I really want to make myself one of these external shunts, but I'm having trouble picturing how calibrate it. In the OP, you wrote:
From the resistance chart, it works out that 4.75 inches of 14 gauge is about .001 ohm. So I cut off a 5 inch length of wire and soldered leads a little greater that 4.75 inches apart. I calibrated it with another HF meter on the 10A scale. At first it read a little high as it was a little too long, but I reduced the resistance slightly by allowing solder to wick into the wire until is was exact.
Could you elaborate on how you did the Bold texted part above?
Thank you HKJ. I think I'm following you. I will try that out this weekend.
My power supply is low end, but I have hooked up DMM's directly to it and have had matching readings on current (within a few milliamps). I don't have a good low voltage DMM, but the rest of what you say should get me most of the way there. Basically, I think you're telling me I should be able to get separate readings from 3 different units (Power Supply Meters current and voltage, DMM current, DMM voltage) to get some comfort with the external shunt.
Yes, when you know the current in the shunt you can use the DMM to verify the reading.
With a 10mOhm shunt you need a meter that is fairly good at low voltages..
I do most of my chargers test with a 10mOhm shunt and a 6½ digit bench DMM, but because I do not maintain calibration factors for the shunts (or calibrate them) I have a few percent tolerance on the current measurement. If I had done some calibration I could get below 0.5%.
BTW, I did not make a .001 ohm resistor, what I did was take a piece of wire and solder connections to it, that after fine tuning, those 2 connections became exactly (well not exactly) .001 ohms apart. There is a subtle difference between the 2, but that difference makes this possible.
I am assuming you know how to make the shunt, what you are asking is how to calibrate it, right?
What I did is put the shunt in series with a HF meter while passing current through both. Then fine tuned the resistance of the shunt so that the voltage read across it corresponded to the current given by the HF meter on the 10A scale. Of course it would only be as accurate as the HF meter. If you have a known current source, there is an easier and better way.
The key to this whole idea is that in constructing the shunt, the first step is soldering test leads to the shunt, before any calibration. This is what makes all this possible with cheap meters. This eliminates contact connection and exact location issues with the shunt resistor and BTW this is a 1 milli ohm resistor, for the lowest possible voltage drop while measuring current.
Once the initial connections are soldered in place and a dedicated meter is chosen (HF meters are cheap or free) The rest of it is fine tuning the resistance between the connection points.
I don’t see why a very good low voltage readout meter is necessary as during the tuning (calibration) process the meter - shunt resistor combo will be calibrated together. If the meter was off in the first place it wouldn’t even matter. Calibrate the combo for the known current flowing through the shunt. Now if a cheap meter is sensitive to a weak battery, that would be another matter.
If the shunt was a separate piece of equipment, then that would be a different matter and every piece of equipment would have to be of high quality.
Because I prefer a resolution in mA, not in Ampere. At 1mOhm 1mV is 1A, i.e. I would want a resolution of 10uV or better (10uV means 10mA in resolution) for most of my measurement.
If I am going to measure above 10A I would probably use a current clamp. If the current was supplied by my (now dead) power supply I could get the value directly from it and not worry about clamp or small value resistor.
Thank you both for your advice. This is on my list of things to try to get done today.
I hope you all don't mind another question. I would like to make one of these shunts for testing current flowing to the emitter. So I will need some extra length to string it from the emitter base to the driver. So the shunt will have the leads for reading voltage somewhere along the shunt wire well away from the ends. When I calibrate, I will be sending x amps through the whole length via the bench top power supply. Do I need to compensate for the extra length of shunt wire outside the voltage meter lead attachment points? I think the answer is no, I don't need to compensate for that.
Another related question. As I use this external shunt, I will periodically need to trim off some of the ends as they will deteriorate from repeated soldering. Will the trimmings require recalibration? I think the answer is no.
Actually, even bigger related question. Then I am calibrating the shunt, I will have like 3/8" or so of each end clamped in the bench top power supply jacks. But when I use the shunt, the ends of the shunt wire will be soldered to a driver and emitter base with no clamps. Do need to alter how I attach the shunt wire to the bench top power supply to make it more similar to how it will be used for measuring current to the emitter? Again, I think the answer is no.
Really, all the 3 questions above are the same question. I think the important part is area between and including where the voltage measurement leads attach to the shunt wire.
Now a somewhat less related question. Why are we not concerned about voltage drop along the leads that go to the DMM that will be measuring the voltage drop along the shunt?
You are correct, you do not have to worry about the part outside the DMM probes (Except it adds some voltage drop, but this has nothing with the current measurement to do).
Depending on the material the shunt is made of, the temperature may also affect the measurements.
We are, but generally the voltage drop will be negligible when the current is 1uA or lower. Remember a DMM usual has 10Mohm input impedance on voltage ranges.
ImA4, you asked me what I meant when I wrote what is in bold in my quote.
You have brought up an old thread and I just now realize that I never fully explained all that was intended when I wrote that.
On the surface it means that I used the only thing I had during the calibration process, a HF meter on the 10A scale. The “10A” scale part is important, but I never explained just what I meant by that.
The purpose for making the shunt resistor is not to use it in measuring 1 to 3 amps. A stock meter with 14 gauge leads is OK for that. This shunt resistor is best used when reading current levels of say, 10A and above. As we all talked about, even though cheaper meters say they have a 10A scale for reading current, most will actually read and display up to 20A. When you calibrate this shunt resistor, you want to calibrate it using a current close to the current levels you will be using it for. In my case it was at some point between 10 and 20A. It was the current flowing through my 6-XM-L2 SRK.
Remember now, the primary reason for constructing the shunt resistor wasn’t necessarily to read currents of 10 to 20A, but was done to lower the resistance in the current measuring path. Measuring 15A with a stock meter and 14 gauge leads has a big impact on the current flowing through LED’s. Pass 15A through that .01 ohm shunt resistor* in a stock meter* and you drop 0.15V across it. The LED’s then only “see” 4.05V instead of 4.20V. I first realized there was a problem with stock meters and their .01 ohm shunt when I first charged up a set of Samsung 28A’s to 4.35V. That extra 0.15V gave me 3A more draw! That means in the case of my 6-XM-L light, when I took a current reading with a stock meter and 14 gauge leads, that current reading would be about 3A less than it would actually be when the light was put together!
Use an expensive and highly accurate meter with it’s 0.01 ohm internal shunt to measure the current, and it too will read 3A short of the actual current when the light is put together.
The reason why that 0.15V drop across the internal shunt has such a big impact is because an LED is a non-linear device and small changes in voltage have a disproportional effect on current draw.
That is the real reason for using this external shunt. It just so happens that larger currents can be measured with cheap meter also.
To fine tune or calibrate the shunt resistor, we need to make adjustments to the resistance.
While a known amount of current is flowing through the shunt wire, and the leads are hooked up to a volt meter we will use the meter reading to make the following adjustments to the shunt wire.
1 Coarse adjustment: Cut the wire a little long and solder the leads a little farther apart than .001 ohm. So we start off a little high. The reason to be a little high is because the next adjustment is perfect for lowering it a little.
2. Medium adjustment: Let solder wick into the stranded wire to reduce resistance a little. Add solder until is a little too low. The reason to be a little low at this point is because the next adjustment is perfect for raising it just a tiny bit.
3. Fine adjustment: Use a pair of diagonal cutters and gently squeeze a notch in the wire where the solder is. With a little patience, just the right number of notches can be created to zero in on an exact reading.
ImA4,
WOW, I just realized than in the 30 minutes it took me to write the above, you asked more questions… HKJ I see, already answered them. What service BLF provides!
Excellent dchomak. Thanks for the added info in Post 28. Nice approach with the excess solder to fine tune.
This may be an old thread, but it is quite relevant. I have read the OP many times trying to figure out how to make one of these. I think it would be nice for future readers if you update the OP. I'm asking as I know I will be one of the future readers.
Thanks, ImA4.
I did a poor job explaining just how to go about creating one of these in the OP.
It takes a LOT of work to write something that is EASY to read.
I can’t imagine how much effort HKJ puts into his write ups, over and above all the work in the actual testing and documenting. It should be greatly appreciated.
I know I often spend MORE time writing about a mod than I do building it!
I will ad an addendum to the OP, a cut a paste of the instructions on how to fine tune the shunt. And a link to my post 28 which I hope explains my thoughts more clearly.
Just wanted to post a big thank you. I finally tried my external shunt in a high power light yesterday. It is so cool. I would have had to purchase a new DMM and probably would not have achieved the same accuracy if it were not for you posting this thread dchomak. Thanks again :)
I should have said this back when I made the shunt, but forgot to make the post. Not sure if the OP says so, but letting the shunt cool before testing (when calibrating) is very important.
I have several DC Volt/Ammeters that don't include shunts for measuring current. This one on Ebay is similar, but the pcb (and maybe some components) appears lightly different. It does appear to use the same shunt spec wise. It calls for a 50A/75mV shunt. Ammeters have labels that either specify 50A/75mV or 100A/75mV.
Since, I don't have a way to push an accurate 100 amps DC through a shunt, I would like to make the shunt by flowing 10 amps through a shunt and measure for a 7.5mV drop. So, my question is:
If I wanted to make my own shunt (or buy one and test it for accuracy), do I need to flow 100 amps through it to measure the 75mV drop? Or does the spec have some other meaning like 75mV drop per Amp of current?
EDIT: I seem to be running into a problem if I go DYI shunt. To get a shunt that can handle 100 amps and have 75mV of drop, I will need a very long shunt. To get a shunt in the length I would want, I would need to go all the way up to 22 gauge wire (5.47 inches long at 77F). I haven't tried to figure out if 22 gauge wire can handle 100 amps, but my intuition says it can't. Even if it can, it will rise in temp at high amps which will throw off the measurement. Guess I need to buy some of these to get these meters working. Am I missing something.
At $3.39 that current shunt is definitely the way to go.
6 Gauge copper wire will handle 100A and you would need just under a 2 foot length of it for it to drop 75mV while conducting 100A. (From ohms law, and the resistance tables of copper wire.)
BTW, that shunt will only need to dissipate 7.5W. P=IxE 100A x .075V = 7.5W
And I can’t wait to see this light that draws 100A you’re working on
Ha! I'd like to see it too. Maybe when emitters have 1Vf at 100 amps.
Thank you. Went ahead and ordered some of those shunts.
EDIT: I still haven't used the shunts I ordered above, but have been using bare 10 gauge wire with a fan blowing air over them to keep the temperature from rising.
Hello friend. Could you explain what is meant by cutting notches in the solder to adjust resistance? Do you cut a v out of some of the exposed solder or do you just squeeze it? Thanks
