Contact resistance in ATX power connectors causing grief

[Jay W.]

Hello folks,

I am having a rather unusual problem with about six hundred embedded
motherboards in the field, and was hoping someone might have an idea (or
two) on what might be happening, and possibly what the fix might be.

These are embedded low power Via C3 based motherboards that are deployed
in the factory automation industry. Power is distributed from the power
supply to the motherboard through a 12" power cable which consists of (2)
18ga. ground wires and (2) 18ga. +5VDC wires to an ATX style connector on
the motherboard; the board draws about 3 amps of +5VDC under maximum CPU
load. After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!) There is an onboard
voltage monitor on the CPU board that triggers a reset at about 4.7VDC,
the end result being that the processor gets stuck in a reset loop - it
will run for a minute or so and then the CPU load momentarily increases
and resets again. The voltage drop is *usually* greater across the ground
lines than the +5 lines for some reason. Maybe there's a clue there?

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

If the connector is reseated here at the factory and "repaired", we cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

We've tried using contact lube - no luck.

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

At first (a number of months ago) we were seeing only a .2 to .3VDC drop,
so we increased the power supplies output to 5.15VDC hoping to compensate
for the drop (a Band-aide, I know), but as you can see above, the contact
resistance kept increasing only to have the same problem occur.

We cannot increase the gauge of the wire, nor can we add more +5 and
ground wires due to there being only 2 pins available for each on the
power supply. The crimps will not accept double crimping.

I know the problem could be solved by soldering the wires between the
power supply and MB, but this would be a huge task due to the number of
units in the field, and the difficulty of disassembling the systems,
pulling the boards, and reworking them.

Does anyone have any thoughts on what is happening here, and possibly an
easier solution?

Thanks very much for any help.

 

[Adrian C]

Hello folks,

I am having a rather unusual problem with about six hundred embedded
motherboards in the field, and was hoping someone might have an idea (or
two) on what might be happening, and possibly what the fix might be.

What is the response from the motherboard manufacturers? Have you
approached them?

These are embedded low power Via C3 based motherboards that are deployed
in the factory automation industry. Power is distributed from the power
supply to the motherboard through a 12" power cable which consists of (2)
18ga. ground wires and (2) 18ga. +5VDC wires to an ATX style connector on
the motherboard; the board draws about 3 amps of +5VDC under maximum CPU
load. After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!) There is an onboard
voltage monitor on the CPU board that triggers a reset at about 4.7VDC,
the end result being that the processor gets stuck in a reset loop - it
will run for a minute or so and then the CPU load momentarily increases
and resets again. The voltage drop is *usually* greater across the ground
lines than the +5 lines for some reason. Maybe there's a clue there?

Hmmm, there is more current on the return from motherboard ground back
to the supply (roughly equal to sum of currents in 5V, 3.3V and 12V
rails) but I'd expect some of this to also travel back to the PSU
through motherboard ground contact points (screws) and the metal case of
the PSU. Check those grounding points?

 

[David C. Partridge]

What chemicals are these guys using in their factory? I'd lay some guesses
that there's something "horrid" flying around in either vapour or droplet
form.

Is that all hapening at one clients?

Dave

 

[Dave D]

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

Gas tight? Any chance that moisture is being trapped inside on assembly?

Dave

 

[GregS]

Hello folks,

I am having a rather unusual problem with about six hundred embedded
motherboards in the field, and was hoping someone might have an idea (or
two) on what might be happening, and possibly what the fix might be.

These are embedded low power Via C3 based motherboards that are deployed
in the factory automation industry. Power is distributed from the power
supply to the motherboard through a 12" power cable which consists of (2)
18ga. ground wires and (2) 18ga. +5VDC wires to an ATX style connector on
the motherboard; the board draws about 3 amps of +5VDC under maximum CPU
load. After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!) There is an onboard
voltage monitor on the CPU board that triggers a reset at about 4.7VDC,
the end result being that the processor gets stuck in a reset loop - it
will run for a minute or so and then the CPU load momentarily increases
and resets again. The voltage drop is *usually* greater across the ground
lines than the +5 lines for some reason. Maybe there's a clue there?

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

If the connector is reseated here at the factory and "repaired", we cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

We've tried using contact lube - no luck.

