Soldering chips and transistors. How hot? How fast? Sockets?

Started by elenore19, December 23, 2010, 10:02:09 PM

Previous topic - Next topic

elenore19

Hey everyone -
My order from Pedal parts plus got lost by USPS.
So I'm in this dilemma....I have always used sockets for transistors and chips because I always felt like I'd fry them, and if I use sockets it's one less thing I can mess up. My sockets were in my PPP order...
That being said.
I have a digital soldering iron (got a sweet deal on it), so I can adjust the exact temperature of my iron.

Is there a suggested temperature for soldering these guys? I've always used 600F for soldering everything, but I've never just straight soldered chips and transistors to a board. The chips I'm soldering are 16 pin and 8 pinned chips (in a rebote delay 2.5) and the transistors are for a fuzz face.

Is there a maximum of time I should have the iron on the leads before I'll burn out the transistors/chips?
Or should I just stop thinking about it and wait for my sockets to get here?
Any help would be greatly appreciated as always. Thanks everyone.

-Elliot

Jhouse

As long as you're careful and don't leave the soldering iron on there forever, you'll be fine. It doesn't take much solder to get the part to stick to the board and that means that the iron doesn't have to be on there very long.

That being said, having sockets is always nice so you can make substitutions or replace parts easier.

Personally, I would try to wait for everything to get sorted out, but if you absolutely HAVE TO HAVE this pedal put together, you'll be fine if you don't go over board. I usually set my soldering iron to about 25 watts. I think that is about 460 degrees.  :icon_confused:

phector2004

#2
Hmmm I'm aching to test out my new soldering station! But I read the temperature should be higher, 700-800F... is it the temperature or the heating duration doing the damage? Can anyone confirm?? Seems like a hotter iron will cut the time down, but it doesn't take too long to heat up IC legs anyways. I did a 22 pin Atmel chip with my obsolete radio shack iron a week ago, didn't take more than a second per pin as there's not much metal there... I've over-heated other stuff, and its more resilient than you'd think.

My only advice here is to solder in the resistors, caps, etc. to "warm up" before you tackle the sensitive stuff. Don't sweat it, you'll do great

Good luck!



EDIT: Might wanna check your soldering station manual/the company's website... apparently for 63/37 I should set mine to 650 F

Jhouse

I know most 60/40 rosin core solders have a melting point of around 350F-370F. 

caress

solder pins 1 and 8 then put in and solder a few other components.
solder pins 2 and 7 etc.

if you just yake a minute or two between 1-3 solders on an ic you'll definitely be fine. just get in with a hot iron and get out quick. 1 sec each time - 2 sec at most

Zapp Brannigan

Quote from: elenore19 on December 23, 2010, 10:02:09 PM
I have a digital soldering iron (got a sweet deal on it), so I can adjust the exact temperature of my iron.

Is there a suggested temperature for soldering these guys?

250 C (approx 482 F) is the most suitable temperature for almost everithing except may be some bigger stuff. Absolutely no harm to the component, no overheating of soldering alloy, it stays shiny. Your 600 F is too hot for semiconductors, I think.

soupbone

I use a small copper alligator clip to clip on to the leads,so the heat will go straight to the clip.

Hides-His-Eyes

Quote from: soupbone on December 24, 2010, 06:29:31 AM
I use a small copper alligator clip to clip on to the leads,so the heat will go straight to the clip.

This is what I do, only time I've killed a semiconductor with heat was the first time I tried perfboarding, I had the legs of the transistor connected to every component, so it was getting heat every single component I put in...

jasperoosthoek

You said that the transistors are for a Fuzz Face? If they are silicon that you'll fine but for germanium you have to be super carefull. Always, clamp the transistor side of the lead wire with some pliers. This will act as a heat sink and prevents heat from entering the transistor. Also wait until it cooles down before you release it.
[DIYStompbox user name]@hotmail.com

R.G.

I've been soldering for over 40 years now. It's not technically correct, but I generally ignore the specific temperature, and only go by my view of how the solder behaves.

