Announcement

Collapse
No announcement yet.

Variations in battery cable resistance?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Variations in battery cable resistance?

    One of the things I did recently with my overpriced science fair project (Outback hybrid grid-tie with 24 panels and 410 Ah AGM battery) was to make a current-measuring shunt using the 5-foot 4/0 negative battery cable. I have a very sensitive USB-connected voltage measuring device (Yocto-milliVolt-RX in case anyone feels like looking it up) that I connected to some wire leads at each end of the cable. I put a 27 Ohm resistor between the wire leads where they reach the voltage sensor (low impedance to minimize inductive coupling effects), followed by an RC lowpass consisting of two 22K resistors and a 1 uF film capacitor across the sensor inputs.

    With some conversion based on the standard 4/0 copper wire resistance at room temperature and the actual measured temperature near the battery, it provides very stable current readings that match up with what my FlexNet DC current meter is showing, with one significant exception: The resistance of my battery cable appears to be about 14% higher than what all the tables say is the resistance (for that length and temperature) of 4/0 annealed copper wire! There's no way the wire was so much hotter than its surrounding, with modest current flow (0-30A), as to explain that discrepancy.

    My wire leads are attached right at the lugs on the ends of the cable, so not even contact resistance to the battery terminals explains it.

    Could there really be that much variation in resistance of a not-cheap cable that was factory-made, crimps and all, for connection to high-current batteries? And if so, from what? Isn't the copper used in these cables all the same? Is it possible that the factory-crimped connections at each end are adding a consistent 14% to the total resistance?

  • #2
    If you want accuracy, use a real 4 wire shunt with temp compensated resistance. Copper is
    not so stable,
    and plain wiring is rather sensitive to the physical arrangement. Bruce Roe

    Comment


    • #3
      Are you measuring the cable resistance (ohms) or the voltage across it from amps ?

      It appears you are using the cable as a shunt and are measuring voltage across it, which only needs 2 wires. So either the cable is not really 4/O or the crimps are no good. The calibration on your meter could be off too, do you have any sort of standard to check it against ?
      Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
      || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
      || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

      solar: http://tinyurl.com/LMR-Solar
      gen: http://tinyurl.com/LMR-Lister

      Comment


      • #4
        Originally posted by Mike90250 View Post
        Are you measuring the cable resistance (ohms) or the voltage across it from amps ?

        It appears you are using the cable as a shunt and are measuring voltage across it, which only needs 2 wires. So either the cable is not really 4/O or the crimps are no good. The calibration on your meter could be off too, do you have any sort of standard to check it against ?
        Just the voltage across it, given the amount of current being reported by the FlexNet DC. I don't have access to a micro-ohmmeter. My next step will be to put a DC load on the battery with an ammeter that tells me exactly how much current the DC load is adding and then note the increase in reported amps from my shunt. That should tell me if the FlexNet DC is reporting things wrong.

        Good point about the calibration of the voltage sensor. I can make a standard by putting a small amount of current through a precision (and precisely measured) resistor from a measured voltage source and seeing how much voltage it reports across that

        Comment


        • #5
          OK this is just light ironic humor from one train driver to another train driver. Way to many Chefs in the kitchen stirring the pot and over cooked it.

          What you overlooked is Contact Resistance and have not accounted for it. How much Contact Resistance BTF out of me, I would have to use my 10-amp DRLO to answer that question. Here is what I can tell to give you an idea of what you are up against.

          In telecom we use 2-volt batteries at 48 volts or 24 cells. Each cel can be as much as 4000 AH. Those batteries have huge Term Plates, large enough to attach 4 x 2-Hole with 1/2-inch bolt hole with 2-inch spacing to facilitate connecting 4 parallel 750 MCM conductors. Basically a lead plated copper buss bar on each battery term post. Point is a lotta current and contact resistance is important. So in the specs, the Contractor is required to use a 10-Amp DRLO with a witness (typically a telco employee) and every connection point must be 25 micro-ohms or less, recorded, and required required in closing documents.

          One would think using big of 750 MCM lugs with the surface area of your hand bolted together with 2 x 1/2-ich bolting hardware torqued down to 40 ft-lbs on a buss bar is easy peazy. Think again it is no accident and takes skill and work to get Contact Resistance down to 25 micro-ohms or less. The buss bar and lug contact are needs buffed off with with green Scotch Bright and cleaned with a cleaning solvent like acetone, then a very lite coat of NO-OX-ID Special-"A" conductive grease applied to contact surfaces. Then Torqued properly to recommended torque values of 40 to 50 ft-lbs.

          OK more to the point of too many chefs in the kitchen. You are way over thinking this, and using new technology that is just not required. IMO you over engineered it re-inventing the wheel. This is one case where New Technology offers no improvement and just cost more more prone to errors. When it comes to measuring DC Current with good accuracy, reliable, and bullet proof is using ole school tech a 20, 50, or 100 mv current shunt. That is why Industry still uses them when precision and reliability are priority #1. Yep a good ole fashion precision temperature compensated power resistor is what you are looking for instead of trying to modern modern tech. Even a good Hall Effect Transducer is better than what you are trying to do if you do not want to sacrifice 20 to 100 mv to get a Current Measurement. Hint, the higher the mv shunt you use, the more accurate your measurement will be.

