reflow flux and other shit

[Uranium-235]

so I got a Sony Viao, I fixed once by keeping the heatsink off the southbridge for a few minutes. Its barely broken but I want to try to do a good reflow (she just texted me it won't come on again). What reflow flux do you suggest? just any ebay flux? Not sure if I want to go as far as reballing or not. I have a good heatgun, should I bother getting an actual hot air rework station? I'm sure I can reflow this with tin foil and my heatgun, but I want to use some flux

[speedypro]

Re: reflow flux and other shit

Its best to reball; or remove any epoxy (glue) around the chip completely then reflow up to 220 degrees celsius. That works for me. If issue returns then you must reball.

[momaka]

Re: reflow flux and other shit

I've used MG Chemicals rosin liquid flux before (not entirely sure on details, though) and Kingbo gel-type RMA-218 flux. The Kingbo seems to work pretty well.
Genuine Amtech is supposed to be really really good. But most on eBay are fake. They will work too, but probably nowhere near as good.

In any case, any flux is probably better than no flux at all.

[Uranium-235]

Re: reflow flux and other shit

yeah I use MG chemicals paste flux for regular soldering. Too pasty to pour under the chip obviously.

I have an ir temp gun coming. I've been thinking about getting a rework station. Newegg has like two brands, and they all look pretty much the exact same casing, just with different wording outside. Whatya reccomend?

I've also seen these cheap, small IR rework stations, I can't find the damn links now, they have a single, small IR heater that more or less stays still, and I don't think does an immediate melt. they were like less than $100 bucks or so

[momaka]

Re: reflow flux and other shit

yeah I use MG chemicals paste flux for regular soldering. Too pasty to pour under the chip obviously.

Not really.

Just line the edges of the chip you're trying to reflow with the paste flux and start heating the board. As the chip gets hotter and the flux starts "melting", the chip will suck the flux under it. I don't know what are the physics behind this for this to happen, but I know it works.

Unless I am mistaken, the stations I see on newegg are all of the hand-held hot air types. Not sure how easy it would be to do a reflow just with that or if it is even possible. If it's a small and very thin laptop motherboard - maybe. Anything bigger usually requires a little more heat flow (not necessarily hotter temperatures, but just bigger quantity of heat per unit of time).

If you have a heatgun, you can use that on the low setting to heat the board from the under side and then use hand-held hot air to heat the chip on top.

An alternative to the heatgun would be an electric stove. Just place the motherboard so that it is above the heating element on the stove (but not touching it), and when you hit around 150C, start using the hot air to heat the motherboard from the top.

If anything, one of the things you should definitely get before attempting any of that above is a type-K thermocouple. If you're willing to wait a little, you can get them on China-eBay as low as $5. The TM-902C is probably the most common one you'll see on there. They are inexpensive, but work pretty good and come with everything you need. The thermocouple probe they come with is a little bit big, so the temperature readings on the meter usually lag a little behind the actual temperature the probe is trying to measure. I know my friend has the smaller thermocouple probes, but I don't know where he got them from. They respond much quicker. But either way, a thermocouple would be very very helpful.

What you will want to avoid is heating up the board too quickly or overheating the chip - those will usually cause delamination, at which point the chip and/or board can be considered dead (whichever develops the delaminations, that is). Another cause for delamination is humidity. If the board has been staying in a fairly humid environment, I would recommend baking or heating the board to around 100C for at least 30 minutes. A few hours would be better, though.

[Uranium-235]

Re: reflow flux and other shit

http://www.youtube.com/watch?v=VTAU647jzzk

fascinating. Seemingly tedious, but fascinating.

[momaka]

Re: reflow flux and other shit

If he had the proper reballing stencils that extend all the way, this could have taken a lot shorted - at least for the reballing part. With those shims, you just pour the balls on the chip, move the entire thing back and forth a few times, then angle, and remove excess balls. About 2 minutes at the most to do. And those bigger reballing stencils are usually similarly priced to the ones he has too.

The lifting/flowing part isn't that hard. I've done RAM chip removal on a nVidia 7600GT on an electric stove top. Plenty of heat and if I had a proper hot air station with me at that time rather than a clumsy heatgun, it could probably have been even easier.

[Uranium-235]

Re: reflow flux and other shit

how did you know what chip it was?

[momaka]

Re: reflow flux and other shit

For the 7600GT?
All 4 of the RAM chips were bad except for 1. Shorted Vdd to ground. Sacon FZ caps can obviously cause quite a bit of havok. The RAM Vdd was only filtered by one Sacon FZ cap, and when it went bad, the buck regulator circuit probably sent a huge voltage spike to the RAM.
Still don't know if that could have damaged the GPU chip as well. Resistances seemed okay with the RAM chips removed, though. So I am hopeful.

[Uranium-235]

Re: reflow flux and other shit

well I practiced on a bad GF 9400, heated under the chip to 100C, got foil, and cut the square for the chip, slowly got it to 220C...but i'm not sure I did it right. I'm not sure the flux got even under the chip, there was probably a few gaps. In the end, a small ball rolled out under the chip. Probably not a good sign. just guessing.

I have a few other bad laptop motherboards to mess with.

what it is is the MG chemicals rosin flux mentioned above, put in a syringe, and injected to one side of the chip, with the board leaning a little to let it move down, then putting it the next side and let it go down again

all I know is it takes only a few seconds for the flux to smoke into nothing. Should I have the foil block the air from getting under the chip?

[momaka]

Re: reflow flux and other shit

well I practiced on a bad GF 9400, heated under the chip to 100C, got foil, and cut the square for the chip, slowly got it to 220C...but i'm not sure I did it right.

220C is a little low, especially with those hand-held hot air wands. If you move the wand away from the chip, did the temperature stay at 220C or did it drop a few degrees immediately? If it drops, it means you need higher heat flow (i.e. more air moving at the same temperature). Having some type of bottom heater for the board is crucial. Also, try going for 225-230C. Do not go over 240C, though, because around 250C you'll cook the chip. Also, you don't really need to put foil around the chip - in fact, I advise against doing so unless you have electrolytic capacitors that will get popped from the heat. Not having the foil allows the board to heat more evenly.

I'm not sure the flux got even under the chip, there was probably a few gaps.

When you heated the chip, did you see the flux melt? If yes, it probably went under.

In the end, a small ball rolled out under the chip. Probably not a good sign. just guessing.

I've never seen that happen before. I'm not even sure if the balls can get detached (and if so, how). I guess lifting the chip will confirm it. If you do it really carefully, you'll be able to do a clean lift (i.e. the balls won't clump to each other).

Before doing that, though - did the board work? Or is it still dead?

I have a few other bad laptop motherboards to mess with.

If the 9400 board is still dead, and you think you messed it up even more, then keep practicing on it. This will get you more experienced for the other boards.

all I know is it takes only a few seconds for the flux to smoke into nothing. Should I have the foil block the air from getting under the chip?

Interesting. My MG Chemicals liquid flux also evaporates fairly quickly. I've done reflows with it, but only on leaded boards where the temperature was much lower (200C tops). Perhaps yours isn't suitable for reflow either.

[Uranium-235]

Re: reflow flux and other shit

maybe that one wasn't suitable. I still have a DV 9000 given to me with pieces lost that i'm sure isn't leaded.

problem with not using foil, sometimes there are plastic connectors close enough to the NB to melt

I still haven't touched the sony yet

and yet the temp drops immediately after

[Uranium-235]

Re: reflow flux and other shit

well I got the motherboard out. There seems to be an electrolytic cap on it, it looks like a polymer casing but its got a vent. Probably 105 max. Not sure what I should do. Still cooking it at 100C is awful close to max, and its close enough to the chip to get hit by the gun

also, what is the melting point of all these plastic connectors?

[Uranium-235]

Re: reflow flux and other shit

helloo?

[momaka]

Re: reflow flux and other shit

100C is quite safe as long as the oven doesn't go much over that. It takes a lot more to pop a cap. At 105C, you will just shorten the life of the cap by a small amount (probably too small to even matter in the useful lifetime of the laptop). I think 150C and over is where they may start to pressurize quite a bit.

If it's just that cap, you can try removing it for the reflow and then putting in back on after it. Normally when there are a lot of caps around, I just cover them with aluminum foil and/or aluminum tape, along with any large copper traces they are connected to on the bottom side.

[sasser]

Re: reflow flux and other shit

For reballing some method need bake the board fisrt before lift and place back bga. If use ir rework machie this must do for remove bga moisture. Different with hot air rework machine no need bake the board. The airflie from hot air remove moisture bga and board. Agree ?

[Th3_uN1Qu3]

Re: reflow flux and other shit

Cover the capacitor with aluminum foil if you're worried about it.

[momaka]

Re: reflow flux and other shit

For reballing some method need bake the board fisrt before lift and place back bga. If use ir rework machie this must do for remove bga moisture. Different with hot air rework machine no need bake the board. The airflie from hot air remove moisture bga and board. Agree ?

I don't know. My thinking is, regardless of whether you use IR or hot air, the board will be heated above 100C, and when that happens, moisture should evaporate all by itself (unless there is just too much).
But again, I don't know. What do you think of that theory?

Also, for really small chips, I don't think it matters. I just removed 4 dead RAM chips from another E-VGA 7600 GT on the stove yesterday. With the hot air wand blowing on top of them for a few seconds, the slid off like butter. With bigger chips, you need more heat, so I can see how that can be a problem.

Cover the capacitor with aluminum foil if you're worried about it.

Yes, but don't squish the aluminum foil tightly on them. In fact, don't squish it at all. Just cover so it deflects hot air and heat out of the way.

[Th3_uN1Qu3]

Re: reflow flux and other shit

I don't know. My thinking is, regardless of whether you use IR or hot air, the board will be heated above 100C, and when that happens, moisture should evaporate all by itself (unless there is just too much).

If there's any significant moisture in it, heat it up too quickly and it will go POP... That's why it is recommended to "bake" boards and chips that have been stored in unknown conditions before attempting a reflow or reball.

[momaka]

Re: reflow flux and other shit

^Yes, I've seen it happen too on some boards I worked on. The state I am in, Virginia, it is quite humid in the Summer.

But my question was, why would hot air not do that vs. IR?