You have to know a little thermodynamics (very little!) to understand skilled soldering. Every material reacts to heat by two processes - thermal resistance and thermal capacitance. The resistance/capacitance makes a small time lag for the heat to get through each little piece of the material. The lower the resistance (that is, the material the heat is conducting through makes a difference) and smaller the capacitance (lower mass) of each section, the faster heat gets from one end of the chunk o' stuff to the other end. The more mass involved, the slower everything works. The way this works out is that every wire is a chain of these little conduction/slowing sections.

If you have to reach a specific temperature at one spot (the joint you're trying to solder), then you have to put in enough heat to get the joint's mass up to temperature; in addition, you have to supply enough heat energy to offset the heat that's flowing away down the wires. Heat flowing away from the joint on the wires is wasted for your purposes in soldering. All it does is heat up other things on the other end of the wires, and cools down your joint, making soldering harder.

A lot depends on the temperature of your soldering iron. If the iron is just barely hotter than the temp to melt the solder, then the joint temperature only approaches soldering temperature slowly, as the long term end result of the joint "filling up" with heat. If the iron is significantly hotter than the melting point of solder, then the joint passes through soldering temperature with less elapsed time, on its way to a way hotter temperature. The reason this matters is that the heat conducting down the wires to other places has less time to heat up, because the short time the joint heats gives the heat less time to travel. So you can heat the joint to soldering, get the solder to wet and flow, and get the soldering iron off the completed joint before heat has time to travel very far down the wires.

The higher temperature of the soldering iron lets the joint get to temperature, flow and wet the solder, and let you remove the iron **before the heat can travel to and heat anything else**.  I can hold a snipped-off resistor lead 1" / 25mm long with my thumb and index finger and get a good solder joint on the other end of the lead before the end I'm holding gets too hot to hold. Well, OK, mostly. Nine times out of ten.  :icon_biggrin:

Think of the difference between grilling steaks and boiling stew meat. If you grill a steak, the outside can be charred before the inside is fully cooked. If you boil a piece of stew meat, the inside and outside of the chunks of meat get cooked at about the same time. Heat takes time to travel. Likewise, you can burn rice in the bottom of a pan before the water boils if you have the heat too high under the pan. If you turn the heat down to low, the heat has time to travel to all parts of the water before the bottom layer scorches. Same principle: the heat gradient is steep with big temperature differences.

So to solder well by hand: use a temperature that's just low enough to not cause massive oxidation on your soldering iron tip. Get the joint ready to solder (clean, all parts on place, wires pre-tinned, etc.) and then touch the smooth-mirror-shiney tip of the iron to the joint, adding solder just as the heat comes up. You can **see* the solder temperature rise to molten as it travels across the few mm of the joint. When it's all flowed smoothly, get the iron off the joint. A good solder joint on resistor leads takes about two seconds.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

twabelljr

I use R.G.'s Rule#4: Get in fast with high heat and get out. I set my soldering station on 600 F. I have even used 700 F. I also use 63/37 eutectic solder as it melts and hardens quicker. I have never "killed a component by soldering it. http://www.geofex.com/article_folders/how_to_solder.htm
I like to use sockets for easy parts swapping like mentioned above, but if I wanted to build something "right now", I would just go for it.
Shine On !!!

PRR

Silicon transistors will WORK while so HOT that the solder on their legs is molten.

Probably not very good or very long. But I saw it work for several minutes.

Germanium-- treat like the demon inside is made of wax. Long leads with heat-clip. Let the body cool between legs.

What R.G. and others say. Prepared parts (clean, tinned), HOT iron, get in and out quickly.

> I can hold a snipped-off resistor lead 1" / 25mm long with my thumb and index finger and get a good solder joint on the other end of the lead before the end I'm holding gets too hot to hold. Well, OK, mostly. Nine times out of ten.

That's very good practice. (See my thumbs?) Test your joint. Rip it apart. An excellent joint will break in the wire not the solder. Generally it is OK if the solder breaks first but not easily. If you tug and it falls off, it's not a solder joint but a failure (today or in the future).

I don't think I've ever been happy under 650F. But it depends a lot on the relative size of joint and iron, and if it is constant-power (dumb), weakly regulated (low-price "digital"), or WELL regulated (the good irons with sensor IN the tip and lots of reserve power).

The only thing I would add is, like shooting tigers or making music, is PRACTICE! PRACTICE! PRACTICE! Don't go out in the tiger-woods, or the Whisky a Go-Go, if you can't nail your shot/note EVERY time. That's what tincans and bedrooms are for: PRACTICE.

Cut-up an old VCR and then solder all the bits back together GOOD.

WATCH how solder melts, FLOWS, and WETS (or not!) both sides of a joint.

Use a GOOD light. Hand-soldering is visual. You need to see the several states of solder (mushy, balled-up, wet, hard).

Wear safety or reading glasses. Keen-eye youths should protect the eyes with safety-glasses or weak +0.75 or +1.0 reading glasses ($10 at the drugstore). Us guys who have 40 years know we need +3 or more to see up close (in addition to any far-vision correction). Parts smaller than TO92/DIP, you may need actual magnification with built-in lamp.

On Silicon devices, I only use sockets when I know I will be swapping the part. Last time I did that, I used TL072s from Radio Shed but planned to swap to a sexier chip once debugged. The time before that, I was working a 28V part at 34V (and in fact I had to replace it twice in 25 years, so was glad of the socket).
  • SUPPORTER

emstin1

ICs not too sure, but I always put an alligator clip on the leads of transistors if I'm not using sockets.  Whether it actually acts as an effective heat sink is another matter entirely, but I've yet to fry one from soldering. 

stringsthings

Quote from: elenore19 on December 23, 2010, 10:02:09 PM
... but I've never just straight soldered chips and transistors to a board. The chips I'm soldering are 16 pin and 8 pinned chips (in a rebote delay 2.5) ...

are you using the PT2399 in your rebote delay build?  if so, according to the datasheet, the PT2399 is a CMOS chip ... sensitive to static electricity ... my suggestion would be to practice direct IC soldering on less sensitive parts first ... ( and i would also recommend having spare chips around )

StereoKills

Quote from: phector2004 on December 24, 2010, 12:04:08 AM
But I read the temperature should be higher, 700-800F...

The soldiering tips designed for these temperatures are usually for lead-free soldering. Depending on the part, 600-650 should be plenty!
"Sometimes it takes a thousand notes to make one sound"

petemoore

  For the trouble it is to assemble a board, socket soldering guarantees all including the chip isn't wasted by heat stroke. For consideration of effecient debugging, sockets are required. That said: if you're real good, without socket is possible.
 Fast, especially where the solder tends to not want to go, once around the entire pad-ring of copper is sometimes the stubborn part to get going.  
 The wiped off leads seem to welcome quickly hot solder/iron tip, when solder flows on all the adjoining surfaces add only enough more [because it cools the joint some] to create sufficient physical rigidity. Retreat the tip using care and the chosen route you find works, I prefer straight up off of the board, toward one of it's connections, away from all others. Just enough fresh tinning of the tip [right before the joint heating begins] and not too much solder added later seems to make the solder less stretchy-drippy during tip removal procedure.
 How the heat transfers can vary in application time needed, depending on the constants you can see which determine thermal mass [how fat is the wire, how many are there, how large is the lug they're going into, how much solder is also added joint, anything else that would impede heating or suck up heat] and possible unknown factors such as dust, debri, or oil layer on the copper pad.
 All the cleaning techniques work, cleaned metals of everything involved in solder-joining makes faster and more consistant heat up rates.
 I like anything which cleans the copper board pads to shiny, such a very fine sandpaper then wipe with rag [fingernail polish is stinky option-read label/use with gloves etc., but it cuts through oil instantly] iron preparation: wetwipe and re-tin soldering tip.
 If everything is flowing nicely, 1 tip cleaning may do well for many nodes.
 Heres how I do the initial start in perfboard, clean/hot iron, board pinned down, lay the thincore solder across the pad and against the lead, heat-cut that bit off with the pin-tip of the iron, let the heat soak for a second and bring the flow-portion of the tinning solder amount I cut off down to the pad and around the wire, using only little bits more solder. My solder tends to seek the wire and sometimes avoid the pad, ensuring there is 'flow' [only enough solder to conduct heat quickly] from the tip to the stubborn surface I want to get hot enough to 'grab solder', the tip side of the joint is always the hottest.
 The solder is one of the heat conductors, and can span across where two surfaces barely touch [wire and pad for instance], improving reliability of the heat transfer.
 The solder is also can be thought of as a 'cooler tube full of metal', it has thermal mass and requires heat/cools where added.
 There's a 'right amount' of solder for heat transfer [location is everything, touching the 'stubborn' surface that refuses to heat fast, is recommended], a right amount for flow to all surfaces, sufficient rigidity of the joint requires only a teeny bit more if any than what creates flow and complete surface 'grab' coverage.
 About a 1 second pause to allow the 'flow solder' to transfer heat into the metal-mass, then less than a seconds worth of new solder melting, heating the solder core works most reliably when started at the iron tip, asking a wire to pre-heat the end of the strand and send heat up the solder core can allow the joint to overheat while the tip works on heating the solder up...slowly...as the heat must first travel through the wire and the wire/cold solder touch point is tiny.
 With the tip tinned and hot, flow to cold new solder is instant, against a wire the heating is unpredictable. Once the core-end is molten it can be drawn in the direction where additional solder is needed.
 Trying to preheat a cold solder core by holding it to the other side of a 'dry wire' which is being heated by the iron tip is mistake, the teeny heat transfer touch-point between the wire and the solder can take days to become a melt-point which begins sending heat up the solder core, while the phelonic board starts baking, always use the tinned hot tip or already flowing point to preheat the new solder on the reel, flowpoints near the tip don't have constant high temperature like the iron.
Convention creates following, following creates convention.

