EEVBlog
ฝัง
- เผยแพร่เมื่อ 6 ม.ค. 2013
- Mailbag continued.
Soldering a 0.4mm QFN chipscale SMD package onto a Schmart Board using 0.3mm JBC and Hakko tips.
Also a demonstration of drag soldering with a chisel point tip.
Schmart board: schmartboard.com/
Green Arrays Multi-Processor Chip: www.greenarraychips.com
Forum Topic: www.eevblog.com/forum/blog/eev...
EEVblog Main Web Site:
www.eevblog.com
EEVblog Amazon Store:
astore.amazon.com/eevblogstore-20
Donations:
www.eevblog.com/donations/
Projects:
www.eevblog.com/projects/
Electronics Info Wiki:
www.eevblog.com/wiki/ - วิทยาศาสตร์และเทคโนโลยี
I am quite surprised that anyone even considered to solder this chip pin by pin using conical tip. Thirst thing that came to my mind was drag soldering as this most reliable method if you don't run out of flux.
You can never have too much flux!
I laughed when you said "Crikey," always entertaining Dave.
I have soldered over ten of these chips and I think that the schmartboard method is nothing short of disastrous for me at least. After ditching so many of them that didn't work (5) I just took one of that lot that I was going to throw out, doused it with solderflux, baked it on an electric skillet until it reflowed, and it worked! I took one of the remaining ones that I had yet to solder, flipped it over, tinned the pads (with the exception of the center paddle) centered, taped, and reflowed it and that one worked as well. I didn't touch it after tinning the backside with a soldering iron.
Temp-controlled toaster oven can work too.
Was going to say drag soldering is the best and recommended way of doing it, but then saw your message pop up.
That's the entire point of the mailbag! People or companies send in stuff and I take a look at it. Even *you* can send in stuff to show off! People like seeing stuff they wouldn't ordinarily get to play with themselves. It's the same things for regular reviews. So the answer to your question is yes, people like to see this stuff, Mailbag is an incredibly popular segment!
I want to make my own SBC single board computer with certain parts and as ambitious as it seems, it feels less impossible watching your videos!
+Anne Blackwood
What's the most complex single part you would use on your board?
This is very good tutor! I now understand how to do those. Thank EEVBlog!
Seems like what you always taught us is true - conical = crap! I purchased a Hakko FX 888 and the chisel tip is magic on all stuff I have ever soldered including drag soldering.
Good point. But the bottom contacts still didn't wet either. And if you have pins extended outside the chips, those side contacts usually do wet, as I demonstrated with the drag soldering.
I was following the Schmart board instruction and using the recommended size tip.
Actually, taping the components while soldering smd is quite helpful!
You are soo good at soldering. I sometiems have problems with 1.27mm spacings without a camera in front of me, in a relaxed position.
The Hakko i bought came with the conical tip fitted as standard as you say and as a newbie I might have gone on to use it had I not seen Dave's vids warning against its poor performance.
I learn more and more with each video, thanks Dave :o). Oh and yes, you do sound like Steve Erwin to us Brits :o)
man i would at least need to use my magnifer for this. my eyes would go crossed trying to do this without my 3x glass. but then again i never wear my corrective lenses which makes my it worse than ever. you got some skill.
The grooves probably increase the surface tension, making the drag system work better. Good review! Luckily I have no need for 0.4 mm pitch!
It looked like from 20:53 that some of the wells had actually shorted together in the middle and at the end of the track; I wonder how much extra effort it would be to rework those as opposed to a pin short.
That defeats the purpose of testing the Schmart Board system. I only did the drag solder test to see what would happen.
Dave, look at the Linear's "QFN DFN Solder Joint Package Inspection" appnote for example.
They clearly state that "Terminal sides not intended to be a wettable by design"
And no need to use those uberfine 0.3mm tips and solder by one pin at a time. 1-1.5mm works better and faster.
They were just sleeping, they're back!
it is really helpful for us other in the buisness to have a non partial person like dave to review lots of stuff and give us his oppinion on it. and the way that is possible is by manufacturers sending him stuff to test out.
do you sugesst he would start blurring the name of the stuff he likes, like his scope, soldering iron, multimeter and so on and not mension them because someone might think its an ad?
dave speaks his mind and if you don belive him, buy the stuff and try for yourself.
Damn Dave! You didn't check the polarity!! You placed it arse-about!! Lucky, it's a breakout board, you can still compute with it!
Good job dave
This episode is just perfect. I happen to sit on a project that really only works with one of these QFN buggers. The chip is unavailable in SMD or DIP (ha, ha!), and there aren't attractive options around. The chip is just right... I tried dead bug soldering it, but it's just too tiresome.
So... after watching this video, I'm sold. Great stuff.
The first soldering iron didn't look like it could deliver enough heat to the fine tip. This is usually where the JBC, Metcal and Ersa irons excel.
Lead-free solder on the Schmart Board compared to leaded for your drag soldering?
actually it is very useful for prototyping, you can get stuff incredibly fast .. I do solder 0.4mm with hand on a normal board (without solder mask between pins) but still using schmartbord is super useful when doing quick prototyping as really few boards allow for proper prototyping as schmartbord do
Just think about IR reflow or even hot air rework stations. They heat the whole device to 250'c or more.
A tip from JBC? That's a spanish manufacturer! I'm glad it was known outside Spain :D
Why not do a bit of soldering paste on this chip?....Maybe that's a dumb question but i'm just learning...Wouldn't it flow underneath the pins a bit easier with a bit of paste?
just wondering if soaking the chip in coke/cola soften the corrosion/ rust away would work???
I feel like they definitely soldered some of those chips before sending it to be reviewed and put in front of such a wide audience. So I'd doubt it's the IC itself. From the looks of things, like everyone that makes something, they think their product is much more groundbreaking and useful than it actually is.
Would running the flux-pen along the chip's contacts helped? (before placing it down of course)
the latter technique is also what i use.. the bridges can be removed with flux. others prefer wick.
My guess is that is an early prototype chip. I work with chips like that at work. The number on it is probably the sequence number of the chip. We use those to track which pass early testing. In our case, all our chips are BGAs, some as large as over 2000 balls.
YES, YES, that is exactly who you sound like!!
Conical tips work well
Looks to me like not enough heat applied to the pins with the first technique.
With the second attempt with the larger tip you were heating them directly, rather than relying on the channels conducting it via the solder .
Wow.. so the drag technique never lets u down... i rarely use this one... im curious if u gets the same thing with a 45 deg angled tip for the qfn?
Those QFN chips may not have tin on the pin sides, only at bottoms. So they can look like bad reflowed outside, but they're actually perfect at the bottom.
Steve Irwin voice xD. schmartboard guys are the best :3
@ everyone saying that QFN sides aren't wettable: Dave did try to lift the chip to see if the solder had taken on the bottom, and it didn't.
is there a second board for part 2 with air gun?:)
Was there nothing you could do to clean the oxidation on the chip itself? And if the drag soldering worked, doesn't this mean that their board is useless ?
Why do irons ship with conical tips as standard?
He put the chip on the wrong way
Sahko123 Its just soldering tutorial, but I was about to write the same comment too.
I'd guess that the conical tip isn't giving enough heat to the pin - it heats the solder fine because the tip sinks in, but when it touches the pin it's only a tiny spot, whereas the chisel tip is both larger (more stored heat to dump in) and has a line of contact with the pin rather than a point. Might have been interesting to hold the conical tip there for a good 5-10 seconds to see what happened.
The GA32 was a prototype chip, never intended for production use- that's why it says "Internal Use Only".
always rince pads before applying reflux with acetone.
11:58 HAHAHA Dave,you are the best :D
the more the better
Dave, I think it wasn't enough heat.. Those little 0.3mm tips have a rather low thermal mass. Cheers !
parallel FORTH engine ... nice chip
If you had polished the copper pads on the outside edges of the package and the soldered/tinned pads on the bottom, with 2000grit wet & dry sand paper, you would have cleaned off any oxidation, and it should've soldered so much better :)
I really wanna try soldering....
I have been looking for one of those WELL-BASED tips ---FOREVER--- with absolutely NO luck! Can anybody help? PLEASE? I have HAKKO and TS100 type irons, BUT I am desperate: If I could find some well-based tip, I would get the proper soldering station just to fit the tip!!!
I think the whole point of this technology was, that you don't need to use hot air reflow to solder that small pitch chip. Only soldering iron and a bit of flux.
Time honoured drag solder: 1
Schmartboards: 0
Not saying those boards are pointless. It certainly makes aligning pins more easy.
where do
you get it
Well, just apply flux, tin the pads with a huge chisel tip, clean up, apply some tacky flux like from Amtech, then put the chip on there are use a hot air gun. Worked even on 0.4mm pitch BGA perfectly. :)
No hand soldering should be done on any QFN really. They are designed for reflow!
re flow solder it, the solder guard on the pcb should prevent it from making shortable connections
You didn't align pin 1. It won't work now.
If the pins were oxidized, why did a different technique work. Can you get some 2000 sandpaper and place it on the bench and just slide the chip lightly across to take the oxidation of the pins? I don't think it's oxidation though. I think the drag soldering is bridging over to the tops of the pins, where the flux is. Try fluxing the grooves then putting the chip into the flux.
+alex tworkowski i guess because when using the drag soldering technique he was using a much bigger iron tip so it pushed more of the solder flux in as well as applying heat around more of the board which would have helped fight the oxidisation
I looked at the price on the DIT-1B...US$450 is not bad at all for what you're getting...
Like Dave, I also discovered the hard way not to set iron temp too high and that drag soldering works really well with these boards and QFN chips.
FWIW the Greenarray GA144 holds 144 CPU cores and their instruction set is a subset of Forth.
why not try using hot air solder?
I blame it on the lead free gunk. It just doesn't whet or flow like good old 63/47 blend.
maybe some weight on the chip?
I don't know why Dave could not make it work, I use it for a while and they always work perfectly.. maybe that chip was too dirty / oxidized ..
i use hot air :)
"What? I sound like Steve Irwin? Crikey!"
plz detail microscope
After that sickness soldering the chip works?
Yey!
the pads might be too oxidised, I never had issues with 0.4mm qfn+schmart ... would be cool if you cleaned and tinned the pins before you started the experiment as if pads were tinned with lead free that's then left to oxidize there's no way you can push enough heat trough the channel alone .. especially not with the pen flux you are using
That chip is ancient in SMD solderability terms. The leads are incredibly oxidized. Not so Schmart.
I think Dave sounds like Murry from Flight of the Conchords. Sorry if I offended anyone.
@ 1:00 i seriously lmao
Dave, I think your tongue angle was off.
it most likely didn't have enough solder in the groove
thatswhy I only use solderpaste and hot air ;)
the contact of the molten blob of solder with the pin was too little, u need a pressured hot blob of solder to burn throught that layer of sh!t on those pins, that's why the bigger tip works better.
Often the side of this kind of chip is not solderable, it does not have the same kind of surface as the underside of the chip, and so if any solder got under the IC the chip may well be soldered properly, with the solder under the IC.
Ah Dave, you know where you went wrong?
You were using SOLDER and SOLDERING flux...
You should have been using SODDER and SODDERING flux!
flux the thing underneath, pin down the corners, swipe across and wick the excess... not very difficult...
09:00 It's not aligned properly.
with such steady hands you should have been a surgeon, not an engineer.
circuit pad stencil mask with solder paste and an oven or hot air dude. iron is just a bad times for ya.
Derp! 63/37. . . 110% solder, what's that? XD
Dat Flux
Lost my shit when I saw the greenarrays chip... That's a forth cpu!!
It is more like a review than an ad...
Sad when engineers make pads longer than needed. Those with the task of hand soldering find uniformity an issue.
Picked up a bunch of these Schmartboard | EZ PCBs when Rat Shack went tits up a while back. Used several size tips. Several temps. Several pressures. Found them to be total crap only suitable for cheap parts during a throwaway prototype stage. Like other comments mention, learning to properly drag solder is the way to go. Even though I picked up these boards for pennies on the dollar, when the time machine is finally invented, I'm going back to get my money back for this garbage. Garbage... Total crap.
Even contacted their customer support for comments or suggestions... only thing they suggested was to use "water-soluble flux" (which I already was!).
Yeah... great idea (if it worked flawlessly), but unfortunately that's just not the case. I agree with Dave that it seems if there were more solder in the groove that you'd have a winner winner chicken dinner. But nope... instead, we have loser loser crack abuser. Cheers.
calm down.. you state you got them for few pennies on the dollar ref close-down-sale but still bitchlng like you have lost your family to this maker..
get a grip, and just be happy that this maker ewen was willing to put you on there shoulder to burp you off when you contacted them with your whiining, and partically.when you have in mind that you havent put any money behind there work, other then some few pennies ..
Ha. The kind of person that "Likes" their own post. LMFAO... The product still sucks ass. Wouldn't take them for free.
like stated' calm down instead of all this editing of your post..
and if the product "sucks ass" like you wrote.. tjahh cant say' havent own it, -I can only relate to this above-video from EEVblog and the video Dave linked in the vido from the maker where they in an older video from 2012 display that drag-solder with extra solder is the way to go to get solder-attachment with these restricted grooved-pins.
Ni! Ni! And if you Reply again, I'll say Ni! to you.
304 views and no thumbs down. Looks like the serial haters are still on vacation
the chip probably was dirty from the mail..
look the dot maybe it won't work bcos xD it wrong direction hehehe
you need the flux with a needle. you cant apply it with a brush properly
I saw it at around 14 min nevermind but yeah flux was the issue haha
Like.
32cores?! wtf... skynet is not far off now...