That was probably endemic in cars in the late 80's/early 90's - I got an 87 RX7 cheap because the clock and display module in the center of the dash was completely dead. Clearly it wasn;t something liek the computer or sensors, because the clock wouldn;t use any of that. I took it apart, and saw the problem pretty quickly - the pins for the docket that the wire harness plugged in to were for the most part all broken joints. Reheat, reflow, add some fresh solder - it worked perfectly until the car died 5 years later.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
AttuvianPerhaps I might have gotten off a bit easier with you had I toned down my post and added the "your mileage may vary" caveat.
I'm just giving you the gears a bit (it's that New Jersey in me). I know how good and competent you are.
For almost every kind of connection other than the right kind and gauge of wire into the right kind of socket -- it's hard to beat a good solder connection or a good braze. That goes double for any sort of "3D" connection to things like multiple-pole multiple-throw or rotary switches.
Of course there are also places where you need a little more than just canonical 'soldering' knowledge (as from those YouTube videos where the perps continually pronounce the "L" in the word). One of them is the joyful sort of situation I had with the windshield-washer control on my mid-Nineties diesel Suburban -- this had three real stout pins from a component, I believe a relay, that made absolute, positive contact via huge solder blobs with the pads on the corresponding board. The problem, as it developed, was that vibration and shock would break the perhaps Friday-afternoon or Monday-morning GM fabrication quality, so the pins would come away from the solder blob, and subsequent road vibration would beat the stiff pins against the softer solder until they had 'peened' it away all the way around. At which point the wipers would abruptly quit, until perhaps some chance bang re-established contact for a while. This was completely invisible if you took the component out and eyeballed it, unless you had good light and knew exactly what to look for -- after which a very quick proper flux and reflow would solve your problem (and lower your Rain-X bill substantially)
There is also the joy of implementing Parylene on your circuit boards post-fabrication. That stuff is a wonder material ... if you do the right defluxing steps, or use a good no-clean flux. It is not quite as wonderful if you get a cracked trace or pad delamination and you have to try fixing the trace with something like nickel print...
THANK YOU for not bringing up suitcase connectors. I never see these without thinking about how my father carefully taught me to hate expediency in all its myriad forms. Telephone-company punch blocks were poor enough! now let's see how badly we can nick the wire in mid-span sight unseen!
Charlie Brown said it best. AAUGH!
(Should we bring up the joys of 'matchlight solder' and a proper hot-air gun, for those who can't work the iron, the flux, and the solder pieces simultaneously?)
of course we all need to solder wires to various types of switches and connectors.
i don't think it's fair to compare soldering a circuit on a non-etched board to an etched board, much less to various switches or other wired connectors.
greg - Philadelphia & Reading / Reading
5-10 years ago, I'd be right there. But it is now so cheap to get a PCB made, it's almost pointless not to. At least not after you've proven the circuit to work on a breadboard.
I just drew up a board to mount my pushbuttons, and I till probably do another to be the 'panel' part that you see. Because the buttons I have chosen are panel mount, so I can;t attach the wires to them before putting them in the panel, and they are small enough that trying to attach the wires behind the panel, under the layout, is more of a challenge with my vision issues than I want to deal with. It was hard enough making up a test cable on the workbench. So, PCB mount them, with an RJ45 connector, and I can use the cheapest patch cables I can find to connect each button set to my controller board.
Since you can get all sorts of colors of solder mask now, I can make the actual panel part as a PCB too, just plain non-plated holes of the correct diameter, and some lines silkscreened on, and I have a very neat panel.
Given the size these come out to, the 10 boards for $2 price nets me - a whole lot, I haven't finalized them to see just how many can be panelized on a single sheet of the maximum size allowed for the low price - but it's goign to be a dozen or more.
