On Mon, 03 Nov 2014 08:44:48 -0800, wrote:


Soft start on a welder takes longer than for a motor. This is an
adjustment on the welder itself. And even though it is pretty fast it
is not ideal. Especially when welding aluminum because you want the
metal to get hot as fast as possible. So the technique is to stomp on
the pedal all the way to get the maximum current. Then as soon as a
puddle starts to form the current is decreased. If the aluminum is
heated slowly too much if the heat spreads into the surrounding metal.
After several inches of welding in a localized area the work gets
really hot until it is too hot to handle. Heating the metal fast,
applying the filler fast, and getting off of the pedal makes for
higher throughput. Today my friend is emailing me the graph from the
power company so I will learn more and will post results here. I will
also post the results of the final fix, whatever it is.

Hi Eric.

IF you'd like to email me a copy of the chart ), I can
probably help you. I'm a retired engineer who spent most of his
career in the utility biz so I speak "utility-eze". Also I have a
little consultancy on power quality so I'm familiar with these types
of problems.

A welder can have two separate problems that can cause light flicker.
One is the magnetic core saturation inrush when the ON switch is
flipped. The cause has been explained elsewhere in this thread. The
other is due to the sudden application of load when the arc is struck.

Since the welder has soft starting for the arc, I suspect that the
problem is with core saturation when the welder is turned on. If
that's the problem (it will be glaring on the chart they provide),
there is a very easy and inexpensive solution. It involves
introducing some temporary resistance in series with the line voltage.
This limits the current for a few seconds until the transformer
stabilizes.

It involves a contactor suitable to handle the load, a resistor and a
time delay relay set to 2-3 seconds. Cost should be under $100. If
saturation inrush is the problem, I'll be happy to spec out the parts
needed.

The resistor is typically 2 or 3 stainless steel TIG rods bundled
together in parallel and connected to using split bolt electrical
connectors.. The resistor won't be in the circuit long enough to get
hot so it can be insulated with electrical tape or heat shrink tubing,
coiled up and it and the other parts placed in the welder.

The contactor is connected in parallel with the resistor. The time
delay relay is connected in series with the contactor coil and both
ends are connected to the line downstream of the "on" switch.

When the "on" switch is flipped, the contactor is de-energized and the
resistor is in series with the transformer, limiting the saturation
current. In two or three seconds, the time delay relay trips,
energizing the contactor and shorting the resistor. Normal full power
is applied to the now-stabilized transformer.

On the regulatory side, your friend has the law on his side. There
are both FERC (federal energy regulatory commission) and ANSI specs on
power quality. If your friend has a 200 amp service then the utility
is required to supply 200 amps from a circuit stiff enough to limit
the voltage dip to, normally -5%. That much dip will cause slight
dimming of lights but not an annoying amount.

If the utility isn't meeting that requirement then they are obligated
to install a sufficiently large transformer and service drop to do so
and at their cost.

It is common to have the utility try to bluff the customer into paying
for the new apparatus. That is happening up here in this remote
mountain-top fishing resort. A number of people who have had weekend
cabins are now retiring and moving up here permanently. The
infrastructure can't handle the addition of things like central air
conditioning. One transformer has (probably a record setting) 13
feeds attached to it.

The utility has thrown up roadblocks at every step of the way and it
looks like legal action may be necessary. But my neighbors will
ultimately win because they have law and regulation on their side.

If your friend's welder is drawing more than 200 amps during the
inrush period then the utility CAN force him to pay for remediation.
We'll know when we get the chart. If that's the case then the above
described inrush limiter is the inexpensive solution.

If it comes down to a ****ing contest, I have a spare Dranetz power
quality analyzer that I could loan you. That would give you hard data
to fight back with. Contact me off-line if it comes to that.

John
John DeArmond
http://www.neon-john.com
http://www.fluxeon.com
Tellico Plains, Occupied TN
See website for email address