MANY MANY MANY thanks, I looked for the schematic using a search engine=20 and did not find it!! I am considering buying a power meter that has a current transformer, to=20 make it 10 times as sensitive, it's just a matter of making the coil 10=20 turns instead of one. There are a few similar devices on that auction website...Chinese=20 unbranded, but for half the price of the P4400 Kill a Wart. I'm thinking=20 about taking a chance on one of those. On 11/27/2016 06:28 AM, RussellMc wrote: > Summary: > > Kill-a-Watt P4400 / P4460 circuit diagram > here: > > http://bit.ly/killawattp4400 > > 1. Replacing the current shunt in the Killawatt with 2 x 0.1 Ohm, 1% 1/4 > Watt resistors in parallel should make it 10 x more sensitive and better > usable with low loads. > (My wanderings below suggest that the meter thatv the circuit referenced > below relates to is 10A, 230V rated with a 5 milliOhm shunt.) > > 2. Replacing Rsense with a CT with a suitable burden resistor to give > 2Vrms at max desired rated current should allow it to be used with the > circuit in refence with burden R being changed as desired to get lower > power operation. > > There are numerous low cost CTs available which should allow this. > > _______________________ > > > You should be able to modify your P4400 Killawatt with relative ease [tm] > to allow it to be more useful at low power levels. > > A possibly correct (looks OK at a glance) circuit diagram of the P4400 / > P4600 Kill-a-Watt is here > > > http://2.bp.blogspot.com/-jxGvzdj4ykQ/UU2u8sAZy5I/AAAAAAAAAJI/r0q3BQoRIqU= /s1600/KAW_LV.JPG > > =3D http://bit.ly/killawattp4400 > > U3D appears to be the voltage monitor (gain of ~ 10k/1M =3D 0.01) and U3A= + > U3C the current shunt monitor (gain of 100k/2k55 x 1M/100K =3D 392) > > To provide greter sensivity you can increase current channel gain or > increase the current shunt resistance. > (I assume this works with a shunt rather than a CT but maybe not. ) > > Vdd iis 5V (U5) and Vref for the amplifiers is 2.33V (5V x R12 / (R12+R27= ) > So say 1.5V RMS swing max. > So V_Isnese max =3D 1.5/392 =3D 3.8 mV. > For I max =3D 10A, Rshunt =3D V/I =3D 3.8mV/10A =3D 0.00038 Ohm. > Rather low. > > NO!!! > > I had another look nd saw that the current amplifier's two gain stages > boith feed the procesor. > So gain for current =3D 39.21 OR 392.1 > ie they have a low and high current setting - so have already done what I > was suggesting :-). > > At gain =3D 39.21, for 2V RMS to uP, Vshunt =3D 50 mV. > So 5 milliOhm at 10A or 2.5 milliOhm at 20A. > > Assume 10A max. > On low I setting you get 2 Vrms at uP for 1A load current. > > That SHOULD easily allow half OK accuracy at the 10 mA range or 2. Watts > on 230 VAC. > > Increasing Rshunt to 50 milliOhms (assuming the above is correct) decreas= es > meter Imax to 1A and SHOULD make very low current loads reasonably > mesureable. > > Using eg 2 x 1%, 0.1 Ohm in parallel. > At 1A max or 500 mA/resistor power per R is 0.5^2 x 0.1 =3D 0.025W. > At 10A max that rises to 2.5W/resistor so you may not wish to use a > modified meter at 10A :-). > > > CT: Use of a current transformer in place of the assumed sense resistor > above would allow easy safish changing of sensitivity. > > > Note that most small energy meters use "real" energy meter ICs which are > essentially the same functionally as used in commercial energy meters. > > I have seen DS5460 / DS5461 ICs (& others) in meters sold here. > > ________________ > > *Files on my disk. Will put on dropbox after a reboot:* > > Microchip MCP39F501 > > DS CS5460A from CL site - energy meter IC as per Jaycar MS6115 energy met= er > www.cirrus.com_en_pubs_proDatasheet_CS5460A_F5.pdf > > DS Energy Meter IC Cirrus Logic CS5490 - highly capable - _Q96 - _ERGE > www.cirrus.com_en_pubs_proDatasheet_CS5490_F3.pdf > > DS Energy Meter IC uChip MCP3909 nice to use _q95 > ww1.microchip.com_downloads_en_DeviceDoc_22025C.pdf > > DS Energy Meter like CS5460 in Jaycar MS6115 energy meter CS5461 pdf, > CS5461 description, CS5461 datasheets, CS5461 view ALLDATASHEET .h= tml > DS Energy Meter like CS5460 in Jaycar MS6115 energy meter CS5461 pdf, > CS5461 description, CS5461 datasheets, CS5461 view ALLDATASHEET > _files > DS Energy Power meter IC superb _Q96 Microchip MCP39F501 Digikey pricing > page - onchip rms power calc & v & I - serial async bidi interface > ww1.microchip.com_downloads_en_DeviceDoc_20005256A.pdf > DS Energy Power meter IC superb _Q96 Microchip MCP39F501 onchip rms power > calc & v & I - serial async bidi interface > ww1.microchip.com_downloads_en_DeviceDoc_20005256A.pdf > DS Maxim 78M6610+PSU energy meter IC - 20 page design guidelines excellen= t > _q95 July 2012 pdfserv.maximintegrated.com_en_an_UG5475.pdf > DS Maxim 78M6610+PSU energy meter IC - excellent _q95 > datasheets.maximintegrated.com_en_ds_78M6610+PSU.pdf > DS notes CS5460a PINOUT ENERGy METER.jpg > > > > > > > > On 26 November 2016 at 16:57, Art wrote: > >> I have a Landis & Gyr E350 FOCUS AX-SD smart meter that I'd like to use >> to do some work on power factor correction-namely, how to inexpensively >> improve the power factor for LED light bulbs. They have outrageously >> poor power factors, if the utilities begin billing for reactive power..I >> think there might be a move back to incandescent lighting::> >> >> I got a neat little portable power meter called the Kill-a-Watt (model >> P4400) from my local library, and already found some big power suckers >> in the house! But it's pretty crude, especially for low power things >> like LED lights and it fails miserably at measuring standby power levels= .. >> >> But, I'd like to access the Landis & Gyr smart meter to get more >> detailed information from it and to continue research into power factor >> correction for these new fangled LED bulbs. >> >> I did obtain the manual for the smart meter, despite the manufacturers >> best efforts to deny me a copy::> It has an optical port that can be >> accessed without taking the cover off the meter, which makes it ideal >> for me. But, the manufacturer will not share any of the details on the >> software or hardware needed to extract the data I need from the optical >> port. The optical port is an "ANSI C12.18 type optical port". I did >> figure out that the optical interface/computer cable can be bought for >> reasonable $$. >> >> I also thought about building an interface to a computer sound card, >> which isn't a big deal technically, but I am not sure I could handle the >> software to extract the real time data from the voltage and current >> transformers output. But, sampling at 48K (or higher) would yield some >> nice high resolution V/A plots! >> >> Has anyone here hacked one of these smart meters or have any experience >> with them?? Email me off-list if you do not want to make a public respon= se. >> >> GL. >> >> Art >> >> >> -- >> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive >> View/change your membership options at >> http://mailman.mit.edu/mailman/listinfo/piclist >> --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .