To date I've just prototyped boards that were then handed off to experienced manufacturing operations for actual building, or I've build things at such small volumes that I would do all the work here.
Now I'm helping a customer with managing a small-lot manufacturing line, using contract help, and an issue came up with board cleaning. Our proposed vendors are new to lead-free assembly, and one of them always cleans boards with an ultrasonic cleaner. I asked one of the part vendors about this, and was told that ultrasonic cleaning isn't compatible with lead-free solders, because the solder fatigues and cracks (don't be too hard on me if I have this wrong -- you're getting it third hand).
Do you do lead-free soldering on a production basis? By hand? How do you clean your boards? When folks talk about "water" cleaning, that can't just mean holding it under cold running tap water -- what is actually done?
Thanks.
--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?
Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com
We do leaded and lead-free. We always use rosin flux (both paste and hand soldered stuff) and a solvent wash. The boards are submerged for about a minute in a boiling solvent tank that has a deflux agent added, and are then removed and sprayed with hot pure (distilled) solvent. That works great. As a bonus, residual rosin flux wouldn't be conductive.
Most contract assemblers use water-soluble flux. After soldering, the boards go through a conveyer water spray line to clean off the flux. It's important that the spray be vigorous and that the final spray be very clean (deionized, high resistance) water. The hazard of water processing is that ionic, hygroscopic crud can be left on the board, especally under parts. Digital stuff will be less affected than analog.
We had to reclean some boards that had a dirty water wash and were leaky at the nA and pA level. We used a Water Pic and distilled water, and blasted under all the ICs by hand, then baked dry. The Brat did it, and got very wet and mad.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Select a vendor which uses a nitrogen atmosphere for lead free soldering. It makes much nicer joints. Either way I wouldn't select a vendor who is new to lead free soldering unless you can get them to work for free until they get it right.
--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
How do you know that they got it right? Sometimes the failures take years to develop, with heat/cool cycling.
Look at the formatter boards on HP P2015DN printers, for example.
Lead-free BGA joints are (I think) what's failing, and lots and lots of people have been able to resurrect them by baking them in a domestic oven (hardly a controlled method). Mine failed (it is maybe five years old) and I put it through a full lead-free reflow cycle on my little desktop oven- came back to life.
Consistent good soldering quality over a batch of 100+ boards.
Would leaded solder do better under mechanical stress? TO220 devices are notoriously good at wiggling themselves free from a solder joint.
--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
HP formatters and JetDirect cards are good candidates for toaster oven based repair. I've done fairly well with the oven method, and very badly with my hot aid desoldering machine. There are a few tricks, involving temperature accuracy and timing in order for it to work. I was doing rather badly baking Jetdirect boards until I began following other users instructions. 10 minutes at 180C (350F) and 20 minute cool down, is what worked for 5 out of 7 JetDirect cards. Somewhat longer cool down for the larger formatter cards.
(This starts off with some formatter diagnostics, but soon switches to "baking" various PCB's).
Hint: Clean the toaster oven before attempting unless you want the printer to smell like pizza.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
I forgot to mumble something about why. Based on my tinkering with a heat gun and thermometer, HP probably uses either a 58% Bismuth 42% Tin, or a 57% Bismuth 42% Tin 1% Silver solder mix which melts at
138C. This low melting temperature is why reflow in a toaster oven at
180C works, which is not hot enough to melt typical Tin Copper lead free mixes which require at least 200C. The problem is that these bismuth formulations are strong, but brittle. Flex the board and the BGA solder balls crack.
See comments under Bismuth Tin mixes.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Funny thing is that it fails just sitting quietly on a desk in a climate-controlled office with the printer powered 24/7. Only 13,500 pages total (it's rated for up to 3,000 pages per month, so only 4 months equivalent heavy usage).
I guess the board itself powers up from standby and then down whenever a set of prints is made so you get thermal fatigue. Early SSR makers got bitten bad by this- the things would fail in months or even weeks if you got the on/off cycles just right.
Best regards, Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
I think (not sure) that the power is always applied to the network card. Same as in Wake-on-LAN in a PC. A networked printer would not know when to wake up from standby if the network card wasn't powered.
I've had two fairly new, in the box, but out of warranty, J4169A (610n) JetDirect carts fail while in the box. I had tested, updated firmware, and pre-configured them on arrival. Two or three months later, both were dead. Both were later revived with a toaster oven. No thermal cycling required to kill the cards. However, there was a third card from the same lot that was installed in a printer, and has been working without incident for about 5 years.
I agree that the thermal cycles are probably responsible for most of the in the machine failures.
Here's the lunatic fringe, using a propane torch to fix a J6057A (615n) card.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
I don't know. My microscope and pen camera can't see under the BGA chips. When I unsolder a BGA chip for rework, the heat melts the evidence.
A bit of Googling excavated this article: "At the solder joint level, the damage is seen as a crack at the package/PCB boundary, although there are other possible points of failure in the solder joint network." Also see "solder joint failures" section in the above URL. Lots of places for things to break.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
Water based cleaning with an aggressive detergent like Safewash works pretty well but in my experience you need Ultrasonics to clean underneath QFN and large TQFP type chips. I would expect BGA to be worse but I haven't done enough to know. Big fancy spray type machines may be as good but I've never had one to try.
"Drag-out", contamination of the water by the things being cleaned, can be a problem. And it seems to me that any additive that is water soluble must leave conductive residue.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Even though it's an "N" model, I use it in USB mode, but I think you're right. I put a temperature sensor on the big ASIC and it didn't change more than about +1°C from the idle temperature of 44°C Tc when printing 20 sheets of paper.
Must be fairly mild cycling.
I suggest a cement kiln for HP printers.
Best regards, Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
I just checked my HP2300DTN printer. The LAN card returns pings from the network when in standby (LCD display dark). I was also able to access the configuration web page, which implies that the printer controller card is also powered in standby. Therefore, the JetDirect card is very much alive in standby.
The HP2300DTN has a convenient vent hole in the JetDirect shield, which allowed me to measure the chip temp. Ambient is 20.8C. Chip is running at 35C as measured with a thermocouple.
My guess(tm) is that the low temp solder was pre-stressed during initial soldering. The mild thermal cycling simply provided sufficient additional stress to crack the solder connection. Once it's cracked, it's just a matter of time before an air gap forms.
I beg to differ. I like HP printers. Well, I don't like all of them. My favorite is the HP2300DN series. I have 3 of my own, and have resold about 8 more to various customers. Once I've disarmed the HP premature failure acceleration features: they work reliably and apparently forever. Unfortunately, I can't say the same for all the HP printers. I have several P2015 series printers, which are a nightmare to work on. For example, a typical 4 minute fuser assembly replacement, becomes a 90 minute ordeal with these printers. Rube Goldberg would instantly recognize the internal linkage and interlock designs. There are others with similar design issues. However, they all have one thing in common in that they *CAN* be repaired. This is in contrast to other brands that can easily be declared eWaste for want of a trivial part that is unobtainable. For HP, I can get almost any part I require, much of it from HP. Despite efforts by HP to destroy their own product line by converting it into a throw away product, the high volume of HP printer sales is sufficient to attract 3rd party parts manufacturers and used printer cannibals on eBay.
Incidentally, if you want quality advice on printer repair, see:
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.