Post by thread:Ref AC Current reading

I'm concerned about the safety aspects of the comments so far on this thread. Even though it's only 120 volts I guess it still bites and means your PIC is live. The real way to do this is with a current transformer which provides safety isolation and is easily made from a miniature audio transformer salvaged from and old fashioned transistor radio (Germanium transistors usually). These are nicer to work with as they are usually only wax impregnated and come apart easily. Pass a piece of insulated wire through the core and put this in series with the load. Experiment by connecting the existing windings directly to the + and - inputs of an OP-amp. bias the + input with equal resistors say 10K ohms across the supply to create a mid rail and decouple to 0v with a 10 microfarad cap. Fit a low value resistor between the op and - input say 100 ohms to start with. The output should give a voltage proportional to instantaneous current in the load. Your PIC can then find the peak or work out the RMS values as required. Current transformers are easy if you stick to the rules. The secondary needs to work into a virtual earth (hence the op amp) The core must not saturate - unlikely with VE as there is little power transfer Primary current * no of primary turns = Secondary current* no of secondary turns NB the voltage from the op amp is simply Secondary current * feedback resistor. Safe Experimenting Bob Minchin Romsey UK bob.minchin.roke.co.uk

>I'm concerned about the safety aspects of the comments so far on this thread. >Even though it's only 120 volts I guess it still bites and means your PIC is >live. . . . > >Safe Experimenting > >Bob Minchin I have to agree with what Bob is saying. I'm not comfortable with projects that sense Line Voltages directly. What I have wanted to try is a few loops of the line cord around something like a nail and attatched to a Hall-Effect Sensor. Has anybody tried this? Myke Myke "We're Starfleet officers, weird is part of the job." Capt. Catherine Janeway

Bob Minchin wrote, >The real way to do this is with a current transformer which provides >safety isolation and is easily made from a miniature audio >transformer salvaged from and old fashioned transistor Amen, Bob, this is the way to go. I would go further to say that I hope the line current is that of a somewhat pure reactance, that is, sinusoidal in nature. If you are monitoring the current into an off-line rectified switching supply, say, you will have huge current spikes only at the peaks of the AC voltage. Standard formulas for RMS values which would apply to a sinusoidal current, don't work for this case. Perhaps some code which detects the frequency, thus the period, and roughly integrates the current waveform would help. Mike

Bob Minchin said: >Pass a piece of insulated wire through the core and put this in series with the >load. Experiment by connecting the existing windings directly to the + and - >inputs of an OP-amp. bias the + input with equal resistors say 10K ohms across >the supply to create a mid rail and decouple to 0v with a 10 microfarad cap. Bob, This is a very interesting idea. Are you describing the use of an audio OUTPUT transformer? By "existing windings" do you mean the high impedance primary, and you are ignoring the existing low-impedance secondary? Also this appears to assume that you already have a power supply that is floating with respect to the power line, which is of course logical but may be worth making explicit. Fit {Quote hidden}>a low value resistor between the op and - input say 100 ohms to start with. >The output should give a voltage proportional to instantaneous current in the >load. Your PIC can then find the peak or work out the RMS values as required. > > >Current transformers are easy if you stick to the rules. > >The secondary needs to work into a virtual earth (hence the op amp) > >The core must not saturate - unlikely with VE as there is little power transfer > >Primary current * no of primary turns = Secondary current* no of secondary >turns > >NB the voltage from the op amp is simply Secondary current * feedback resistor. > >Safe Experimenting > >Bob Minchin >Romsey UK > >bob.minchin.roke.co.uk Reg .....................Reg Neale..................... "Ignorance is a renewable resource" P.J. O'Rourke

> >I'm concerned about the safety aspects of the comments so far on this thread. Wise to be concerned. > I have to agree with what Bob is saying. I'm not comfortable with projects > that sense Line Voltages directly. What I have wanted to try is a few loops > of the line cord around something like a nail and attatched to a Hall-Effect > Sensor. You must remember that you have to make use of the unbalanced current of only one of the mains conductors, the mag field created by winding both comductors around/through anything is zero. A easy way to get a neat pickup is to take out one of the torroidal filter inductors from a powersupply, use the windings as is, series them if there are two or make your own and leave enough room for one pass of a good quality mains conductor. If using a Current Transformer (CT) ALWAYS remember to place a Burden (load) across the secobdary winding to prevent the open circuit voltage from reaching close to the theoretical infinity. Make sure that it has a high enough power rating for worst case loads. Turns ratios of 1:5 and 1:30 are possible so a 1 Amp load will give 33 mA and with a 33 Ohm resistor will give 1.1 V (AC RMS = 1.55 V peak) and give you a bit of over range capacity of about 3A to get to 5V peak. If you want to do real power metering for consumers then a dedicated metering IC is a better option. A local IC manufacturer makes a range of real neat metering ICs with async serial and pulse outputs that give you Wh with Class C (I think it was, better than 1% or something) accuracy over the whole range. They are used by the 100 thousand in prepayment electricity meters here and abroad. If I get email requests I will post some details. Requests to the list will just irritate me and waste bandwidth. Cheers -- Kalle Pihlajasaari spam_OUTkalleTakeThisOuTdata.co.za Interface Products Box 15775, Doornfontein, 2028, South Africa +27 (11) 402-7750 Fax: +27 (11) 402-7751

didn't quite follow the verbal schematic.. more verbage or a drawing, please. Thanks. ---------- From: .....Bob.MinchinKILLspam@spam@SAGE.ROKE.CO.UK[SMTP:Bob.MinchinKILLspamSAGE.ROKE.CO.UK] Sent: Wednesday, April 10, 1996 1:59 AM To: Multiple recipients of list PICLIST Subject: Ref AC Current reading I'm concerned about the safety aspects of the comments so far on this thread. Even though it's only 120 volts I guess it still bites and means your PIC is live. The real way to do this is with a current transformer which provides safety isolation and is easily made from a miniature audio transformer salvaged from and old fashioned transistor radio (Germanium transistors usually). These are nicer to work with as they are usually only wax impregnated and come apart easily. Pass a piece of insulated wire through the core and put this in series with the load. Experiment by connecting the existing windings directly to the + and - inputs of an OP-amp. bias the + input with equal resistors say 10K ohms across the supply to create a mid rail and decouple to 0v with a 10 microfarad cap. Fit a low value resistor between the op and - input say 100 ohms to start with. The output should give a voltage proportional to instantaneous current in the load. Your PIC can then find the peak or work out the RMS values as required. Current transformers are easy if you stick to the rules. The secondary needs to work into a virtual earth (hence the op amp) The core must not saturate - unlikely with VE as there is little power transfer Primary current * no of primary turns = Secondary current* no of secondary turns NB the voltage from the op amp is simply Secondary current * feedback resistor. Safe Experimenting Bob Minchin Romsey UK bob.minchin.roke.co.uk M&^ @8AN!!" `<'="H > !B)Y"&$XLD?0<W\*A2& '' @-A_0-W(<L7?W,6 < M($5@9$5@1?-&$0/PZQPP'I%C/@5T'M `@ (0]G('@ 7 =RN0$7 ?0 -@_G8; M``>10V0*A00`!O =,?]%P4J21A$?<0,00[ `P VPMQNP`V$>D6T+@ <P= AP M>1N@874:\$XA3\E#8&S^=AXQ0J #4@J%2I(&\%:A_Q^ *Y "($*13[,X@06Q M'M"953(H1P20`X%I=4(`8UB8!"!U<W4'0$A0*?\?L@>05/$6$ J%`P!"44W2 MVG<%L&M.M48B94.Q'<%76Q5()$V@>$?0;1-09?QG;ATP0I)*H43!5/$*L>YT M"H531" ]4!^ 2G$?0/\(D$)0.#(+@%L@4F%"D0/POQW!1D%'DT2$5.)*H7!" M\W].8@.@$; (@4ZE'#%+YF_[1H%'$'A#\ 401/)*($.PGT(A,2 D$!\!'"-E M>#B!OVHB'O$^H@0@&O WD71(4?M-X1PR*TJ#)-8+@&91."/A`Y%/4"UA7\ = M@$HP[UWD;+%MHTZT91M@'+$6$"=:ED-@2&$P2QY@:&T[2G$%`&\$$ J%0R-U M</\+4$W#!0`?<!N15%)"H![0/P,12I(%@G- -V%-X3!V_TZV&3!487(Q18$^ MD$(080#]'8!&&X!7!FAA!^!68 I!-W"G2B 1P'<)X1PC;W#_;-1OE7%42*!Q MTTWA$\!A(;].LTR:0N%[(T>!5\%G3&'O'I%(PAXR(\%P%