          So what is the mv rating of your shunt, and how are you compensating for Contact Resistance that is unknown? See anything wrong? Contact resistance is going to change with temps and throw off any calibration.

          The joys of engineering. We measure things with a micrometer, mark the spot with chalk, cut it with an axe, and hope it works.

          EDIT NOTE:

          Backwoods had a thought I want to share. I still think you have a Connection Resistance issue and not accounting for it. But you can eliminate Connection Resistance by not measuring through it. It depends on where you take the sample voltage and where you attach the sense leads. Is it on the Supply Side or Load Side of the connection point. If on the Suppply Side, then you are measuring on the Supply Side and seeing Connection Resistance in your measurement and have to account for it. On the Load Side you are not measuring through the Connection Resistance and can ignore it.

          Example say when you terminated the battery lugs onto the 4/0, you stuff say a couple of inches of your stripped 30 AWG sense wire into the Compression Terminal along with the 4/0 skinner and sqoosh them together in the crimper. You would be on the Load Side not incurring the added voltage drop of Connection Resistance.

          Does that turn a light on?

          If it were me. I would use a 100 mv shunt scaled to whatever current you want. Where I attach the two sense leads, use 1/10 amp fuse (safety safety) on each conductor as close as possible to connection point, and run twisted lead for decoupling. Between the resistance of the sense wire and fuses coupled with twisted wire makes enough inductance and capacitance makes an adequate LPF. The method also makes Current Balun Choke that eliminates any Common Mode Noise. If you want Monster Brute Force Filtering still nothing better than ole school. Use Material 31 Ferrite Cores and run the same UTP wire and loop it through a few times. You can operate in High RF Environments without interference. KF5LJW out.
          Last edited by Sunking; 09-26-2018, 07:19 PM.
          MSEE, PE

          Comment


          • #6
            Originally posted by Sunking View Post
            OK more to the point of too many chefs in the kitchen. You are way over thinking this, and using new technology that is just not required. IMO you over engineered it re-inventing the wheel.
            Over-thinking and over-engineering, me??? Never! (Replies the guy who sunk $20K+ and months of happy play/work into an elaborate system that yields $3.50 in electric bill savings on a good day.)

            Originally posted by Sunking View Post
            I still think you have a Connection Resistance issue and not accounting for it. But you can eliminate Connection Resistance by not measuring through it. It depends on where you take the sample voltage and where you attach the sense leads. Is it on the Supply Side or Load Side of the connection point. If on the Suppply Side, then you are measuring on the Supply Side and seeing Connection Resistance in your measurement and have to account for it. On the Load Side you are not measuring through the Connection Resistance and can ignore it.

            Example say when you terminated the battery lugs onto the 4/0, you stuff say a couple of inches of your stripped 30 AWG sense wire into the Compression Terminal along with the 4/0 skinner and sqoosh them together in the crimper. You would be on the Load Side not incurring the added voltage drop of Connection Resistance.
            Yeah, I've been (over)thinking of something along those lines. The cable came with factory crimps that I have no intention of messing with, but I was considering exposing a tiny amount of copper on the surface of the 4/0 wire near the crimp with some careful X-acto knife surgery, then jamming a piece of through-hole component lead or something else stiff into/onto that spot, with the sense wire soldered to the lead. I have some conductive epoxy that would work as well to just attach the wire at that point on the cable.

            Using a factory shunt would make the most sense, I suppose. But this is an overpriced science fair project, after all...

            Comment


            • #7
              Originally posted by BackwoodsEE View Post
              Over-thinking and over-engineering, me??? Never! (Replies the guy who sunk $20K+ and months of happy play/work into an elaborate system that yields $3.50 in electric bill savings on a good day.)
              Been there done that. One thing that never changes over generations, boys and toys. As we get older our toys get more expensive like wives. My thing is racing golf carts, ham radio, and GOLF. Had to give up women, they got too expensive to play with.

              Originally posted by BackwoodsEE View Post
              Yeah, I've been (over)thinking of something along those lines. The cable came with factory crimps that I have no intention of messing with, but I was considering exposing a tiny amount of copper on the surface of the 4/0 wire near the crimp with some careful X-acto knife surgery, then jamming a piece of through-hole component lead or something else stiff into/onto that spot, with the sense wire soldered to the lead. I have some conductive epoxy that would work as well to just attach the wire at that point on the cable.
              I would never ever recommend you do this because it does not comply with codes, wink wink, but if the Compression Terminals are either Lead plated or Tin plated one could very easily just solder a wire to the tongue or barrel. But you would never do that as an engineer knowing there might be some code and safety issues doing so.

              Been a while since I used them, but another idea for the Sense Wire is they make Ring Terminal where the Ring is also either a Dragon Tooth Lock Washer or External/Internal Tooth Lock Washer. You can kill two birds with one rock. You want to use Locking Hardware for the bolt, and a Tap for the sense lead. Never tried it and would likely never use it except on an application that requires no meaningful current is they do make a solder type of conductive glue. I should not have to say this, and I am not doing so to tell you anything, rather DIY's, when working with Raw Unfused Battery, you had better know exactly what you are dealing with. Just one small mistake of less than a blink of an eye, you are surrounded by plasma, fire, acidic smoke, and boiling hot acid spewing everywhere. Not good for the ole hair cut .
              Last edited by Sunking; 09-27-2018, 11:22 AM.
              MSEE, PE

              Comment

              Working...
              X