intersting data?

jpf321

For your viewing pleasure, I attach below some interesting data of my oil consumption (GAL), runtime in hrs. (HR) and Heating Degree Days (HDD) The pink and blue vertical shaded areas denote the differences in the Thermostat Settings .. my HDD station is KLGA,



<img src="http://content.screencast.com/users/JPF321/folders/Jing/media/523ccb2a-712c-4f00-b92c-d8bc07747ae9/2010-02-04_2208.png" />



<img src="http://content.screencast.com/users/JPF321/folders/Jing/media/7fcff9a9-a959-4de0-bb24-4fe5676f9542/2010-02-04_2202.png" />

haaljo

Yes, it is interesting

Your data shows ~25% savings between constant 68 and 71F? No savings from 68F with a "bump to 71 " compared to constant 71F?  I think you have suggested night setback may not be advantageous in previous posts.

I have a tough time with the graph but your spreadsheet is clear. Maybe you could put in a column calculating differences. If you continue to monitor, the confidence of the sample will increase.

You may come to a hypothesis that an old steam boiler with its large volume of water, should not  have setback; and that the sweetspot is 68 setpoint (although that may be more to do with your home's envelope).

Thanks for the info.

jpf321

updated data ...

I just got another oil drop yesterday .. my T-stat setting has not changed .. nothing else in system has changed ..

lutorm

Interesting, but...

The sampling is so rare that it's hard to say what's going on. The spline smoothing of the curves also makes things appear that aren't really there.



But my question is: why is the gal/hr not constant? I assume gallons and hours are since the last fill? If so, and if the burner only has one setting, then gallons of oil should exactly track runtime, right? I guess it could indicate that the tank does not get filled up to the same level every time, in which case it will cast a doubt on the gal/hdd too. The bump in gal/hdd is perfectly mimicked by the gal/hr bump, so that seems to indicate that it's just measurement error.



I'm considering making my own system, it'll be a home-made thing of Dallas DS18B20 probes and an Arduino board. Might not get to it this season, though.

jpf321

i justified a non-flat GAL/HR as

if i understand correctly, long burns are more efficient than many short ones .. and although I do have the actual burn data (from my dataloggers) i haven't looked at it in a while .. but I can see why GAL/HR is not necessarily a constant based on number of cycles and cycle length... for example .. when I was running my 68/71 t-stat and i cycled quite a bit on pressure it caused many short cycles and i think each time the burner starts there is waste or at least great inefficiency. maybe a more experienced burner pro can comment on these thoughts.



a note about my methods .. I always stick my tank with measuring stick just before delivery and then again just after, same with the hour meter .. so delivery levels are always measured not simply "did they fill the same amount" ... also i get my HDD data from http://www.degreedays.net/



thanks for the comments..

lutorm

Maybe...

I think you're right about the inefficiency, but I thought that was in the heat gotten out of a certain amount of fuel. Unless there is something more than a simple on/off switch for the fuel supply, I don't see how that can affect the gal/h.

jpf321

maybe gas is different ...

but with oil .. i usually smell it in the first few seconds and then the smell goes away .. this must mean that there is a difference in the first few seconds vs all the other burn time .. which would also mean that many cycles which contain "the first few seconds" would be a factor in overall consumption. i think there may be a few seconds before the pump is fully up to speed or something .. hopefully someone can weigh in.

Steamhead

You're right

I'll bet your burner doesn't have valve-on delay, which holds back the oil until the air flow, pump pressure and spark are established. The valve also closes instantly on shutdown rather than waiting for the motor to coast to a stop. I think all new burners have this feature, and it can be added to older ones.



Startup and shutdown are the dirtiest periods of burner operation. Valve-on delay helps prevent this. In some cases, the valve also closes before the motor stops- this is called motor-off delay, and is used on some boilers with tight flue passages.

jpf321

so steamhead ...

please confirm that it is likely that my number and length of cycles is affecting the overall GPH consumption data? assuming that my burner is an "old" type that doesn't have all the modern bells and whistles. 

Mike Kusiak_2

Elapsed time meter error?

What is the minimum time increment that your elapsed time meter can register? Does it register time accurately when you are rapidly pressure cycling?



I have a feeling that the meter is not registering the short time periods of firing when pressure cycling. If this is so, you are actually consuming oil during those periods, but not registering the corresponding elapsed time, causing the apparent Gal/Hr to increase.



How is the time meter connected? If it is connected to the burner motor circuit, then it should indicate the oil flow accurately, provided there is no valve on delay, as Steamhead mentioned. The only other source of error would be the time meter and how accurately it measures short time intervals.

jpf321

hour meter ...

the smallest increment is 1/10 of an hr. (6mins) .. the meter is connected into the burner wiring .. which wires I'm not exactly clear on .. as soon as the burner starts the meter flickers and shows me a reading by the time i have a flame. (within 3-5secs) .. so i really don't think there is much error being introduced by the meter ... 

Mike Kusiak_2

Burn time less than minimum increment

OK, so say your burn cycle time is 3 min when pressure cycling and the minimum meter increment is 6 min. Since the time is less than the increment, you will never register the elapsed 3 min, yet the oil  is still being consumed. Over a period of rapid cycling, you may not record a considerable number of unregistered 3 min periods, which may ultimately accumulate a considerable loss of elapsed time.

TomM

burn time less than minimum increment

Just because the minimum increment is six minutes, that doesn't necessarily mean that it is not measuring cycles less than six minutes.

.

my watch only reads down to seconds, but it is still measuring milliseconds no matter how you look at it. 

.

also, if there is a problem measuring oil loss during short cycles, then it would show that a setback is even more inefficient (or less efficient) than the data suggests, since there are more short cycles when recovering from a large setback.  Correct?

.

TomM

jpf321

no...that's not true...

it only displays in 6min increments .. but it captures and saves fractions thereof. if i turn it on and off for 2secs 180 times, it will register the 6mins eventually.



jpf