Maker Pro
Maker Pro

Methods for potting connectors?

M

Michael

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

Thanks!

-Michael
 
G

GregS

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

Depending on the temperature and opperating conditions, I have
had good luck with hot melt glue in ordinary conditions.
The hot melt is simi flexable. Plummers Goop also is a much
stronger and less flexable glue. I do use a flexable epoxy sometimes.
Hardman 04007

greg
 
M

MooseFET

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

You can do low volume production with gravity pour casting instead of
injection molding. You can get the potting material put up in bipacks
that you use with a gun and mixing nozzle. The mold can be made from
plastic.

You can also use heat shrunk boots instead of potting.
 
J

John Larkin

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

Thanks!

-Michael

A 2-part pourable silicone is ideal. Get the official stuff, or go to
Tap Plastics or equivalent and get the hobbyist mold-making liquid
silicone. There are some cheaper rubberey arylic casting compounds,
too.

Master Bond is a good source for official electronic casting stuff.
They are helpful.

I'd avoid electrical tape; the adhesive might react with the potting
stuff. Maybe wrap some mylar to form the tube, and tape that.

John



John
 
M

Michael

Jan 1, 1970
0
Depending on the temperature and opperating conditions, I have
had good luck with hot melt glue in ordinary conditions.
The hot melt is simi flexable. Plummers Goop also is a much
stronger and less flexable glue. I do use a flexable epoxy sometimes.
Hardman 04007

greg- Hide quoted text -

- Show quoted text -

Temperature is about 25-60 degrees C. Hot melt glue (same thing as hot
glue, right?) is too rigid for our purposes and also makes a lousy
connection with the Harwin crimp housing.

How flexible is the Hardman epoxy? I'm hoping to find something as
flexible as the PVC jacket of the cable.

Thanks!

-Michael
 
G

GregS

Jan 1, 1970
0
Temperature is about 25-60 degrees C. Hot melt glue (same thing as hot
glue, right?) is too rigid for our purposes and also makes a lousy
connection with the Harwin crimp housing.

How flexible is the Hardman epoxy? I'm hoping to find something as
flexible as the PVC jacket of the cable.

Thanks!

Its pretty flexible, and I would say very similar to PVC.

I have done some stuff with Sylgard a silicone product, but
it takes a long time to set up. The Hardman stufff is pretty thick, and I
add heat to speed cure. Its does not run too much, but sometimes
I make a little form out of tape to control flow.

greg
 
D

D from BC

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

Thanks!

-Michael

Last time I got into stress relief I was doing stuff like:
springs
plastic or rubber tubes
silicone molding
using high strand count wire
doubling up stranded wires
using rubber insulated wire
using a guitar string
plastic clamps


D from BC
British Columbia
Canada.
 
Q

qrk

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Any thoughts on my predicament? Are there any good materials out there
for this application?

Thanks!

-Michael

I've used test points
<http://www.componentscorp.com/details/tp107.html> and tied the cable
bundle to the test point with small-sized lacing tape
<http://www.mouser.com/Search/Refine.aspx?Ntt=602-LC134-BK> or a
tie-wrap. Mount this test point with the long dimension parallel to
the cable run. This prevented pullout and wire breakage on these very
Harwin connectors. The pcb footprint for this test point can be made
smaller than recommended.

Helpful hint with these Harwin connectors: On your test cables, cut
the little latch knobs off the connector body with an Xacto knife.
This will prevent the metal latch on the pcb connector from getting
bent out of shape when you do repeated unmates. The metal latch is
fragile.
 
M

Michael

Jan 1, 1970
0
I've used test points
<http://www.componentscorp.com/details/tp107.html> and tied the cable
bundle to the test point with small-sized lacing tape
<http://www.mouser.com/Search/Refine.aspx?Ntt=602-LC134-BK> or a
tie-wrap. Mount this test point with the long dimension parallel to
the cable run. This prevented pullout and wire breakage on these very
Harwin connectors. The pcb footprint for this test point can be made
smaller than recommended.

So you're saying you put a hard right angle bend in the connector
(assuming you're using vertical connectors like us) and then tied it
down to the PCB? Seems like a decent way of handling it, though in our
application if the cables were mounted to the PCBs in such a manner it
would be hard to remove the PCBs from their enclosures...
Helpful hint with these Harwin connectors: On your test cables, cut
the little latch knobs off the connector body with an Xacto knife.
This will prevent the metal latch on the pcb connector from getting
bent out of shape when you do repeated unmates. The metal latch is
fragile.

