Copper bar soldering problem

[s3rkan]

Hello everyone. I have just received this copper bar from AliExpress:

 http://s.aliexpress.com/3iQrQFBz

I want to make a mosfet switched spot welder and will use this bar for mosfet drain and source. But even I have tried all I have, like soldering paste, Flux, grinding the surface and HCl, I couldn't make stick the solder to the "copper" bar. I tried it between 320-350 C degrees. Does any one have any idea what I am doing wrong?

[Dave]

Copper is an outstanding conductor of heat. My bet is that even though your iron is set to 350°C, the copper bar does not get anywhere close to that temperature.
A really big iron would be my only idea.

[Gyro]

Don't forget to check them with a magnet. 

You probably have issues with oxide formation during the heating time (you are heating the whole bar aren't you?) try abrading and fluxing while hot. If the temperatures you're quoting are just for a soldering iron, then you need to pre-heat the bar too.

[The Soulman]

there is a lot of thermal mass and on top of that copper is a great heat conductor.
In order make a solder connection almost the entire thing needs to be brought up to temperature.
Best chance is a oven and low melting temp. solder.
Last option is blowtorch, but your mosfets probably wont like that.

[filssavi]

It’s hard to tell for shure without seeing anything but my guess is not even remotely enough heat/power

Try with 400 degC, if it still doesn’t work you could try to preheat the bar with a heat gun so that the delta T is lower

[sokoloff]

If pre-heating (oven or torch) the bar doesn't work, I'd consider wrapping the MOSFET in Kapton/other shield and trying a torch.

[Twoflower]

I would be a bit careful and do some tests on cheap parts. If you preheat the copper bar it stays hot for some time. That might be enough to roast the devices. It would be very interesting to see to maintain an acceptable temperature curve.

I think you have to mount all devices in one shot to minimize the thermal stress. My idea would be to use solder paste, fix all the devices to the copper bar (e.g. metal clamps) and heat the bar from the backside using a hot air gun until the solder paste melts. Now using a fan to cool it down (not too fast, not to slow).

[Karlo_Moharic]

use propane TORCH !!!!!!!
Soldering irons aren't made to give out that much heat.

[james_s]

When I made my capacitive discharge spot welder, I ran into a similar problem trying to solder the capacitors to the bus bars. What I ended up doing is wrapping two turns of the heaviest wire I had around the core of a toroidal power transformer and then touching one wire to each side of the place I was trying to solder. A few hundred Amps heated the heavy bus wire directly to the point where a soldering iron could finish the job.

[dmills]

The way the RF power fraternity tend to handle this when soldering big RF power FETs to heat spreaders is with a hot plate, a big chunk of ally (as a heat store) and a spare large heatsink (for cooldown).

There are youtube videos showing the process, search for something like "soldering ldmos" from W6PQL or similar.

Regards, Dan.

[s3rkan]

Thank you all guys, I'll try all the advices. I want to make like this guy (after 2:30)

https://youtu.be/pQ53sk9ksxQ

He does the soldering with out any problem. I can screw the drain to the bar but source pins need to be soldered.

[Gyro]

It looks as if he's using much thinner copper.

[BravoV]

Don't forget to check them with a magnet. 

+1 

[s3rkan]

Don't forget to check them with a magnet. 

Sorry for this noob question, for what I'll check with magnet?

[helius]

If the bars are fake (iron alloy covered in copper) they will be attracted to a magnet.

[jmelson]

Hello everyone. I have just received this copper bar from AliExpress:

 http://s.aliexpress.com/3iQrQFBz

I want to make a mosfet switched spot welder and will use this bar for mosfet drain and source. But even I have tried all I have, like soldering paste, Flux, grinding the surface and HCl, I couldn't make stick the solder to the "copper" bar. I tried it between 320-350 C degrees. Does any one have any idea what I am doing wrong?

No electronics soldering iron can possibly solder to this type of bar.  You need either a "plumber's" soldering iron with 100 - 200 W or a Weller-style soldering gun (transformer-type).  I have used those tools to solder to bar like that, and it demands the full output of these high-power irons to heat the copper enough to make proper solder joints.

Jon

[Jwillis]

Drill holes in the copper, tap some threads.solder eyelets on the mosfets and screw them on to the copper bars.Minimal heat required.Simple chip replacement .

[GerryBags]

I have one of these cheap butane irons for soldering to thick stuff. I've used it to solder heavy guage earth cable to an A4 sized 2mm thick copper sheet without any difficulty at all. It can make a fine flame that you can be fairly precise with too.

[JS]

Drill holes in the copper, tap some threads.solder eyelets on the mosfets and screw them on to the copper bars.Minimal heat required.Simple chip replacement .

Something like that was my first thought, I don't know if I'd even try the soldering. I don't know your mosfets, but they might have the drain connected to the tab, there goes your screw, now you have to cool the bar, or make the heatsink the bus.

JS

[cncjerry]

Drill holes in the copper, tap some threads.solder eyelets on the mosfets and screw them on to the copper bars.Minimal heat required.Simple chip replacement .

use this ^^^^^^^ or low temp solder.

[Bud]

when i did not have a propane blow torch i did some plumbing repair using a  blow heat gun, it heated copper pipes enough to resolder joints and valves.

[IDEngineer]

There's a reason they use propane or MAPP gas torches when soldering copper plumbing pipes. All of the above comments about the heat being conducted away are exactly correct. And your bus bar has far more cross sectional area, and is thus a better heat sink, than copper pipe.

When soldering copper pipe with a torch, I usually protect the flow of heat away from the joint with wet rags. Granted in that case I'm trying to prevent the heat from spreading to and damaging sensitive and expensive components like ball valves, and what's really happening is that the water in the rags converts to steam which carries away the heat. But thinking about it, wet rags might also create a region of colder temperature and the resulting gradient might prevent heat from conducting away so easily from your work area.

No matter what, though, you're going to need much more heat. And by that, I mean you need a HOTTER heat source. Equal area under the curve is not equivalent here; slowly raising the temperature with an insufficient heat source just allows the flux to evaporate, and as you've discovered solder doesn't flow well onto unfluxed bare copper. You need to properly prepare the surfaces, then quickly raise them to the necessary temperature to flow the solder so the flux is still "acting" when the solder begins to melt.

I'd treat this exactly like soldering copper pipe: Sand or otherwise mechanically clean the surfaces, apply lots of flux, apply intense heat for a short time, flow the solder, and allow the joint to cool.

[s3rkan]

Thank you all friends. I got the point. First I have tried to make a solder mask on the surface with a 200w soldering iron and thought maybe it can be more easy to solder in that way but doing this didn't make any difference. Also soldering the mosfets with a high power soldering iron could be a detrimental process for the FETs so I made the system with M3 tapping the holes and screwing the mosfets. Relevant pictures are attached.

[JS]

Nicely done! check contact resistance if you have a way to do so, (4W measurement) so you know if you have some oxidation increasing the contact resistance or poor pressure in some contact.

JS

[Ian.M]

Solder cold-flows which will remove most of the contact pressure with time, so take it apart, heat up each copper bar with a torch or hotplate, flux it and wipe as much solder off the surface as possible so you are only left with the thinnest possible tinned surface.   Then reassemble using Belleville spring washers on top of your plain washers to maintain contact pressure.  Also, you need to pack^* under the opposite side of the washers holding the MOSFET leads to the same thickness as the lead so they aren't sitting at an angle. 

* e.g an off-cut of wire, with a drop of glue to hold it in place during assembly.