What's inside a DC welder
Don
it must have some diodes, but wonder what the configuration is. For a
unit that runs off 220V single phase current, it seems there could be
several possibilities:
One diode...half wave rectification... (not likely?, but would it
work?)
Two diodes...full wave rectification..requires center tapped
secondary.
Four diodes...full wave rectification...only needs single secondary
winding.
After the diodes, is there a filter capacitor to smooth the DC, or is
it just used in the rectified state (ie, pulsed at 60 Hz).
What are the advantages of being able to weld with DC rather than AC?
I have a 230 A ac welder, and wonder if I can dig up some diodes and
make it work for DC. I have 2 huge diodes in the junk box (Motorola
1127 SB...I think theyre rated 450A, about the size of a hen egg, look
like about a 3/4 inch stud mount), but I guess I would need to find 2
more to get full wave, assuming that half wave wouldn't work. I know
I'll need to find a pretty huge heat sink to mount the diodes on.
Haven't though about the Capacitor if one is needed. Is this a crazy
Idea?
Thanks
Don
Mike Graham
>After the diodes, is there a filter capacitor to smooth the DC, or is
>it just used in the rectified state (ie, pulsed at 60 Hz).
Not a capacitor. Choke coil, IIRC.
>What are the advantages of being able to weld with DC rather than AC?
Smoother arc, and usability of many specialized rods for stainless,
aluminum, hardfacing, etc. etc. etc.
>Haven't though about the Capacitor if one is needed. Is this a crazy
>Idea?
If it is then there are a *lot* of crazy people out there. 8-) Everybody
and their pet hamster sells boxes of various complexity that claim to do
what you're doing. Usually they are simply a single diode, and are, for all
intents and purposes, crap. Done properly, it's a beautiful thing.
--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Mike Graham | Metalworker by trade
mikegraham at sprint dot ca | Rancher by choice
Caledon, Ontario, Canada | Weird by nature
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Grant Erwin
Mike Graham wrote:
> >What are the advantages of being able to weld with DC rather than AC?
AC current goes to zero 120 times per second. That makes it much more prone
to having your electrode "stick" to the work. Most HSM types either don't
get enough practice to maintain their welding skills to a high level of
proficiency, or never were there in the first place - guys like that have
a lot easier time welding with DC.
Grant
Edward Haas
: Mike Graham wrote:
: AC current goes to zero 120 times per second. That makes it much more prone
: to having your electrode "stick" to the work. Most HSM types either don't
economics, in terms of welding machine manufacturing/selling price?
--
"Steamboat Ed" Haas : My millenial resolution:
Watch link rot in action! : to cast no stones...
http://www.nmpproducts.com
---Decks a-wash in a sea of words---
Gary Coffman
>What are the components and details of a 220V DC arc welder. I assume
>it must have some diodes, but wonder what the configuration is. For a
>unit that runs off 220V single phase current, it seems there could be
>several possibilities:
>
>One diode...half wave rectification... (not likely?, but would it
>work?)
Unlikely. You'd only be able to use half the power of the welder
and the arc would be rough.
>Two diodes...full wave rectification..requires center tapped
>secondary.
Again, unlikely. Welding transformers generally don't have a
center tapped secondary.
>Four diodes...full wave rectification...only needs single secondary
>winding.
Yes, that's pretty standard.
>After the diodes, is there a filter capacitor to smooth the DC, or is
>it just used in the rectified state (ie, pulsed at 60 Hz).
With a full wave bridge, you get 120 Hz pulsation, which is usually
plenty smooth for welding. Capacitors, and/or an inductor, may be
used to produce DC with even less ripple. Keep the capacitor size
moderate to avoid very high starting surges through the diodes.
Any capacitor used should be a low ESR type.
>What are the advantages of being able to weld with DC rather than AC?
>I have a 230 A ac welder, and wonder if I can dig up some diodes and
>make it work for DC. I have 2 huge diodes in the junk box (Motorola
>1127 SB...I think theyre rated 450A, about the size of a hen egg, look
>like about a 3/4 inch stud mount), but I guess I would need to find 2
>more to get full wave, assuming that half wave wouldn't work. I know
>I'll need to find a pretty huge heat sink to mount the diodes on.
>Haven't though about the Capacitor if one is needed. Is this a crazy
>Idea?
