[517]

My homemade bridge rectifier to convert my AC stick welder into a DC source using SCR's is working. The SCR's were bought surplus from Princess Auto for $9.99 each. I used a 300 ohm resistor and a small diode to connect the gate of each SCR to the anode to convert them into simple diodes. I mounted them to 2.5" aluminum channel for a heat sink and have temporarily mounted it to a board for testing.

I ran a bead with some 3/32 6013 rod first to see if the diodes heated up. At 80 amps, no trace of heat. I turned it up to 100 amps and burned a full rod. Barely perceptable that it was warming up the heat sinks. I then switched to a 1/8 7014 rod and turned the welder up to 130 amps input. After burning the rod up, I checked the heat sinks. They were now warm, and the diodes were hot to the touch. That is probably the limits. I will have to use some heat sink compound on the base of the diodes to make sure they are conducting heat will to the aluminum channel. I ran the stick to the "-" side and the ground clamp to the positive output. I sure seemed effortless to start the arc and there was no tendancy of the rod to stick to the work on first striking the arc. I wondering if this is a result of the reverse polarity setup?

Next step is to experiment with using a lamp dimmer to control the input of the gate of an SCR to control the output. The lamp dimmer will chop up the ac phase to produce ac pulses at 60 hz, but of varying duration. I will run this thru a bridge rectifier to get dc pulses to input to the gate.

Ultimately the goal is to power an air cooled tig torch.

Anyone know the limits of material I can weld using 110 amps? I'm thinking that should be good to around 3/16 on mild steel or stainless.

Here's a picture of the bridge as tested this afternoon.

 

[Dirt Dog]

Looks like your getting it.

I'm interested to see how the thing as a whole ends up doing.

 

[010]

what is a scr?

 

[517]

It's a semiconductor, similar to a diode in that it only flows current in one direction. It is different in that it has a third connection, called the gate, which must have positive bias in order for the scr to function. So, essentially, it's like a solid state relay, but it only flows in one direction. You can use it as a very dependable electrical switch, that operates at high speeds. They use them on power control circuits. Sorry, I'm not an electrical engineer, just a backyard mechanic, so that's as good as my explanation gets.

High current diodes are expensive new. You can sometimes get them or scr cheap if they are surplus or pulled from service. I'm trying to have fun and scrounge together something usefull on the cheap.

 

[Jim_Lou]

With 110 amps you'll be pushing it to do 3/16"; a more realistic goal would be 1/8". Here's the chart out of a Miller handbook:

And just for good measure, the chart for Stainless:

You can push it some, but it gets slow; you have to slowly walk the puddle back and forth across the joint. The time and heat worsen the warping and distortion.

 

[Dex]

Hey Tim, what does a rectifier do? hey and O/T ,,i been way up there several times to Winnipeg back in late '90s.. stayed in that tallest building beside both rivers a close to the beer store. hey she was very nice lady her name was Mona.
ok very nice work dude!!!
DEX

 

[517]

A rectifier is a device that converts alternating current (AC) into Direct Current (DC). For some welding applications, DC is necessary. DC welders tend to cost a lot more and are harder to find used. I already have an old AC unit, and am interested in trying tig welding steel, which requires DC.

Sounds like you had a good time in Winnipeg. Not quite sure if you met Mona in Winnipeg or brought her along. You must have stayed close to the Forks. Nice area to go for a walk by the river, do some shopping or have a drink / dinner.

 

[Dex]

ill show my dad your hook up. heck im still doing kinda good at stainless steal rods, self tought, at my factory when ever i can get over to shop
[email protected] stay in touch
i make glass ber bottles owens illinose

 

[JimmyZ]

Could you show a wiring diagram plzzzzzzzzzzzzzzzzz?

 

[517]

I'd like to make a drawing, but I'm not sure how to do it. Are there any programs out there for drawing schematics easily? Hoping there is something shareware or at least free. I usually just draw it on paper. I did, but it look pretty ugly.

 

[251]

ad a couple of cheapy muffin fans to that little array there, and increase the duty cycle!

Ive already got the DC machine, when you start attacking the Hi freq arc starter Ill follow more closely.

