Post by thread:[EE]:Maxim 472 / 4377 Current sense amps

Hi list Due to the discussion on the current sense amps a few days ago ( week or so ? ) I have some samples. However my application calls for measuring > 10 amps at 24v which would require a sense resistor of 5 mOhm. How on earth (apart from a pcb track which I don't think I will be able to measure or make) do I get 5 mOhm ? If anybody has worked with these IC's would they be kind enough to help me out with the resistor values for the above parameters ? Thanks Steve ----------------------------------------------------------------------- Information in this email may be privileged, confidential and is intended exclusively for the addressee. The views expressed may not be official policy, but the personal views of the originator. If you have received it in error, please notify the sender by return e-mail and delete it from your system. You should not reproduce, distribute, store, retransmit, use or disclose its contents to anyone. Please note we reserve the right to monitor all e-mail communication through our internal and external networks. ----------------------------------------------------------------------- -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

>However my application calls for measuring > 10 amps at 24v which would require a sense resistor of 5 mOhm. How >on earth (apart from a pcb track which I don't think >I will be able to measure or make) do I get 5 mOhm ? This sounds like you are into territory where you start with a piece of copper bar or tube as the sense resistor, and solder (or more likely bolt) a couple on wires on it at appropriate tap points. If you do take this route, then you may need to think about the temp co-eff of the material used for the bar. If using solid bar, then you can probably make it in a manner that minimises heat rise due to current flow. Personally I would start with a piece of copper 1/2" water pipe, work out from theoretical resistance of copper how far apart the tap points should be, and solder a couple of wires on and give it a try. Doing things this way also has the advantage that it is almost impossible to burn out the current shunt :) -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body

I've used the MAX 472 to measure 100 amps, using a 0.5 milliohm sense resistor. These are commercially available products commonly called "shunts". Take a look at the Blue Seas Systems web site: http://www.bluesea.com/Products/ME/ME.html I'm sure there are a lot of other suppliers as well. Shunts are typically rated as "amps for 50 mv", since they were originally designed for use with analog meters, which typically required 50 mv for full scale deflection. My shunt is 100 amps for 50 mv (thus the 0.5 milliohms). With the MAX 472, you can set the gain with the resistors to basically any shunt value. When I first started playing with measuring current, I didn't have a proper shunt, so I made one out of a few feet of 8 or 10 gauge wire. Using copper wire tables, I figured out about how long a piece I needed, and cut the wire a bit longer. Then I put a measured 10 amps through the wire. Using a digital voltmeter, I clipped on lead to one end of the "shunt" and clipped a needle to the other end of a probe, and pushed it through the insulation on the shunt in various places until the DVM read the proper voltage (in my case I wanted 100 millivolts for 10 amps). Then I stripped a bit of insulation off the shunt at that point and soldered a wire to it to use as a sense wire. The problem with using copper wire (or a copper trace) is that copper's resistance changes with temperature - a shunt is built to have a close-to-zero temperature coefficient. But for most hobby purposes, it isn't a problem as long as the wire gauge used is large enough that the current flowing through it does not cause it to heat up. Larry At 12:10 PM 11/21/2002 +0000, you wrote: {Quote hidden}>Hi list > >Due to the discussion on the current sense amps a few days ago ( week or so >? ) >I have some samples. > >However my application calls for measuring > 10 amps at 24v which would >require a >sense resistor of 5 mOhm. How on earth (apart from a pcb track which I don't >think >I will be able to measure or make) do I get 5 mOhm ? > >If anybody has worked with these IC's would they be kind enough to help me >out >with the resistor values for the above parameters ? > >Thanks > >Steve Larry Bradley Orleans (Ottawa), Ontario, CANADA -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

> Personally I would start with a piece of copper 1/2" water pipe, work out > from theoretical resistance of copper how far apart the tap points should > be, and solder a couple of wires on and give it a try. A piece of #12 wire will be easier to deal with, and the much smaller outer diameter will make it easier to compute the expected resistance. Computing the resistance of a hollow tube between two point connections to the outside of the tube is not a trivial matter. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam.....mitvma.mit.edu with SET PICList DIGEST in the body

{Quote hidden}> -----Original Message----- > From: pic microcontroller discussion list > [EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU]On Behalf Of Goring, Steve > Sent: Thursday, 21 November 2002 8:10 PM > To: PICLISTspam_OUTMITVMA.MIT.EDU > Subject: [EE]:Maxim 472 / 4377 Current sense amps > > > Hi list > > Due to the discussion on the current sense amps a few days ago > week or so > ? ) > I have some samples. > > However my application calls for measuring > 10 amps at 24v which would > require a > sense resistor of 5 mOhm. How on earth (apart from a pcb track > which I don't > think > I will be able to measure or make) do I get 5 mOhm ? > > If anybody has worked with these IC's would they be kind enough to help me > out > with the resistor values for the above parameters ? > > Thanks > > Steve I usually calculate it with a length of cable, and place the two sense points say 6" or so apart. Any good spec sheet for cable will give you the resistance per 1000', so you could divide this down by 1000 to get the resistance per foot. Regards, Kat. --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.419 / Virus Database: 235 - Release Date: 13/11/2002 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