No, it isn't a crazy idea at all. Using DC gives you much more control
of the arc, and produces nicer welds. Adding a full wave rectifier to a
buzzbox is a very practical project. It is even halfway safe as high power
electrical work goes because you're dealing only with relatively low
voltage circuitry. You can make the diode pack and any smoothing
capacitor/choke external to the welder (in their own box, you wouldn't
want to drop a wrench across them) so there's no chance you'll stick
your fingers on 240 VAC by mistake. This leaves the welder unmodified,
so you can still use it for AC welding when that's more appropriate, ie
welding grungy stuff with 6011 rods.
Gary
Gary Coffman KE4ZV | You make it |mail to ke...@bellsouth.net
534 Shannon Way | We break it |
Lawrenceville, GA | Guaranteed |
Don Foreman
wrote:
>What are the components and details of a 220V DC arc welder.
I made a fullwave rectifier with 4 big mot diodes and two
heatsinks, for use on a 220 vac buzzbox. It worked OK.
Most commercially-made DC welders have a choke coil
--smoothing inductor.
Watch your grounds. If the welder's groundlead is tied to
earth ground (mine was) then with a fullwave rectifier both
leads will be "hot" with respect to ground.
Regards from Foreman, dfor...@Ugoldengate.net.
Please remove U from email address to reply
Wayne Cook
wrote:
>Grant Erwin (grant...@halcyon.com) wrote:
>: Mike Graham wrote:
>: AC current goes to zero 120 times per second. That makes it much more prone
>: to having your electrode "stick" to the work. Most HSM types either don't
> --Good point. Sooooo, why use AC at all? Or is it a question of
>economics, in terms of welding machine manufacturing/selling price?
For the buzz boxes it's economics.
For the real welders there are times that AC is better than DC.
First in DC there's two ways to hook it up. Straight polarity and
Reverse polarity. One will give you more melting of the rod than the
work piece (better for building up stuff) and the other will give more
melting of the work (for better penetration). AC will give you
something in the middle.
Next there's a phenomena that effects DC called arc wander. The
magnetic field caused from the current running through the work piece
can push the arc away from where you're welding. AC is much more
resistant to this. On the same note some materials become magnetized
(common in drill pipe and old pipeline pipe) so that it'll push the
arc away giving you fits when you try to weld it. The AC helps here as
well.
Another reason is the reason it's used when TIGing aluminum. AC gives
a better cleaning action to old dirty metal or clean aluminum.
Wayne Cook
Shamrock TX
Paul Koning
> >After the diodes, is there a filter capacitor to smooth the DC, or is
> >it just used in the rectified state (ie, pulsed at 60 Hz).
>
> With a full wave bridge, you get 120 Hz pulsation, which is usually
> plenty smooth for welding.
And that's exactly what you get with AC. The instantaneous power
delivered to the arc is the same whether there are polarity
inversions (AC) or not (full wave rectified unfiltered DC).
If you want smoother, you have to do some filtering, so the
voltage/current doesn't cross zero 120 times per second.
> Capacitors, and/or an inductor, may be
> used to produce DC with even less ripple. Keep the capacitor size
> moderate to avoid very high starting surges through the diodes.
> Any capacitor used should be a low ESR type.
What for? Low ESR capacitors are needed in switching
regulators where you're filtering ripple frequencies in
the megahertz range. For filtering line power ripple, any
old ESR will do.
paul
Don
> Gary Coffman wrote:
> >
> > On Fri, 31 Dec 1999 01:46:21 GMT, bjmc...@usit.net (Don) wrote:
> > ...
> > >After the diodes, is there a filter capacitor to smooth the DC, or is
> > >it just used in the rectified state (ie, pulsed at 60 Hz).
> >
> > With a full wave bridge, you get 120 Hz pulsation, which is usually
> > plenty smooth for welding.
>
> And that's exactly what you get with AC. The instantaneous power
> delivered to the arc is the same whether there are polarity
> inversions (AC) or not (full wave rectified unfiltered DC).
Thanks.
>
> If you want smoother, you have to do some filtering, so the
> voltage/current doesn't cross zero 120 times per second.
What amount of ripple is considered acceptable? I now realize that a
series inductor is apparently the method used in the filtering to keep
the arc at dc, but I do not know what the current characteristics
should be. What percentage should the DC be allowed to drop to? (this
is a function of the inductance). With no inductor, you would have
pulsed DC, half sine waves, dropping to zero current every 1/120
second. With an inductor of sufficient (infinite) inductance, the dc
current would stay constant at the welding level. Since we arent
talking about infinity, what are the practical specs on the current
waveform? Does anyone have any Ideas of this, and hopefully the
inductance required?
