Register now to get rid of these ads!

Cheap Tech AC welder to DC

Discussion in 'The Hokey Ass Message Board' started by dare-to-be-different, Jan 17, 2008.

  1. In answer to some questions I have been asked, I am going to explain how I turned my AC welder into a DC welder for the price of a couple pizzas.

    I don't have much time to put any polish on this, and at the moment I do not know how to put text in between the pic attachments (the attachments seem to be grouped all together), so the pics may run together and the text come before or after the pics.

    I may even have to put up several posts in this thread to explain a detail and prevent something simple from looking confusing.

    Here goes.

    First (before ebay existed) I went to an electrical equipment repair company that repairs large motors for factories and underground coal mine equipment to see what heavy duty diodes they had in their discard pile. They are usually used to run the dc motors or charging equipment for the battery powered underground coal mining carts.

    Here is an ebay auction that has the pictures of what you want to buy... I searched for "DIODES" this guy has a few of these auctions going regularly.

    http://cgi.ebay.com/Lot-of-5-1N4045-100-Volt-275-Amp-Power-Rectifier-Diodes_W0QQitemZ200192081250QQihZ010QQcategoryZ73142QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

    Just make sure the amp rating is higher than your expected welding amps. Mine is a 180 amp welder, and I usually weld around 90 amps, so a 200 amp or larger diode is more than I need, and will run cool for a long life.

    I don't worry about the diode voltage rating because most arc welders are about 24 volts output or something like that. You wont need 220 volt diodes because you won't be using them on the high voltage input side... just on the low voltage out put side.

    Long ago I ran for many years with 125 amp diodes since I never cranked up my machine that high anyway. That gave me a 6-8 year lifespan before a diode failed. I never knew the voltage rating of the old diodes I used for many years.

    OK, no more chattering and I'll get right to it.

    You need to get 4 diodes to make a full wave rectifier.

    ********************************************

    FULL WAVE RECTIFIER- simple explanation- if you want to change AC into one direction only (DC), you can put in a one way valve (diode).

    USING ONE DIODE- When AC flows one direction it flows right past the diode, then when it changes direction it is blocked by the diode. The problem with this method is that it gives a pulsating DC that is not smooth at all.

    Simply put- we will use 4 diodes in a criss-cross fashion to make the current absolutely smooth.

    If you like fancy scientific-sounding titles, that is known as a full-wave rectifier.
    ***************************************

    Each diode will have a positive connection and a negative connection. Some will use the pigtail wire as the positive and the case bolt as the neg, and some will use the pigtail as the NEG and the case as POSitive.
    Either type will do, but for simplicity sake make sure all your diodes are made with the same direction of flow or it could get confusing figuring out which ones to turn backwards.

    For easy explanation lets assume you want the results to be DC REVERSE (you can change it later) Ground clamp NEG and electrode-rod-holder as POS-- just for now.
    Lets also pretend that you have the kind of diodes with the case as POS and the pigtail wire as NEG -- just for now.

    What you want to do is connect TWO of the 4 diodes with both pigtails (one from each of the two diodes) hooked to the cable going to your GROUND CLAMP. (neg ground for DC reverse)

    Take two more diodes and connect both CASES (positive) to the OTHER cable - the one going to your ELECTRODE-ROD HOLDER.



    From here on, whatever you hook to the other (loose) ends of the diodes the current will only be able to flow as NEG to the ground clamp and POS to the welding rod.

    So far so good. Now to make it into a full-wave.......
    The first pic shows what you have now.
    Two pigtails hooked to one cable and two diode cases hooked to the other wire.

    Z ddiodesfirst.jpg


    One cable (DC OUTPUT TO ROD HOLDER) with two diode CASES (POS) and also one cable (DC OUTPUT TO GROUND CLAMP- NEG) that is hooked to two diode PIGTAILS. Thats the first pic.

    The next pic shows one of the AC OUTPUT wires from the welder (the wires that USED to go from the welder to your ground clamp and rod holder) at the lower part of the picture. This step is to connect ONE of your AC WELDER OUTPUT wires to ONE of the loose pigtails and ONE of the loose CASE BOLTS on a different diode.

    X ddiodepath2.jpg

    ddiodepath1.jpg



    So now you have ONE of the AC WELDER OUTPUT cables hooked to BOTH the DC OUT cables thru two diodes. Don't worry if it now looks like your are about to short-out something, those diodes are one way valves to prevent that. When your AC wire is positive the diodes will send it to the electrode, when the AC cable is NEG it will go straight to the ground clamp.

