How bad is this solder?

[komatsu]

How bad is this soldering job?

[PeterFW]

How bad is this soldering job?

I am torn between "looks pretty bad" and "looks worse then it actually is".
The angle of the picture is not that good, get one from the side, at a slight angle.

But i would guess, you should re solder the joints.
Put a bunch of flux on the pads and touch them with the soldering iron.

Keep an eye on the small resistors, do not overheat the pads or you could loose one, there is allready one missing but that may be intentional

[komatsu]

Thanks Peter.

The joints certainly don't look very nice. 

But from a technical point of view - what is the main priority?

[RJFreeman]

two main priorities of equal importance:
1) No solder bridges between adjacent pins - this looks ok
2) solder continuity between device pin and pcb pads - this is a bit harder to tell from the picture, but the pins on the left hand side warrant a closer look

looks like it has been soldered or re-soldered by hand and the brown stuff is flux from the solder while not critical cleaning off the solder flux would make it easier to see how the solder has flowed underneath

[SL4P]

It looks like the chip was sitting on the shelf for a couple of years and the pins have oxidized.
They should have been cleaned before soldering - then at least a chance of getting a decent joint.

[komatsu]

ok thanks for the replies so far.

So far, this is the story. I soldered ( using a hot-air gun and flux) on a ROM chip to this board.

In my after-soldering test, the disk which this board belongs to worked perfectly. Happy days.

That was 5 days ago...today I went to turn on the drive and it will not even turn on.

Looks like its back to the drawing board...

[komatsu]

It looks like the chip was sitting on the shelf for a couple of years and the pins have oxidized.
They should have been cleaned before soldering - then at least a chance of getting a decent joint.

cleaned with something like Isopropyl alcohol?

[SL4P]

It looks like the chip was sitting on the shelf for a couple of years and the pins have oxidized.
They should have been cleaned before soldering - then at least a chance of getting a decent joint.

cleaned with something like Isopropyl alcohol?

No - you'll probably need something more aggressive.
Not sure what it was but a local SMT house gave me a small bottle of very corrosive flux - and that worked quite well.
The kind of stuff you don't want o spill on clothes or table tops!

[wraper]

The kind of stuff you don't want o spill on clothes or table tops!

So you used it on the PCB without any cleaning 

[SL4P]

No - I put the chips in a small glass bottle with the 'solvent/flux', and shook it around for a few minutes - and presto - all the oxidation was gone (floating around in the liquid).
Then extract the parts, and rinse - literally - in IPA or similar to remove the flux & crud residue.

[komatsu]

No - I put the chips in a small glass bottle with the 'solvent/flux', and shook it around for a few minutes - and presto - all the oxidation was gone (floating around in the liquid).
Then extract the parts, and rinse - literally - in IPA or similar to remove the flux & crud residue.

Would you not be worried that your "corrosive" flux would damage the chip?

[BradC]

It looks like the chip was sitting on the shelf for a couple of years and the pins have oxidized.
They should have been cleaned before soldering - then at least a chance of getting a decent joint.

cleaned with something like Isopropyl alcohol?

Cleaned by something like a dremel with a brass brush.

[nanofrog]

Pins on the chip definitely didn't wet properly (solder should flow over the top of the pins as well). Bit harder to tell for me after that due to flux, but I suspect you ran into a combination of oxidation on the chip legs as SL4P mentioned, and not enough heat for the solder to truly flow properly (ground plane sucking heat off on you).

Might want to "soak" some heat into the board before trying to solder (set the hot air station to say 125C & hit the board before placing the chip). Then add your paste really quickly, place the chip, and solder. An iron may also be of use here (drag method would likely yield better results over pin-by-pin as it's quicker).

There are tools that help with this, such as a hot air plate or a hot plate/skillet can do wonders (can add temp control to one you find in a local store for say ~$30 max; garage sale or thrift store find would be ideal).

As per cleaning oxidation off of pins and pads, a fiberglass scratch brush is really good for that, and they happen to be inexpensive as well (even a German made one is ~$11 shipped). Just scrub at the pins on both sides for a few seconds, then wipe with some alcohol to clean off whatever is scratched off as well as any dirt & oils that may also be on there (i.e. from your fingers). Rather quick and easy actually. 

[Shock]

Remove that flux with IPA and the softist toothbrush you can find. Presoak both the board and brush and be gentle scrubbing. Then take photos again and we should have a better idea.

To get the best result you need to start with a clean PCB, good flux, solder and fresh components.

[komatsu]

Thanks for the replies.

No "paste" or "solder" was used when attaching this chip to the PCB.

The chip seemed to make a pretty solid connection to the board without either.

Is that ok?

[nanofrog]

No "paste" or "solder" was used when attaching this chip to the PCB.

The chip seemed to make a pretty solid connection to the board without either.

Is that ok?

Answer = NO. 

And this ^ helps explain things a fair bit...  

Essentially what you did is tack every pin down and stop (handy for SMD work; usually on a pair of opposing corners). That in itself is fine, but you're supposed to go back over every joint with fresh solder (pin-by-pin is one way, drag soldering is another; youtube has plenty of videos on SMD soldering  ). So you just skipped the back half.

If you use paste, you apply it to the pads before placing the chip, then either use hot air or put it in an oven (activates the flux, then melts the solder to make the joints).

Also, that dark of a color on your flux means you either heated it too long, and/or used too high a temperature (i.e. you "cooked" it). It should retain a honey color for rosin, clear for no-clean (read the datasheet for the solder/flux being used).

Although they're not SMD specific, Pace's Basic Soldering Lessons 1 - 9 and their Rework & Repair Lessons 1 - 8 would be worth your time to pick up the proper basics (lot of crap on YT). 

[komatsu]

No "paste" or "solder" was used when attaching this chip to the PCB.

The chip seemed to make a pretty solid connection to the board without either.

Is that ok?

Answer = NO. 

And this ^ helps explain things a fair bit...  

Essentially what you did is tack every pin down and stop (handy for SMD work; usually on a pair of opposing corners). That in itself is fine, but you're supposed to go back over every joint with fresh solder (pin-by-pin is one way, drag soldering is another; youtube has plenty of videos on SMD soldering  ). So you just skipped the back half.

If you use paste, you apply it to the pads before placing the chip, then either use hot air or put it in an oven (activates the flux, then melts the solder to make the joints).

Also, that dark of a color on your flux means you either heated it too long, and/or used too high a temperature (i.e. you "cooked" it). It should retain a honey color for rosin, clear for no-clean (read the datasheet for the solder/flux being used).

Although they're not SMD specific, Pace's Basic Soldering Lessons 1 - 9 and their Rework & Repair Lessons 1 - 8 would be worth your time to pick up the proper basics (lot of crap on YT). 

This project was put on ice for a couple of months and now I'm back to it!

In this particular scenario, what soldering technique would you recommend?

[nanofrog]

No "paste" or "solder" was used when attaching this chip to the PCB.

The chip seemed to make a pretty solid connection to the board without either.

Is that ok?

Answer = NO. 

And this ^ helps explain things a fair bit...  

Essentially what you did is tack every pin down and stop (handy for SMD work; usually on a pair of opposing corners). That in itself is fine, but you're supposed to go back over every joint with fresh solder (pin-by-pin is one way, drag soldering is another; youtube has plenty of videos on SMD soldering  ). So you just skipped the back half.

If you use paste, you apply it to the pads before placing the chip, then either use hot air or put it in an oven (activates the flux, then melts the solder to make the joints).

Also, that dark of a color on your flux means you either heated it too long, and/or used too high a temperature (i.e. you "cooked" it). It should retain a honey color for rosin, clear for no-clean (read the datasheet for the solder/flux being used).

Although they're not SMD specific, Pace's Basic Soldering Lessons 1 - 9 and their Rework & Repair Lessons 1 - 8 would be worth your time to pick up the proper basics (lot of crap on YT). 

This project was put on ice for a couple of months and now I'm back to it!

In this particular scenario, what soldering technique would you recommend?

I'd drag solder the IC (tack opposing corners first, then drag solder all the pins). Might require you to buy a dedicated tip designed for the task, but well worth it.

[komatsu]

ok thanks for that.

stupid question here but when you have an IC and you just place it on the PCB without solder and heat it with hot air gun...it does stay in place.

So what is the actual function of solder? (when an apparent connection seems to be already made?)

[nanofrog]

stupid question here but when you have an IC and you just place it on the PCB without solder and heat it with hot air gun...it does stay in place.

Instances where a part might adhere to the PCB without adding any solder, are with boards that have an HAL/HASL finish on the exposed pads/holes (Hot Air Solder Level/Hot Air Level respectively) . Even if the part "sticks", the HASL surface is NOT intended to make proper solder joints (not enough solder & zero flux).

You must add solder and flux and flow it with heat in order to form proper joints.

So what is the actual function of solder? (when an apparent connection seems to be already made?)

It makes both a proper electrical connection as well as a mechanical one.

[fivefish]

pins 1, 2, 3, 4, 5, 7 seems just sitting on top of the solder pad. Doesn't look like a reliable connection. A few heat/cool cycles and it could lose connection and become open connection with the pads underneath it. 

pins 6, 8 could use more solder.

the pins look dirty and most probably that's why solder didn't stick on the top.

[komatsu]

Thanks nanofrog and fivefish for those detailed answers.

This is the video I used as an example before I undertook this job. Obviously, it was the wrong video to learn from?

[nanofrog]

Obviously, it was the wrong video to learn from?

Definitely.

Go back and check out the soldering tutorials I linked previously (created by PACE; links reposted below).  They might be a bit on the old side, and not explicitly geared towards SMD packages, but the information is still applicable to this day.

You can also find their SMD soldering techniques on Marc Siegel's page (see below).

Personally, I'd recommend watching the videos in the following order:

• Basic Soldering Lessons 1 - 9
• Rework & Repair Lessons 1 - 8
• Marc Siegel's SMD Soldering Videos

Additional resources:

• PACE Video Search
• Manual for Rework/Repair & Assembly (.pdf)
• NASA Workmanship Standards: Pictorial Reference
• Micheal Hobson's YouTube Page (soldering & other videos that might be of use)

[Aodhan145]

Remove that flux with IPA and the softist toothbrush you can find. Presoak both the board and brush and be gentle scrubbing. Then take photos again and we should have a better idea.

To get the best result you need to start with a clean PCB, good flux, solder and fresh components.

Best soldering video. It looks so clean and nice compared to the shit job I do.