Does hump affect average fuel consumption significantly?

Discussion in 'Powerboats' started by Nidza, Oct 5, 2017.

  1. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    At first, I will explain how I measure the fuel consumption, because maybe I am mistaking there, but not seeing it. When the fuel tank was new and I have put the level sensor in, I started measuring the voltage on the sensor after pouring each 20 liters of fuel until the tank was full (like some calibration). At the same time I have measured the battery voltage (because sensor voltage is derived from battery voltage so to offset the battery voltage variation). I do have a gauge, but the gauge is showing very roughly amount of fuel and you can know exact amount only at some specifically marked point, like reserve or one half, but not practical for measuring fuel consumption (small differences). So, I have put the data in software and have drawn the diagram and it turned out that the sensor voltage divided with battery voltage is very linear with amount of fuel (as expected). So when I want to measure the fuel consumption at some predefined RPM, I measure the voltages of sensor and battery at the beginning and at the end of test run and calculate the fuel consumed in software (easy calculations). To save fuel and time and get the results fast I do a test run for ten minutes and then multiply the amount of consumed fuel with 6 to get the result in liters/hour. Of course, I let the boat stop and calm down to level off for measurement at the beginning and the end of test run.

    Now, why the question. I got very accurate results at displacement speeds and in accordance with two year experience with the boat and consumed fuel on longer trips. But, the question is should the getting over the hump be excluded for this type of measurement? Because with 195HP diesel engine, which at maximum power consumes approximately 37 liters/hour per the book, I have measured 90 liters/hour (getting over the hump included), so I was a bit confused. That was 15 liters in ten minutes. The boat is heavy, but not overloaded, no smoke, RPM can go a bit above maximum declared at WOT. Nothing is malfunctioning. If I presume that engine does consume 37 liters/hour without going over the hump, that means that in ten minutes it consumed 6.2 liters on planning and 8.8 liters to get over the hump? Is this reasonable thinking? Is it possible that engine consumes more than declared at maximum power when getting over the hump (that much more)? I would have measured average without the hump, but the boat is not leveled, fuel is not calm and a bit awkward concentrating on that during planing. Just to add, during this test run, I was planing upstream the river, no wind, no waves.

    I would appreciate your opinions/knowledge on this.

    Regards,
    Nidza
     
    Last edited: Oct 6, 2017
  2. baeckmo
    Joined: Jun 2009
    Posts: 1,652
    Likes: 668, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    Depending on the shape of the fuel tank and where the signal source is placed, this measuring method is very sensitive to trim change. This method is not even close to a reasonable precision. So, your instrument is sensitive to trim, and thus only calibrated for one position; ie one speed.

    Since the engine power required is proportional to speed x resistance, the consumption at hump speed will be high in terms of liter/distance. In addition to that, the engine efficiency will vary over the operating "map".
     
  3. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    Fuel tank is almost in the middle of the boat (almost at the center longitudinally (a bit more to the stern), exactly in the center per latitude). Sender unit is in tube, again almost in the middle of the tank. The fiberglass tank is almost perfectly squared, so it was easy to confirm calibration even measuring the height of fuel. Due to position of tank and sender unit it is somewhat resistant to trim, but it is definitively dependent of trim and that is why I measure voltage on the sender only when the boat is stopped and not at motion so "just" the weight distribution dependent result, not speed dependent. Measuring resolution that I use on the voltmeter is 10mV, and change of trim due to rocking the boat (which is very stable) by walking on the deck is +-20mv maximum, which could make a mistake around +-1.5 liters after the conversions, but definitively not as much as 8 liters. I do agree that precision is not reasonable as you say, but informative of some consumption range. Actually I have done this calibration to know what is the exact amount of fuel in tank and for that purpose the precision is more than enough, but I was curious about consumption per different specific RPMs without buying expensive measurement equipment or disassembling the hoses and using the bucket. Precision of this method would be increasing with increasing the time of test run (error decreasing with time), but then again, time and fuel are exactly what I wanted to save.

    I need to go through engine fuel maps one more time for the sake of better understanding. Basically, the shorter the hump (resistance peak) the better and critical distance shorter, that is why full throttle is the best for overcoming the hump as soon as possible? If efficiency is decreasing, I thought that engine loses power at the same amount of fuel, not increasing fuel consumption to compensate for bad efficiency, what is exactly happening (although that is same thing said in two different ways)? If engine is increasing the consumption to compensate efficiency losses, then my "informative" measurement would have sense and obviously prevent me even more from going over the hump.
     
  4. baeckmo
    Joined: Jun 2009
    Posts: 1,652
    Likes: 668, Points: 113, Legacy Rep: 1165
    Location: Sweden

    baeckmo Hydrodynamics

    There is no way your 195 hp engine could "swallow" 90 l/h, no matter what. The 37 l/h you mention is probably at the best efficiency point, representing about 160 g/Hph, which is quite low. From there you often have a slight increase to maximum power, say to 180 g/Hph. When decreasing rpm (and thus power according to the propeller law) the specific consumption in many engines increases to something like 200 to 230 g/Hph, meaning that at 50% Power the consumption would be 97.5 x 0.23 / 0.84 = 27 l/h. The figure 0.84 is the fuel density in kg/l.

    But even if you demand full Power to overcome hump speed, your Engine will not drink more than just below 40 l/h. So there is something wrong in your measurements or your calculations.
     
  5. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    Fair enough, I will go through calculations and do another test as soon as possible. It is common sense to me, too, but I was on vacation and have done only one test with explained result, was skeptic, but fuel level could change my range intended for vacation, so gave up testing it further at that time. Anyway, wanted to consult with more experienced than me. I will also try to think of some more precise method because you made me calculate the measurement precision and I am not satisfied with 1.5 liter error because I want to know more precisely consumption at a couple of different RPMs at displacement speeds.
     
    Last edited: Oct 6, 2017
  6. Mr Efficiency
    Joined: Oct 2010
    Posts: 10,386
    Likes: 1,042, Points: 113, Legacy Rep: 702
    Location: Australia

    Mr Efficiency Senior Member

    Surely a fuel flow meter that works on volume passing through it, is to be preferred.
     
  7. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    Flow meter would not be that cheap since diesel fuel is circulating - two flow senders (one in line from engine to pump and the other for return line from engine to tank) and control unit measuring the difference, so three components plus cables, stainless mounting plates, hose clamps, ... the list goes on... It would be enough for me to know approximate consumption at specific RPMs, of course there will be some errors due to weight changes according to one time test, but precise enough for planning routes prior to traveling.
     
  8. Mr Efficiency
    Joined: Oct 2010
    Posts: 10,386
    Likes: 1,042, Points: 113, Legacy Rep: 702
    Location: Australia

    Mr Efficiency Senior Member

    Further to your original question, though, hump speed will increase fuel consumption significantly, if you motor along under that condition, the nmpg will be well short of a clean planing cruise speed. How much depends largely on the boat.
     
  9. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    If the return line is just for leakage of the injectors, one flow sensor in the supply line is enough for your calculations. If your engine is a common rail diesel, the returned amount is more substantial, so one sensor is not enough, but in that case a signal is available from the ECU.
     
  10. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    @CDK
    It is Cummins PT fuel system (old V-504-M) so more similar to common rail, not possible without 2 sensors (both lines to engine and from engine are 16mm in diameter because of high flowrate).

    @Mr Efficiency
    Of course it will be high during the hump, but I was confused with the amount measured which was so high and out of any specs. I need to measure once more, but I will do a longer run (much longer than 10 minutes) for better precision (higher amount of fuel consumed) and since I do plan a short trip soon, fuel will not be wasted just for the test.

    @baeckmo
    You were correct about a bit too good consumption value at WOT, I have not memorized it very good, here are the diagrams of engine in the first two links (same engine, different configurations) and in the third link at the bottom of the document can be found factory measured consumption at different loads for the same engine in fire pump application. Although the first diagram shows about 45 l/h (12 gal/h), the firepump is using 36.7 l/h @ 3000RPM, but then again it is 185 SHP, 195BHP, so when SHP taken it falls close to 200 grams/HPh. The links:

    https://gce.cummins.com/gce_hist/curves/3800_389.9/C3802A.PDF

    https://gce.cummins.com/gce_hist/curves/3000_329.9/C3131C.PDF

    http://manuals.chudov.com/Grove-Rough-Terrain-Crane/Cummins-504-Engine-1.pdf
     
  11. arkie
    Joined: Mar 2016
    Posts: 4
    Likes: 1, Points: 3, Legacy Rep: 10
    Location: St. Petersburg, Russia

    arkie New Member

    Anyway, the consumption at WOT corresponds to the curve #4 at the first PDF. Put hump RPM to that diagram and find max fuel consumption possible.
     
  12. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    Thanks Arkie.

    Finally did the test today. This time it was better correlating with the engine data. I had two errors in test described in the beginning of the thread, which I have taken in consideration this time. I have remembered that beside stopping the boat, I need to stop the engine, too, because of circulating fuel which maybe bubbles or does who knows what inside the tank. The second was not obvious and after the engine is stopped, battery voltage is decreasing due to disabled charging from alternator for about two minutes and during that time it turned out that sender voltage does not correlate with battery voltage during the whole transition. So before measuring, I had to wait for the transition of battery voltage to stabilize. Probably the combination of these two errors made unrealistically high values in my initial test.

    So the average result of a couple of measurements are 41.68 liters/hour at WOT (27km/h) and between 4.5 and 5.5 liters/hour at the speed of hull before the hump (11km/h / 1500RPM). Of course, these measures are not very precise, but usable for voyage planning. It is semi-displacement boat, therefore the planning speed numbers are a bit lower. The engine can go above maximum RPM, probably could have higher pitch, but since the boat is heavy, I prefer it this way, especially since I am driving it mostly at displacement speed.
     
  13. Barry
    Joined: Mar 2002
    Posts: 1,847
    Likes: 501, Points: 113, Legacy Rep: 158

    Barry Senior Member

    I am curious as to where you are attaching the probes off your multimeter to get a voltage from the sending unit.
    Without a guage you should not have any voltage showing at the sending unit.
    Ie a guage has a couple of coils and a direct path to ground through the guage and a tap between the two coils that goes to the sending unit. Ie there is always a path for the current to travel in case the variable resistor on the sending unit gets damaged. Ie no current would flow into the tank, cause a spark and blow up the tank.

    ie the sending unit measures resistance between the tap and ground in parallel with the main current to ground circuit.

    If you have hooked up a hot wire to the sending unit and measuring voltage, something is wrong.

    On the surface, the variable resistor between a voltage source and ground can change a read voltage, potentiometer, but in a fuel gauge, there are two paths to ground to inhibit a spark if the variable resistor fails.

    So the questions are, 1) have you hooked up a wire to the sending unit? and 2) where are you taking the voltage readings?
     
  14. Nidza
    Joined: Nov 2016
    Posts: 106
    Likes: 28, Points: 28, Legacy Rep: 10
    Location: Belgrade, Serbia

    Nidza Senior Member

    @Barry
    That explanation about the coils to prevent sparks in case of broken resistor is very interesting to me, it always seemed awkward to me putting the resistor with electric current inside the fuel tank, but I have never seen the circuit behind the gauge so the simplest possible solutions that came to my mind are:

    1) injecting constant current in the resistor and measuring the voltage on it (that would be independent of battery voltage) or

    2) resistor voltage divider - one resistor inside gauge serially connected to sender, then supplied from battery and voltage measured from sender and according to measurements this second solution seemed implemented since sensor voltage varies linearly with battery voltage (although I was confused why it had some delays only during the battery voltage decrease period, but without the schematic I would not even try to guess why).

    Therefore I have measured voltage on the sender resistor (on the top of fuel tank) with voltmeter while the gauge is attached, otherwise I would have to measure resistance since there would be no electrical source in that case, obviously. And the battery voltage I have measured directly on start battery poles with the voltmeter since that gauge is supplied from that battery. Voltage on the sender resistor is approximately in the range of 1.5V to 4.5V from full to empty tank, respectively.

    Sender unit and gauge are typical stuff from VDO.
     

  15. Barry
    Joined: Mar 2002
    Posts: 1,847
    Likes: 501, Points: 113, Legacy Rep: 158

    Barry Senior Member

    FuelGaugeCircuit-01.jpg
    My understanding is that the coils are used to create an electromagnet which moves the needle. One coil ( on one side of the needle) with a fixed resistance is in series with the source, then the tap to the variable resistor in the sender, and the second coil is in parallel (on the other side of the coil) with the variable resistor in the sender. So the coil voltage changes and the electromagnets change and due to the difference in the strength the needle moves.

    The new question then is "why if you have a working gauge do you want to measure the voltage differential through the sender, when the gauge is already giving you feed back?" You did not mention that you had a gauge in the circuit until this post.


    I just added a sender circuit.
    BUT note that this circuit is for a specific sender. Ie where it mentions that the resistance is low at low fuel is not necessarily correct for every sender. The senders have various ranges of resistance AND also a polarity difference. Not sure if polarity is the correct term here but with some senders, at full fuel the resistance is at the low end of the its resistance range and with some senders at full fuel the resistance is at the high end of its resistance range.
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.