    Next step - next pic (above I think). The next pick shows the OTHER AC cable being hooked up to the remaining pigtail of one diode and the last remaining case of the last diode on the OTHER side. Again, when this AC wire is POS the diodes send it to the electrode holder, and when the AC wire becomes NEG the diodes send it to the GROUND CLAMP.

    Quick summary-- The left side AC cable gets hooked up to one pigtail and one diode case. Then the right side AC cable gets hooked up to the last remaining ONE PIGTAIL AND ONE DIODE CASE.



    I am going to attach two more pics showing some RED wire being used to illustrate how the current flows as the AC cycles back and forth. I hope it shows it clearly enough to understand. No matter which AC wire is POS it will be directed to the welding rod. No matter which AC cable goes NEG it will be channelled to the ground clamp.

    I sure hope this comes through well enough to make sense to those who see it. The red line in the first pic below shows the path when the right AC wire goes positive. The red line in second pic below shows the path when the left AC wire goes positive.

    X ddiodeacpatha.jpg

    ddiodepathb.jpg


    Hooked up this way, no matter which AC wire turns POSitive, it will be channeled to the electrode, and no matter which AC wire turns NEG, it will be channelled to the ground clamp. That is what is called a FULL WAVE RECTIFIER because it takes ALL the electrical pulses and redirects them into a smooth DC current.



    I will put up this one post, look at the results, and put up another for the next steps I used on my welder. I'll be right back.
     
  2. pitman
    Joined: May 14, 2006
    Posts: 5,148

    pitman

    Very clear info, thanks and looking forward to your next entries.
     
  3. The setup I showed above is what I have used for many years on an AC arc welder I bought new in 1968. Back then I used whatever diodes I could get my hands on.

    Since I didn't use any heat sink to keep the diodes cool, and they didn't seem to get overly hot when I welded only 10-15 minutes at a time, I didn't pay any attention to them for quite a few years.

    One day after about 8-10 years of reliable use, I flipped on the welder, it buzzed real loud and sounded like a dead short, as if I had the electrode holder laying on top of the ground clamp by accident.

    Nope, that wasn't it. Maybe there was damage in one of the cables I had running across the shop. Nope, not that either.

    I ended up pulling all the diodes apart and finding out that one of them had failed. When a diode fails it doesn't fail OPEN, it fails as a SHORT. In other words when it fails, it doesn't block the current, it will allow current to go in both directions and act as a short across the two AC OUTPUT lines coming out of the welder.

    That is no big deal really, since that happens to your welder all the time when a rod sticks, but it can sound alarming if it catches you by surprise and you don't know what's doing it. Ten years from now if it ever happens, you will know to check the diodes to see which one to replace. No big deal.

    Diodes last a whole lot longer if you keep them cool. For years and years I never used any heat sinks. If you do a lot of welding for long periods or if you feel the diodes getting warm at all, you will want to get a piece of aluminum, drill a hole and mount a diode to it. That is all it takes to keep a diode cool. Don't mount all the diodes to the same plate of course or you will short-out your rectifier circuits. You can mount them each on their own plates and wire them exactly as described above.

    8 or ten years on some used diodes I got for almost free wasn't bad.

    Well, after a number of years of using my diodes simply wired and bolted to the cables, without heat sinks, I had a diode go bad recently. I went to the coal mine repair shop, and asked for any discarded diodes. Instead of giving me a couple diodes (one for a spare), they showed me some large diodes mounted on some large heat sinks they had in their "spare parts" pile. They said "help yourself", so I sent them a couple Pizza Hut pizzas the next day.

    I brought home some heavy duty parts that were big enough to last a lifetime with my light-duty use.

    I will put up some pics of my NEW setup. It will look more invloved than the "old reliable" setup I described in my last post, but it is wired exactly the same, only with bigger parts.

    If you find some bigger parts and want to copy my new setup, please refer to my first post for the wiring diagram first. If you look at this post first, the procedures may not look so clear.