>
Ted Edwards
> And that's exactly what you get with AC. The instantaneous power
> delivered to the arc is the same whether there are polarity
> inversions (AC) or not (full wave rectified unfiltered DC).
When the electrode is -ve, most of the heating is in the work piece.
Electrode +ve puts most of the heating into the electrode.
Ted
James W. Swonger
Paul Koning <pko...@lucent.com> wrote:
>Gary Coffman wrote:
>> Capacitors, and/or an inductor, may be
>> used to produce DC with even less ripple. Keep the capacitor size
>> moderate to avoid very high starting surges through the diodes.
>> Any capacitor used should be a low ESR type.
>
>What for? Low ESR capacitors are needed in switching
>regulators where you're filtering ripple frequencies in
>the megahertz range. For filtering line power ripple, any
>old ESR will do.
return to the output during the "valley" period to at most
Vpeak/ESR. If you had (say) 50V peak and were running 50A,
to have a chance of holding that current up you'd need an
ESR below 1ohm. Besides which, if you're cycling big current
then ESR determines the power lost as heat internal to the caps
(which they don't like, especially electrolytics - and they
will be electrolytics, in the size you'll need).
If, by way of example, you want a "valley" voltage no less than
10V less than peak at 100A arc current, you would need
dV=10V
dT=6mS (approx time for doubled half-sine to return above
sagged capacitor voltage)
I=100
I=CdV/dT -> C=IdT/dV=(100)(6E-3)/(10)=6E-2F (60,000uF)
In this case you'd want an ESR of 0.1 ohm (10 volt drop at 100A),
actually much less. An ESR of 1 would give you a time constant of
16Hz, so it would be a poor energy "pogo stick" - more like
trying to pogo on a shock absorber. Even an ESR of 0.1 puts you
at 160Hz, pretty near the base frequency (120Hz), so you could
expect some losses.
--
##########################################################################
#Irresponsible rantings of the author alone. Any resemblance to persons #
#living or dead then yer bummin. May cause drowsiness. Alcohol may inten-#
#sify this effect. Pay no attention to the man behind the curtain. Billy!#
Gary Coffman
>> ...
>> >After the diodes, is there a filter capacitor to smooth the DC, or is
>> >it just used in the rectified state (ie, pulsed at 60 Hz).
>>
>> With a full wave bridge, you get 120 Hz pulsation, which is usually
>> plenty smooth for welding.
>
>delivered to the arc is the same whether there are polarity
>inversions (AC) or not (full wave rectified unfiltered DC).
No, it isn't the same. Polarity does matter. With full wave rectification,
the waveform never crosses zero. Electrode negative and electrode
positive have different welding characteristics. Even without a smoothing
filter, full wave rectified current welds differently from AC.
With halfwave rectified AC, there is a half cycle gap between current
pulsations, which allows the arc to quench. With fullwave rectified AC,
there is no gap. The arc will ripple, but not quench. Using a smoothing
choke or a capacitor will reduce the ripple, but that isn't essential.
>If you want smoother, you have to do some filtering, so the
>voltage/current doesn't cross zero 120 times per second.
>
>> Capacitors, and/or an inductor, may be
>> used to produce DC with even less ripple. Keep the capacitor size
>> moderate to avoid very high starting surges through the diodes.
>> Any capacitor used should be a low ESR type.
>
>What for? Low ESR capacitors are needed in switching
>regulators where you're filtering ripple frequencies in
>the megahertz range. For filtering line power ripple, any
>old ESR will do.
Wrong! Low ESR is required any time large ripple currents must
be tamed. Otherwise the capacitor will overheat and rapidly fail.
Note, switching power supplies do *NOT* operate at MHz frequencies.
They operate from a few kilohertz to a few tens of kilohertz, ie from
the audible range up into the ultrasonic range. It isn't practical to build
a switcher that operates at MHz rates (you'd have to use an air core
transformer, and flux leakage would kill efficiency). But even at 60 Hz,
a high ESR will cause the capacitor to heat if substantial currents are
drawn from it. ESR is a consideration for any high current supply.
Eastburn
They have a control pot for voltage and current adjust - lots of others as well.
e.g. foldback current - where there is either current or voltage or both.
Not suited for this change. typical changes are +/- 1v to allow for some age/burnin.