We do exactly that on our test and programming cables. It also helps
since some of the Harwins are in incredibly hard to reach places. The
metal latch problem is one that I found a half decent workaround for:
I replace them when they break. You can actually order just the
latches themselves. Removing the remains of the old ones are hard, but
once you get the hang of it is not terrible. Installing the new ones
just takes a careful application of pressure.

-Michael
 
U

UltimatePatriot

Jan 1, 1970
0
Hi there - we are seeing an unfortunately high failure rate on the
wires going into our Harwin Datamate L-Tek connectors. We are using
Harwin M80-8881005 (http://www.harwin.com/catalogue/search/index.html?
familynum=M80-888&PartNumber=M80-8881005) connectors. It seems the
wires going into them are being bent fairly repeatedly right at the
crimp connection due to the lack of strain relief on the jacket of the
cable. This is causing huge problems for us, no big surprise!

Anyways - there isn't really a nice traditional way of providing
strain relief that I can see due to our space constraints. So what I'm
hoping to do is pot the end of the connector (the wire side) in some
sort of fairly soft material. Due to our low volume, instead of making
a mold I'd probably just wrap electrical tape around the top of the
connector and fill that in with some sort of rubbery material. Maybe
fill it up to about 4-5mm above the back of the connector. The idea is
that the potting will prevent the wires from bending right at the
crimp, and instead have a much larger bend radius.

Part of the problem may be that you crimpers are crimping too tightly.
The reason I say this is because you related that you wire breaks are
occurring at the crimp location. Over crimping is the common reason for
this.

The wire may pass some lame, undeterministic "pull test", but that
doesn't mean the crimping was done right. In a crimped connection, the
crimp has to walk the fine line between being tight enough for a
connection that won't be bothered by oxygen, and being so tight that the
wire itself gets damaged, even if it is not visibly apparent.

Backfilling a connector is not easy as the connector system usually has
to be designed for the purpose. Otherwise you will undoubtedly end up
with a connector that has been populated with conductors, but the back
fill has invaded the mating area.

Also, you would have to choose a potting compound that will stick to
the wires. This is because a simple potting compound will likely NOT
stick to the wires, which would thereby NOT provide any real strain
relief.

It still sounds to me as if the crimps are too tight, because you
should not be having broken wires at the crimp, and all your broken wires
being there is a strong indicator that the crimp tool is not adjusted
properly.
 
J

JosephKK

Jan 1, 1970
0
Depending on the temperature and opperating conditions, I have
had good luck with hot melt glue in ordinary conditions.
The hot melt is simi flexable. Plummers Goop also is a much
stronger and less flexable glue. I do use a flexable epoxy sometimes.
Hardman 04007

greg

Reasonable choices. Personally i would thixotropic silicone rubber
baked at 85C for 45 minutes.
 
J

JosephKK

Jan 1, 1970
0
You can do low volume production with gravity pour casting instead of
injection molding. You can get the potting material put up in bipacks
that you use with a gun and mixing nozzle. The mold can be made from
plastic.

You can also use heat shrunk boots instead of potting.

With some careful use of hot air you can combine the methods.
 
Q

qrk

Jan 1, 1970
0
So you're saying you put a hard right angle bend in the connector
(assuming you're using vertical connectors like us) and then tied it
down to the PCB? Seems like a decent way of handling it, though in our
application if the cables were mounted to the PCBs in such a manner it
would be hard to remove the PCBs from their enclosures...


We do exactly that on our test and programming cables. It also helps
since some of the Harwins are in incredibly hard to reach places. The
metal latch problem is one that I found a half decent workaround for:
I replace them when they break. You can actually order just the
latches themselves. Removing the remains of the old ones are hard, but
once you get the hang of it is not terrible. Installing the new ones
just takes a careful application of pressure.

-Michael

We used the right angle versions with the idea I mentioned above.

It just occurred to me, is the insulation of your wire sliding into
the connector pin a tad? If I remember correctly, this is necessary
for strain relief. The strip-length and max insulation diameter is
critical for Harwin pins.

Didn't realize you could replace the latch. Good to know.
 
J

JosephKK

Jan 1, 1970
0
We used the right angle versions with the idea I mentioned above.

It just occurred to me, is the insulation of your wire sliding into
the connector pin a tad? If I remember correctly, this is necessary
for strain relief. The strip-length and max insulation diameter is
critical for Harwin pins.

Yes, good of you to mention that, as it is true of most crimp contacts
and lugs.
 
Top