# Using microwave xmfr for resistance soldering?

Greetings to all the microwave oven scroungers, I have a job coming up that requires soldering the corners of brass frames. These frames are made of 1 1/2 x 1 1/2 x 1/4 brass angle. In
the past I've used a torch to heat the brass. This leads to a little distortion. This can be troublesome because the brass expands so much that it moves the joint even though everything is clamped. The parts must then be straightened when cool. Because of the application and fit the frames must be straight within .010" in 36". I saw some resistance soldering units in a catalog and the description of how they operate says that only a small area is heated and gets to heat fast. This would be ideal. However, the ones I saw were too small and the price too large for me. But I've got several microwave transformers and they seem like they might be perfect. A rough calculation from the specs and pictures given in the catalog leads me to believe that they output about 12 volts open circuit. Some have variable outputs. So I have a few questions: 1) Does 12 volts sound reasonable? Would a different voltage be better? 2) Is DC better than AC? Does it matter? 3)What would be good ways to limit the current? Would a lamp dimmer on the input side of the xmfr work? Wouldn't that also lower the voltage? Would that matter? 4) I have a timer that pulses a relay on and off. I can set the length of the pulses. Sort of what a lamp dimmer does but much longer pulses (1 second and up) and the voltage would be the same out as in. But the brass would average the heating. Would this work almost as well as lowering the current? Better? Any other input is much appreciated. Thank You, Eric R Snow, E T Precision Machine
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
wrote:

How are you planning to connect the transformer to get 12 volts? Put power to the high voltage secondary?
If you put power to the normal primary you get 10s of thousands of volts at relatively low current. Only way to use a microwave transformer for resistance soldering is to remove the HV secondary and install a very husky secondary of several windings - to give you a couple volts at very high current.
AC or DC is not important and a"motor duty" dimmer can be used to control output. A variac works better.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 14 Jul 2004 02:01:46 GMT, snipped-for-privacy@snyder.on.ca wrote:

I will remove the secondary and wind my own. I have done this to make a spot welder. It works well. ERS
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
on Tuesday 13 July 2004 06:02 pm, Eric R Snow wrote:

All you need is to work on your torch technique. Try preheating. There's a NG - news:sci.engr.joining.welding where they discuss this sort of thing. And don't use OA - use propane/air. (i.e. Bernz-O- Matic o.e.)
I've seen spot welders (well, web pages with them) that used a microwave transformer, and they replaced the secondary with enough turns of wire to get 4V. That sounds like a good starting ballpark. Use #2 or #4 weld cable, and big copper electrodes.
A simple timed on-off switch should work, like a 555 one-shot, tranny, and relay (or SSR). You control the heat by on-time. You might even be able to use the SSR out of the microwave itself. :-)
--
Have Fun!
Rich

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

My technique is not the problem. Even preheating the whole frame does not solve the problem. I'm good at this. Had lots of practice. It's because of the really close straightness tolerance. ERS
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
I dunno much about the specific outputs/capacity of the resistance soldering units, but I'll throw a few observations into the thread. The units that I've seen in use in a starter/generator repair shop were fairly compact units with a transformer housing of about a 6" cube. The plier/tweezer-style handpiece appeared to have carbon jaws that made the contact to the joint to be soldered. The heat was rapid, and they used heavy gauge solder, 1/8" maybe.
Some butt welding machines that I used to repair were used to weld ends of heavy steel wire together (similar to a bandsaw blade welder). The sizes ranged up to about 1/4" diameter. The secondary of the transformer was only a couple of turns of flat braided cable securely clamped at the ends. The cable was the type that was used as engine ground strap in autos decades ago. This stuff would be good for using as a secondary winding in a modified transformer, easy to thread thru the frame aand flat for fitting into a square shape. For insulation, a good product would be fiberglas tape.. thin, high temp resistant and an effective barrier/insulator.
WB .................

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
resistance solder units that I have output between 1 and 2.5 V

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 14 Jul 2004 22:14:16 -0700, "william_b_noble"

Thank You. That's just what I was looking for. ERS
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 14 Jul 2004 22:14:16 -0700, "william_b_noble"

Reading the original post it seems that it's soldering that you need rather than flash or resistance butt welding which would need a much higher peak power input.
It takes time for heat to travel and distribute itself into a workpiece. The shorter the heating time the smaller the volume of the workpiece affected and, for a defined temperature rise, the smallest total heat input and smallest heat affected volume. This means that the aim should be for a high peak power input to permit a very short heating time.
Some resistance soldering units use either one or a pair of carbon electrodes sharpened to a point. These can apply intense local heat but it flows non uniformly into the brass and can only melt the solder after heating a relatively large volume of brass.
A better approach (and probably the one you're already intending to try) is to resistance melt the solder directly by clamping the transformer output leads to the brass frame just either side of the joint and applying pressure and time controlled current pulse for a second or so. If you can succeed in mostly filling with copper the vacant space left by the high voltage secondary AND removing any magnetic shunt pieces, a single microwave oven transformer should be enough. You probably need about three volts. These transformers are typically about 1 turn per volt so two to four turns is the right range.
If there's a choice solder should be in the form of flat preform lightly fluxed on both sides.
This is all partly informed guesswork so let us know how you get on. Jim
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 21 Jul 2004 12:33:50 +0000 (UTC), snipped-for-privacy@yahoo.com wrote:

Thanks for the input Jim. You have hit exactly on the head what I'm trying to do. Often it is hard, for me, to get an idea across consisely. And I end up clarifying over and over. It must be hard to be a teacher. Since I have already stripped out the secondaries and shunts of two xmfrs I'll be wiring them in parallel. I will then machine two copper electrodes to match the shape of the pieces to be soldered and clamp the wires to them. I hope that getting the electrodes within 1/4" of the joint will get me lots of heat fast. The method of joining will be to use small .002" thick shim pieces in the joint to provide capillary action. I have done this when using a torch and it works very well. Eric
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Eric, I had a simillar idea a few months back when I needed a spot welder. The main thing with using a microwave xformer based welder is the power output, the "big" xformers are rated for around 1500 VAs. That's O.K. for spot welding thin sheet metal & thats what the plans available on the web seem to be aimed at, such as
http://www.5bears.com/welder.htm
1/4" brass is probably more than a single transformer home built will handle. As a comparison, commercial units rated for spot welding 3/16" steel are rated around 2500 VAs w/ a 250 V primary. I ended up buying a unit off of Ebay, they seem to go for about \$70. Thats probably a lot cheaper & certainly quicker than trying to roll my own.
If you're trying to butt weld the brass as I suspect then I really have no idea how much juice you'll need but you're certainly talking about a lot more than you'll get from a m/w trans. or the 70 bucks on Ebay.
Howard.
wrote:

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
1. I suspect 12 volts is a bit high. 2. AC ought to be as good as DC. 3. Use fewer turns on the secondary. 4 Set the length of the pulse so one pulse is the right amount of heat. 5. Use very heavy wire. 6 gauge or bigger. Might be easier to wind three or more #10 wires in parallel.
Dan

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Required voltage depends on whether the heat is produced by the resistance of the workpiece (brass) or by the contact which might be copper, SS, tungsten, or even carbon.
I have a Wassco Glo-Melt resistance soldering unit model 105-B2, 500 watts. It's output is variable in steps from 1.0 to 5.8 volts AC.
A microwave transformer's capacity will depend on the power rating of the microwave from which it came. They vary from 600 watts to 1200 watts. I don't know if that power rating is input to or output from the magnetron. A given sized core can only handle so many VA at given frequency (60 Hz), but I'd think a few experiments with such a trannie and your own lowvoltage secondary would have you knowing a good deal more than the "experts", at least about your particular application!
I'd wind a one-turn secondary and measure that voltage as a reference point for volts per turn. Secondary material is up to the guy winding it, and if you built a spot welder than you know the drill.
I'd then experiment with electrode material and configuration. That may affect your results as much as available power.
wrote:

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

No, it can handle a certain number of turns per volt (which of course has a factor of frequency, yadda yadda, but we're at 50/60Hz only so forget about it). Current depends on the wire. That's why you can short a transformer and get the windings smokin'...

Personally I'd probably melt some aluminum wire into bars (I've got 13 pounds as weighed, soft crap to burn.... literally? HMMMM.....) and wrap that around the core, a turn or two. Some duct tape would make good insulation I bet. Most MOT welders have uh...3-5T? Maybe pile up four, run them in series-parallel off 240V and loop some heavy welding cable through the space where the secondaries used to be.
Tim
-- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Tue, 20 Jul 2004 02:12:47 -0500, "Tim Williams"

Well, golly, do you suppose I've been designing transformers wrong for 35 years?
Given 60 Hz, and given max flux density determined by the core material , volts per turn (or turns per volt if you prefer) are a fn of cross sectional area enclosed by the winding. However, for more VA (at given voltage) you need to use larger wire for those turns. This dictates more "window" area so the overall dimensions (envelope), volume and weight of the core must be larger to make a larger window.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

^^^^^^^^^^^ Given the duty cycle, it can be ran "safely" at a rather abhorrent current density. By "safely" I mean, if it melts, hell with it. Make another.
Tim
-- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Ya got me there, Tiim. I've never designed one-shot transformers.
On Thu, 22 Jul 2004 04:22:38 -0500, "Tim Williams"

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

How about superconducting transformers? I've got to imagine it's been done by now, unless iron loses its ferromagnetic properties at low temperatures (reverse-curie temp? heh). In that case, the wire can be microns in diameter, thousands of turns can be applied in a very small area, while still maintaining a high current density due to the material.
Tim
-- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Tim Williams wrote:

Umm, no. There's a surprisingly low critical current density where the superconductivity disappears.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

As I recall the critical field is up around 20T in the tougher superconductors, I don't know what field is present at and slightly inside the conductor's surface for a given current and diameter though. Suppose I should drag out our old friends Biot-Savart and Mr. Integral...
Tim
-- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

## Site Timeline

• ### v-up 7018 fillet - stringer more productive than weave?

• - last updated thread in ⏣ General Metalworking
• ### Inclusive Urbanism - Ep. 3 - Livecast with Juma Assiago - 8th April 2020

• - the site's newest thread. Posted in AutoCAD discussions
• Share To

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.