Wire preference - stranded or solid

[Teched3]

Al,
Excellent point made on stranded wire compression at a terminal connection.

[MikeE]

There is no reason to use solid core for any of the "large wires" i.e. 22 AWG or larger, but it's fairly common to see some of the bundled data wires (RJ12 telephone cable for CAN bus or ethernet cabling) using solid core rather than stranded. I'd still prefer to use stranded for these applications but IMHO it's less critical.

[MagiChau]

If my memory serves me correctly the required power convertor for 12V to 5V solely for the bridge's power supply has solid wires for connecting it to the power supply & bridge.

[Al Skierkiewicz]

Chau,
Those are pre-tinned stranded wire. You insert one side in the connector for the PD and solder the output to the power cable for the radio. Pre-tinned seems to indicate the primary or intended customer for this regulator would solder it to a circuit board.

[wireties]

FYI - When electricity flows through a wire, it mostly flows on the surface of the wire, not through the middle. This effect is more pronounced on high frequency AC than it is on DC or low frequency AC. This means that a "wire" of a given size that made up of many smaller strands can carry more power than a solid wire - simply because the stranded wire has more surface area. Plus flexibilty is key on our robots and all the termination mechanisms we are given work best on stranded wire.

So... my professional opinion is there are no compelling reasons to use solid-core wire on a FRC robot (even if it was donated).

HTH

[Al Skierkiewicz]

Originally Posted by wireties

FYI - When electricity flows through a wire, it mostly flows on the surface of the wire, not through the middle.

Not exactly, the "skin effect" is more pronounced by higher voltages and higher frequencies. While the Jaguars may introduce this effect, I am guessing it is very small compared to the other losses in the system. There is very little effect at DC.

[Jon Stratis]

Also keep in mind that there are different kinds of stranded wire. The number of strands and the insulation material can bake it hold a bend more or less easily. On our robot this year, we've got two different types of power wire. One is very stiff, and holds a bend very nicely. We use that for a vast majority of our wiring, so we can get nice, clean runs that aren't going to flop loose over time. The second type is much, much more flexible. It won't hold a bend at all. We use that for our elevator, since we require that wire to be constantly moving. Stiffer wire in this situation would be more prone to breaking over time.

[Al Skierkiewicz]

Originally Posted by eagle33199

The second type is much, much more flexible. It won't hold a bend at all.

This is sometimes marketed as "rope lay" or "superflex".

[wireties]

Originally Posted by Al Skierkiewicz

Not exactly, the "skin effect" is more pronounced by higher voltages and higher frequencies. While the Jaguars may introduce this effect, I am guessing it is very small compared to the other losses in the system. There is very little effect at DC.

Jeez Al - did you miss the "This effect is more pronounced on high frequency AC than it is on DC or low frequency AC" caveat or the "flexibility is key" line in the original post? And "Very little" is not accurate. The robot is NOT a static load. The load is variable with components (of the load) related to the rotation of the motors etc. People confuse this all the time, DC means Direct Current, not Direct Voltage. And direct current does not always imply constant current. Nor is the robot's ground reference an infinite sink. Just because the source is a battery does not mean the load can't vary, perhaps with some periodicity. This effect in the robot is important enough to be aware of.

HTH

[Al Skierkiewicz]

Keith,
The skin effect caused by AC currents in copper wire is about 8.5mm at 60 Hz or about twice the diameter of #10 wire (~4mm). At the switching frequency of the Victor of 150Hz, skin effect is not much different and at full throttle surely does not exist. At 15kHz, (the switching frequency of Jaguars) the skin effect is about the diameter of #17 wire. The difference between solid and stranded conductors is still minimal.

[wireties]

Originally Posted by Al Skierkiewicz

Keith,
The skin effect caused by AC currents in copper wire is about 8.5mm at 60 Hz or about twice the diameter of #10 wire (~4mm). At the switching frequency of the Victor of 150Hz, skin effect is not much different and at full throttle surely does not exist. At 15kHz, (the switching frequency of Jaguars) the skin effect is about the diameter of #17 wire. The difference between solid and stranded conductors is still minimal.

