So I just had my meter go funky again. Could not measure amp draw reliably at all.
This is not the first meter I have had go bad. More like the third or fourth. So frustrating.
Inspired by dchomac's tread about making a shunt I did the same thing. First i tried using math and cut lengths of wire to the exact length of 84 mm. Did not work right. Measured too high and gave me grief. Then it ocurred to me that it was probably the soldering of the wires and such that messed with the accuracy due to something Texaspyro wrote in that thread about the shunt resistor.
So I decided to do it the other way around. Take a solid copper wire and solder measuring wires to it and then adjust their positions until it read 2,8 mV when measuring a nanjg driver and 3,0 mV when measuring a qlite driver. So this is what I came up with and it works as far as I can tell.
Please comment if you know something that would mess with the accuracy of this (I have already taken into account that the temperature will affect it but it was made in a 18'C room and I will try to use it at that temperature as far as practically possible)
Here it is anyway. I did not really take any pictures along the way because once I got the ball rolling it was done in 5 minutes.
You can see how I had to reheat the solder and slide the measuring wires further up the legs of the copper wire in order to adjust the measurement.
You need a way to fine tune it after you solder the lead wires onto it.
I used stranded wire and on purpose I cut it a little long. I did that so I could lower the resistance, after soldering the test leads, by letting solder wick into the wire between the 2 connections little by little. If it is lowered too much, then I could squeeze a notch in the wicked portion with my diagonal cutters, introducing a small restriction at that point and thereby slightly increasing the resistance. It wasn’t easy.
I used 14 gauge stranded wire, however it would have been easier had I used 12 gauge or better still, 10 gauge. Heavier gauge wire means that you would be working with longer pieces and would have greater “resolution” to work with. Solid wire could work too.
Measuring high current levels in our flashlights is not easy in and of itself, but when you introduce the resistance of the .01 ohm shunt resistor in most DMM’s AND the resistance of the test leads AND the resistance of the contact points, the reading that you get will NOT be what is actually being delivered to the light in real life.
Modded SRK type flashlights with multiple emitters could use 20A or more. My modded 6 - XM-L2 is pulling in excess of 16A.
I am working on integrating a shunt resistor to a replacement tail cap contact board complete with braided springs so that I can install it onto the back of the light. Only then, when exact operating conditions are met will I be able to get a true measure of the amps drawn.
I first calibrated a shunt resistor in this post, It was difficult but not as difficult as the .001 ohm one.
The resistance of copper has a rather large temperature coefficent… as it heats up, its resistance changes. I have some test equipment that uses pure copper wire as a thermometer. Good resistance shunts are made out of special alloys… generically called shunt wire. A common alloy is manganin. Another is constantan. BTW, many shunt wire alloys don’t solder well at all… and can lose their magic foo powers if they get too hot.
What I did to calibrate was a 3 step process, coarse, medium and fine.
Coarse - Cut the length of wire a little long.
Medium - Allow the soldier to wick. This step used heat and as you mention, it affects the resistance.
Fine - Use diagonal cutters to make the notches.
In my thread on constructing the shunt wire, I used the story of David and Goliath as an example. In the story, David picks up 5 stones for his sling as he goes to confront Goliath. Why 5?, he knew it might take more than 1 shot. What’s really funny is that after you posted how hard it is to actually pull it off, I looked in the trash by my bench. In trying to calibrate, soldering and notching, I had to scrap 4 pieces of wire and start all over again. It was the 5th try that was successful.
My first name just happens to be David
the “d” in dchomak.
@ dchomak : The fine tuning was done a little step at a time. Each time allowing the thing to go back to room temperature and then remeasuring. Then aply heat and move the measuring wire a mm or two up the leg again. Let solder cool and test again.
@ Texaspyro: I have considered that temperature coefficient issue already bu I figured that as long as the measuring is only for a few seconds - maybe as long as 15 seconds at most - then I'll be allright. Even at higher amps the watts of heat being generated are really quite small. I think it is more likely that I heat it as much by holding it with my clammy hands :-) That is why I have now added a whole lot more of that shrink wrap to insulate it from the heat of my fingers...
Anyway as you point out the alloys normally used for these resistors does not like to be soldered at all. Plus where should I get it in small quantity? (maybe ebay - that just occured to me, DOH) and how should I prepare it.
Do you happen to think that this is so amateurish that it is introducing more errors or could we be on the right path. As David points out it is not lab grade I am trying to achieve. Just a lot better than using the meters directly. So what is your verdict - success or fail ? :-)
