... a longer wire will have more heat loss because of resistance
... a longer wire will have more heat loss because it weighs more and requires more amps to lift it
which sounds additive and is if trying to calculate total heat loss and nothing else.
But what we are doing is comparing two cases, each with the same lengths but using different size wire. It's the comparison where we take one case and subtract from it the other case to compare. So length, the same in both cases, is added for case A and subtracted for case B. So length washes out in the comparison.
Shp Wgt Gain - The total length of that particular wire type on the ship x the wire weight. There is a table that you need to fill out that contains the wire weights for the complete list of AWG wire size.
Watts Lost to Weight - For each wire type, this is grams of that wire times the watts per gram it takes to lift that weight. The value is approximately 0.16 grams per watt, but it varies by ship. An easy way to derive this number is weight your ship, calibrate the battery factors in Mission Planner, then hover the ship, get the watts from the log, and then divide the watts by grams.
Also, while probably less important for the lengths we're talking about, longer wire = more voltage drop which in turn means more throttle required to get prop rotation rate where it needs to be. In other words, the wattage drop due to voltage drop from lengthening a wire has to be made up with current (W=V*A). so it's a double-whammy: longer wire creates more resistance which generates heat (also known as wasted energy) and also decreases voltage causing more current pull which feeds back to the resistance issue. it's a pretty negative feedback loop. I think with DC current, best bet is shortest wires possible. if developing a unit, when finished, go back and shorten up as much excess wiring as possible of the wires that carry a lot of current.
I don't quite understand your suggestion here. Your values don't cancel out, they are added, making things worse!
In simplistic terms: As wire length increases, so does the overall resistance (this is linear). This means that for a specific current draw, power is consumed by the cable W = I^2*R, which produces heat in the cable. The higher the current, the more heat. The longer the cable, the higher the resistance = more power loss = more heat. In order to restrict the power loss (voltage loss/drop really) then we are limited to a specific current in this cable to keep the heat loss to a minimum. We can improve this by either: 1. using a thicker cable (so less resistance per metre), or by reducing the length of the cable. A shorter cable will therefore allow a higher current to flow before its volt drop causes power losses in the cable causing heat.
See the posts starting on this page with the user 'splitboarder'.
Another article here: http://www.4-max.co.uk/silicone-wire.htm suggests that 14awg silicone RC cable should take 45amps (doesn't mention lenght though, but for short lengths I would consider this to be fine.
I just checked my Hobbywing X-Rotor 40A ESCs and they come with 16awg power cable by default. 14awg is obviously thicker and able to handle more current.
Forrest thanks for your help, what your table says reassures me, even though I have already flown a lot with my exa.
Anyway I'd like to understand better what you have listed on your table, expecially:
Battery o PDB
1) How do you get Ship Wgt Gain (g)
2) How do you get Battery Wire Watts Lost to Weight?
PDB to motors:
1) How do you get Ship Wgt Gain (g)
2) How do you get Ship ESC Wire Watts Lost to Weight?
what I understand it is that your table is focused on watts lost in heat but How and where do you get the weight?
Bests...
Forrest Frantz said:
For your application, hex at 22 amps at hover:
Battery to PDB--As you can see from below, 14 AWG if optimal for photography (normally relatively low speeds).
For PDB to the ESCs and motors--As you can see, 22 AWG is optimal. All that means is that if you are using something larger then you are using up more amps to lift the wire than necessary and if you use something smaller, then you are burning up too much in heat loss. It is better to error on larger wire size.
I naively put it into the table too thinking that it was necessary. Then found out that when changing the length numbers, that the critical AWG size never changed. What???
As wire length increases, more heat is lost in watts (linear).
As wire length increases, more weight is added, so more watts are needed to lift the wire (linear).
So the two cancel each other out ... one of those embarrassing realization moments. Oh well.
P.S. I left wire length in the worksheet (it is off to the left) only because the worksheet also tells you how much weight you are adding to the ship. It also is important because different wire has different isolation weights (typically equal to the copper weight) so the trade changes. So there is also a table of the wire weights you are planning to use.
