Behmor Vs Kill-A-Watt

[187]

On my last roast the Kill A Watt showed a draw olf 1795 Watts with elements,light, and afterburner going. Roast times were OK and everything seemed normal. V was 128 when I started and dropped 4V when running.Is this the way this meter fails or do I have some other problem?

[Ascholten]

That might be about right there.  The kill a watt is rated at about 1800 Watts MAX I believe so you are really pushing it to begin with.   Also, the plug in the kill a watt itself which you plugged the behmore into, that can get loose over time and the large current draw can cause the voltage drop you seen.   I had one I was using on my still to monitor heat input and I ended up melting the front a bit because of a loose connection in the plug head itself after hours of use.

This could also be a problem with your house wiring or the plug you put the kill a watt into, or any of other numerous things that can happen.  Check to make sure things are plugged in tightly and a big give away of an impending problem would be heat where it really does not belong, ie at the plug heads etc.

Aaron

[ButtWhiskers]

I wonder what the effect of heat is on the Kill A Watt itself?

[Ascholten]

BW as I said earlier, I melted the plastic around the plug in a bit on my Kill A Watt, and kept it running at high wattage, probably in the 1500's for several hours.  It discolored but otherwise the meter kept running just fine, even now a good year later.  Electronics of that nature, when they die, generally don't just get 'innacurate' in the way he described, most the time they just quit working or go totally haywire.  While of course I do not recommend it being treated this way if you can avoid it, the Kwatt otherwise if it still turns on, probably is working correctly.  Luckily they are fairly cheap, and if it does end up finally letting the magic smoke out of the wires, they can be had for about twenty bucks for a new one.

Aaron

[ButtWhiskers]

Sorry for my lack of clarity...  I meant, I wonder how heat effects it's reading.  If the device is at room temperature will the readout be significantly different than if the device is at 180?F?

[Ascholten]

Whiskers,  given it's a digital device it *shouldn't* matter significantly at all what the ambient temperature of the killerwatt is.  The only thing to worry about is excessive heat killing the watt  but as far as accuracy, it should not really change it that much at all.

TOmorrow if I remember Ill bring my kilawatt to work and throw it on one of our nist traceable standards and track voltage in to it compared to what it reads and run it from room temp to about 150 degrees or until the display blacks out and see if there is a huge deviation...  again, i dont think there will be, the digital stuff don't have all those resistors and capacitors to change values a ton as they heat like the old simpsons or other midevial equipment did.

I will run various voltages through the thing too (within reason, i dont want to fry my meter, however could go from a few thousandths of a volt to 1.4 million volts) to see if there is a spot where it has problems, without the probs being the voltages getting way out of op range.

Aaron

[Ascholten]

Ok folks, here's what I came up with.

I ran the Kill a watt (referred to at this point as the kw) through a few tests to see how it would perform and survive.

I held it at 125 volts and put a 1000 watt resistive load at first on it.  I used a dedicated calibrated power supply and as the wattage started climbing, adjusted the voltage accordingly to put 125 volts out TO the kw meter to see how it read it.   Overall though the voltage drop was maybe a volt from 0 to 1000 watts,

the meter tracked pretty linear and accurately, being off by maybe a few tenths of a volt here and there.  Nothing significant given the specs on the meter.

I then ran the watts down to 500 watts and ran the power factor from -.4 to +.4 to see how it tracked under inductive and capacitive loads.  (Note: I didnt run it at the full 1000 watts because I was too damned lazy to go look for the cap bank and the other transformer to hook through to control and bring the power factor down that far.   Again it tracked voltage and power pretty close.

We have a small heating oven which we use for bearings sometimes, and next step (without getting into all kinds of geekly how I set it up stuff) I placed the kw into the oven to start heating it a bit.  I set it to 1000 watts of load through a resistive load which would be a unity  1.0 Power Factor and let it run.   I ran it up in 5 degree increments holding for 5 minutes at each step to 180 degrees F.   At about 165 degrees the meter started goofing and acting strange so I stopped the test, however up until the point where it started blacking and scrambling it was tracking pretty much dead on. 

Conclusion, the meter will track accurately through temperature changes and load changes and the only way heat seems to affect it is when it gets wayyyy too hot,  much hotter than you really should be running the thing, it will start freezing, scrambling or blacking out on the display.

Nother note, at the higher temps I used an IR thermometer and found that around the plug was getting MUCH hotter than the rest of the meter,  perhaps the heat was causing the metal to flex a bit and a looser connection at the plug in, hence the extra heat..... also not checked BUT if that did happen, the heat could have easily radiated up the metal prongs of the plug, into the meter workings and caused much hotter temps than the 165 ambient I was subjecting it to and giving it problems there too.

Again though the thing worked as it should until I just ... stopped.  In other words there was no 'drift' in the readings or gross inaccuracies as it went on.

For what this was worth, it was fun to do at least and I hope this experiment helps somehow.

Aaron

[Ascholten]

Let me add one more thing,  IF you overload the meter, it will start beeping and the display will start flashing.  I took mine up to 2000 watts... well 1987 watts to be honest and held it for about a minute....

Not sure how far it will allow the overload and wouldn't recommend testing it lest you want a nice meltdown.

Aaron