My actual controlelr boards are much bigger, but still, in quantities of 10, they are $1 per board, les sif I order more. You can;t buy the nasty chemicals to etch a board yourself for that price, the laser toner transfer method is less than ideal, and in both cases you have to still drill out the holes and neither method makes plated through holes, so you have to solder jumpers and/or both sides of a component to link both sides of a double sided board. Drilling is the worst - FR4 is tough on drill bits, and you want good ventilation and a dust mask when doing it. It's just not worth it any more.
gregc Attuvian if the circuitry is settled and its installation is intended to be permanent - I strongly lean to soldered connections. have you ever soldered up so many circuit boards not using an etched PCB for the circuit?
if the circuitry is settled and its installation is intended to be permanent - I strongly lean to soldered connections.
have you ever soldered up so many circuit boards not using an etched PCB for the circuit?
Attuvianif the circuitry is settled and its installation is intended to be permanent - I strongly lean to soldered connections.
Overmod,
Thanks for your extensive, point-by-point reply. And for throwing me a bone or two. I have been properly schooled.
On the other hand, I am not unaware of the wide variety of applications for making unsoldered connections and the means of doing so. I employ them where called for (I'll leave a discussion of suitcase connectors for others). It's just that, for the type of work displayed above - if the circuitry is settled and its installation is intended to be permanent - I strongly lean to soldered connections. I have at hand the necessary tools and materials. It is one of the things that I do best, reaching back to my Navy days and then having built specialized test equipment from scratch for a major defense contractor. Old school has served me well and, at my age, I am surely not as technologically adventuresome as some in the hobby.
Perhaps I might have gotten off a bit easier with you had I toned down my post and added the "your mileage may vary" caveat. Thanks again.
John
His rant is over, but mine is just getting under way...
AttuvianMay I suggest that though a particular un-soldered board is not likely to be snagged or bumped does not guarantee that its connections are permanently secure.
Nobody is talking about 'permanently secure' like nature's way of eliminating embryonic DNA: the idea is that, even if you have to use some kind of spring tool to release the deep contacts holding the 'patch' wires in the board, the connection is reversible and 're-makable' in the field, at will, without special tools, materials, access to electricity or fuel, etc.
Are you of the opinion that all connectors with pins, or headers, should be carefully soldered on their 'open' sides to assure that they stay forever conductive? I doubt you will find many who'd agree, or wouldn't use something like conductive grease to assure continued conductivity in corrosive or 'weather' conditions rather than introducing heat-flowed metallurgical connection.
The frame to which the board is attached could be bumped, jarred or vibrated unexpectedly - from a distance.
You must not be familiar with the contact system in these prototyping 'breadboards'. It is possible that a really good bang, or just the wrong sort of acyclic vibration or shock, could knock something like a transistor with a big floating heat sink out of contact 'inertially.' The number of Gs needed to do that to a small-gauge wire run is almost unthinkable; damage to 'other things' would be far more significant long before you'd reach such a point. And if pulling wires out accidentally is, in fact, a concern, a few strategic dabs of silicone or acrylic caulk, or perhaps even something like blue-tac poster adhesive, would keep the wires 'enough' in place. And still removable and reconfigurable without tools...
Frankly, aging itself affects joints and contact surfaces as well: think of the possibilities of dust or surface changes from tarnish, dissimilar material contact, arcing and pitting, even variations in rates of expansion and contraction.
All relatively tiny, in the context of these actual 'breadboards', unless you are modeling somewhere like the inside a smelter or a shack along the route of the proposed Yukon container railway. If you're concerned with contact corrosion you can choose your wire or contact-plating systems accordingly, or use conductive grease or other compound, or periodically pull and reinsert any 'trouble' joint. The sheer contact area between wire and contact socket is so large that 'arcing' or 'pitting' would be slight, as would be material transfer from socket to wire or vice versa from current effects. Since insertion and removal are 'normal to' the clamping force of the socket, there is little consequence from expansion or contraction even if the wiring puts erroneous 'preload' on the joint to pull it out. Periodic eyeball inspection would catch most potential problems long, long, long before 'just before the nuts work loose. (And of course the 'solution' is to stick the wire back in the socket again, something remarkably not available to most soldered joints or splices!)
Why propose that electrical contact for the applications that are being considered here NOT be soldered?
There are plenty of times and places for this. Indeed, if complex wiring to some sort of backplane or switch matrix is desired, one could even use wavesoldering techniques to ensure permanent contact at minimum individual cost and expense. But that's almost manifestly not what the people proposing use of the prototyping 'boards' are intending: they are building up wiring piecemeal, sometimes rebuilding parts of the layout or revising large chunks of the wiring layout or construction. Breaking large numbers of soldered joints, many insulated with 'nonmaintainable' heat-shrink tubing, is a very poor experience, particularly on your back under benchwork reaching up over your head in half-dark.
Meanwhile I refer you to the substantial, and very-long-established, community using wirewrap techniques for assembly. I don't think many people have credibly questioned the integrity of proper wrapped joints, even though solder is not involved, and in fact would be difficult and perhaps dangerous to involve. Note the technology that has replaced much of wire-wrap for prototyping...
... the tools and materials are readily at hand to [solder joints].
You'd be surprised at how much complexity in materials, flux removal, and so forth is required for the 'range' of expectations even in electrical connection. And perhaps how much of it poses burn dangers to the elderly, or less-coordinated -- or that isn't RoHS compliant or produces fumes in confined spaces.
Meanwhile, the skill involved in stripping a wire end to gauge and inserting it fully by pushing into a breadboard socket is much, much less, even if 'special tools' that avoid nicking the wire, that straighten the wire end, or even clean or tin the contact area are being considered.
Surely a person in this hobby should be capable of at least modest (if not expert) soldering technique and results.
Now, there I'll agree with you. But there is a long, long history of articles in MR, how-tos on the Internet and other media, and various posts here that indicate soldering is still often an arcane and ifrequently incorrectly- or even ineptly-performed operation. The right choice and technique for fluxing alone is a complicated and often nonintuitive learning curve -- assuming you have someone other than Mr. Murphy and Mr. Finagle as instructors.
We have a thread going right now in which basically everyone appears to be carefully avoiding any discussion of how to relate precise or even approximate tip temperature to solder composition, type of tip or joint, method of preheat or tinning, or temperature control within the iron system ... do they know that stuff? quite possibly. Could they learn what they don't yet know? Probably. Will they have the motivation? I'd be getting less optimistic. Would they prefer a less complex alternative? I'm not at all sure I'd bet agin' it.
Should MR have effective soldering 'clinic' content -- and an effective and directed community search to find that content specifically and easily when it is needed -- I'd be among the first to recommend everyone learn. And there are, in fact, many places where soldering is preferable to breadboarding -- anywhere, for example, high current or large-diameter conductor size is desirable, or when stranded or Litz wire is 'in play'. The techniques for making proper types of splice or connection, pretinning, and cleaning vs. using no-clean flux are relatively easy to learn ... but they have to be learned, and they are really far better 'learned right' than discovered via trial and error, particularly when the 'errors' may not reveal themselves immediately.
Theoretically, someone might want to suggest that we dismiss bolts and welds and begin to glue our cars largely together with space-age adhesives. To my notion, it's just uncalled for.
Evidently you would be surprised if you actually studied modern automotive fabrication and see how much of this there is. And in fact how many different nontraditional welding or joining technogies and techniques other than bolted or threaded fasteners or traditional fusion or blanketed welding are in use. Keep in mind that a far better way to join small-diameter wires than 'soldering' is proper electrical spot- or shot-welding (as seen repeatedly, for example, in the YouTube video on modern fabrication of Nixie tubes) and at least in theory (although with more than a little awe at the potential terrors) this could be extended to larger and stranded conductors. So why bother with kludging around with non-sticking low-melting alloys any more?
Oh, one more thing: soldering connections is virtually as "low-tech" as doing without. IMO, of course.
I spoze you have the right to retain low-tech soldering methods. There is much more rich opportunity, in so many ways, by embracing better technique. When practically called for, that is, in a given hobbyist's opinion. Joseph Kennedy Junior would have made an infinitely better President than the Princeton flunkout, but crude soldering methods alone made that impossible.
Perhaps I am misreading something here.
May I suggest that though a particular un-soldered board is not likely to be snagged or bumped does not guarantee that its connections are permanently secure. By his very nature, Murphy shows up without warning. The frame to which the board is attached could be bumped, jarred or vibrated unexpectedly - from a distance. Frankly, aging itself affects joints and contact surfaces as well: think of the possibilities of dust or surface changes from tarnish, dissimilar material contact, arcing and pitting, even variations in rates of expansion and contraction. The possibilities, though perhaps very minor, are endless.
Why propose that electrical contact for the applications that are being considered here NOT be soldered? Surely the tools and materials are readily at hand to do so. And surely a person in this hobby should be capable of at least modest (if not expert) soldering technique and results.
There are, indeed, places where successful hobbyists "stretch the envelope". But this isn't one of them. Theoretically, someone might want to suggest that we dismiss bolts and welds and begin to glue our cars largely together with space-age adhesives. To my notion, it's just uncalled for. Carl425 (above) is correct.
Oh, one more thing: soldering connections is virtually as "low-tech" as doing without. IMO, of course. Rant is over.
wvg_caI don't feel this would be a permanent solution, temporary only ... just my opinion...
The way he has these oriented, like tiny versions of patch bays, it would be comparatively unlikely that something would snag components in a way that would irrecoverably damage them, or pull out connections in a way that damages the contacts in the 'sockets'. You could, of course, strain-relieve or clamp the wiring in any appropriate way, or use a small dot of adhesive or sealant to secure the wire more firmly when it has been inserted.
I don't remember the number of 'cycles' of wire insertion and removal that these things are rated for, but it is certainly orders of magnitude more than typical 'experimentation' or periodic rewiring or changes to connections at appropriate wattage for this relatively small size. Supposedly the socket connection is long enough to give good long-term contact to the stripped periphery of the wire when insulation is stripped 'to gage' keeping the wire straight and unnicked and the straight end is then carefully inserted. At least in theory the sockets can be made 'comparable' to those used in ordinary push-to-lock connectors or headers, and the individual wires provided with soldered or crimped-on contact pins ... and I don't remember anyone claiming those have particularly limited utility or lifetime in normal modeling work.
A comparable board made with wirewrap pins would suffer much more from casual impact, as bent pins will be highly likely to short, and fractured pins may not be cost-effectively or even functionally repairable. It's certainly a technology with greater mechanical reliability for nonsoldered connections. You'll need very special tools, though, to get the connection density out of wirewrap that one of these solderless breadboards affords at small permissible pitch size.
yet another example is i'm helping someone put together about a dozen "simple" detection circuits (3 components) for panel display. the owner has used solderless breadboards to mount relays and diodes for single button switch machine control in yard ladders.
while i've made custom PCB for more complicated circuits and have soldered up custom circuits on partially etched prototyping boards, using solderless proto boards seems like a reasonable approach for the typical modeler who it willing to build such circuits.
Using these proto board removes a huge impediment to building your own circuits for your layout.
here's another example. a combination of custom PCB that is used in multiple locations elsewhere on the layout, and a breadboard for some unique circuity.
i don't feel this would be a permanent solution, temporary only ...
it doesn't have the stability of a permanent [read soldered] solution, it's too easy to pull wires out of the breadboard ..
just my opinion ..
gregcknowlegable modelers are successful at pushing the limits.
I strongly disagree.
Knowledgable modelers are successful because they don't push the limits. They are where ideas like "best practices" come from.
I have the right to remain silent. By posting here I have given up that right and accept that anything I say can and will be used as evidence to critique me.
rrinker I know you said the guy's an MIT PhD and all that, but solderless breadboards just were not designed for permanent circuits.
not a PhD, but a switch designer who understands issues involving electrical contacts. He has reliable operating sessions every week.
knowlegable modelers are successful at pushing the limits.
Frankly, I think that is a disaster waiting to happen. If the form factor and layout are important, there are plater perf boards laid out exactly the same as soldeless breadboard, with the linked sets of holes and even the power rails, so you can take a design both physically and electrically from a breadboard and make it permanent.
I know you said the guy's an MIT PhD and all that, but solderless breadboards just were not designed for permanent circuits. A solderless solution for permanent circuits is wire wrapping.
If it works, great, but it's all too easy to lift a wire or a passive from a breadboard by accident. The ICs are probably fine, they aren't goign to fall out any time soon.
a while back there was a discussion about using solderless proto breadboard for permant circuity