Maker Pro
Maker Pro

Need inductive loop/clamp for A0 input to measure -30 to 80 amps dc

dpuklicz

Apr 11, 2020
6
Joined
Apr 11, 2020
Messages
6
I desire to measure whether a solar battery bank is currently being charged (+ amps) or discharged (- amps) which changes based on sun /charge (max 80a) and load draw overnight (max -30a). The A0 (analog) input feed is to an Arduino NANO which feeds Node Red (serial, 0 -1023) running on an Rpi3. The NANO is already in place, monitors the battery bank voltage and is successfully feeding data to the Rpi3 where it is processed.

The cables between the battery bank and the charge controller (and inverter) are thick so any inductive loop should be at least 20 - 25mm inside diameter.

Additionally, I'm not sure what voltage to expect from an inductive loop. The NANO input can be 0 - 5v (I doubt a voltage divider is needed). So any idea where I can source an appropriate inductive loop and any help on how best to use the loop output as a feed to the NANO would be appreciated.
 

dpuklicz

Apr 11, 2020
6
Joined
Apr 11, 2020
Messages
6
Thanks so much Bluejets but the cables between the battery bank and loads is 0 AWG, I could never manage to connect the ACS758 in circuit for wire that size. That is why I was seeking an inductive pickup.
 

dpuklicz

Apr 11, 2020
6
Joined
Apr 11, 2020
Messages
6
Thanks again but I found these as well and the inside diameter does not meet the size I mentioned (min 20-25mm).
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
You could just purchase a small Hall-effect sensor such as this one and roll your own with a toroid ferrite core of appropriate hole diameter to pass your humongous conductor through. (Really? 0 AWG for a measly 80 amperes?) Use a diamond abrasive saw to cut a radial slot in the toroid for the Hall sensor to slide into. Add some signal conditioning and voila! Bob's yer unkle.

Or secure an old TV fly-back transformer and salvage the ferrite core, which is generally in two pieces held together with a steel wire spring. Slip the Hall sensor between the two halves and add a spacer of the same thickness as the Hall sensor (or use two identical Hall sensors) so the two pieces will fit back together. I wouldn't worry too much about magnetic fringe fields since you will need to calibrate the current sensor with known currents, perhaps by using a clamp-on meter with a Hall sensor built in. This is a hobbyist forum, so jump on your hobby horse and ride your project to completion. Not everything has to be off the shelf with a manufacturer's warranty.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Really? 0 AWG for a measly 80 amperes

Maybe for voltage drop at the low level.
Still, it's only 50 sq mm so probably no more than about 12mm od.(with insulation)
Don't know where the 20mm comes from.
 
Last edited:

dpuklicz

Apr 11, 2020
6
Joined
Apr 11, 2020
Messages
6
The 0 AWG cables were supplied by the solar distributor I used and yes, I suspect it is primarily to minimize resistance between the battery bank and the inverter (reduce any losses/ inefficiencies as much as possible is a solar mantra). There are actually 2 wires, the 0AWG which connects to the battery bank to the inverter (6') and a 10AWG wire that runs directly off the batteries to support 12 v devices (audio amp for ceiling speakers, automation computer, doorbell, 12 to 5v converter (Arduino stuff), CCTV system, etc..).Whenever a device used a 120vac/12vdc (or 5v) power adapter I just connect it to the batteries. There is a 2AWG cable between the solar charger and inverter (they are parralel across the batteries) and when the solar charger is dumping 80amps for 3 hours it does get fairly warm, and its only 2 ft long.

Thanks for the feedback Hevans1944, I'll look into your suggestions.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
I really envy those of you who can afford an 80 A solar photovoltaic system of any voltage or peak power capability. As soon as we retired to the sunshine state of Florida we began to be bombarded with offers of "free" solar energy to be installed on the roof of our (relatively) new house. However, upon carefully reading the "mouse print" details, it turns out we would not be buying anything but instead would be leasing the solar equipment. The "free" part was supposed to be the savings created by selling power back to the grid utility, said savings going directly to pay the monthly lease fee. No mention of what effect a solar panel array would have on the future value of our house, or the cost to remove it and repair the roof afterward. And no batteries to store energy for use overnight. That would surely keep the installation cost down and decrease the risk associated with defaulting on the lease arrangement. Given these negative factors we have (so far) been able to resist all of the "free" solar offers.

