RE: [CAD_CAM_EDM_DRO] Variable Transformer

Jon:

Thanks, this is very interesting.

Can you tell me why there is an AC component across the cap when the "-" is
not grounded, and not when it is? I have seeen this, but am not sure why it
happens.

Also, the current-limiting capability of the transformer is something I
hadn't seen mentioned before today-also very interesting.

Thanks again for taking the time to write this up!

Andy Wander
Verrex Corporation

-----Original Message-----
From: Jon Elson [mailto:elson@...]
Sent: Wednesday, June 29, 2005 1:38 PM
To: CAD_CAM_EDM_DRO@yahoogroups.com
Subject: Re: [CAD_CAM_EDM_DRO] Variable Transformer

Andy Wander wrote:

>Tony:
>
>I think that a person who makes a statement should be prepared to back the
>statement up.
>
>My statement has been that I don't know why this might be dangerous, and
>that I would like to know. I stand by that.
>
>
>And, BTW, I have looked into this quite extensively, and have not gotten

any

>definitive answers, which is why I keep hoping someone who "knows" this so
>well, will be able to explain it to me.
>
>

OK, let me give it a try, specifically aimed at the CNC power supply we are
all thinking about, here.

First, there's NOTHING wrong with a Variac (tm) per se, but the use of ONLY
an adjustable autotransformer in a system where the drives may not have been
designed for direct line connection.

When a regular (isolating) transformer is used, the DC - terminal of the
capacitor
can be grounded. Touching any point in the drive circuitry will get you
DC only
(except in the case of a stepper drive, which will give some AC when the
motor is
spinning). DC can give you burns, but in most cases is not
"tetanizing", ie. causing such
muscle stiffness you cannot tear yourself away from the contact. 50 -
60 Cycle AC just
happens to be a REALLY bad choice of frequency, as it has the strongest
tetanizing
effect, where the shocked person is unable to pull away from the
electrical contact.
It is also really good at causing ventricular fibrillation of the heart,
which is nearly
always fatal, as CPR is relatively ineffective at providing minimal
blood flow until a
defibrillator can be brought to the scene.

When an autotransformer (or no transformer at all) is used, the DC -
terminal of the
cap must not be grounded. Both cap terminals will have a DC voltage
plus the AC line
voltage superimposed, and these voltages can be quite high. At the full
120 V input,
you would have 168 V DC across the cap terminals, with a 120 V sine wave
added
to that. At the + peak of the line, the + terminal of the cap would be
+168 V, and the
- terminal would be at ground. At the - peak of the line, the +
terminal of the cap would
be at ground, and the - terminal would be at -168 V. The cap would
cycle between these
extremes 60 times a second (in the US). This is all assuming a bridge
rectifier.
With an autotransformer, the voltages would be proportionally reduced by
the setting of the
dial.

So, touching anything in the drive system, either the + or - terminals
of the cap, or any
motor terminal, would be about the same as plugging your finger in a
light socket!

Some servo drives are actually designed to operate this way, but they DO
make special
provisions in the design of the drive to assure the best isolation
between control circuits
and power transistors, so that this hazard is reduced. But, you need to
be aware which
drives work like this, so you don't touch anything on the power side of
the drive.

So, with an isolated power supply up to 70 V, with the motors at idle,
most persons can pretty
safely touch anything on the motor with dry hands without any
surprises. I'm not suggesting
this as standard practice, but just indicating that there would not be
very hazardous voltages
present.

With a Variac (tm) type transformer ONLY, and set for an output of 70 V
DC on the capacitor,
You would have an AC component of 70 V peak, or ~50 V RMS on BOTH
terminals of the
capacitor all the time. That is if the common terminal of the variac
were connected properly to
the neutral with a polarized plug. The AC component would be higher if
the variac were
connected with the common to the hot wire, by mistake. This 70 V peak
AC component would
not be as dangerous as the full 168 V from the line, but it would still
be more dangerous than
the grounded minus terminal and no AC component that you get with an
isolation transformer.

One other concern is leakage inductance and its effect on surge
current. Due to the way
autotransformers work (variable or fixed) they provide a direct current
path from input to
output, and therefore are not good at limiting fault currents.
Isolating-type transformers
have an additional effect when run well outside their ratings, such as
in short-circuit conditions.
Above a certain point, perhaps 10 to 50 times their current rating, they
will deliver no greater
current. Shorting the ordinary 120 V wall socket will permit a current
of about 1000 Amps
to flow for a couple ms until the breaker trips. An ideal 4:1 step-down
transformer would
produce 4000 amps at 30 V, or 120 KW, during a short circuit event.
Wow! But, a real
transformer with a 10 - 20 A secondary rating may only produce 100 to
1000 A into a
short, and most probably toward the lower end of this scale, perhaps 7.5
KW, causing much
less damage when something goes really wrong. This current limiting
effect is something that
power engineers study with care all the time, and is the reason that
your house doesn't explode
when you plug in a shorted appliance. Only the copper resistance of an
autotransformer limits
the current in a short, where an isolating transformer uses both copper
resistance and leakage
inductance to limit fault current.

I hope this dissertation helps clear the waters a bit!

Jon