What kind?? In tough spots I grab my old Cramolin vial, and or Tweek or Stabilant.
The pure form of Deoxit in the tube, or Caig greese would do.
And have you tried tightening the female contact going to the board.
I suppose the female pin is crimped to the wire, could be faulty crimp, of
course many crimps are faulty from the getgo.

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

At first (a number of months ago) we were seeing only a .2 to .3VDC drop,
so we increased the power supplies output to 5.15VDC hoping to compensate
for the drop (a Band-aide, I know), but as you can see above, the contact
resistance kept increasing only to have the same problem occur.

We cannot increase the gauge of the wire, nor can we add more +5 and
ground wires due to there being only 2 pins available for each on the
power supply. The crimps will not accept double crimping.

I know the problem could be solved by soldering the wires between the
power supply and MB, but this would be a huge task due to the number of
units in the field, and the difficulty of disassembling the systems,
pulling the boards, and reworking them.

Does anyone have any thoughts on what is happening here, and possibly an
easier solution?

Stop the fault from tripping the CPU. Bios?

 

[Bill Davidsen]

Hello folks,

I am having a rather unusual problem with about six hundred embedded
motherboards in the field, and was hoping someone might have an idea (or
two) on what might be happening, and possibly what the fix might be.

These are embedded low power Via C3 based motherboards that are deployed
in the factory automation industry. Power is distributed from the power
supply to the motherboard through a 12" power cable which consists of (2)
18ga. ground wires and (2) 18ga. +5VDC wires to an ATX style connector on
the motherboard; the board draws about 3 amps of +5VDC under maximum CPU
load. After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!) There is an onboard
voltage monitor on the CPU board that triggers a reset at about 4.7VDC,
the end result being that the processor gets stuck in a reset loop - it
will run for a minute or so and then the CPU load momentarily increases
and resets again. The voltage drop is *usually* greater across the ground
lines than the +5 lines for some reason. Maybe there's a clue there?

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

If the connector is reseated here at the factory and "repaired", we cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

We've tried using contact lube - no luck.

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

At first (a number of months ago) we were seeing only a .2 to .3VDC drop,
so we increased the power supplies output to 5.15VDC hoping to compensate
for the drop (a Band-aide, I know), but as you can see above, the contact
resistance kept increasing only to have the same problem occur.

We cannot increase the gauge of the wire, nor can we add more +5 and
ground wires due to there being only 2 pins available for each on the
power supply. The crimps will not accept double crimping.

I know the problem could be solved by soldering the wires between the
power supply and MB, but this would be a huge task due to the number of
units in the field, and the difficulty of disassembling the systems,
pulling the boards, and reworking them.

Does anyone have any thoughts on what is happening here, and possibly an
easier solution?

Thanks very much for any help.

Clearly it's environmental, since you can't get it to happen in the lab.
This suggests something nasty in the environment. However, one thing you
can check is if everything is properly grounded and you're not getting
any ground loops. While it's unlikely, I can just barely envision the
manipulation of replugging the connector regrounding the board.

The term gas tight is reassuring, but I wouldn't take it on faith. If
the failure vs. time is relatively constant, sealing the top and bottom
of the connector would be an interesting data point. A bead around the
bottom where the connector meets the board, and also one at the top
where the wires enter the connector. If it's a unit assembly, can you
shrink wrap where the cable enters the connector?

Is there any visible change in the pins to indicate corrosion?

 

[Jeff Liebermann]

After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!)

Are you sure it's the connector contact resistance and not the wire
crimp on the ATX connector? I've seen something like this inside
wafer fab tunnels, where the rather caustic chemicals attack the
copper wire to solder plate junction on the connector pins. I've
never seen a 0.5VDC drop. More like 200-300mv at worst. Egads,
0.5VDC at 3A is 1.5 watts. The ATX connector should melt with that
much power being dissipated. Any discoloring of the nylon connector
shell?

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

Well, that certainly sounds like its the connector contact and not the
crimp. However, I'm suspicious and will pretend that it might still
be the crimp.

