Searching \ for '[EE] FireWire polarity reversal' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=firewire+polarity
Search entire site for: 'FireWire polarity reversal'.

Exact match. Not showing close matches.
PICList Thread
'[EE] FireWire polarity reversal'
2007\02\16@164608 by Mike Hord

picon face
Has anyone had any experience with polarity reversal of
1394 cables?

We've had a couple of instances on different products with
people plugging in the cables "backwards".  As one might
expect, connecting a 24V supply to what is supposed to
be a tenderly shepherded high speed data line has, ah,
exciting results.

Our general feeling is to make some small mention of it
in our documentation and then give up on trying to
outwit the nitwits.  If anyone has heard of any other
approach, though, I'd welcome input.

I'd also welcome similar stories, and general comment
on the sort of person who is willing to push hard enough
to mate a FireWire cable backwards (it takes more
pressure than I would EVER apply to a cable in an
attempt to plug it in).

My personal feeling is that this is similar to thread
stripping on a screw- some people will NEVER strip
a screw, others have simply no feel for how much
torque is too much to apply to a tiny little screw.

Mike H.

2007\02\16@191824 by Peter P.

picon face
Mike Hord <mike.hord <at> gmail.com> writes:

> Has anyone had any experience with polarity reversal of
> 1394 cables?
>
> We've had a couple of instances on different products with
> people plugging in the cables "backwards".  As one might
> expect, connecting a 24V supply to what is supposed to
> be a tenderly shepherded high speed data line has, ah,
> exciting results.
>
> Our general feeling is to make some small mention of it
> in our documentation and then give up on trying to
> outwit the nitwits.  If anyone has heard of any other
> approach, though, I'd welcome input.
>
> I'd also welcome similar stories, and general comment
> on the sort of person who is willing to push hard enough
> to mate a FireWire cable backwards (it takes more
> pressure than I would EVER apply to a cable in an
> attempt to plug it in).

The easiest fix to increase the skills of your audience pool and lengthen their
attention span is to raise the price ;-) Anyway the mini-FW connectors are
fairly hard to put in in reverse (due to metal shell). I.e. if someone did put
something in in reverse then there should be clear mechanical signs that this
has happened and warranty would not be an issue. No ?

> My personal feeling is that this is similar to thread
> stripping on a screw- some people will NEVER strip
> a screw, others have simply no feel for how much
> torque is too much to apply to a tiny little screw.

A few years ago there was an advisory on no-name FW cables which were wired
wrong. I don't know if this is related. I never saw one but others did.

Peter P.


2007\02\16@194824 by peter green

flavicon
face

> I'd also welcome similar stories, and general comment
> on the sort of person who is willing to push hard enough
> to mate a FireWire cable backwards (it takes more
> pressure than I would EVER apply to a cable in an
> attempt to plug it in).
does this damage the socket in any way (e.g. is there any easy way to see when a retard doing this was the cause of the failure)

whats the details on the firewire power supply? could some strategically placed diodes to the power rails combined with a zener between the power rails absorb this without too much trouble?


2007\02\16@214017 by Peter Todd

picon face
On Fri, Feb 16, 2007 at 03:46:00PM -0600, Mike Hord wrote:
> My personal feeling is that this is similar to thread
> stripping on a screw- some people will NEVER strip
> a screw, others have simply no feel for how much
> torque is too much to apply to a tiny little screw.

I'll say... When I was a fair bit younger I once got to fix a computer
where my client had decided to install some more SIMM ram, you know, the
type where you put it in at an angle and tilt it into place. He had
tried to install the ram vertically, and when that didn't work, he got
out a hammer...

--
http://www.petertodd.ca

2007\02\17@022152 by Randy Glenn

picon face
With cheaper FireWire connectors, it's pretty easy to stick 'em in the
wrong way.

That's my story, and I'm sticking to it. I needed to upgrade that
enclosure anyways...

-Randy

On 2/16/07, Mike Hord <spam_OUTmike.hordTakeThisOuTspamgmail.com> wrote:
{Quote hidden}

> -

2007\02\17@023126 by Robert Rolf

picon face
peter green wrote:
>>I'd also welcome similar stories, and general comment
>>on the sort of person who is willing to push hard enough
>>to mate a FireWire cable backwards (it takes more
>>pressure than I would EVER apply to a cable in an
>>attempt to plug it in).
>
> does this damage the socket in any way (e.g. is there any easy way to see when a retard doing this was the cause of the failure)
>
> whats the details on the firewire power supply? could some strategically placed diodes to the power rails combined with a zener between the power rails absorb this without too much trouble?
>
>
www.circuitprotection.com/techpapers/1394Firewire.pdf
1) A primary power provider must deliver between 20 and 33 V under full load conditions.
2) A primary power provider shall implement current limit circuitry to comply with all appropriate
regulatory agency specifications.
3) Current draw through any single port shall not exceed 1.5 A.

I would suggest hefty crowbar 5.1V zeners to force the power source to go into limiting mode
if the connector is reversed. (How the HECK do they manage that? 4pin FW has no power).
I had no idea FW could pump out so much power (1.5A x 33V~=50W). {other specs I found say
40V max, 8V min)

Since the zener would be forward biased when 'protecting' it only needs to be rated for 2A
or so. (0.8V x1.5A=1.2W rated).

Pinouts and links to reference material
http://www.interfacebus.com/Design_Connector_Firewire.html

Discussion of the protocol and throughput issues.
http://www.ausairpower.net/OSR-0201.html

Robert

2007\02\17@032055 by Tony Smith

picon face
> www.circuitprotection.com/techpapers/1394Firewire.pdf
> 1) A primary power provider must deliver between 20 and 33 V
> under full load conditions.
> 2) A primary power provider shall implement current limit
> circuitry to comply with all appropriate regulatory agency
> specifications.
> 3) Current draw through any single port shall not exceed 1.5 A.
>
> I would suggest hefty crowbar 5.1V zeners to force the power
> source to go into limiting mode if the connector is reversed.
> (How the HECK do they manage that? 4pin FW has no power).
> I had no idea FW could pump out so much power (1.5A x
> 33V~=50W). {other specs I found say 40V max, 8V min)


I always thought FW was 5v - 40v.  Upping the voltage is an easy way to get
more power, increasing the amps means thicker wires.  Some Dell laptops have
a double USB port to enable devices to use more than 0.5A.  The 2nd port
just supplies power (no data), I'm not sure how much.

Cars are supposed to be going to 42v for the same reason, there's that much
stuff in the average car now the wiring is getting a bit tricky.

Imagine the desktop coffee warmer you can have when more than 2.5W is
available...

Tony

2007\02\19@093510 by Mike Hord

picon face
Everyone-

Thanks for commiserating with me on this.  Unfortunately,
I have no control over any of the devices on the bus: they
are an OTS hub, card, and camera.

Mini-1394 isn't really an option, as it was hard enough for
us to find an almost acceptable 6-pin hub with 6 ports for
this application.  Plus, the mini connection doesn't
deliver power, and this system is bus powered.

Even if I could redesign the hub/camera/card, I don't think
there is a way to protect that data line from excessive
applied voltage.  The 1394 signal is just a little too
tenuous to be hanging diodes and such off of it.

I think the bottom line here is that there isn't any way to
fix this short of user education.

Mike H.

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...