 

[KEG]

Sounds like everything is going just as you hoped. The 110 amp rating you quoted earlier is for 100% duty cycle. If you exceed this rating you must let the device cool accordingly. I talked to another guy at work this morning about what you were doing and he reminded me about how SCRs fail. I told you that I had only had to replace 2 or 3 in my experience with DC drives that use SCRs for power conversion. It kinda slipped my mind that in all of these cases the SCRs had shorted. All drive systems that I work on have protection in case of this event, a shorted SCR. I would suggest you install some kind of IOC protection after your AC welder and before you bridge. This would protect your welder from a possible melt down before the breaker in your service entrance decided to kick out.

 

[517]

Thanks for the idea Keg, some kind of fuse or link.

After this, it's on to the power control of the SCR's.

Once that is complete, the hi frequency unit is next. I have seen a schematic that uses a neon sign transformer, some hv caps, a spark gap and a hand made air core transformer to couple the high frequ into the welder cable. It's do-able.

 

[517]

Yes, muffin fans. There are some at the surplus store as well.

Here's the hi-freq link

web page

 

[014]

You could try www.orcad.com

I believe that they have a student version of their software that is free. It is not the easiest thing to use though. I guess I would suggest using powerpoint, that's the way I do the simpler electronics schematics.

 

[KEG]

I hadn't thought about it when I posted yesterday but doesn't a welder have some way of controlling current. I mean when they are in operation don't they just supply x amount of current depending on the front panel setpoint. If that is the case one could short the leads together and not burn one up. I guess I am just fuzzy on how a welder works. I have never given much thought as to how they operate, I just worried about making a nice weld.

 

[517]

I believe that stick welders have the transformer designed in such a manner as to be current limiting. The voltage is high on an open circuit, but once connected to a load or short some aspect of the transformer design limits current flow. I'm using one with a variable core to select current flow. I really don't know, but I suspect that if I shorted the leads, it would only flow the selected current, and the danger would be exceeding the duty cycle.

 

[KEG]

Sounds reasonable enough. I guess since the welder is only going to supply X amount of current, then a fuse would not help because current should not exceed the welders max output current. I do not think one would want to fuse for less than max output of the welder. Does this sound right? The welder probably has some sort of thermal trip incase the duty cycle is exceeeded. Would this provide enough protection?

 

[517]

Perhaps some sort of thermally based cutoff on one of the heat sinks? At least it would kill the power if an SCR is overheating and before it melts down, for whatever cause, failure of the device or excessive current consumption. Any idea what exists? I've seen devices stuck to the side of transformers for this purpose. I'm handling a lot of current and they usually put them on the primary side.

I picked up another scr and a slider type dimmer switch. I'm still looking for an inexpensive isolation transformer so that I'm not coupling line current into the welding line via the scrs.

My plan is to run 120volts into the dimmer, reduce the voltage via a transformer for safety reasons (how will the dimmer like feeding into a transformer??), then rectify the output. Then limit current to approx 100 ma using a resistor and bias it accross the scr's gate and cathode. I'll see how this works on my spare scr before hooking it up to the bridge and ruining anything. If the slide dimmer works, I'll build a pedal assembly around it.

 

[KEG]

I looked in the Newark catalog and they have thermal trip units for 5 or 6 bucks. Newark will sell retail but the only bummer is they have a minimum order limit of $25. I forgot the copy of the page at work but if you are interested I will attach a scan next time.

I thought you were going to power the bridge through an AC welder if so you wouldn’t need an isolation transformer would you? If you are not using an AC welder then the fuse dealy might not be a bad idea.

The dimmer that was in the drawing was a triac which had an adjustable firing angle. Wasn’t it used to supply an adjustable firing angle to the main SCRs? Voltage supplied to the gate is not what makes an SCR supply different current levels it is the firing angle supplied to the gate. Its more a condition of time instead of voltage, when the firing pulse arrives is more important than the level of the pulse. I think the gate is listed as having a minimum firing voltage I do not know about maximum. I would not think ½ cycle peak of 120vac supplied to the gate would be excessive on a 600v SCR.