Have you tried to take an ohms-guesser or ESR meter and measure the
contact resistance (or crimp resistance) of the connections? Also,
measure the voltage drop when running between:
motherboard to ATX pin
ATX pin to copper wire (across crimp)
Case ground to motheboard ground (just for fun).

Instead of removing and re-inserting the connector, can you get it to
recover by merely wiggling the ATX connector from side to side, or
doing the same with individual pins by wiggling the wires sideways? If
yes, then you punching through a layer of surface crud, corrosion,
oxidation, or chemical buildup of some sorts. If no, then inspect the
crimp for insulation crimps, dissimilar metal issues, and corrosion.

Any evidence of copper wire embrittlement at the crimp? Heating of
the connection can do that.

We've tried using contact lube - no luck.

No. That's just solvent and light oil. Try Cramoline, De-Oxit or
some oxidation prevention chemical. Also try grease or silicon
grease. Anything to provide a seal.

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

Gas tight over what area? Verify that you have a large mating surface
area and not a tapered loose fit. I've seen connector tolerances
create a loose connection that eventually fails. (Hint: I designed
marine radios in a previous life). Sacrifice one power supply
connector and remove the nylon shell with a pair of wire cutters. Do
NOT use an extraction tool for this. Take each connection and
*MEASURE* the insertion and extraction force. If the power supply
manufacturers crimping machine screwed up the connector fit, you have
a potential cause.

Also, tin oxide is almost transparent and rather difficult to see.
However, you can detect it with an ohms-guesser. Try passing the
probe tips over the ATX plug surface, without punching through, and
see if you get any insulated areas.

I think the best approach is to identify whatever chemical is
attacking the connectors. I'm not sure how to do this. Also, I'm not
convinced that the connectors pins are as great as you claim. Are
they tin, solder plate, or lead-free solder plate? If lead-free, what
substitute solder plating are they using?

 

[M.Joshi]

If the connector is reseated here at the factory and "repaired", w
cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

What do you mean by -repaired-

 

[Franc Zabkar]

Hello folks,

I am having a rather unusual problem with about six hundred embedded
motherboards in the field, and was hoping someone might have an idea (or
two) on what might be happening, and possibly what the fix might be.

These are embedded low power Via C3 based motherboards that are deployed
in the factory automation industry. Power is distributed from the power
supply to the motherboard through a 12" power cable which consists of (2)
18ga. ground wires and (2) 18ga. +5VDC wires to an ATX style connector on
the motherboard; the board draws about 3 amps of +5VDC under maximum CPU
load. After several months in the field, contact resistance on the ATX
connector increases for some reason, and is causing a voltage drop as seen
on the motherboard (sometimes as much as .5VDC!) There is an onboard
voltage monitor on the CPU board that triggers a reset at about 4.7VDC,
the end result being that the processor gets stuck in a reset loop - it
will run for a minute or so and then the CPU load momentarily increases
and resets again. The voltage drop is *usually* greater across the ground
lines than the +5 lines for some reason. Maybe there's a clue there?

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

If the connector is reseated here at the factory and "repaired", we cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

We've tried using contact lube - no luck.

If it's an environmental issue, try flooding the connector with the
same gel (?) that telco's use for their wire joins. Or try vaseline.

- Franc Zabkar

 

[Lance]

Just a thought, I imagine each embedded motherboard has several other
connectors (IDE, fan, whatever). If it's a environmental problem like a
lot of people seem to think it is, shouldn't all the connectors be failing?

Lance
*****

Jay W. thought carefully and wrote on 5/11/2006 3:18 AM:

 

[Jason D.]

If it's an environmental issue, try flooding the connector with the
same gel (?) that telco's use for their wire joins. Or try vaseline.

- Franc Zabkar

Agreed, silicone grease fill the ATX female and fill the male side as
well, this will be messy but should stop the corrsion.

On next design, make sure the crimped connections are CORRECT or use
screw terminals, preferably. ITC0008 pincushion thick 3 wire plugs
had mis-match gauge crimp that it is just enough to develop resistance
over time or tin pins/contacts corroding.
Soldering wires directly to PCB was the solution on ITC008.

Cheers, Wizard

 

[Mark]

and check the solder joints to the PWB...I have seen this kind of int.
behavior due to marginal solder joints...

but it could also be the connector itself

Mark

 

[Jeff Liebermann]

Just a thought, I imagine each embedded motherboard has several other
connectors (IDE, fan, whatever). If it's a environmental problem like a
lot of people seem to think it is, shouldn't all the connectors be failing?

No. The ones carrying high currents will tend to fail first. Also,
connectors that carry very little current can tolerate far more
contact resistance than connectors that carry high currents.

 

[Peter Dettmann]

Just a thought, I imagine each embedded motherboard has several other
connectors (IDE, fan, whatever). If it's a environmental problem like a
lot of people seem to think it is, shouldn't all the connectors be failing?
Lance
*****

Probably right Lance, but maybe a bit of extra resistance does not
matter so critically in all circuits. Also the discussion has not
mentioned the necessity for good mechanical pressure in the
connectors. I had extensive problems with otherwise good connectors
in another set up where the spring fingers were not having or
retaining their intended springiness.
Also the contact material may not be as good or pure as hoped for.

Peter Dettmann

 

[Leythos]

On a system that is failing, unplugging the ATX connector and reseating it
will "fix" the problem and the system will continue to work for several
months until once again the resistance increases in the contacts and...
well, you get the picture.

If the connector is reseated here at the factory and "repaired", we cannot
get it to fail again under any conditions: vibration, humidity,
temperature cycling in a environmental chamber, etc.

We've tried using contact lube - no luck.

Thinking that there might be a problem with dissimilar metals, we checked
that the contacts on the cable and ATX connector on the board are tin. The
contact specification claims that they are gas tight.

At first (a number of months ago) we were seeing only a .2 to .3VDC drop,
so we increased the power supplies output to 5.15VDC hoping to compensate
for the drop (a Band-aide, I know), but as you can see above, the contact
resistance kept increasing only to have the same problem occur.

Seen this many times near plating machines and other corrosive areas
with computer hardware. You have two choices: enclose and seal the
hardware so that it can't be exposed to the vapor, or find a better
contact coating. You have to purchase a contact coating that is specific
to the metal you are protecting - lube is not what I'm talking about. In
the old days, we purchase one for gold and another for silver. Now we
just move the computers into safe rooms and then use
video/keyboard/mouse extenders to put the display and kbd/ms in the bad
areas....

I found this, it looks promising:
http://www.actonrust.com/electrical.htm

 

[Leythos]

Just a thought, I imagine each embedded motherboard has several other
connectors (IDE, fan, whatever). If it's a environmental problem like a
lot of people seem to think it is, shouldn't all the connectors be failing?

Different metals corrode at different rates and based on the amount of
current they are handling.

Vasoline is not a good choice in an industrial location that gets any
heat, it will tend to ooze/flow.

 

[Jeff Liebermann]

Also the discussion has not
mentioned the necessity for good mechanical pressure in the
connectors. I had extensive problems with otherwise good connectors
in another set up where the spring fingers were not having or
retaining their intended springiness.
Also the contact material may not be as good or pure as hoped for.
Peter Dettmann

Someone most certainly did mention mechanical pressure, which methinks
is a big part of the problem. See:
http://groups.google.com/group/comp.sys.intel/msg/4cd035216258215
starting with "Gas tight over what area...". I recommended he
sacrifice one power supply connector end and measure the retention and
extraction force on the mating connectors without the shell. My
guess(tm) is that they're fairly loose.

 

[Peter Dettmann]

Someone most certainly did mention mechanical pressure, which methinks
is a big part of the problem. See:
http://groups.google.com/group/comp.sys.intel/msg/4cd035216258215
starting with "Gas tight over what area...". I recommended he
sacrifice one power supply connector end and measure the retention and
extraction force on the mating connectors without the shell. My
guess(tm) is that they're fairly loose.

Great Jeff, I missed that, but agree with your analysis.

Peter Dettmann

 

[Bennett Price]

It sounds as though the females are shoddy; perhaps losing their
springiness and becoming loose. I'd suggest getting new sockets and
shrouds from Molex or similar and equip them with 6" pigtails. In the
field, cut off the defective females and splice the new pigtails on.
(Or a bit more expensive but saving you labor, buy extension cords with
suitable connectors and discard the end you don't need.)