Welcome back, it has been a little over a year since I started this crazy project. Over the last year I've found three occasions to actually NEED the generator. All three were local utility outages, where the entire house was ran off the generator. All in all, I've put about 20hrs on the generator and I'm Very happy with it!

         Of course, I have been a little negligent with this project. Once the generator was able to reliably make electricity, I pretty much moved onto other projects instead of finishing up all the things I wanted to do. Hey, hurricane season had passed, and I found other things in need of my time :) {shrug} In any event, the '06 hurricane season has made itself known, so I've found new motivation to get working on the generator!

First order of business, fuel consumption testing (click for larger images)

Fuel Dispenser Steam Powered? Dsc04397.jpg

The first image above on the left is my homemade fuel dispenser and timing system. I started with a clean paint can, and brazed a 3/8 hose barb to the bottom. Some 3/8" clear vinyl tubing connected my new "fuel tank" to the motor. The method used for timing was pretty simple.

  #1 I added enough fuel to give me enough time to get the generator warmed up and running on the load I was measuring. (all loads were resitance heaters, in my case a 6 element pottery kiln, and when needed, my powder coat oven.)

  #2 Once I was ready, and every thing is in a steady state, I grabbed my stop watch and waited (hoping I didn't add to much fuel to the paint can)

  #3 In the picture top left, the paint can is empty, and you can see the fuel level lowering in the 3/8" inside diameter tube. When the fuel level gets to the writing on the side of the tubing, I started my stop watch

  #4 Now that the stop watch is running, I can leisurly refill the paint can with one quart of fuel, and commence to baby sitting.

  #5 While babysitting, every five minutes I would log the voltage, amperage and frequency. I used these values to create an average load for each test.

  #6 Finally, after a period of time, the paint can would run empty. As it did so, I could watch the fuel level start decreasing in the fuel line. When the fuel level reached the writing on the tube, I stopped my stop watch and recorded the time it took to consume the quart of fuel

  #7 I repeated this test for each of 5 load levels (all consuming 1 quart of diesel @ 3.02us / gallon)
    a) No Load = 62 minutes 12 seconds = 0.24 gallons ($0.72) per hour = 5.76 gallons ($17.40) per day!
    b) 3Kw = 37 minutes 22 seconds = 0.4 gallons ($1.21) per hour = 9.6 gallons ($28.99) per day!
    c) 6Kw = 24 minutes 43 seconds = 0.61 gallons ($1.84) per hour = 14.64 gallons ($44.21) per day!
    d) 9.2Kw = 17 minutes 48 seconds = 0.84 gallons ($2.54) per hour = 20.16 gallons ($60.88) per day!
    e) 11.2Kw = 13 minutes 2 seconds = 1.15 gallons ($3.47) per hour = 27.6 gallons ($83.35) per day!

*** Not so big grin, but during this test I found out my generator is only good for 11.2Kw of electrical power output. I was a little surprised at first, but one thing is for certain, MATH still works. In this case, my diesel is rated for 14.7Kw at 2200rpm. Since I'm direct drive, I run the generator at 1800rpm, or 82% rated output rpm. 82% of 14.7Kw happens to be 12.1Kw *at the Flywheel*!! Assuming no one lied, the ST generator head must convert 93% of that energy to produce 11.2Kw of usefull electricity.

So, the question is, "Is it better to have a generator head rated at more power than the engine can provide, or vice versa?"

Now, lets make some comparisons based advertised fuel consuptions on other generators. (click for larger images)

#1 Dollars per killowatt hour #2 Gallons per killowatt hour

#3 Liters per killowatt hour #4 BTUs per killowatt hour

         OK, now this is getting interesting! I found as much published data on as many generators as I could. After crunching all the numbers and figuring out how to plot the data (*1) the above is the combined result.

  #1 Assuming diesel = $3.02/gallon and gasoline = $2.88 per gallon, this is how it works out. Of course, this is a worthless graph is the fuel cost changes....

  #2 & 3 These two graphs break out the liquid volume of fuel consumed per kilowatt hour. I find this handy as a consumer, because I purchase fuel by the gallon or liter.