Martin
--
NRA LOH, NRA Life
NRA Second Amendment Task Force Charter Founder
Martin Eastburn, Barbara Eastburn
@ home on our computer old...@pacbell.net
kenneth knaell
This is somewhat off topic for this news group but please forgive my
transgression; I want to know the answer. Maybe it will be useful to some
of the electroplaters, resistance welders, or motor runners on the group. I
see alot of junked switching power supplies at the scrap yard. Some are
just PC supplies but some are heavy duty industrial types also.
Does anyone know if it is reasonably conveniently possible to change the
output on such beasts. I remember seeing some 5V, 200AMP Digital Eqipment
Corp switching supplies that looked nice but were going to end up being
chewed up for their copper or aluminum or whatever. These were about the
size of a loaf of bread (heavier though of course). Could such a supply be
reasonably converted to, say, a 24V supply at some lesser current or a 12V
battery charger? How about a variable voltage supply for plating? I know
basically how they work but there is one heck of a lot of impressive coils
and "stuff" in those things. I can reason this out on basics but this is
not a good answer. There is more to these things than the basics. Anybody
really know?
ken knaell
Gary Coffman wrote in message ...
Spehro Pefhany
> Does anyone know if it is reasonably conveniently possible to change the
> output on such beasts. I remember seeing some 5V, 200AMP Digital Eqipment
> Corp switching supplies that looked nice but were going to end up being
> chewed up for their copper or aluminum or whatever. These were about the
> size of a loaf of bread (heavier though of course). Could such a supply be
> reasonably converted to, say, a 24V supply at some lesser current or a 12V
> battery charger?
Not very easily. You'd have to add turns to the transformer secondary
most likely (big heavy turns) and change a lot of the other stuff on the
output side.
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany "The Journey is the reward"
sp...@interlog.com
Fax:(905) 271-9838 (small micro system devt hw/sw + mfg)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
M.Simon
wrote:
>
>This is somewhat off topic for this news group but please forgive my
>transgression; I want to know the answer. Maybe it will be useful to some
>of the electroplaters, resistance welders, or motor runners on the group. I
>see alot of junked switching power supplies at the scrap yard. Some are
>just PC supplies but some are heavy duty industrial types also.
>
>Does anyone know if it is reasonably conveniently possible to change the
>output on such beasts. I remember seeing some 5V, 200AMP Digital Eqipment
>Corp switching supplies that looked nice but were going to end up being
>chewed up for their copper or aluminum or whatever. These were about the
>size of a loaf of bread (heavier though of course). Could such a supply be
>reasonably converted to, say, a 24V
No.
Simon - http://www.spacetimepro.com Free CNC machine tool software
KD6JDJ
>What amount of ripple is considered acceptable? I now realize that a
>series inductor is apparently the method used in the filtering to keep
>the arc at dc, but I do not know what the current characteristics
>should be.
>What percentage should the DC be allowed to drop to? (this
>is a function of the inductance). With no inductor, you would have
>pulsed DC, half sine waves, dropping to zero current every 1/120
>second. With an inductor of sufficient (infinite) inductance, the dc
>current would stay constant at the welding level. Since we arent
>talking about infinity, what are the practical specs on the current
>waveform?
(Don)
As I read this thread, there may be some confusion generated by the
responses.
OR, maybe I am the person who is confused. But, I suggest that you will
really enjoy the information you will get about welder design , from a book
that Grainger sells for less than $35. It is Modern Welding.
It is my experience that stick welders dont improve their ease of use by
having a capacitor in their output. A capacitor is used in their primary,
cheifly to improve their power factor. Additional benefits from the capacitor
are explained in the Modern Welding book.
If you are considering building a DC welder from an AC stick welder, without
buying a book, I suggest that you construct the 4 diode bridge, and try using
it with no series inductor. The series inductor will be real helpful when you
are welding at low currents. But you will probably not need the series
inductance for welding at currents over 125 amps.
I make my recommendations based on the idea that it may be impractical to
build a single phase welder with all the ripple smoothing and 'slope control'.
Buy the book or E-mail me if you want more of the information that I have.
Jerry
Don
> Something just occurred to me related to this thread. There is a probably a
> complete difference in philosophy of design of the series inductor needed
> to smooth the arc in a DC welder and that which is used in the design of a
> typical so called "choke input" pwer supply. In a choke input power supply
> one does not want the iron to saturate resulting in a drop in incremental
> inductance and therefore current spikes to the load. This requires large
> amounts of iron, air gaps and 'swinging chokes' and stuff about which I am
> not completely informed.
