Here’s a little movie of a solderjoint being made.
Beside being nice to watch, I think it is also showing very clearly how the solder should flow around the parts by the temperature...
https://twitter.com/TechAmazing/status/1309664895902928897
Beside being nice to watch, I think it is also showing very clearly how the solder should flow around the parts by the temperature...
https://twitter.com/TechAmazing/status/1309664895902928897
Yeah I thought so when you can see the solder being smeared around the part.
But after that, the connecting parts seem to be heated up enough so that the solder flows around on itself—not good enough?
Actually, I had to teach that method when we had to use tin/silver eutectic for all our cryogenic connections, as well as teach a power supply vendor how to solder correctly
. The old method of heating the work tends to bring the work temp far too high as the tip temp is well above most plastic parts temperature range, typically to increase production speed.
Examine a small spst switch that has 3 tabs out the back, 1/4 inch hole mount. Notice that the connections are actually epoxy mounted on the switch. The two outer ones are single formed pieces that become the contacts. If those terminals are overheated, the epoxy will go through the glass transition temperature (Tg),(also called heat distortion temperature). At that time, it will soften, and the coefficient of expansion will climb. (The epoxies I've used typically run 30 PPM/degree C because they are alumina loaded). Above Tg, expansion is in the 120 PPM range. So it swells, you solder the terminal, and as it cools it pulls away from the metal. Because flux is on all the surfaces as it cools, it can wick into the switch and compromise the actual contact surface.
We had 600 power supplies where rack fan vibration was causing intermittent switch bounce as a result. I had to establish a procedure to do it right, show the Manu how to do it, and rework a whole lot of supplies. In the ensuing 20 years of operation, we do not find the replaced switches causing problems.
It requires they use higher tip mass at a lower temperature. By using the technique shown on first post, the bulk of the heat transfer to the part will be at the melting temp of the solder, not the much higher tip temperature.
It is a tradeoff of course, the way mentioned is very good for a novice to learn, but as experience is gained and solder joints understood, this method is very good for long term reliability of the components....of course given a proper wet/fillet.
Jn
. The old method of heating the work tends to bring the work temp far too high as the tip temp is well above most plastic parts temperature range, typically to increase production speed.
Examine a small spst switch that has 3 tabs out the back, 1/4 inch hole mount. Notice that the connections are actually epoxy mounted on the switch. The two outer ones are single formed pieces that become the contacts. If those terminals are overheated, the epoxy will go through the glass transition temperature (Tg),(also called heat distortion temperature). At that time, it will soften, and the coefficient of expansion will climb. (The epoxies I've used typically run 30 PPM/degree C because they are alumina loaded). Above Tg, expansion is in the 120 PPM range. So it swells, you solder the terminal, and as it cools it pulls away from the metal. Because flux is on all the surfaces as it cools, it can wick into the switch and compromise the actual contact surface.
We had 600 power supplies where rack fan vibration was causing intermittent switch bounce as a result. I had to establish a procedure to do it right, show the Manu how to do it, and rework a whole lot of supplies. In the ensuing 20 years of operation, we do not find the replaced switches causing problems.
It requires they use higher tip mass at a lower temperature. By using the technique shown on first post, the bulk of the heat transfer to the part will be at the melting temp of the solder, not the much higher tip temperature.
It is a tradeoff of course, the way mentioned is very good for a novice to learn, but as experience is gained and solder joints understood, this method is very good for long term reliability of the components....of course given a proper wet/fillet.
Jn
Last edited:
I thought that best practice was to wet the tip of the iron with solder -- not have it dry or dripping -- to facilitate heat transfer (and speed) from the iron to the part and trace.
Then apply solder to the other side of the part so that the solder flowed into the joint when it got hot enough and wouldn't cool unevenly (aka a cold solder joint) because everything was at the same temperature.
Then apply solder to the other side of the part so that the solder flowed into the joint when it got hot enough and wouldn't cool unevenly (aka a cold solder joint) because everything was at the same temperature.
Last edited:
You are correct in that it aids heat transfer. However, while that transfer is occurring, the part and trace will very quickly rise to the tip temperature. Moving solder application to the other side of the part will not allow the heat capacity or the energy required to create the phase change of the solder to prevent the part from going to tip temp.
By directly putting the solder between tip and part, you are basically guaranteeing that the part is seeing exactly 183 C or 221 C molten metal, depending on alloy. Think about using a torch to melt and drip solder over a surface... the surface only sees the exact melt temp. This is just like how people melt chocolate in a double pot setup with water in the bottom one. No matter how the stove is set, the chocolate will never see more than 100C.
I find too many people will use tip temperature to compensate for oxidized flux, incorrect flux, tarnished parts, or even because of lack of patience.
25 years ago, before ROHS, I had to pioneer tin/silver soldering techniques for wires from AWG 30 to 4/0 and copper buss bars. With over 20,000 joints made by the techs, no failures over 20 years.(note that all our tin/silver joints are cooled to liquid helium temperature and cycled to liquid nitrogen several times a year..room temp joints are always tin/lead for reliability.)
Jn
Oh, btw. Cold solder joints are not made as a result of a cooling gradient. Poor technique, bad flux, insufficient heat are basically what causes them, however, I have indeed come across times when excessive assymetric cooling sets up internal forces within the solder joint, and that can result in a peeling failure between the intermetallic layers of copper/tin. But you make a good point.
Pps..you both do make very good recommendations, other than some details I've learned over the years, we are in violent agreement..
Last edited:
Absolutely. For the fine tips, I sometimes find I need a set point of 350C to get something done. Unfortunately, that is a temp many epoxies cannot do well with.
Simply by going with a larger mass, I can drop the temp, but unfortunately for fine detail work, the tip can be larger than the part.
A few days ago I had to replace a micro SD socket on my 3D printer main board, no way to go big tip, I had to use a microscope to even see the terminals.. and yes, it was very difficult to get the solder to reflow because the shell is soldered to a high fill pour of copper. Took temp up to that needed, lost one pad as the glue holding the copper to board let go. Luckily, it wasn't a signal trace.
Jn
Simply by going with a larger mass, I can drop the temp, but unfortunately for fine detail work, the tip can be larger than the part.
A few days ago I had to replace a micro SD socket on my 3D printer main board, no way to go big tip, I had to use a microscope to even see the terminals.. and yes, it was very difficult to get the solder to reflow because the shell is soldered to a high fill pour of copper. Took temp up to that needed, lost one pad as the glue holding the copper to board let go. Luckily, it wasn't a signal trace.
Jn
That took WAY too long to heat up. That length of time would be needed for something with LOTS of thermal mass.
It appears to be an IDC pin on a PCB with no ground plane and should be only a couple of seconds. The tip needs
to be properly tinned - that one is not. It's not possible to tell the tip temperature from a video so I can't tell
if it's a combination of temp AND tinning. My GUESS is the temperature is too high and the tinning is non-existent.
What I find helps sometimes is to clean the tip with rosin flux while hot and then re-tin it. The tips do reach a point
where they're worn out. The end result seems OK but I would not accept it after seeing the gyrations it went through.
G²
I agree that the tip should be slightly tinned for better heat transfer and that it took way too long to heat up. 15 second video for one joint? I think a whole XLR connector with 4 solder joints takes me about 15 seconds. But around half of the "How to solder xyz" videos on Youtube are also crap. Most of the "instructors" are seemingly incapable of making a proper joint first-time and thus tend to reflow everything at least twice. I think this is one of the good ones, must have been a training video for corporate customers back then:
Basic Soldering Lesson 1 - "Solder & Flux" - YouTube
Basic Soldering Lesson 1 - "Solder & Flux" - YouTube
Isn't 'soddering' illegal in Wales?
I've always thought it sounds a little filthy.
another one...
hi friends, again!
Just watched one of those nice 60ies/70ies movies, this time from Mill-Max
at around 40', they show a neat trick: stuff the soldering of a connector with cold solder, melt it, and just stick the wire in. Seems very handy, but... isn't this resulting in a cold solder-joint, too?
hi friends, again!
Just watched one of those nice 60ies/70ies movies, this time from Mill-Max
at around 40', they show a neat trick: stuff the soldering of a connector with cold solder, melt it, and just stick the wire in. Seems very handy, but... isn't this resulting in a cold solder-joint, too?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.