"Maximum amps for chassis wiring" vs "Maximum amps for power transmission"

[JoeN]

Anyone have a good definition for this?  I am wanting to use 28AWG solid wire for a project.   I have used 26AWG quite a bit but I can't get it anymore and I find that 24AWG is bigger than conveniently sized on my project boards and unnecessary at these currents.  Since the current is pretty small on this one-off hobby project (LED matrices, 20mA LEDs 1:8 duty cycle, probably 20 in total) I think either of these values works for me just fine and I'll use 22AWG for VCC and GND to the drivers themselves as I calculate 3.2A worst case on the total power draw, but probably no more than 1.6A (1/2 LEDs lit at any time with the 1:8 duty cycle).   I see 22 gauge gives 7A max chassis wiring, 0.92A max power transmission, so I hope that LED matrices on a board count as chassis wiring, otherwise I am over for VCC distribution to the driver ICs.   I think what they are saying is that chassis wiring means that it is mostly open air wiring on a board in an enclosure and power transmission is for multiple lengths in additional insulator, or what?  Just trying to understand the difference.

http://www.powerstream.com/Wire_Size.htm

Thanks!

[German_EE]

Chassis wiring is small lengths between components, probably with a resistance of milliohms.

Power transmission is between equipment or connecting your equipment to the mains supply. It also covers wiring between power sockets.

[Monkeh]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

[suicidaleggroll]

How to pick a wire size:

1) Determine the max current that will be passing through that wire
2) Determine the max acceptable voltage drop that you can tolerate in that wire run (make sure to divide by 2 if this is a power/ground pair)
3) Determine the length that the wire needs to be to get from point A to B

From those three, calculate the maximum resistance you can tolerate in ohms, convert to ohms per 1000', and find a wire gauge that meets or beats it.  Make sure you stay well below that table's "Chassis Wiring" numbers as well to make sure you don't burn up the wire.

If the calculated wire size is too big for what you want to do (can't fit in conduit, etc.) then re-visit steps 1-3 and see what you can do to lighten the requirement.  If, on the other hand, you have extra room available, then feel free to go bigger to improve current capacity and voltage drop/efficiency.

[sparx]

Online Calculator

[JoeN]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

[Monkeh]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

These are sane numbers, subject to volt drop: http://www.engineeringtoolbox.com/wire-gauges-d_419.html

Perhaps a tad high on the small stuff, but if you're actually driving current you should be avoiding piddly little wires for mechanical reasons as well as thermal.

[Simon]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

Look up AWG on wkipedia, it's not rocket science

[Monkeh]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

Look up AWG on wkipedia, it's not rocket science

Where you will find only ratings for 18AWG and up. Thanks for playing, though.

[Simon]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

Look up AWG on wkipedia, it's not rocket science

Where you will find only ratings for 18AWG and up. Thanks for playing, though.

 

http://en.wikipedia.org/wiki/American_wire_gauge

from 0000 to 40, I made a spreadsheet version to keep on my computer

[Monkeh]

Their 'power transmission' numbers are hilariously low, and their chassis wiring figures are high.

Ignore those people entirely.

Well, OK, then link to an acceptable chart since you obviously know what you are talking about and I admit that I do not.

Look up AWG on wkipedia, it's not rocket science

Where you will find only ratings for 18AWG and up. Thanks for playing, though.

 

http://en.wikipedia.org/wiki/American_wire_gauge

from 0000 to 40, I made a spreadsheet version to keep on my computer

And now if you actually read the tables you'll find ampacity listed only for 18AWG and up. It's not rocket science!

[Simon]

And the last time I checked I was more concerned about voltage drop than so called ampacity. "ammpacity" is based on how many volts your willing to loose and how much heat your cable will generate, how much temperature the insulation can withstand. Realistically you can't specify the amp rate of a cable unless you know the insulation spec and you know what you want.

Last time i speced some 20AWG it was based on the known cable run length and the voltage drop I was willing to have, I also totally disregarded the 20A rating of 16AWG because I needed less voltage drop and went with 10A per 16AWG core (much to the surprise of the loom subcontractor) but based on real tests I carried out. In this case I'm talking about Raychem 44 cable which is rated up to 150C so not your average auto cable......

[Monkeh]

And the last time I checked I was more concerned about voltage drop than so called ampacity. "ammpacity" is based on how many volts your willing to loose and how much heat your cable will generate, how much temperature the insulation can withstand. Realistically you can't specify the amp rate of a cable unless you know the insulation spec and you know what you want.

Last time i speced some 20AWG it was based on the known cable run length and the voltage drop I was willing to have, I also totally disregarded the 20A rating of 16AWG because I needed less voltage drop and went with 10A per 16AWG core (much to the surprise of the loom subcontractor) but based on real tests I carried out. In this case I'm talking about Raychem 44 cable which is rated up to 150C so not your average auto cable......

*sigh*

We're talking ordinary bog standard every day 60C PVC cable. Like most people use.

[Simon]

I'd be more worried about voltage drop in which case unless your making a heater your never going to exceed basic insulation temperature, few tables that are not for the specific cable you want to use are going to be so precise. Of course you'd also need to consider if your bunching and sleeving a number of cables in which case you loose capacity that is usually specified for a single cable in free air (and hence i derated a 20A to 10A) so again either you work it out or get specific instructions from the cable manufacturer or based on a particular insulation spec. that you know your using.

[Monkeh]

Or you could just get the ballpark figures he's asking for as a rough guide of what size of wire to use for a short run in a project.

We're in the Beginners section, Simon, not the 'I need to know how to do this because my boss wants me to do it for a contract' section.

[Simon]

Well you could take the amp value of one wire and work out how many amps per ohm of wire are admissable and then scale it for the other size although I'm sure that going up in size you'd need to derate slightly as the surface area does not go up as fast as cross section so it heats quicker but for going smaller no problem (reason i used 4x16AWG instead of a larger wire)

[Monkeh]

Well you could take the amp value of one wire and work out how many amps per ohm of wire are admissable and then scale it for the other size although I'm sure that going up in size you'd need to derate slightly as the surface area does not go up as fast as cross section so it heats quicker but for going smaller no problem (reason i used 4x16AWG instead of a larger wire)

Or you could just supply the guy a reasonably sane reference table for generic wire. Which I did. Ages ago. It even has derating for bunching, and temperature factors.

[Simon]

Sorry did't realize that link had more detail than the wikipedia one.

[Simon]

Although due to the obvious misconceptions shown by the Op it won't hurt for him to sit down and work out voltage drops for various cables to get an idea of what he is dealing with as there is clearly a concept shown that wire just has an amperage capacity and the notion of voltage drop is not really being considered.

[Richard Crowley]

They generally mean that "chassis wiring" is short-lengths and "open" (i.e. air-cooled, not bundled together and overwrapped with lots of insulation which prevents the wire from air-cooling).  And "power transmission" means long distances, and bundled together (or pulled with other wires through a conduit pipe) which has a whole different set of factors to consider.

My rule of thumb is that for "chassis wiring", 30 AWG ("wire-wrap" wire) can safely handle 1 Amp (1000mA). 
For casual, hobby use, I would consider 28 AWG suitable for the currents you are planning.

[Simon]

I haven't forgotten trying to help a waterways trust draw up loom diagrams for a canal boat. First the guy in charge of commissioning the job who came from the semiconductor industry thought he knew everything and got me to do diagrams that were up to the electrician to do, then he decided that he wanted 12V and didn't understand the concept of just how thick his cables would get (9 KW at 12V yea right....) then he decided that he could loose 3V in the lighting loom just because the LED lighting would work down to 9V, I gave up and the electrician who he had decided was trying to cheat them by using too bigger wire fortunately told the guy to get stuffed and somehow they managed to get it up to 24V as i originally suggested.

[Zero999]

Wire sizing isn't always easy.

I agree, most of the time, the voltage drop is probably the most important thing but it's very important to check the temperature rise isn't excessive, especially given that the resistance of copper has a positive temperature coefficient which could cause thermal runaway under some circumstances.

When you get into mains wiring the resistance also needs to be low enough to keep the voltage on the earth below 50V when shorted to phase and the fault current high enough to trip the breaker before anyone is shocked.

[JoeN]

Thanks to everyone for input here.  I did see a lot of these charts online and none of them completely agree so I was looking for someone here to call balls and strikes (in other words, be the judge or referee) on these charts and I think I have gotten some good information from several posters so thanks to everyone who replied.

[Simon]

They will be different as people may have different ideas on acceptable voltage drop and better insulations allow for more power dissipation and voltage drop. Exact figures are up to the cable manufacturer.