greigoroth

Hi!

Any tips about tempterature from guys using lead free solder? I just completed my first build (GGG green ringer - thanks JD!) but my tips oxidised pretty much straight away.

What is the "normal" amount of time to solder a leg of a component to a pad? How much time warming the pad, how much time the component?
Built: GGG Green Ringer

R.G.

Quote from: greigoroth on December 29, 2010, 04:06:17 AM
Any tips about tempterature from guys using lead free solder?

Lead free solder should not be heated above room temperature.  :icon_biggrin:
Using lead free solder is a bad idea unless you are (a) a corporation and (b) forced to do so by your political heroes.

QuoteI just completed my first build (GGG green ringer - thanks JD!) but my tips oxidised pretty much straight away.
Yes. This is only one of the many joys of using the existing incarnations of lead free.

QuoteWhat is the "normal" amount of time to solder a leg of a component to a pad? How much time warming the pad, how much time the component?
One second warming the lead if it's a resistor or cap lead, then touch the pad. If it's an IC lead, touch the lead and pad simultaneously, and solder should flow within one second. These times are for 60-40 electronic solder with a flux core. If you insist on or are forced to use lead free, the times will vary hugely depending on the alloy (and there are many). If you are forced to use lead free, use tin-copper-silver, especially not tin-copper, and be prepared to rework/re-tin your soldering iron tip repeatedly. SnCuAg is the best of the lead frees, and it still uses a much higher temperature than tin-lead. So problems with overheating components are worse, as are flowing/wetting problems, and the increased frequency of cold joints.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

greigoroth

Thanks RG, me thinks I'll un-unlead myself. The lead free took a lot of work, and with my unskilled hand I'm pretty sure I cooked the life out of every lead, component and pad on that PCB!

Thanks for the detailed answer about time warming stuff - it is really hard to guage this sort of stuff as a total beginner.
Built: GGG Green Ringer

R.G.

Soldering is much like playing guitar - you're going to make mistakes early on as you develop the skills and experience. The trick with soldering, just like guitar, is to make your mistakes in places where it doesn't matter much. Having exactly enough parts, especially expensive parts, to make a pedal when you do not yet have the manual dexterity and experience to solder them correctly is a guaranteed recipe for frustration.

Count on messing some things up. Go DELIBERATELY learn to solder on things you PLAN to throw away. This is NOT wasteful, because you are not trying to build something you're going to keep, you're trying to develop the skill. Once the skill is developed, then make things you plan to keep.

Develop the attitude that the finished product is just the garbage left over from the enjoyment of doing the work to make it. Developing the skill is a goal itself, just as much as having a working pedal is.

If all you want it a working pedal, it's cheaper and faster to just work more hours to get the money to buy it than it is to buy the tools, build the skills and then build the pedal. It's only if you want to make pedals that it's worth it to put in all the time to learn to do it well. And worst of all is neither getting the pedals you want nor learning to build them well, and getting hugely frustrated in the bargain.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.