    The first pic shows my most recent find.
    I picked up three heat sinks with two diodes in each heat sink.
    The pic shows two of them. I will use these two for my new rectifier.

    d1diodestwo.jpg

    OK, so I am now once again using two pair of diodes just like shown in the first post. Looking at the first pic in the first post, you can see that one of the DC cables has two diode pigtails hooked to it, and one DC cable has two CASES hooked to it.

    ddiodesfirst.jpg

    I will now duplicate that with my new heat sink mounted diodes.

    d1diodepigtails.jpg

    On this one (above) I disconnected the copper bar that connected the two pigtails together. This way I ended up with a diode pair that has both CASES connected together by the heat sink. If I bolt one DC welder cable to the body of the heat sink, I will have an exact duplicate of the the first pic that shows two CASES hooked up to one DC cable. Since these diodes also have a positive case and negative pigtails, this cable from this heat sink (case) will run to my POS electrode holder to give me DC REVERSE POLARITY to go with my favorite welding rods.

    According to the very first pic in the very first post, the next hookup needs to be the "other" (ground) DC cable hooked up to a pair of diodes that have two pigtails together and the CASES kept apart.

    Like this.... I hacksawed the case in two and kept the two halves apart... pigtails connected together (bolted to brass bar), and cases kept apart....

    ddiodescut1.jpg

    I left the brass bar connecting the two pigtails together and then cut the heat sink to separate the two diode CASES from each other.
    Now I have the "TWO PIGTAILS TOGETHER, AND SEPARATED CASES" that the "other" DC cable needs hooked to it. Don't worry, the brass bar is NOT bolted to the heat sink case. If you look closely you can see the insulator under the brass bar so the brass bar is hooked ONLY to the two pigtails. This pair of pigtails will connect to the DC cable going to my GROUND CLAMP. I bolt that ground cable to the brass bar.

    Neat, huh?

    So then I mount the heat sinks and diodes to a piece of 3/4 plywood that I mounted on the garage wall.

    I will go and look for more pics to show how I hooked these up. I do want to point out that this new set is what I am currently using on my welder today. I did a couple hours of welding today so the circuit I will show in my next post is indeed a correctly wired duplicate of the full wave rectifier I have been using for many years.
     
  4. Here's what I mean...

    Here is one of the new diodes pair and a pic of the equivalent wiring explanation below it...

    The diode pair and heat sink with the case CUT in two, and pigtails together is equivalent to the DC cable with two pigtails attached to it in the first pic in the first post.

    The heat sink in one piece and the pigtails separated is equivalent to the DC cable in the first pic with two cases bolted to the cable.


    ddiodeequivalentbar.jpg

    ddiodeequivalentgrnd.jpg

    Am I being clear enough or am I trying to be too simple?
     

  5. Thats cool as hell.
     
  6. I don't know if this will help or if this will make it look more confused, but this is a pic of the finished "RIG" I have on my welder right now.

    Again- remember that this one is massive overkill simply because I got these oversized parts for almost free.
    For many years I ran with a much smaller version with smaller diodes.
    That smaller version only took up a few inches of room, while this oversize one takes up quite a bit of room.
    Yours doesn't have to be this huge.


    dd1rectifier.jpg

    It may look like a bit of a mess right now, but it was put together fast to get my welder up and running again quickly.

    The AC from the welder is coming up from the bottom.
    The two leads going upwards to the ceiling are the ones going to my welding leads for the ground clamp and the rod holder.

    All three copper connector bars you see with the different cables bolted to are on insulators and do not make contact with the CASES (heat sinks) of any diodes.

    One of the DC OUTPUT cables that you see going upwards (to the welding leads) is bolted to the diode case (heat sink) on the left. That is the positive output to my rod holder.

    The other DC OUT cable at the top is bolted to the brass bar that has two diode pigtails bolted to it on the right side of the pic.
    That is the NEG that goes to my ground clamp.

    So far that part is easy to follow.

    The two AC leads coming up from the bottom of the pic are the output leads from the AC welder.

    According to my previous posts describing the circuit, one AC cable needs to be hooked to one of the unused pigtails (on the left), and one of the unused diode cases (on the right) (the heat sink I cut in half to make two has no connections to any cable so far).

    Then the other AC cable coming up from underneath, will get hooked up to the last free pigtail (on the left) and the last remaining heat sink case (on the right).

    So in the pic you can see one AC cable bolted to the lower left brass bar which is mounted on an insulator, also bolted to the bar (as an electrical connection) is one pigtail on the left diode and a short jumper cable going to the diode heat-sink-diode-case on the lower right.

    The OTHER AC cable is bolted to the copper bar on the upper left (also bolted to an insulator so it does not contact the heat sink). That bar also has one pigtail connected to it from the upper left diode, and a short jumper cable running to the case (heat sink) of the upper right diode.

    The copper bar on the right side is also insulated from the heat sink, and has the two "pigtails to the ground clamp" supplying power to the DC cable running upward.

