Wire guage

Ok, so I tried doing some searching online and found conflicting information. I thought that I would ask you fine folks for help yet again. I have learned a lot form you all, you are a great source of information! So far there hasn't been a problem that I haven't worked out yet.

Well this isn't really a problem per say, but I want advice on what gauge wire to use in my project. Basically what I have in a nutshell is a bioreactor. The reactor controls pH, dissolved oxygen, feeds a concentrated solution, removes treated water and adds dilution water. All of this works (hurray!) it was quite an ambitious first project.

Now that I'm done with the prototyping I have to wrap it up all nice and fancy, it is for a customer after all.

I'm looking at what gauge wire I should be using. Here is what I have:

  • 2x peristaltic pumps: 12VDC 100mA
  • 3x peristaltic pumps: 12VDC - The spec sheet says 2.5A max, I have never seen it draw more then 1A
  • 1x air pump: 110VAC 500mA
  • 1x Motor rated at 12VDC 25W so just under 2.1A

Some more information:

  • All power is supplied by a 110VAC to 12VDC power supply rated at 20A
  • Everything is controlled by a 8 channel relay
  • Only one pump will ever be on at one time
  • The air, mixer and a pump may all be on at one time
  • The longest a pump will be on is 5 minutes, 3x a day (12VDC 2.5A max one)

My concerns are with the influx current of the DC motor.

I was hoping to use 18AWG solid core wire, I have a bunch of it.

Let me know if any further information is needed.

18AWG should be fine. There are two issues with current and wire gauge -

I found [u]this chart[/u] which says 18AWG is good for 16 Amps when used in "chassis wiring" and 2.3 Amps when used for "power transmission". You are "chassis wiring". I didn't dig-in to see what lengths they are using for "power transmission", but that would be related to voltage drop.

The resistance is proportional to length and inversely proportional to the cross-sectional area. For long runs, you can get a voltage-drop across the resistance. But, this is usually only a concern for electricians wiring a building. (I know you can find building codes for electrical wiring that will give you the minimum wire gauge for a given current, and there's probably a length specified too.).

For the kind of things most of us are doing, any wire will have insignificant resistance (a fraction of an Ohm) and therefore insignificant voltage drop.

The other issue is heat which is also related to resistance and current.* You can probably find a chart for that too. In the case of heat, length isn't important because the heat is distributed over the length of the wire.... A wire that's twice as long will have twice the resistance and twice the total-heat, but the heat is spread-out.

Usually when a wire burns-up, it's at or near a connection because all of the resistance in the connection is concentrated at one spot.

  • One way to calculate power is Current2 x Resistance, and for a wire that's the power "lost" in the wire and converted to heat.

Thanks DVDdoug, that was one of the charts I was looking at.

As far as length goes the longest wire would be about 3 feet in total length. I'm not as concerned about the voltage drop, I learned a bit about that. I'd imagine the voltage drop would be consistent once its wired up. All my pumps are proportional to the voltage, so if they run a bit slower it's no problem it will just have to run a bit longer. I have a way to calibrate the flow rate of the pumps written into my program.

Thanks for the response!

Voltage drop is not constant. If two pumps run at the same time, they will draw more current and the voltage drop will be more. They will output a different amount to when you calibrated them individually. If this calibration is important to you, then you may have to do it with all reasonable combinations of pumps and other devices running.

Voltage drop is not usually a problem, until it is a problem. I ran into this recently with a project where I ended up pushing wires 40ft long through an inaccessible place. Not fun.

Don't use solid wire if it's going to be flexed like a power cord, the wire will break pretty quickly, use fine stranded (soft) wire.

At 3' max wire length, you are in the chassis wiring range. Use those figures.
18 guage solid wire is way overkill for the combined loads (everything starting at the same instant) that you list. No safety issue there. cost effective only if you have a surplus on hand (as you state that you do).
you should not have any flexing issues once everything is wired up.

You're good. go with it.

MorganS:
Voltage drop is not constant. If two pumps run at the same time, they will draw more current and the voltage drop will be more. They will output a different amount to when you calibrated them individually. If this calibration is important to you, then you may have to do it with all reasonable combinations of pumps and other devices running.

Voltage drop is not usually a problem, until it is a problem. I ran into this recently with a project where I ended up pushing wires 40ft long through an inaccessible place. Not fun.

This will never happen. Only one pump will ever be on at a time. Well the air pump and one of the other five pumps may be on at the same time, but this isnt an issue. the air pump is AC and the rest are DC. They draw from different supplies.

Flexibility is not an issue. Once in place nothing should be moving.

[various things using < 3A]
My concerns are with the influx current of the DC motor.
I was hoping to use 18AWG solid core wire, I have a bunch of it.

18G wire should be fine for that sort of power consumption. After all, 14-16G is a "100ft heavy duty extension cord" suitable for running something like an electric leaf blower...

All power is supplied by a 110VAC to 12VDC power supply rated at 20A

Got fuses? They might be a good idea. I DON'T think the 18g wire is good for 20A, which could happen if there is some sort of failure somewhere. 20A would probably melt the insulation and could start a fire. A 5A slow-blow fuse in each high-power circuit would probably be sufficient.

Warning! I see great confusion relation to tables of wire current carrying capability and wire for your project. The tables are for wire that is not insulated, just suspended in air. Insulated wire must be rated based on the type of insulation and how easily it melts. You need to find a table of wire sizes with your type of insulation. Most new wire will show the ration on the spool.

If your wiring has more than one wire in a single insulating sheath, the rating is reduced still further because heat is generated by each of the wires and must be added together.

In the US, number 12 house wiring with PVC insulation is rated for 20 amps. Put the same wire in a conduit where the heat cannot be dissipated and the rating is reduced to, I think, 15 amps.

Paul