@Forrest, The table looks good, except, I don't see wire length in there?
Length is very important in terms of what wire is optimal as well, as the longer the wire, the thicker the gauge is needed. For a 3-4 inch wire, you can use much smaller gauge than say a 8 inch wire.
The only thing I would add is from a practical perspective, feel the wires (and more importantly the connection points all the way from battery leads to main power lead to PDB to ESCs) after a standard flight operation and note whether anything is hot to the point of feeling uncomfortable when you touch it. The temperatures should not feel more than mildly warm.
Assuming the result is mildly warm, then I would do the same test after running the vehicle hard for a few mins in the air (or secure it and run the motors strong on the ground - though less optimal because you lose some of the natural airflow that you would get in the air) and then see if any of the connection points or wiring become hot.
If hot, then I think you have to make some decisions: The issue with hot is that, (1) like the table above shows, you're wasting battery capacity on heat generation which reduces flight time, but worse (2) heat becomes the enemy of heat because heat increases resistance (lowers flight time more) and generates even more heat, which can eventually cause solder to melt, and if you haven't experienced that before, I can tell you that rarely ends well.
If you're the only one flying it and you know this limitation and can reasonably expect that you won't have to draw high current for more than a few seconds at a time, and 95% of the time you're just hovering or flying slowly and not in windy conditions, then leave it. If other people are flying it who may not be as careful, or if you are flying it in windy conditions occasionally where it could have to fly into a headwind for extended periods and draw higher power, or occasionally putting a larger camera on (heavier payload) then I'd consider taking corrective actions.
Besides moving to a lower AWG wire, if the higher current draw scenarios are still more of a rare occurrence than a regular one, and you're interested in minimizing weight to maximize flight time for the vast majority of the use scenarios, then you can address the danger from heat by moving as many connection points and as much wiring as is practical to areas that will be below the prop wash when flying. This is a great way to keep components and wire connections cool because you have a natural "fan". Keeping them cool not only lowers resistance that would otherwise be there, but also minimizes the risk of solder melting. If you do this, especially with ESCs, try to position them in ways that expose as little flat surface area straight up to minimize airflow resistance from the prop wash that pushes down on those components, effectively making the vehicle "heavier" and requiring more current to offset that force pushing down on the aircraft.
We spent a lot of time optimizing all of this with our quad. But it may not be directly relevant to you, because ours is designed for a variety of uses and payloads/battery packs. With the smallest pack (6S 13AH) hover current is about 33 AMPS, but max draw with payload and maneuvering can get up to 200 AMPS. Connection points like the current meter (eg the Pixhawk power module - though we use something else since that can't handle 6S), are particularly vulnerable to solder melting. By placing that, battery lead connectors, and ESCs in places where they remain under the prop wash, we effectively eliminated this problem, and connectors that were sometimes too hot to touch before, are only slightly above ambient temperature now even after sustained high current draw.
Look at the current draw in the APM flight log after a flight if you haven't. You may be surprised how high the current spikes can be, especially if you were flying in windy/gusty conditions. The flight controller can request very high current levels repeatedly to keep the vehicle stabilized or to fly into headwinds. We had some unfortunate situations in the early days of testing where good solder points melted during those types of maneuvers because either the heat dissipation wasn't managed well or the wire gauge was just too high. We're using a mix of 12 and 10 AWG now, so again, I know it's probably a different situation than yours. I just want to spare you and others the pain of cutting it too close for a few extra seconds or (or a minute or 2) of flight time. Those rare circumstances, which you can't always control, can come back to bite!
Battery to PDB--As you can see from below, 14 AWG if optimal for photography (normally relatively low speeds).
For PDB to the ESCs and motors--As you can see, 22 AWG is optimal. All that means is that if you are using something larger then you are using up more amps to lift the wire than necessary and if you use something smaller, then you are burning up too much in heat loss. It is better to error on larger wire size.
15 amps is cited as the limit for household wiring applications, but this is due to the long cable runs in house wiring. In drones we are talking very short runs so proportionally lower volt drop, so in this case the same cable can handle a lot more current. 75amps is suggested for short runs in the articles I have read.