    This is a DC REVERSE setup.
    If you want to reverse it to get a DC STRAIGHT POLARITY (I think thats what they call it), simply trade the two DC cables going out the top. DO NOT change the AC cables as that will have no affect at all.
    Switch only the top two cables to switch polarity.
     
  7. So this massive thing

    dd1rectifier.jpg


    is exactly the same as this little thing

    ddiodepathb.jpg

    that I have used for many years.

    I think I am finally finished.
     
  8. Dan10
    Joined: Aug 14, 2007
    Posts: 386

    Dan10
    Member
    from Joplin

    Being that I know very little about welding, can you explain why you do this?
     
  9. Good thing I went back to look at my post for possible spelling errors.
    I usually don't go back and see these replies.

    I like the smoothness of DC welding.

    When I used to weld with AC, I didn't like the degree of difficulty due to the "rough" sort of electrical current coming out, or the fact that I couldn't use certain welding rods.

    With the ability to use DC you can select from a wide variety of rods that do not work well with AC.

    Maybe it's just me, but my welds are a whole lot smoother when I use DC.
     
  10. INXS
    Joined: Dec 3, 2005
    Posts: 348

    INXS
    Member

    Awesome stuff!

    I nominate you for the "Home Hobblers of the year" award.
     
  11. I know I said I was finished, but there is one more thing...

    By the way, thanks for the good comments.

    If you do happen upon some good heavy duty diodes and have no way to translate the numbers printed on them to figure out if the amp ratings are heavy enough to work on a welder, you can get an idea by looking at the diam of the mounting post built in to the diode.

    I know, I know, right now there may be someone reading this that wants to jump in to say that there is such a wide range of amp values that it is stupid to go by bolt size. How dumb etc etc etc.

    But the fact remains that if you see an axle shaft that is 1/8 inch diam it probably belongs to a toy slot car, and if you see one that is 4" diam it likely belongs to something like a dump truck.

    Now that we know that this is very unscientific, quite redneck, and beneath the dignity of just about any "expert", I want to say that the 275 amp diodes in one auction on ebay look to be about 5/8 diam on that fuzzy ruler they used. My 300-plus amp diodes have a 3/4 stud, and that the smaller diodes I used to use for years had something like a 7/16 stud.

    Maybe it is safe to say that if the diodes you find have 1/4 inch or 3/8 studs, you will likely overpower them and burn them out quickly. look for something bigger.
    - If they have 1/2 inch studs they will probably be fine if you keep them reasonably cool, And if they have 5/8 or 3/4 studs, they will probaly be more than big enough to handle the job.

    None of this is an absolute certainty in every single case, but if you don't have access to your Orphan Annie secret decoder ring to translate the diode part numbers, this will give you an idea whether it is safe to gamble your pizza money on a certain set of diodes.
    If you lose your guess, and burn one out early, you are only out the price of a cheeseburger.
     
  12. J'st Wandering
    Joined: Jan 28, 2004
    Posts: 1,772

    J'st Wandering
    Member

    When welding with AC, the current flow is running both direction, from electrode to what you are welding and also the reverse direction. AC --> alternating current

    When welding with DC, the current flow is always from the electrode. Therefore much easier to well overhead and vertical. Also gives you a smoother flow. You can hear the difference.

    Neal
     
  13. Awesome info and a great inexpensive upgrade.

    Question - could this be used to convert an AC Tig to square wave DC output to weld aluminum?

    Steve
     
  14. I don't think this will give you a square wave. In fact I am 99% sure it won't.
    I am no electrical expert, but I think it will look like sine waves that are all pushed closely together (yeah go ahead and laugh at the language).

    This full-wave DC won't be nearly as choppy as AC or crudely-intermittent (not full-wave) DC, but it will still have the waves that are associated with changing currents. A lot more waves that are run together, so it is less choppy, but still waves.

    Doesn't a square wave require an instant-on, a level power value, and then instant-off? How they do that I don't really know... charge-up and discharge capacitors in the circuit, or ???

    This full wave rectifier will give multiple waves that are smoother than the chopped up AC.

    ***************

    I have no familiarity with TIG either, so I can't answer any TIG questions.
     

Share This Page

Register now to get rid of these ads!

Archive

Copyright © 1995-2021 The Jalopy Journal: Steal our stuff, we'll kick your teeth in. Terms of Service. Privacy Policy.

Atomic Industry
Forum software by XenForo™ ©2010-2014 XenForo Ltd.