resistance question

[Lolaralph]

I recently completed my first maglite mod, an SST50 on a britelumens heatsink, powered by a taskled H6flex driver, set a 5 amps, running off two Eagletac 18650's. I used 24 gauge teflon wire for all connections, the longest being 3 inches from switch to driver. My question is, did I use too small of wire, if I did how much is it limiting my light. I used my cheap DMM across the batt. neg. and body and get between 3.75 and 4.4 amps. Does this indicate wire limitations or possibly less than ideal batteries. Would IMR batteries be a better choice, or simply AW cells. Any opinions/advice is greatly appreciated, thanks

 

[45/70]

Real quick, yes 24 ga. wire is too small for 5 Amps, and probably has to do with why your readings are low, provided your cells are in good shape and capable of delivering 5A. Most likely the light would be noticeably brighter as well, with heavier wire. For 5 Amps, I would use 18 ga. wire, that's what I used in my ROP (4.3A). You could probably get away with 20 ga. though.

IMR's would probably be a better choice and last longer, but good 18650 LiCo's should be able to handle it. That's about the limit though.

Dave

 

[Mr Happy]

According to a handy web site (http://www.cirris.com/testing/resistance/wire.html), a 3 inch length of 24 AWG wire has a resistance of about 0.006 ohms. IMHO, this is way down in contact resistance territory, so I would not worry too much about the wire being too thin. Before swapping out the wire I would look for other sources of resistance such as switch contacts, poor solder joints, steel springs, etc. One thing to note about solder joints is that solder is a poor conductor of electricity. When making high current joints a good tight copper/copper connection is required underneath the solder -- the solder is there to provide mechanical strength, keep the joint together, and keep the air out -- not to conduct electricity.

 

[45/70]

IMHO, this is way down in contact resistance territory, so I would not worry too much about the wire being too thin.

Well, I did say "real quick"!

Before swapping out the wire I would look for other sources of resistance such as switch contacts, poor solder joints, steel springs, etc

This is a good point. Stock M@gs have a lot of areas where resistance can be improved. When I made my first ROP, I used something like 24 ga. wire for the resistance mods (tailspring also), as it was easier to route etc. It worked OK, but after replacing the smaller wire with 18 ga, the light was noticeably brighter. I don't know whether it was because the small wire was heating up (which increases resistance), or maybe it was that it didn't make as solid a connection, as you mentioned, but the heavier wire definitely made a big difference.

When making high current joints a good tight copper/copper connection is required underneath the solder -- the solder is there to provide mechanical strength, keep the joint together, and keep the air out -- not to conduct electricity.

So much for pretinning then. I never thought about this, concerning high current loads. I always thought connections, such as in house wiring, were not soldered because of the risk of it melting and coming apart. I never considered the solder not making as low resistance a connection as copper to copper.

Dave

 

[Aquanaut]

One thing to note about solder joints is that solder is a poor conductor of electricity.

That is one of the reasons why I use solder with a high silver content, since silver has the highest electrical conductivity of any element. It also provides a stronger bond. However, it does require more heat to melt.

Thanks for your wire resistance URL. Quite useful.

 

[DIWdiver]

So much for pretinning then. I never thought about this, concerning high current loads. I always thought connections, such as in house wiring, were not soldered because of the risk of it melting and coming apart. I never considered the solder not making as low resistance a connection as copper to copper.

Dave

Actually the industry standard (J-STD-001) requires that stranded wire be tinned before assembly. It also requires that wires be wrapped around posts, eyelets, etc to provide mechanical strength before soldering.

While it's true that the conductivity of solder is about 1/10th that of copper, a decent joint is going to have much larger area and vastly shorter length than the wire, so in practice we never worry about the thickness of the solder layer. By the time it becomes important you will have experienced other problems much worse than high resistance joints.

Another purpose of the solder, in addition to sticking everything together and keeping the air out as MrH points out, is to provide a large contact area, filling in all the gaps that would otherwise be filled with air (and the conductivity of solder is WAY higher than air). The beauty is that it can do all this in places where almost no other method is practical.

D

 

[DIWdiver]

I recently completed my first maglite mod, an SST50 on a britelumens heatsink, powered by a taskled H6flex driver, set a 5 amps, running off two Eagletac 18650's. I used 24 gauge teflon wire for all connections, the longest being 3 inches from switch to driver. My question is, did I use too small of wire, if I did how much is it limiting my light. I used my cheap DMM across the batt. neg. and body and get between 3.75 and 4.4 amps. Does this indicate wire limitations or possibly less than ideal batteries. Would IMR batteries be a better choice, or simply AW cells. Any opinions/advice is greatly appreciated, thanks

Since you are using a buck regulator, you would expect the battery current to be lower than the LED current, so the readings you are getting don't indicate a problem. If you want to measure the LED current, you have to measure it between the driver and the LED.

I agree 24 ga is a little thin for 5A, and if you needed every millivolt from the battery (like in direct drive), or every second of runtime, I'd recommend changing it. But since you have plenty of voltage, and the regulator takes care of giving you a regulated current, I wouldn't bother.

D

 

[MrGman]

you should change the wire. In an ideal world with no nicks or imperfections yes that wire is good to 0.006 ohms. Copper resistance goes up with temperature. If the flashlight is getting warm and the wire with it not because of any wire problem, resistance is going up, if the wire is getting warm on its own, resistance is going up. 4 amps at 0.006 ohms is already a 24 mV drop (0.024V). If for whatever reason that goes up to 0.012 ohms (and that is not unreasonable when inside a hot flashlight) then boom you have a 48mV drop. So go with at least 20 gauge stranded wire and minimize your losses.

 

[Mr Happy]

Actually the industry standard (J-STD-001) requires that stranded wire be tinned before assembly. It also requires that wires be wrapped around posts, eyelets, etc to provide mechanical strength before soldering.

I may be showing my ignorance here, but how can these two sentences be reconciled with each other? In my experience, bare stranded wire is nice and bendy and can easily be wrapped around posts, eyelets etc. before soldering. But once you tin it that stranded wire becomes stiffer than a corpse in a mortuary, and no way can you bend it afterwards, let alone wrap it round things.

 

[DIWdiver]

I may be showing my ignorance here, but how can these two sentences be reconciled with each other? In my experience, bare stranded wire is nice and bendy and can easily be wrapped around posts, eyelets etc. before soldering. But once you tin it that stranded wire becomes stiffer than a corpse in a mortuary, and no way can you bend it afterwards, let alone wrap it round things.

Even 12 AWG solid wire can be bent back flat on itself, but not with bare hands. I typically use a pair of needle-nose pliers, though chain-nose are really better. To be quite honest though, unless I'm working on a product for sale, I often don't tin first, especially with thicker wires. And it's much easier if you have a PCB with a hole in it, so the wire can just go straight in.

D