DIY strain-relief on wires to PCBs ?

I have found on some PCB device that wire leads attached directly to the board eventually break at the point of soldering. Has anyone had any drawbacks to putting a dab of silicone sealant
right there to give a little strain relief ? Better ideas ? Seems like a no-brainer but I thought I'd ask ...
I'm talking about such things as a MiniSSC-II, etc.
Thanks ! JCD
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
you could also drill a hole through the pcb, and route your wire through the hole before soldering it to the pad. Then the strain is on the wire and not the solder joint. or route it through two holes. or one hole, but a knot of wire on either side of the hole.
I've seen sealant/glue used before in devices i take apart so I'm sure that works too
pogo wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

That's a good idea! Thanks !

relief
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
pogo wrote:

Since silicone sealant requires a setting and curing time (and contain acetic acid, formaldehyde, and other icky stuff) why not get one of those cheap $1 glue guns and some glue sticks. These heat up in a minute, and the glue sets in 30 seconds. The "glue" is not really glue, but a thermoplastic. But test first to make sure the glue has no electrical conductivity. Most don't, but obviously the stuff with glitter mixed in is not for PCBs!
Obviously you don't want to use this near components that will get really warm, like a power transistor.
You may also find it helpful that if there is more than one wire from the PCB to bunch them all together, and tie them with a tie-wrap. That way all the wires act as their own strain relief. Sort of share the strain thing. And as Andrew notes, if there's a free hole in the PCB, or you can drill one without wrecking it, that's a good choice, too.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

relief
Thanks - good tips! I have a small glue gun so that sounds like a better idea than silicon sealant - which seems to dry up on me the same day I open it.
Thanks again ! JCD
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
pogo wrote:

Generally, it's better to use header pins for connections rather than soldering directly to the board, if there's any wire motion expected. You can use stranded wire into a header connector. (If you're using solid wire and are having wire break problems, going stranded is the first step.) Header connectors with crimp pins or insulation displacement connectors are what's generally used to solve this problem.
                John Nagle
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
John Nagle wrote:

Good advice. Another advantage of a header over solder is that it is easier to disconnect and reconnect. That can be important on an experimental robot.
Another option is to use an FFC (flat flexible cable) or some other cable/connector combination that is specifically designed for a high vibration/shock environment. I have used FFCs with locking headers for board interconnects in several designs and have never had a connection fail.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.