The evils of leaving solder flux on boards
|
Yeah! Flux residues and dust are not to be overlooked. Let me tell some of my experiences (though not RF related), just to show how slippery they can be: The onboard regulator on a PLC peripheral module in a machine began refusing to turn on. Taken to the maintenance dept and fed with a 24V PSU it didn't fail. Back to the machine, ok. A few days later it failed again. Back to the lab, good again. Tried feeding with 22 to 26V, applying it at once or with a slow ramp, checked the electrolytics ESR, rotated the module to match its vertical position in the machine, ok. Back to the machine, ok but just for a while. Suspected noise on the 24V bus, decoupled it, no improvement. Back to the lab, I was about to replace the switching regulator IC. In that moment I noticed the shine of rosin around the SMD IC, it seemed it had been manually replaced in the past. I thoroughly cheaned the rosin with IPA (isopropanol = isopropylic alccohol), trying to make it flow through the underside of the IC too. Never failed again. But why it didn't fail at the lab? The only possible explanation: there we have low ambient humidity because of the air conditioner!!! Same machine, we had a random failure of the braking module for a variable frequency drive. A few days after replacing a bad IGBT, it failed again. This happened shortly after the previous experience, so I nailed my eyes on possible flux residues. None, but the board had a thin dust/smoke layer all over it. Cleaned the dirt. It didn't fail again. This module works with rectified triphasic 400Vac, that is, some 690Vdc. When you have a dirty surface and two places with high voltage between them, a "tracing" phenomen may occur: a slightly conductive channel develops and it gets more and more conductive. This is compounded by the that fact high voltages attrack dust. Take ths into account on your power supplies for tubes. Ethylic (medicinal) alcohol can be used instead of IPA, but it tends to leave a white residue. Don't let the rinsed (dirty) alcohol get into components with mechanical contacts, such as potentiometers and dipswitches. Daniel Perez LW1ECP El domingo, 26 de mayo de 2024, 11:38:11 a.?m. ART, Reginald Beardsley via groups.io <pulaskite@...> escribió: When I was living in Houston I had 2 VCRs die,? each after about a year of light use.? I took one apart to strip anything useful and observed that the entire board was covered with flux which is hygroscopic.? Houston is humid.? I was in the process of moving and throwing out junk, so I simply made a mental note of it. I have a 20 year old Aiwa CD/cassette/radio that I use to play CDs while I play guitar.? About 10? years ago the system stopped working.? I opened it up and removed flux residue near the MCU at which point it worked again.? However after a few months it quit with the same problem.? I repeated the process focused on flooding the underside of the MCU with IPA.? Never have had a problem since.? Until I returned from FDIM and found the remote no longer worked.? Worked for 20 years and just quit. Battery swap didn't help.? After cleaning the solder flux from the remote it now works properly with the old cells. My point being that it worked for 20 years and failed because the board had not been cleaned.? Eventually there was enough conductance that the parasitic capacitance degraded the data lines to the MCU.? Three parts, MCU, cap and LED plus the elastomer keypad.? Failed "part" was the flux residue. Now when I open up anything and see flux I wash it off.? It may not be a problem now, but I know from experience it will eventually.? All of the consumer kit repairs I've done in the last 15 years have been flux residue induced.? Sole difficulty has been internal assemblies which are too complex to make it worth the effort of getting access to the board. Have Fun! Reg |
|
I just clean the flux residues with isopropanol and an old toothbrush, trying the (already dirty) liquid does not get into tact switches, trimming potentiometers, trimming capacitors, dip switches, rotary switches, buzzers, connectors. Also clean the edge connector with a cloth before the alcohol evaporates. 19 de junio de 2024, 07:33:41 p.?m. ART, Glenn Little <glennmaillist@...> escribió: Be careful with circuit boards in an ultrasonic cleaner. The bond wires between header and die can be broken by the ultrasonic energy. Glenn On 6/5/2024 7:21 PM, Reginald Beardsley via groups.io wrote:
Well, add another flux repair to the list. LG window AC wouldn't work as in no response at all. On inspection board was saturated with flux. Cleaned it and it immediately started. I'm beginning to think that this is a deliberate design defect in a lot of consumer grade stuff to sell replacement products at 2-3 year intervals. I strongly suspect that if I subjected new kit, still in shipping box to a 100% humidity chamber for a week or two it would not work when powered up for the first time. BTW I bought a 30 liter ultrasonic cleaner for this sort of "repair" work. It will take a full size HPAK PCB. Have Fun! Reg -- ----------------------------------------------------------------------- Glenn Little ARRL Technical Specialist QCWA LM 28417 Amateur Callsign: WB4UIV wb4uiv@... AMSAT LM 2178 QTH: Goose Creek, SC USA (EM92xx) USSVI, FRA, NRA-LM ARRL TAPR "It is not the class of license that the Amateur holds but the class of the Amateur that holds the license" |
Daniel Ricardo Perez LW1ECP