Hello, I am looking for a current sensor with the following specification:
- current te be measured +/- 200 A
- DC current, 12 V
- logger cand feed sensor with 0-5 V or 0-12 V
- logger can accept 0-5 V signal from sensor
- ACCURACY - 0.5 % up to max. 1%
The sensor is going to be used in a racing motorcycle to watch current in and out of the battery. That means that I need a quite light sensor and one that is simple to install, calibrate and change if necesary.
Thank you for your help, Andrei
Dear Andrei:
There is a fusable link between the battery and the rest of the system. Calibrate it for known currents. Measure the voltage drop across it.
How did you intend to use a + only logger to track a +/- current flow? Are you really interested in outbound-from-battery currents? You can probably use an OpAmp powered by the +12vdc, and offset the current to "mid span = 0", but this will be non-trivial.
David A. Smith
I expect there are off the shelf solutions, but if you had an electronics tech on hand, he might find it possible to select a short length of 2 gage welding cable in line with the battery feed, and from each side of the cable (which acts as a low resistance droppper) drive an ordinary op amp - perhaps X2000 gain would transduce a 2 or 3 millivolt drop to a 5 volt output. A capacitor to band limit the amplifier, and you would be in business. Could be built on a miniature perf board....
Brian W
Hmmmm - I put a little more thought into your requirement. As you have seen it's not difficult to get a reading out of some in-situ wiring resistance. But there's a problem - if you want accuracy.
The resistance of copper wire goes up three percent for a ten degrees C rise. So your current readings look higher the warmer it gets. That was a justification of keeping the self-heating of the wire as low as possible, with a beefy conductor section.
If the current heats the wire so much the worse. Metals all have quite a positive temp coefficient of R, or most of them any way. But nichrome wire, like the coil that comes in your wife's hair dryer for example has a ten times smaller temp coefficient than copper. (The carbon/graphite pencil in a battery cell has a smallish negative temp constant but it's harder to make a good connection.... Nichrome wire resistance in the usual wire gage is too high for your purpose. But if you doubled a length of nichrome over a few times - its resistance could be reduced to a useful value, and give you a better shot at an accurate stable reading. You would braze or weld the ends together. To cope with the bidirectional current near one rail, you would want your op amp fed from the middle of a 1K 1Meg pair of resistors each side of the sensor to the other rail,to provide a little headroom for the op amp. You would choose a zero latch up type op amp just in case naturally.
Brian W
Thank you all for your responses. There are 2 problems - the solution has to be accurate 0.5% - 1% and has to be quite simple to fit on the bike. I have found different current sensors from LEM, FW BEll, Honeywell but all aff them have disadvantages. The classic LEM open loop current sensor has a lower accuracy, the MR current sensors are good but are PCB mounted (this is something that I do not know how) and some of them need dual voltage (+/- 12 V), where I am anly able to supply 0-5 V or 0-12 V.
Dear Andrei:
The solutions offered can be arbitrarily accurate, but you will have to calibrate them for a "run". Probably need to measure ambeint temp too.
The accuracy you state is extremely good. Why in heaven's name do you need that?
You don't know how to mount their sensor, or you don't know how to handle the PCB it is mounted on?
Dual voltage means you carry two small battery packs, one for each polarity.
Welcome to the world of "making do".
David A. Smith
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.