Searching \ for '[EE] Power dissipation in a connector' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=power
Search entire site for: 'Power dissipation in a connector'.

Exact match. Not showing close matches.
'[EE] Power dissipation in a connector'
2006\01\17@165009 by

Question:
Is it possible for a connector to produce more heat as a result of power
dissipation at a voltage lower than it's rating?

So say you have a power connector good for 10 A, and 60V.

Will I get more heat off the connector if I run say 10A and 12V through it?

I could see a small possibility if you model the connector with reactive
components, but to me, the obvious answer is no.

If you say yes, please give a good explanation.

I'm speaking of DC only, not AC.

--

Shawn Wilton (b9 Systems)
http://black9.com
The power is I squared R, so only current affects it heating.

----- Original Message -----
From: "Shawn Wilton" <black9gmail.com>
To: "Microcontroller discussion list - Public." <piclistmit.edu>
Sent: Tuesday, January 17, 2006 1:50 PM
Subject: [EE] Power dissipation in a connector

{Quote hidden}

> --
As I expected.  I'm having an argument with a fellow that argues otherwise.
I'll just ignore him.

Thx.

On 1/17/06, Mike Hagen <picdogdslextreme.com> wrote:
>
> The power is I squared R, so only current affects it heating.
>
> {Original Message removed}
Shawn Wilton wrote:
> Is it possible for a connector to produce more heat as a result of
> power dissipation at a voltage lower than it's rating?
>
> So say you have a power connector good for 10 A, and 60V.
>
> Will I get more heat off the connector if I run say 10A and 12V
> through it?

This question is a great example of why using terms properly can be
important.  Current goes thru things, voltage can be accross them.  Voltage
thru something is just gibberish.

A connector is intended to have as low a resistance as possible.  It will be
designed to carry some maximum current.  The voltage accross the connector
at any time is the current thru it times its resistance.  Ohms law states
that V = I * R.  If R is fixed as it is with a connector, you only get to
pick one of V or I.

The voltage spec of a connector refers to the open voltage or the insulation
voltage between adjacent pins or to some other point.

> I could see a small possibility if you model the connector with
> reactive components, but to me, the obvious answer is no.

I don't see how you can answer a question that doesn't make sense in the
first place.

> If you say yes, please give a good explanation.

A good explanation requires a good question.

******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products
Seriously Olin, keep the first and third part of your response stuffed in
your shorts.  So I didn't feel like saying a connector with 12 volts across
it and 6 amps through it.  It's easier to say 12 volts and 6 amps through
it.  You understand what I'm asking.  But instead of being polite, as most
people are to you, you constantly feel the need to respond like you're an
A**.

So do me a favor.  If you can't respond to my questions politely, do not
respond at all.  I really won't be the worst for it.

On 1/17/06, Olin Lathrop <olin_piclistembedinc.com> wrote:
{Quote hidden}

> -
On 1/17/06, Shawn Wilton <black9gmail.com> wrote:
> Seriously Olin, keep the first and third part of your response stuffed in
> your shorts.  So I didn't feel like saying a connector with 12 volts across
> it and 6 amps through it.  It's easier to say 12 volts and 6 amps through
> it.  You understand what I'm asking.  But instead of being polite, as most
> people are to you, you constantly feel the need to respond like you're an
> A**.
>
> So do me a favor.  If you can't respond to my questions politely, do not
> respond at all.  I really won't be the worst for it.
>

I think you are missing what Olin is saying.  He piut his finger right
on the misconception that would make you ask the question in the first
place.  That's a sign of a good teacher.  Try a little humility and
you might learn something.

Regards,
Mark
markrages@gmail
--
You think that it is a secret, but it never has been one.

> Seriously Olin, keep ...

Ignore the rude bits. He can't help it. Not worth the hassle.
Concentrate on the useful technical gems hidden amongst the dross.

> your shorts.  So I didn't feel like saying a connector with 12 volts
> across
> it and 6 amps through it.  It's easier to say 12 volts and 6 amps
> through
> it.  You understand what I'm asking.

He may not.
I take it that you mean that you have two systems which draws 6 amps
from either a 12v or 60v supply respectively. The actual drop across
the connector junctions in such cases will (hopefully!) not be the
system voltage. The connector may "see" the system voltage between a
current carrying connector and an adjacent one but only leakage
current will flow across this path. The actual connector voltage drop
would usually be well under a volt and hopefully only millivolts.

If you assume that leakage current between adjacent contacts or from
contacts to ground is minimal then heating comes essentially only from
the product of current and contact voltage drop or current squared x
contact resistance - as Olin says. The latter is the best way to look
at it  - power = I^2R. Clearly the system voltage is not involved
directly.

There is an important special case in which Olin's answer is quite
wrong technically. It's unlikely to apply in the specific example
voltages that you gave, but, as you were only giving examples, and you
wanted a general answer, then it's a good illustration of how one
needs to keep an open mind and try and understand the general intent
of a question rather than getting all flustered with the terminology
and partially missing the point.

Special case: Contact resistance is *never* constant with current or
even linearly variable with current (for most values of never :-) )
but in the overwhelming majority of the time it's close enough to
constant for practical purposes to allow a maximum value to be used.
Most times you don't care if it's lower than spec. However at very low
voltages *across the contacts* (which is what Olin missed) and/or low
currents through the contacts the contact resistance can be much or
much much much higher than usual. The actual voltages and currents
involved are usually small by practical standards and vary with
contact materials. Some materials which have excellent low contact
resistances "once they get going" are notoriously bad below some
critical threshold. This phenomena is addressed by the notion of
'wetting current" (Google knows). At 10A through or 12V available
system voltage you would be most unlikely to find normal materials
that has a wettting current problem, but it could perhaps happen. In
such cases the power loss in the connector is liable to be higher with
reduced system voltage.

But, the rest of the time, it's likely that current though the contact
points is well correlated with power loss in the connector, regardless
of system voltage.

Russell McMahon

But instead of being polite, as most
{Quote hidden}

>> --
Shawn, Olin and Mark are right. Connectors are spec'd by their
current-carrying
ability. The only reason voltage is mentioned is to assess breakdown in a
specified environment (usually a military salt-spray breakover spec),
and those
variables almost always refer to the distance between conductors and the
dielectric
strength of the material that makes up the connector.

You can learn a lot about connectors by studying the specifications of
AMP and
Bendix military connectors.

--Bob

Mark Rages wrote:

{Quote hidden}

--
Note: To protect our network,
attachments must be sent to
attachengineer.cotse.net .
http://beam.to/azengineer

On 1/18/06, Mike Hagen <picdogdslextreme.com> wrote:
> The power is I squared R, so only current affects it heating.

P = U * I, where I = U/R so P = U^2/R or P = I^2*R

So you probably understand why you wrong with your supposition.
The heating is the effect of the contact resistance. As long there is
an R there is both a voltage drop on the contact and a power
dissipation which depends of course by the current.
But is wrong to say that only the current affects the connector heating.

greetings,
Vasile

>
> {Original Message removed}
Shawn Wilton wrote:
> So I didn't feel like saying a connector with 12 volts
> across it and 6 amps through it.  It's easier to say 12 volts and 6
> amps through it.

I don't see how it's easier but it is certainly incorrect.  Attention to
detail is an important quality of a good engineer.

> You understand what I'm asking.

Not really.  By using the terms incorrectly I concluded you misunderstood
the concepts in the first place.

Also realize that it is not possible to have both 10A thru and 12V accross.
I tried to explain that you don't get to pick both the voltage and the
current simultaneoulsy because their relationship is fixed by the contact
resistance, which you don't get to pick at all.  Note that 10A thru and 12V
acrross implies a contact resistance of 1.2ohms and a dissipation of 120W.
These values are just not credible.

******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products
The heat produced is directly from the current thru the contact and the voltage drop across the connector. THe only time the signal voltage plays a part is if the connector is being connected or disconnected live. the voltage drop on the contact is directly related to the current thru the contact and the contact resistance. The only other way to get heating in the connector is in a failure mode whrer the contacts are arcing but that is a different issue altogether.
Larry

---- Shawn Wilton <black9gmail.com> wrote:
{Quote hidden}

> --
Thread was killed but, it was a hypothetical.  I was not looking for an
exact answer.  I was looking for the possibility that a connector could be
modeled as anything but a resistor.  Instead I get a rhetoric on how to
calculate power..

Now the 12V (across) and 6A (through) was not a theoretical.  That equates
to 70W (6/12 ohms) and the guy who this conversation has been based on
claims that was what he was pulling through a little connector specc'd for
1A.  What he failed to mention when we were having our conversation
regarding greater heat dissipation at lower voltage was that he was assuming
constant power and I was assuming he was referring to constant current (as
connectors are specc'd for a max. current rating).  So instead of using the
connector at the rated 70+ V and 1A, he was drawing 12V (across) and 6A
(through).  Big difference going from 1A on the spec to 6A.  Hence the
heat.  Not wanting to assume I know everything there is (I don't, yet) I
thought I would ask if there was any way that a connector could produce more
heat at lower power.  I didn't see any way, but I have certainly seen and
heard of things more bizarre.

I have a classical education in Electrical Engineering.  In fact I just
finished up my second degree (first in EE) last spring.  However, there are
still numerous things I don't know, and just can't learn in school.  That's
why I sit on this list.  To learn what you can't in school.  Most of these
conversations are quite interesting, and ultimately useful.  However, when
the conversation degrades in to insults and name-calling (Yeah, I'm guilty
of this too.  I suffer from Olin syndrome too ;-) then the useful knowledge
tends to fall between the lines much as detritus falls to the ocean floor.

Hopefully we can kill this thread now.  Please feel free to email me if you
have any further comments or questions.  Olin, I apologize for telling you
to stuff it and calling you an awful name.  I hope you will accept my
apology.

On 1/18/06, Olin Lathrop <olin_piclistembedinc.com> wrote:
{Quote hidden}

> -
Shawn Wilton wrote:

> Now the 12V (across) and 6A (through) was not a theoretical.  That
> equates to 70W (6/12 ohms) [...]

Just a small detail... 12 V across and 6 A through gives 2 ohms (12 V/6 A).
(And 72 W... but you probably knew that :)

Gerhard

I'm going to stop composing email just after waking up.  It's really not
doing me any favors.

On 1/19/06, Gerhard Fiedler <listsconnectionbrazil.com> wrote:
{Quote hidden}

> -
>I'm going to stop composing email just after waking up.
>It's really not doing me any favors.
>
>
>On 1/19/06, Gerhard Fiedler <listsconnectionbrazil.com> wrote:
>>
>> Shawn Wilton wrote:
>>
>> > Now the 12V (across) and 6A (through) was not a theoretical.  That
>> > equates to 70W (6/12 ohms) [...]
>>
>> Just a small detail... 12 V across and 6 A through gives 2 ohms (12 V/6
>> A).
>> (And 72 W... but you probably knew that :)

Well I was happy with it, after all it really is an abbreviation for 6A
through the mated connector (that is what we were talking about) and 12V

The load that the connector connects to the supply may well be 2 ohms ...

Connector current ratings are generally for a given temperature rise.

Radio control guys seem to enjoy pushing the limits on everything,
discharging bateries at 30C rate, charging them significantly beyond maximun
temperature, and believe it or not, soldering the cells in a battery
together end to end.   Then they wonder why the batteries have short