Shunts are available on EBAY occasionally. If you need to make a measurement and cannot wait to get the proper shunt, a reasonable shunt can be patched together with a chunk of wire that can handle the load and a sensitive meter. A meter that will read millivolts is best. The innovative part is how to calibrate the reading. I generally: 1- measure the voltage drop on an existing cable already in the circuit. 2- Add a known load to the measured load & measure again 3- do the arithmetic Don't forget to measure and consider the voltage each time. Hopefully, you will have the ability to measure the load to be added directly. If you find the measurement varying with time you probably are heating the cable you are using as a shunt. Shunts are every where, just look around. A roll of house wiring, a piece of pipe, an auto headlamp(lamps can be tricky), an electric heater, even an aluminum cooking pot! John Ferrell 6241 Phillippi Rd Julian NC 27283 Phone: (336)685-9606 Dixie Competition Products NSRCA 479 AMA 4190 W8CCW "My Competition is Not My Enemy" {Original Message removed}

Olin Lathrop <KILLspamolin_piclistKILLspamEMBEDINC.COM> wrote: > > Personally I would start with a piece of copper 1/2" water pipe, work out > > from theoretical resistance of copper how far apart the tap points should > > be, and solder a couple of wires on and give it a try. > > A piece of #12 wire will be easier to deal with, and the much smaller > outer diameter will make it easier to compute the expected resistance. This is true. > Computing the resistance of a hollow tube between two point connections to > the outside of the tube is not a trivial matter. No, but that shouldn't even be an issue. Assuming he's building a proper 4-wire shunt, the potential around any circumference along the tube should be the same. The load current should be applied at the centers of caps attached to the ends of the tube. The sense wires can be tacked on at points on the outside of the tube. BTW, you can buy SMT milliohm shunts commercially, and they're not all that expensive. You do need to be careful in your layout so that the sense traces come directly off the shunt's pads and not from elsewhere along the load current path. Also, I have reverse-engineered some APC UPS units, and they just use a PCB trace as the shunt. Of course, they aren't going for incredible accuracy; they're more interested in just detecting overloads. -- Dave Tweed -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

> However my application calls for measuring > 10 amps at > 24v which would require a sense resistor of 5 mOhm. How > on earth (apart from a pcb track which I don't think > I will be able to measure or make) do I get 5 mOhm ? My impression has been that at higher currents, you start looking at current transformers or linear hall effect sensors... Still, 10A doesn't seem like it would be THAT high; maybe you're just looking at a sense amp designed for lower currents, and need to find a different part... BillW -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

You can buy a four-point shunt for 10A with 5mohms. You can also do this with wire or PCB: 10A>========*=======*=======>10A |tap1 |tap2 |fixed |move till calibrated v v to circuit With a PCB you would have the trace to tap 2 not connected and fit a jumper on prototypes. Later you will know how to patch the layout to yield the correct shunt (it will still need calibration for accuracy). Last, note that copper is a VERY poor choice for shunt material (high temperature coeff.). Use nichrome or constantan instead. Peter -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email TakeThisOuTlistservEraseMEspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

On Thu, 21 Nov 2002, Olin Lathrop wrote: *>A piece of #12 wire will be easier to deal with, and the much smaller *>outer diameter will make it easier to compute the expected resistance. *>Computing the resistance of a hollow tube between two point connections to *>the outside of the tube is not a trivial matter. It is trivial for a shunt because the tube with uniform axial current flow forms equipotential rings on it and the load (meter) impedance is very high wrt the shunt. You only want the resistance by length of the tube to be able to take a long enough piece. That can be computed from ohms/sq and the weight of the tube per unit length (easier than fumbling with wall thickness etc). All you need is a kitchen scale and a resistivity table. Me, I'd take a 1kW heater band from some appliance (toaster ?) and fold it over N times. That's the proper material for shunts too (manganin or constantan). Connection is best made by crimping. By the time you need to reproduce this there will likely be other options. Peter -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservEraseME.....mitvma.mit.edu with SET PICList DIGEST in the body

>Boy, are you wrong ;-). Need to do more work >involving lead acid batteries imho. Oh I'm quite aware of how beefy Lead Acid batteries are, and always take great care around them because of this. However I don't recall the original poster talking of an automotive app, but may have missed it. OTOH it may just be that I'm too used to dealing with 29V space craft power supplies being imitated by 30A switchmodes which shut down when anything goes awol. But what I was getting at was the possibility of having a shunt that did go o/c if something on the output side went wrong, would the chip stand a full 24V across the terminals that are supposed to have only millivolts across it? I have not checked the data sheet, but ...... -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservmitvma.mit.edu with SET PICList DIGEST in the body

At 12:10 PM 11/21/02 +0000, you wrote: >However my application calls for measuring > 10 amps at 24v which would >require a >sense resistor of 5 mOhm. How on earth (apart from a pcb track which I don't >think >I will be able to measure or make) do I get 5 mOhm ? You could buy two 10m Ohm parts and put them in parallel. Digikey has them for (1W 0815 SMT parts from SSM) for about $2 each one-off; 1% tolerance. If it's a volume application, you can get hairpin type open wire parts for much, much less, but this could get you started. Obviously layout counts at these levels. Use Kelvin connections. If you insist on making home-made resistors, at least try to get ahold of Constantan or Manganin wire which doesn't suffer from the rather large (about +4000ppm/K) tempco of copper (and most other elemental metals). Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" RemoveMEspeffEraseMEEraseMEinterlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservspam_OUTKILLspammitvma.mit.edu with SET PICList DIGEST in the body