Replies
This is what you are missing:
You are correct in that:
... a longer wire will have more heat loss because of resistance
... a longer wire will have more heat loss because it weighs more and requires more amps to lift it
which sounds additive and is if trying to calculate total heat loss and nothing else.
But what we are doing is comparing two cases, each with the same lengths but using different size wire. It's the comparison where we take one case and subtract from it the other case to compare. So length, the same in both cases, is added for case A and subtracted for case B. So length washes out in the comparison.
Optimizing Wire Size, the explanation.
Shp Wgt Gain - The total length of that particular wire type on the ship x the wire weight. There is a table that you need to fill out that contains the wire weights for the complete list of AWG wire size.
Watts Lost to Weight - For each wire type, this is grams of that wire times the watts per gram it takes to lift that weight. The value is approximately 0.16 grams per watt, but it varies by ship. An easy way to derive this number is weight your ship, calibrate the battery factors in Mission Planner, then hover the ship, get the watts from the log, and then divide the watts by grams.
Also, while probably less important for the lengths we're talking about, longer wire = more voltage drop which in turn means more throttle required to get prop rotation rate where it needs to be. In other words, the wattage drop due to voltage drop from lengthening a wire has to be made up with current (W=V*A). so it's a double-whammy: longer wire creates more resistance which generates heat (also known as wasted energy) and also decreases voltage causing more current pull which feeds back to the resistance issue. it's a pretty negative feedback loop. I think with DC current, best bet is shortest wires possible. if developing a unit, when finished, go back and shorten up as much excess wiring as possible of the wires that carry a lot of current.
I don't quite understand your suggestion here. Your values don't cancel out, they are added, making things worse!
In simplistic terms: As wire length increases, so does the overall resistance (this is linear). This means that for a specific current draw, power is consumed by the cable W = I^2*R, which produces heat in the cable. The higher the current, the more heat. The longer the cable, the higher the resistance = more power loss = more heat. In order to restrict the power loss (voltage loss/drop really) then we are limited to a specific current in this cable to keep the heat loss to a minimum. We can improve this by either: 1. using a thicker cable (so less resistance per metre), or by reducing the length of the cable. A shorter cable will therefore allow a higher current to flow before its volt drop causes power losses in the cable causing heat.
The article I originally refereed to which explained this is here: http://rc.runryder.com/helicopter/t685208p1/
See the posts starting on this page with the user 'splitboarder'.
Another article here: http://www.4-max.co.uk/silicone-wire.htm suggests that 14awg silicone RC cable should take 45amps (doesn't mention lenght though, but for short lengths I would consider this to be fine.
I just checked my Hobbywing X-Rotor 40A ESCs and they come with 16awg power cable by default. 14awg is obviously thicker and able to handle more current.
For info on wire resistance see this: http://www.powerstream.com/Wire_Size.htm
Design your system to not lose more than 5W over the length of your cable and this should be fine.
Forrest Frantz said:
Forrest thanks for your help, what your table says reassures me, even though I have already flown a lot with my exa.
Anyway I'd like to understand better what you have listed on your table, expecially:
Battery o PDB
1) How do you get Ship Wgt Gain (g)
2) How do you get Battery Wire Watts Lost to Weight?
PDB to motors:
1) How do you get Ship Wgt Gain (g)
2) How do you get Ship ESC Wire Watts Lost to Weight?
what I understand it is that your table is focused on watts lost in heat but How and where do you get the weight?
Bests...
Forrest Frantz said:
I naively put it into the table too thinking that it was necessary. Then found out that when changing the length numbers, that the critical AWG size never changed. What???
As wire length increases, more heat is lost in watts (linear).
As wire length increases, more weight is added, so more watts are needed to lift the wire (linear).
So the two cancel each other out ... one of those embarrassing realization moments. Oh well.
P.S. I left wire length in the worksheet (it is off to the left) only because the worksheet also tells you how much weight you are adding to the ship. It also is important because different wire has different isolation weights (typically equal to the copper weight) so the trade changes. So there is also a table of the wire weights you are planning to use.
