Author Topic: HASL vs ENIG question  (Read 527 times)

0 Members and 1 Guest are viewing this topic.

Offline Rat_Patrol

  • Contributor
  • Posts: 46
  • Country: us
HASL vs ENIG question
« on: October 07, 2019, 06:53:14 pm »
I have some boards that I designed that are decent sized (150x250mm) and are used in an automotive environment (conformal coated with urethane after finished and tested).

I have been using HASL finished boards to this point, and have had no issues I can think of related to the HASL, but the price for production of ENIG isn't that much more in the grand scheme of things.

Are we missing out on advantages of ENIG? The finest pitch we run are PLCC packages and SOIC-8 type packages with similar pitch.

Better strength of solder joints? Strength of the solder joints is important due to the boards operating in a high vibration environment.
 

Offline wraper

  • Supporter
  • ****
  • Posts: 10786
  • Country: lv
Re: HASL vs ENIG question
« Reply #1 on: October 07, 2019, 07:08:34 pm »
Unless vastly different solders are used for HASL and soldering, I don't see why solder joint strength would be inferior. Packages you use are large and do not benefit brom ENIG. IMHO in your case it comes down to just PCB quality, not HASL or ENIG technology in particular.
 

Offline MR

  • Contributor
  • Posts: 24
  • Country: tw
Re: HASL vs ENIG question
« Reply #2 on: October 07, 2019, 07:21:32 pm »
Shitty Pick and Place machines cannot properly handle HASL Fiducials (eg. those polish Mechatronika Sp.J. Pick and Place machines).

https://www.eevblog.com/forum/manufacture/mechatronika-mx80/msg2717576/#msg2717576

just have a look at the fiducial pictures, it caused our components having an offset. We fixed it by adding some shielding around the camera ring and it reflects the light nicely for having a nicely illuminated PCB..
The surface and shadow parts of such fiducials is not predictable.
I have read quite a few articles about that some people advising to make the fiducial as big as possible to the solder can stick on it but then you might have a "mountain" on it throwing shadows (with poorly illuminated pcbs). I guess professional SMT manufacturers are taking care about such issues. It's absolutely unacceptable not supporting HASL PCBs properly.

It might be difficult for solder to attach small/thin pads, ending up with no solder on the pads. For very fine structures I still prefer ENIG boards (less troublesome in every case)
« Last Edit: October 07, 2019, 07:31:26 pm by MR »
 

Offline KL27x

  • Super Contributor
  • ***
  • Posts: 3820
  • Country: us
Re: HASL vs ENIG question
« Reply #3 on: October 07, 2019, 07:22:07 pm »
The main two things I use ENIG for are boards that will need further soldering after an initial oven reflow. Bare HASL pads can grow a thick oxide layer in the oven. And obviously for boards with exposed contacts.
 

Offline Rat_Patrol

  • Contributor
  • Posts: 46
  • Country: us
Re: HASL vs ENIG question
« Reply #4 on: October 07, 2019, 07:40:29 pm »
There is some hand-soldering of THT components after SMT oven reflow.

We haven't had an issue with soldering them though with HASL.

I may get a small batch of them in EING and see how they hold up.

Is there any solder strength advantage either way?
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12898
  • Country: cn
  • Power Electronics Guy
Re: HASL vs ENIG question
« Reply #5 on: October 07, 2019, 07:52:38 pm »
Theoretically, ENIG is worse due to gold leaching into solder, changing composition. Also, nickel layer may be a source of failure, known as black pads.
ENIG is used in many cases simply because it's overall cheaper if you factor in the cost of storage, limited shelf life and compatibility with different solders.

Also, ENIG is more flat, so if you have tons of fine pitch BGAs , you have to use it. For QFN/SOP boards, I can't see ENIG being anywhere better.
ENIG is also wire bonding compatible, so COB is possible without special treatments (not saying it's proper, you should always use ENEPIG or EN-only for the best quality with Au and Al wires).

Finally there is the psychological thing. People like golden stuff, despite many research have shown OSP/HASL can be more reliable than ENIG if used soon after fabrication and stored properly.
 

Online langwadt

  • Super Contributor
  • ***
  • Posts: 1661
  • Country: dk
Re: HASL vs ENIG question
« Reply #6 on: October 07, 2019, 08:02:57 pm »
Theoretically, ENIG is worse due to gold leaching into solder, changing composition. Also, nickel layer may be a source of failure, known as black pads.
ENIG is used in many cases simply because it's overall cheaper if you factor in the cost of storage, limited shelf life and compatibility with different solders.

Also, ENIG is more flat, so if you have tons of fine pitch BGAs , you have to use it. For QFN/SOP boards, I can't see ENIG being anywhere better.
ENIG is also wire bonding compatible, so COB is possible without special treatments (not saying it's proper, you should always use ENEPIG or EN-only for the best quality with Au and Al wires).

is the ~100nm gold with ENIG really sufficient for wire bonding?

 

Offline KL27x

  • Super Contributor
  • ***
  • Posts: 3820
  • Country: us
Re: HASL vs ENIG question
« Reply #7 on: October 07, 2019, 08:11:52 pm »
Quote
We haven't had an issue with soldering them though with HASL.
Doh, my bad. I mixed up my finishes. I was thinking OSP. I can't imagine HASL being that bad.

Even with OSP, it doesn't make the board completely not solderable, like all or nothing. The thickness of the oxide layer makes it require longer dwell times even if you jack up the temp on the iron. I've had to buff the oxide layer off of test pads, before, and it can get really thick. In the range of a thou. Maybe 3 seconds with a dremel and buffing wheel with compound to even cut through it.

For some reason HASL doesn't appear to be an option on FPC. The cheap finish is OSP, and i'm not sure what that even is. It's kinda salmony color and it is thin and flat; no lump/bump.

« Last Edit: October 07, 2019, 08:20:45 pm by KL27x »
 

Offline MR

  • Contributor
  • Posts: 24
  • Country: tw
Re: HASL vs ENIG question
« Reply #8 on: October 07, 2019, 08:28:26 pm »
I'm using HASL for things that I can see (QFNs only), and ENIG for things I cannot see (at the moment BGA, Dual row QFN, etc.).

ENIG always has been less troublesome for me, but if HASL can do I'll pick it, HASL also seems to be quicker when manufacturing PCBs (because some companies outsource the ENIG process).
 
The following users thanked this post: KL27x

Offline blueskull

  • Supporter
  • ****
  • Posts: 12898
  • Country: cn
  • Power Electronics Guy
Re: HASL vs ENIG question
« Reply #9 on: October 07, 2019, 08:32:38 pm »
is the ~100nm gold with ENIG really sufficient for wire bonding?

For gold, yes. For aluminum, no. That's why it's preferred to bond directly to nickel.
You can still bond aluminum, but the wire scrapes through gold layer first, which is a recipe for purple plague under certain conditions.
 

Offline GerardG

  • Contributor
  • Posts: 30
  • Country: nl
  • Pick and Place support manager
    • electronicatools.com
Re: HASL vs ENIG question
« Reply #10 on: October 08, 2019, 12:47:17 pm »
As I have been working with automotive electronic design in Germany.

The preferred finish is Chem Sn. for most of the boards. Germany is using this because they are using a lot of Press-fit connectors.
ENIG is using a hard nickel layer and is not to be used for press-fit connectors.

When getting a thick layer of gold.
To much gold in dissolved in the soldering can make the connection brittle. This will normally not be a problem on the larger pads.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf