DANGER! 17,000 VOLTS!

From: Max Eskin <maxeskin_at_hotmail.com>
Date: Thu Jun 4 18:24:27 1998

I often work on systems with power on. With PCs it's no big
deal in terms of my safety. The SE, however, is an incredible
pain to repeatedly re/disassemble if my fix doesn't work, so I would
like to work on it cover off/plugged in. I'm only dealing with the
logic board anyway. That's why I originally asked...
By the way, how do people work on machines like the PS/2 Model 25,
in which one can only get to the MB when it's not plugged into the
PSU?
>> > But not everyone has an EHT meter. It's probably OK to use a 10M
resistor
>> > (prefereably a high-voltage one) and a well-insulated screwdriver
to
>> > discharge the CRT. Connect one end of the resistor (use clip leads,
etc)
>> > to the CRT earth and the other one to the screwdriver blade. Then
push
>> > the screwdriver under the edge of the anode cap and hold it in
contact
>> > with the metal contact for about 30s.
>>
>> Most resistors are rated for a 500V or 1kV rather than EHT, so there
is a
>
>That's why I said 'preferably a high-voltage one'. Farnell and Maplin
>both sell the Philips HV resistors good to about 10kV. A couple of
those
>in series.
>
>> small risk that they'll break down or flash over. Better to use two
or
>> more in series if you can.
>>
>
>I'd not trust my life to it, but most resistors break down and go to a
low
>resistance when they flash over, not open.
>
>> It's also worth pointing out that even 1mA at 15kV is a fair wattage

--
>> another reason to use more than one resistor.  A large colour CRT 
might
>
>Yes, but the energy stored in a CRT is not that great. Resistors burn 
out 
>because they overheat, and if there's not enough energy to heat them 
up, 
>then they'll not fail.
>
>> have a 25kV supply.  Of course, as Tony points out, usually the CRT 
will
>> discharge when switched off.  But if it doesn't, that's just when you 
don't
>> want to find out the limitations of ordinary resistors.  (I once 
proved
>> this theory on a Commodore PET).
>
>'Nice' monitors have EHT bleeders anyway, often as part of the EHT 
>reguation circuit (the feedback loop consists of a potential divider 
>across the EHT supply). So the EHT should discharge at switch-off. I 
>check it anyway with the meter, of course.
>
>As this is classiccmp, I think I should mention that some vector 
>displays, like the DEC VR12, VR14, VR17, etc have a mains-derived EHT. 
>There's a sealed (oil-filled?) can containing a transformer and a 
voltage 
>doubler - a schematic is printed on the can. These things can supply 
>considerable current for short periods - easily enough to kill you. 
>They're nasty. 
>
>> 
>> Perhaps this is a good place to mention the "hand in pocket" rule:  
if
>> you're working on a high-voltage system, and one hand is in your 
pocket, it
>> can't be touching a good earth (unless you're even odder than I am 
:-)) so
>> there's a reduced chance of current crossing your body.  You used to 
often
>> see TV engineers with one hand in a pocket and an EHT probe in the 
other.
>
>Nowadays you see computer preservationists with one hand in their 
pocket. 
>I do it whenever I work on HV stuff with the power applied. 
>
>The idea is to prevent the current taking a path through your heart. 
Down 
>one leg is possibly OK, arm-arm is not. 
>
>I didn't mention it, because the original poster didn't say he was 
>planning to do work with the power on. If you are, there's one other 
rule 
>that I'll state - Always have somebody nearby who knows how to turn off 
>the power if anything goes wrong. Never work alone.
>
>As a practical point, it helps if that person is not either of your 
>parents, wife/husband, or girl/boyfriend. The reason is that they tend 
to 
>cut the power on the slightest problem - like a minor flashover to a 
>screwdriver, which is very annoying!
>
>> Pete						Peter Turnbull
>
>-tony
>
>
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

Received on Thu Jun 04 1998 - 18:24:27 BST