@Forrest, The table looks good, except, I don't see wire length in there?
Length is very important in terms of what wire is optimal as well, as the longer the wire, the thicker the gauge is needed. For a 3-4 inch wire, you can use much smaller gauge than say a 8 inch wire.
That looks like a very helpful table.
The only thing I would add is from a practical perspective, feel the wires (and more importantly the connection points all the way from battery leads to main power lead to PDB to ESCs) after a standard flight operation and note whether anything is hot to the point of feeling uncomfortable when you touch it. The temperatures should not feel more than mildly warm.
Assuming the result is mildly warm, then I would do the same test after running the vehicle hard for a few mins in the air (or secure it and run the motors strong on the ground - though less optimal because you lose some of the natural airflow that you would get in the air) and then see if any of the connection points or wiring become hot.
If hot, then I think you have to make some decisions: The issue with hot is that, (1) like the table above shows, you're wasting battery capacity on heat generation which reduces flight time, but worse (2) heat becomes the enemy of heat because heat increases resistance (lowers flight time more) and generates even more heat, which can eventually cause solder to melt, and if you haven't experienced that before, I can tell you that rarely ends well.
If you're the only one flying it and you know this limitation and can reasonably expect that you won't have to draw high current for more than a few seconds at a time, and 95% of the time you're just hovering or flying slowly and not in windy conditions, then leave it. If other people are flying it who may not be as careful, or if you are flying it in windy conditions occasionally where it could have to fly into a headwind for extended periods and draw higher power, or occasionally putting a larger camera on (heavier payload) then I'd consider taking corrective actions.
Besides moving to a lower AWG wire, if the higher current draw scenarios are still more of a rare occurrence than a regular one, and you're interested in minimizing weight to maximize flight time for the vast majority of the use scenarios, then you can address the danger from heat by moving as many connection points and as much wiring as is practical to areas that will be below the prop wash when flying. This is a great way to keep components and wire connections cool because you have a natural "fan". Keeping them cool not only lowers resistance that would otherwise be there, but also minimizes the risk of solder melting. If you do this, especially with ESCs, try to position them in ways that expose as little flat surface area straight up to minimize airflow resistance from the prop wash that pushes down on those components, effectively making the vehicle "heavier" and requiring more current to offset that force pushing down on the aircraft.
We spent a lot of time optimizing all of this with our quad. But it may not be directly relevant to you, because ours is designed for a variety of uses and payloads/battery packs. With the smallest pack (6S 13AH) hover current is about 33 AMPS, but max draw with payload and maneuvering can get up to 200 AMPS. Connection points like the current meter (eg the Pixhawk power module - though we use something else since that can't handle 6S), are particularly vulnerable to solder melting. By placing that, battery lead connectors, and ESCs in places where they remain under the prop wash, we effectively eliminated this problem, and connectors that were sometimes too hot to touch before, are only slightly above ambient temperature now even after sustained high current draw.
Look at the current draw in the APM flight log after a flight if you haven't. You may be surprised how high the current spikes can be, especially if you were flying in windy/gusty conditions. The flight controller can request very high current levels repeatedly to keep the vehicle stabilized or to fly into headwinds. We had some unfortunate situations in the early days of testing where good solder points melted during those types of maneuvers because either the heat dissipation wasn't managed well or the wire gauge was just too high. We're using a mix of 12 and 10 AWG now, so again, I know it's probably a different situation than yours. I just want to spare you and others the pain of cutting it too close for a few extra seconds or (or a minute or 2) of flight time. Those rare circumstances, which you can't always control, can come back to bite!
For your application, hex at 22 amps at hover:
Battery to PDB--As you can see from below, 14 AWG if optimal for photography (normally relatively low speeds).
For PDB to the ESCs and motors--As you can see, 22 AWG is optimal. All that means is that if you are using something larger then you are using up more amps to lift the wire than necessary and if you use something smaller, then you are burning up too much in heat loss. It is better to error on larger wire size.
Thanks Paul,
would you share the article you read.
Best regards.
Marco
Paul Atherton said: