, those bunched up electrons on one plate make their way to the other plate until the number of electrons are equal on both plates. With equal number of electrons the voltage is zero.

The secret to safety is to not let your body be the path for those electrons as they make their way to less crowed quarters.

[Bruce O'D]

I don't understand how the electricity can really go out of the circuit if the amp isn't grounded. Sitting the chassis on a wooden table or such doesn't seem to me to leave anywhere for the the electrons to go.The charge dissipates into the metal chassis. You're taking a large potential voltage with a small surface area (inside the capacitor) and dissipating it into a huge chunk of metal. It doesn't have to be connected to earth ground. When you walk across a carpet in the middle of winter and scuff your feet, you become a capacitor. When you touch a doorknob, that charge dissipates quickly into the doorknob and you get zapped. The doorknob isn't connected to earth ground.

[gdomeier]

Thank you both!

[donnyjaguar]

Wow, interesting descriptions. I was going to simply describe it as "energy" stored in the capacitor is converted to "heat" in the bleeder resistor. If you short the capacitor to ground you get extreme heat in the form of a spark.

[John Phillips]

The charge dissipates into the metal chassis. You're taking a large potential voltage with a small surface area (inside the capacitor) and dissipating it into a huge chunk of metal.No, that isn't right.

The charge flows round the completed circuit via the chassis into the other side of the cap (which is also grounded), until both sides have equal instead of opposite charges. The charge doesn't 'dissipate' anywhere. You can discharge a cap completely without the chassis being involved at all - just linking both its terminals with a resistor will accomplish it.

[Rob F]

What rating of resistor is recomended?

[John Phillips]

100K 1W is OK.

The lower the resistance, the quicker the discharge time but the higher power resistor you need.

Power = Voltage-squared divided by resistance.

You can get away with a lower power than you would expect if it had to deal with that voltage continuously - if you do the math on 100K at a voltage of 450V, you'll find that the initial power exceeds 2W - because as soon as the cap starts to discharge the voltage falls away quickly (and the power even faster, since it's proportional to the square of the voltage), but you don't want to overdo it.

[FFTT]

What is the SAFEST way to discharge caps.

If you don't have any special tools, what common shop items can you use
to do this safely.

Pix might help too.

[John Phillips]

It depends on the amp. With many, you don't have to worry.

If you know what you're doing, it's much less of an issue than it first appears.

Any amp with 'stacked' filter caps (often just the first or first two stages) with divider resistors effectively has built-in discharge resistors. Just turn the amp off and leave the standby switch on (play), and the caps will all discharge through the divider resistors in a couple of minutes. You don't need to do anything else.

Assuming it's not one of these, and it has a tube rectifier, you can simply pull the rectifier tube and leave the amp powered up for a few minutes - that will discharge the caps through the tubes. (There is one exception I know of, the Mesa Blue Angel, which has both tube and solid-state rectifiers operating at the same time - and obviously if it has switchable rectifiers, you need to select 'tube'.)

That leaves just amps with solid-state rectifiers and non-stacked caps. Many (but not all) of these have the standby switch upstream of the caps, so simply leaving the amp on standby for a few minutes will do it.

And for the relatively few cases where none of the above work - eg earlier Marshall JMP50s - the simplest is just to connect pin 1 of the first preamp tube (the first one clockwise from the gap in the ring of pins) to ground - that will drain the caps through the entire B+ chain. Any plain piece of wire will do. For safety, this method works in every other case too, if you're unsure. Leave the standby switch to on (play) to make sure you drain all the caps.


But be careful - I got my first serious shock from a tube amp at about the age of 18 while inside (you've guessed... ;)) an early JMP50 which had been turned off and unplugged, so I thought I was safe.

There is a slight issue with 'memory charging' in amps which don't have resistors internally - where the cap will charge up again a little by chemical action in the electrolyte - but it isn't serious and won't give you a dangerous shock, although it may surprise you if you short something and get a spark. Connecting V1/pin1 to ground will eliminate this too.

Just remember to disconnect it before you power up again.

[jcoloccia]

re: remembering to remove the jumper before powering back up.

Couldn't you just discharge through the power chord, i.e. clip from the cap to the chord or a tube socket to the chord (through the third prong). If you did it like this there's no way you could plug it back in with the jumper in place.

Just a noobie thought.

[Oatman]

I bought some radio shack 'jumpers'. They are just two alligator clips connected by 4" of cheap stranded wire. Will this be sufficient for draining caps as described? Should I solder on a 100k 1w resistor between the alligator clips? Is their a preffered (safer) way to to it? (in other words can I do this by hand? does it matter if I connect to chassis first or pin 1 on the first pre-amp tube first?) thanks.

[rockon1]

I bought some radio shack 'jumpers'. They are just two alligator clips connected by 4" of cheap stranded wire. Will this be sufficient for draining caps as described? Should I solder on a 100k 1w resistor between the alligator clips? Is their a preffered (safer) way to to it? (in other words can I do this by hand? does it matter if I connect to chassis first or pin 1 on the first pre-amp tube first?) thanks.


Read John Philips posts above. He goes thru all the methods of draining the caps. On all my amps shutting them down without using the standby works. I then check the caps for voltage with a multimeter. This method doesnt necessarily work on all amp however.Bob

[mcholley]

John Phillips;3416480]
But be careful - I got my first serious shock from a tube amp at about the age of 18 while inside (you've guessed... ;)) an early JMP50 which had been turned off and unplugged, so I thought I was safe.


Wow, does that bring back memories...it was about '84, and man did I hit the floor, and let at a yell like a banshee I was so suprised. And guess what? It was a JMP 50 Watt.

Memories........:dude

Mark

[donnyjaguar]

I just use a screwdriver with an insulated handle myself. Better safe than sorry (or deceased). The spark will make you jump if you drink too much coffee. :)

Agreed on the capacitor's voltage coming back up. This is caused by a phenomena known as dielectric absorption. Generally not anything to worry about but worth mentioning to young players if they all of a sudden measure three volts where there "should" be zero.