>
> However in a welding circuit the only requirement for the inductance is to
> supply some minimal current when the driving voltage from the bridge
> rectifier is near zero. Keeping the arc going even if not at high power may
> allow the arc to start conducting better at the next rise in votage better
> than would happen if the arc is completely extinguished every zero crossing
> of the AC drive. I'm not saying that the gases in the arc space cool down
> or even that the ions there have become completely deionized in the zero
> crossing intervals. I'm saying that keeping some minimal current flowing
> helps the startup on the next voltage rise.
>
> For this reason the smoothing inductor for welding may not need to be even
> fractionally as large as normally designed into a correctly filtered "choke
> input" DC power supply. In fact if the reasoning above is valid, any
> little bit of saturating inductance would help while a non saturating
> inductor as is used in a properly designed DC "choke input" power supply
> would lower the average voltage to (2/pie)*peak which would probably never
> hold an arc. For this reason building a saturating inductor that could help
> smooth the welding current in the welder modifications we have been talking
> about in this thread should be a much easier thing to do than building the
> non saturating inductor for a "choke input" DC power supply. The amount of
> energy stored before saturation is probably well known to the welder
> designers though. I can't think of any way to calculate what it might need
> to be from basic principles except to say that it might need to be some
> fraction (say like 1/10) of the average arc current.
>
> If anybody can understand what I mean from what I said above, you are doing
> better than me - I understood it before I tried to write about it and now
> I'm not so sure. :o)
> ken knaell
>
>
>
I think I understand it. The theory is that the inductance holds a
small amount of energy then saturates. When the current drops near
zero, this small energy is expended by forcing some current to flow
rather than return to zero current. Perhaps this is indeed the way
these commercial welders operate. Its amazing to me that there are no
specs or mention of this aspect that I've been able to dig up. Maybe
it turns out that only a very tiny amount of current is required to
keep the arc going.
Don
kenneth knaell
kenneth knaell
Don wrote in message ...
>small amount of energy then saturates. When the current drops near
>zero, this small energy is expended by forcing some current to flow
>rather than return to zero current. Perhaps this is indeed the way
>these commercial welders operate. Its amazing to me that there are no
>specs or mention of this aspect that I've been able to dig up. Maybe
>it turns out that only a very tiny amount of current is required to
>keep the arc going.
>
>Don
...................
Exactly!! I guess the manufacturers are not particularly interested in
giving away their design algorithms.
Ken Knaell
tom
> > Capacitors, and/or an inductor, may be
> > used to produce DC with even less ripple. Keep the capacitor size
> > moderate to avoid very high starting surges through the diodes.
> > Any capacitor used should be a low ESR type.
> Paul Koning wrote:
> What for? Low ESR capacitors are needed in switching
> regulators where you're filtering ripple frequencies in
> the megahertz range. For filtering line power ripple, any
____________________________________________
The "R" in ESR is resistance . In a welder there is high current and
low volt' .
All the power that you are trying to store in the cap' ( then dump into
the
weld) is disapated in that R if ESR is high .
BTW Did i see a post on alternators to MIG ? They make very good MIG as
they
are at their best at 20 to 30 volts ( peak power ). I have a 200A and
someday
i'll hook a 10HP Honda to it . The big surprise is they put out most of
their
power at 3000 to 4000 rpm . Intuitive auto mechanics thought they needed
to
spin very high since in a car they are geared up 2.5 or so .
Delta wind is best for welding .
--
..In 1980 , w/ the stroke of a legal pen , US govts
turned 50 mil ppl into felons , to make jobs for judges
, jailers and cops .
tom
level . It is not used to ballast ( provide a high resistance as CC
welders
need ) , cause the windings in a stick welder are loosely coupled and
this
makes the ballast effect . Transformers will try to maintain their
output
voltage only if the windings are close coupled by being wound on top of
each
other . If you stuck a rod on this type , you'd suck 1000 amps !
I made an AC buzz box into DC and it did not make a big deal to me .
Switching chokes saturate to allow all the current to get thru . MIG
does
not use "swinging" chokes as they are called , so they use a large air
gap
which reduces inductance MUCHO , but prevents saturation ( sat' is bad )
.
The diodes have the same current as the choke. Visualize the diodes
conducting
way past peak volt . It's the choke that forces current ( by it's
collapsing
field) thru both the load AND the diodes ( because that is it's only
path )
, so the diodes are conducting at strange angles , like when the
winding
is at zero ! A cap input filter has a very short diode conduction angle
or
time and choke has a very long cond' angle .