  #4 Finally, there are some internet nit pickers that will complain that a straight up comparison between diesel and gasoline motors isn't "fair" as diesel fuel has a higher energy content than gasoline. To satisfy this group of people (and my own curiosity), I created this graph which displays the comparison based on BTUs per kilowatt hour. BTW, the first version of this graph was in grams of fuel per kilowatt hour, and other than the y axis labels, it looked identical (the difference between g/kwhr and btu/kwhr is about 0.5% in this case.)

         To use these graphs, figure out the kilowats per hour you going to use as an example. Find that value on the X axis, find the models of generator you want to compare, and multiply the Y value by the number of kilowats. For instance:

  Suppose your running a 1.5Kw load for 1 hour and you want to know what it will cost.
     The Honda EU2000i consumes $0.5 per Kwhr at a load of 1.5Kw. So $0.5 * 1.5 = $0.75 per hour.
     The 3.5Kw Troy-Bilt consumes $0.54 per Kwhr at a load of 1.5Kw. So $0.54 * 1.5 = $0.81 per hour.

  Suppose your running a 9Kw load for 1 hour and you want to know the cost.
    My ChangFa and The Kohler Marine consume $0.28 per Kwhr at 9Kw. So $0.28 * 9 = $2.52 per hour.
    The VoltMaster 1 cylinder diesel consumes $0.37 per Kwhr at 9Kw. So $0.37 * 9 = $3.33 per hour.
    The Honda EB11000 consumes $0.48 per Kwhr at 9kw. So $0.48 * 9 = $4.32 per hour.
    The Centurion 15Kw consumes $0.58 per Kwhr at 9Kw. So $0.58 * 9 = $5.22 per hour.

*1 (if anyone can show me how to do this in Excel, I'd appreciate it. I wrote my own program to plot this data (aka what a PITA))

Moving on, lets look at the electrical output. (click for larger images)

#1 The 15hz strobe effect #2 Output noise #3 rotor voltage.

         What an exciting time it was when I first ran my entire house of my home made creation! The fist thing I observed was that yes indeed, the generator could run my whole house. The second thing I noticed was that the compact fluorescent lights did a 15hz strobe like effect when the generator was under a heavy load (HVAC + Hot Water heater).

         While doing the recent resistive load tests, I decided to observe the A/C output with my O-Scope. Image #1 shows the 15Hz strobe effect. (BTW, both axis on the O-Scope are in the variable range mode to get the effect to display, so no trying to measure anything from this image!) The reason for the 15hz strobe is that as the frequency (rpm) of the generator drops, so does the voltage. Unfortunately, the best I can do from the diesel is regulate each power stroke. If everything is working correctly, utility voltage is supposed to be 60 cycles per second. To produce 60hz I need to run the generator at 30 cycles per second. Since the motor is a 4 stroke motor, I only get 1 power stroke every 2 cycles, or 15 power strokes per second. So each power stroke has to carry the load thru 4 AC cycles. Image #1 shows the end result, so much so, you can see the voltage decrease after the power stroke, and especially so during the compression stroke. cool :)

         When I first connected the O-Scope to the line voltage, I was rather surprised by the "not so smooth" appearance. Fortunately my SmartUPS 1400 doesn't complain, so it can't be that horrible, but it sure looks ugly in picture #2! At first I thought it might be a problem with the slip ring brushes on the rotor. After inspecting them they seemed ok. Next I thought it might be the rectifier bridge,but no it tested OK as well.

         The third picture is what I saw when I measured across the rotor slip rings. WOW that's weird.. I have no idea why that is what it is. Using my DVOM, I get 63vdc measured, but that doesn't look like DC to me, as according to my O-scope, those are 80+ volt peak to peak measurements...

         I think I'm going to re-do those measurements with more care this time. In addition I'll do them during different load environments to see what I see. One other thought is perhaps I can do something to the voltage regulator (like add a capacitor?) and find some way to improve it. I should also try and measure the current flow thru the rotor and perhaps it's inductance to see if anything of interest shows up.

See you next time! James