Who said anything about the switching frequency of the Jaguars? I don't think I did. I'm talking about the variable load presented by the entire robot. I'll tell you what, bet you a dinner (if we are ever in the same town) that I can vary the load (might cost me a motor or 2) in such a way that there is a nice several volt AC signal modulating the DC supply provided by the battery. And that component (if over 400Hz or so) would travel more on the outside of the #6 AWG wire than the inside, more and more the higher the frequency. I never claimed this is a primary or even secondary factor, just something worth knowing about.

And the skin depth numbers you are talking about refer to the depth at which the current density decays to 1/e (0.37) of the surface density. It is a relative measurement and not a threshold, there is not zero decay under that depth. Indeed there is some decay at lower frequencies and with more constant loads (but not with zero load and/or a perfectly static load). We tend to think of the robot as a static system just because it uses a battery as a power source. That is not strictly true, often times not even close.

HTH

[nuttle]

Skin effect is comparable between solid and stranded and isn't going to be a practical issue in this context anyway. If this were an issue, you'd likely want to use something called Litz wire. The reason this is so is because the fields and forces that produce the skin effect will operate in much the same way in a bundle of stranded wires that are shorted together along their length as in a single solid conductor.

[Al Skierkiewicz]

Keith,
Before you get too much more enraged, let's get on the same page. I agree that there is significant modulation of the DC voltage (when measured at the battery terminals) with varying load. I believe most everyone by now knows that is a factor in much of our power related problems. However, that effect is due more to the resistances encountered internal to the battery, at each termination, and within the wire used on the robot rather than skin effect. While there are short duration pulse type loads put on the electrical delivery system, the majority of time, the power demand is at a steady throttle value predominated by the switching frequency of the relative controllers (at less than full throttle). I have witnessed pulses in the 10-100 micro second range particularly when changing directions or with noise induced from cheap joysticks. Anyone with a scope can see those pulses on their own robot.
This thread is discussing the relative merits of solid vs. stranded wire on our robots. While I agree that skin effect is a serious issue in many aspects of electronics, it is not the major contributor to the losses encountered in our application. While stranded wire as you have pointed out, is better at minimizing skin effects, any solid conductor wire that we see is either #14 or #12 house wire that can be purchased at a local home store. When driving a robot at a steady throttle, the ~.003ohms/ft. of #14 (solid or stranded) added to the termination resistance and internal resistance of the speed controller, is going to be the predominant loss.

[wireties]

Originally Posted by Al Skierkiewicz

Keith,
Before you get too much more enraged, let's get on the same page

Al - over the years you have offered very good advice, saving my team many long hours - and we thank you for it. But your a little off the rails here and I'm not sure where you get the impression I am "enraged". This thread is about stranded vs solid wire on the robots. I pointed out a common reason an engineer might use stranded vs solid wire followed by stating that the effect is diminished at DC and that flexibility and tooling are the primary reasons to use stranded wire on the robot. This is not a very controversial statement. You quoted a single line from the post (leaving out relevant and qualifying conditions), added some skin effect numbers from wikipedia (BTW it is more like 9.8mm for good copper wire at 60Hz), proceeded to correct the post and stated a similar conclusion to my own. I thought it an odd thing to do. Perhaps I am too thin-skinned.

I've politely amended your posts a few times over the years and I apologize if you were/are upset or offended in any way. Next time you amend something I've posted, please (at a minimum) quote the relevant text. Done.

Kind Regards,
Keith

[Al Skierkiewicz]

Keith,
OK, now I am confused. Are we in agreement? What did I amend?
I used the calculator for skin depth on this page...
http://www.daycounter.com/Calculators/ and then did a quick conversion to mm.
I often point ham radio students to pages like this when discussing RF. It is easy to demo reasons for using copperweld for antennas and to discuss why shields on coaxial cable work.