@dpuklicz : I think I will purchase a couple of the inexpensive Hall-effect devices I linked to and see how well they work with some fly-back transformer cores I have saved. That will also give me an excuse to get off my ass and finish repairing the 300+ ampere DC power supply cum battery charger cum car engine starter that is taking up valuable floor space in my garage. Previously I used a few feet of heavy gauge nichrome wire as a "current shunt" to measure the current output of this supply, but the wire gets rather warm and changes its resistance, so the "calibration" leaves a lot to be desired.

I had a commercial shunt, that I have since misplaced, that could measure a few hundred amperes with only a few hundred millivolts of "burden" and practically no change in its resistance as a function of temperature. You might want to investigate obtaining one of these on eBay. Yeah, its very "old school" but they do work just fine after a little signal conditioning to bring the millivolt signal up to the volts level needed for Arduino a/d conversions. These shunts have massive copper terminals for the current to pass through and a pair of small screw terminals for measuring the voltage developed as a result of the current passing through the shunt. So some assembly is required.

Remember to have FUN!
 
Last edited by a moderator:

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
The 0 AWG cables were supplied by the solar distributor I used and yes, I suspect it is primarily to minimize resistance between the battery bank and the inverter (reduce any losses/ inefficiencies as much as possible is a solar mantra). There are actually 2 wires, the 0AWG which connects to the battery bank to the inverter (6') and a 10AWG wire that runs directly off the batteries to support 12 v devices (audio amp for ceiling speakers, automation computer, doorbell, 12 to 5v converter (Arduino stuff), CCTV system, etc..).Whenever a device used a 120vac/12vdc (or 5v) power adapter I just connect it to the batteries. There is a 2AWG cable between the solar charger and inverter (they are parralel across the batteries) and when the solar charger is dumping 80amps for 3 hours it does get fairly warm, and its only 2 ft long.

Thanks for the feedback Hevans1944, I'll look into your suggestions.

You are aware the current transformer or whatever you decide to use only goes in one leg of the circuit...??
If, as you say, the cable size is for voltage drop, it would be quite acceptable to break gauge just for the measuring device, whichever, without any impact on performance.
 
Last edited:

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
... it would be quite acceptable to break gauge just for the measuring device...
Sure, but if you do decide to open one of the heavy current-carrying cables, you can insert into the cut cable a short length, say three or four inches (100mm), of three or four parallel-connected 10AWG insulated wires. Feed one or all of these smaller gauge wires through the hole in a Hall-effect current-measuring coil. You can pass just one, two, three, or all four 10AWG wires through the sensor coil. Leaving the remaining 10AWG wires outside the coil, the total current passing through the 0AWG wire will divide evenly among the smaller 10AWG wires if these wires are parallel-connected with a low-resistance connection technique.

To make a low-resistance transition from a single 0 gauge wire to several parallel-connected 10 gauge wires, I would get two pieces of copper buss bar, secured on opposite sides of the current-sensing coil (perhaps bolted to a plastic or Lexan polycarbonate base), and with cable clamps* attached to secure the wires. Feed one, two, three, or four 10AWG through the Hall-effect current measuring coil. You could get by with passing just one of the wires through the Hall-effect current measuring coil, measuring one third (if using three wires) or one fourth (if using four wires) of the actual current in the 0 gauge wire. The remaining two thirds or three fourths of the total current flows through the two or three 10AWG wires that do not thread through the current measuring coil. You do need really good, low-resistance, parallel connections of the 10AWG wires at the copper buss bars if you don't want to, or cannot, pass all the wires through the sensor coil.

*Actually, I would go further and crimp a wire ferrule onto each wire. Drill holes in the buss bars to accept the ferrules. Drill and tap holes perpendicular to the holes just drilled to accept threaded machine screws, or bolts, or set screws, or whatever that will tighten against the ferrules. If using a threaded bolt, you could add a jam-nut to make sure it stays tight. Set screws are available with nylon inserts that help prevent them from backing out.

@dpuklicz Don't try to cheat and not use the ferrules, for example by tinning the bare ends of stranded wire with solder. Solder creeps under pressure and then will require periodic re-tightening of the connections to the buss bar to maintain low-resistance parallel connections. It's okay to tin the wire with solder after insertion into the ferrule and after the ferrule has been crimped (preferably with a "star" crimp). This extra tinning operation isn't necessary with a proper crimp on the ferrule. Always used stranded wire, never solid wire, with crimped connections, whether ferrules or ring, spade, or forked terminals.

As noted above, you can divide your 0 gauge cable into a short length of several parallel, smaller-gauge, cables for the purpose of measuring current. If you take this far enough, you could (for example) use the original Allegro current sensor suggested in post #2, except I would use one that is recommended for new designs and not mounted on a cheesy Chinese circuit board claiming to be adequate for 100 amperes. Look at this 150 ampere bidirectional model from Digi-Key. You can solder directly to the pins and insulate the soldered connections with shrink tubing. The three smaller terminals for +5V, Signal, and Ground can be twisted, stranded, insulated 24AWG hook-up wire several feet long connecting to the Arduino. No circuit board and no extra op-amps necessary for interfacing to Arduino analog input port.
 
Last edited:

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
An inductive sensor will not work with DC current in your solar installation. Inductive sensors can only detect a change in current, i.e. AC, not static current, i.e. no DC.

A hall sensor will detect DC current. You can buy clamp on type hall sensors (link is just one example) for various wire gauges.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
Hall-effect sensors have been around for a long time, since 1879 when Edwin Hall discovered them. Putting a Hall-effect device to practical use is another matter. The devices are temperature sensitive, the Hall-effect is weak, and output offset, as a result of other magnetic fields, including the Earth's magnetic field, must be "zeroed" out. I have used a "bare" Hall-effect sensor mounted on the end of a plastic stick to probe magnetic fields. This commercial instrument came with the electronics needed to amplify and display the Hall voltage, but in my limited experience it was unsatisfactory in performance. Of course that may have been the result of the cheap price a notably penny-pinching engineer I worked with paid for it. It did work, after a fashion, but it also was not very sensitive and required frequent re-zeroing. I never did trust its calibration. It was also somewhat awkward to use, being mounted on the end of a ten- or twelve-inch long plastic rod, IIRC. So, after "playing" with it for a few days, I gave it back. I also gave it no further thought until this topic came up.

A few years ago, some of us here on EP discovered Allegro current sensors that were based on the Hall effect. One of us successfully used two of them to create a "motor on" sensor for 120 VAC and 240 VAC woodworking shop motors. These Allegro sensors, with the aid of a PIC micro, turned on a solid-state AC switch that operated a vacuum dust collection system whenever a woodworking motor was turned on. The PIC monitored both Allegro sensors and, when all motors were turned off, deactivated the dust collection system after a certain delay to allow time for sawdust to clear the dust collection tubes. The whole circuit was designed to fit inside a standard electrical "handy box." With a deep enough box, the circuit board could be mounted behind the convenience outlets for the tools, one outlet for 120 VAC and the other outlet for 240 VAC. I don't think the builder actually did that because (1) it would have been a tight fit and (2) he would want to "take with" if he and his wife ever sold their Florida home. So better to build it in a separate box with two power cords for 120 VAC and 240 VAC, and two (or more) outlets for the power tools, plus a third outlet for the dust collector motor.

You can purchase a commercial version (120 VAC only) that plugs into an ordinary convenience outlet and provides an outlet for the power woodworking tool and another outlet for the vacuum dust collector. Cost is about fifty bux on Amazon. But where is the fun in that?

What was nice about the Allegro device was Allegro did all the engineering to turn the Hall-effect sensor element into a practical device. Temperature compensation, chopper-stabilized amplifier for signal conditioning, EEPROM look-up table for calibration, ratiometric output referenced to the supply voltage, and (most important) an integrated high-current, low-resistance, circuit that created the magnetic field for the Hall sensor, all built into a small 5-pin plastic package. I don't know how much the Allegro devices sold for when they first became available, but they are now dirt-cheap and have better specs than the earlier models, which apparently have now been dumped on the Asian markets, or perhaps reverse-engineered and "cloned".

From the beginning Allegro has aimed their product at mass markets, such as automobiles. The resulting economies of scale have driven the price down to where just about anybody can afford to play with them today. It isn't necessary to use Allegro sensors with a magnetic flux ring except for very high current applications... 300A and up. But if that range of current is necessary, Allegro sells a version without the built-in "current shunt" that creates the magnetic field their Hall-effect sensor measures. Visit this web page for further details on that.

One of the nicest things about the Allegro current sensors is their electrical (galvanic) isolation from the real world. Check the datasheet to see what voltage they will withstand without arcing. If used with a ring magnetic flux concentrator, the sky's the limit since those things can be potted against high voltage conduction after the sensor is installed in a slot in the ring.
 
Top