[TriEmbed] Power Supply rail current

Tue Jul 8 16:50:13 CDT 2014

So,  let me make sure I understand this...
When you say have the relays energized, I know you mean energize the
solenoid but do you mean by that that the relay contacts are closed (power
flowing) or open (no power)? I'm guessing you mean don't go from open to
closed while there's power flowing since their will be a temporary arc?
Could I just use a normally closed relay and then use the microcontroller
to open the relay and cut the power if I need to.

Do you have any links to those higher cost relays? I would really love to
actually make it so that I could have a button on the front (hooked to the
uC) that would allow me to, using the relay, turn the power on or off to
any of the outputs. So having one of these nicer relays, if not too
expensive, might be nice.

On Tue, Jul 8, 2014 at 4:16 PM, Rodney Radford <ncgadgetry at gmail.com> wrote:

> Realize that the relay drop out at 30A will cause some arcing across the
> contacts, so should be used as a last drop safety effort - not something
> that is routinely done.
>
> And always have the relays energized before hooking up the current drain
> to the supply as you don't want to energize the relays with 30A across the
> contacts - this will eventually lead to welding the contacts closed.
>
> There are higher cost drop out relays designed for this purpose, but if
> only used in the extreme case that you want to drop out an oversupply, it
> is probably not warranted for a 1-off supply like you are building.
>
>
> On Tue, Jul 8, 2014 at 3:10 PM, Adam Haile <email at adamhaile.net> wrote:
>
>> Yeah, I was definitely wanting a separate current sense on each of the
>> front panel outputs at least. I like this plan of having a relay drop
>> out... I was going to need some sort of auto-shutoff anyways for safety.
>> Will look into that. Seems to make more sense then trying to allow 100A on
>> any single output :)
>>
>>
>> On Tue, Jul 8, 2014 at 3:08 PM, Rodney Radford <ncgadgetry at gmail.com>
>> wrote:
>>
>>> What about have a current sense on each of the output rails
>>> independently - and be able to switch from the front to measure total
>>> current draw, or draw on any of the individual rails
>>>
>>> Then, if any one of the rails exceeds some max - say 30A if you have 4
>>> outputs, you cut off the output - either a relay drop out (the easiest) or
>>> an electronic drop out
>>>
>>>
>>> On Tue, Jul 8, 2014 at 3:01 PM, Adam Haile <email at adamhaile.net> wrote:
>>>
>>>> Yeah... I'm trying to build an enclosure around this that adds a few
>>>> features. Mainly high-side current sensing with an LCD for live output of
>>>> the real-time current draw and serial output to my computer for
>>>> data-logging.
>>>> I would actually like to have 3 separate outputs on the outside of
>>>> the enclosure (maybe 4) because I'm usually dealing with large LED displays
>>>> and need to drop power in at multiple points in the matrix.
>>>> So, yes, I could definitely have a limited total current draw on each
>>>> output but that makes me a little nervous because, technically, the supply
>>>> would be capable out throwing 100A at any one of them, probably melting
>>>> something in the process. So, in my usual mantra of the only kind of kill
>>>> is overkill, I figured to just engineer for 100A at any of the outputs even
>>>> if it is unlikely (and probably unwise) that I would ever draw 100A on any
>>>> single output. As I'm finding, this may e impractical. But it is what I am
>>>> shooting for.
>>>>
>>>>
>>>> On Tue, Jul 8, 2014 at 2:56 PM, Rodney Radford <ncgadgetry at gmail.com>
>>>> wrote:
>>>>
>>>>> Looks like you have 3x tie points for +V an 3x tie points for -V, with
>>>>> a total across all three loads as 100A.
>>>>>
>>>>> Since you would not want to run a 100A wire (as you see the size
>>>>> required for that), just run 3x smaller wires.
>>>>>
>>>>> Is all 100A going to the same point, and if so, where?  From your
>>>>> first post, it sounds like you are trying to build a 100A supply around
>>>>> this - is that true? So why not just feed all 3x of each to the outside
>>>>> instead of trying to find a 100A binding post?
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Tue, Jul 8, 2014 at 2:51 PM, Adam Haile <email at adamhaile.net>
>>>>> wrote:
>>>>>
>>>>>> I don't actually have it yet... this is part of why I'm asking, to be
>>>>>> sure that I'm getting the right thing. But the internal schematic on it's
>>>>>> datasheet (http://www.jameco.com/Jameco/Products/ProdDS/374002.pdf)
>>>>>> seems to show it as a single output line.
>>>>>>
>>>>>>
>>>>>> On Tue, Jul 8, 2014 at 2:48 PM, Rodney Radford <ncgadgetry at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Are the three rails really attach to a common point?  Can you
>>>>>>> visually see they are tied together, or ohm them out to verify that?
>>>>>>>
>>>>>>> I have seen power supplies rated on total output, across all
>>>>>>> supplies, but they are not tied together on the output, so before
>>>>>>> recommending tying them together, I would need to know if they are already
>>>>>>> connected.
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Jul 8, 2014 at 2:43 PM, Adam Haile <email at adamhaile.net>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> I'm still diving into my high current supply enclosure build but
>>>>>>>> ran into a question about the supply I'm looking to buy:
>>>>>>>> http://www.jameco.com/1/1/2936-se-600-5-se-600-500w-ac-dc-enclosed-switching-power-supply.html
>>>>>>>>
>>>>>>>> It's rated at 100A and has 3 separate power rails. Each of which
>>>>>>>> seem pretty small for 100A. The datasheet (on the link provided) doesn't
>>>>>>>> mention each power rail being only able to handle a portion of the total
>>>>>>>> rated current so I assume they are just there for convenience. However, my
>>>>>>>> ampacity tables are recommending around a 2 gauge wire for 100A... which is
>>>>>>>> freaking huge. Not even sure I could fit that on the supply terminals.
>>>>>>>>
>>>>>>>> Should I just tie all three outputs into one to step up the gauge?
>>>>>>>>
>>>>>>>> Also, anyone know where you can actually buy a 100A binding post?
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Triangle, NC Embedded Computing mailing list
>>>>>>>> TriEmbed at triembed.org
>>>>>>>> http://mail.triembed.org/mailman/listinfo/triembed_triembed.org
>>>>>>>> TriEmbed web site: http://TriEmbed.org
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20140708/bd838ca8/attachment.htm>