Panbo

Maretron FFM100, positive displacement fuel flow metering

... written for Panbo by Ben Ellison and posted on Aug 22, 2011
Gizmo_NMpG_graph_cPanbo.jpg

As illustrated above, the fuel cost for cruising Gizmo varies enormously with speed. It's fortunate that we like poking along at 6 to 8 knots in most conditions because knocking off miles on a 16-17 knot semi-plane is four or five times more expensive. But as suggested by the lumpy lines, those fuel-flow versus boat-speed numbers are not necessarily precise, and since I was only able to collect them on one day, they definitely don't reflect changes in vessel trim, bottom cleanliness, engine condition, etc. I'd love to have accurate real time fuel flow data flowing around Gizmo's NMEA 2000 network and thus am excited about Maretron's FFM100 system, which was recently introduced with bold claims...

According to the press release (PDF here):

Larry Anderson, Maretron VP of sales & marketing, said "the accuracy of the FFM100 is nearly that of commercial vessel systems costing tens of thousands of dollars, yet the FFM100 costs less than existing recreational systems found on the market today."
   Additional benefits of positive displacement metering technology are: elimination of flow conditioning components such as "straighteners" and pulsation dampers, true temperature compensation with embedded temperature sensors, and detection of momentary reverse flow in the fuel lines due to fluctuating pressure caused by the injection pump.

Maretron_FFM100_Fuel_Flow_Monitor.jpg

I'd say that's a shot across the bow of Floscan, which dominates the complicated world of fuel flow monitoring. Unfortunately, I can't find details on Floscan's site about the technology it uses to actually measure flow or to compensate for temperature differences between the diesel flowing into and out of an engine, but there's definitely a calibration process, and I've heard mixed reports from boaters with Floscan systems (the negative ones possibly due to poor calibration).
   Meanwhile, the Maretron FFM100 manual (PDF here) does a good job of explaining the difficulties of measuring diesel fuel consumption accurately, and purportedly why their MxRSP positive displacement flow sensors with built-in temperature sensors work so well. There are an amazing number of ways to measure fluid flow and Maretron claims a 0.25% accuracy for their sensors though they also explain how that can lead to a worse case 1.75% inaccuracy in a diesel system because typically three times more fuel returns to the tank than is actually burned in the engine. (I never knew the figure was so high, but was not really surprised as I've recently learned how fast I can trim Gizmo by taking fuel from one tank and returning it to the other).

Maretron_M1RSP_flow_sensor.jpg

And mind you that accurate fuel flow measurement may be of interest even to skippers with modern diesels that claim to report it. We had a long discussion about this last year, and I came away pretty much convinced that no diesel actually measures the fuel it burns, though some are very good at internally estimating the burn based on engine load and other factors. Bill Parletore ended that conversation with some valuable research he'd done, and I was pleased to find that his eventual Passage Maker article on the subject can be read online. You'll find that he highly recommends checking fuel flow "totalizers" -- a common feature that keeps track of gallons used -- against actual fuel pumped into your tanks in order to at least refine your sense of accuracy if not improve calibration. It will be interesting to see how Maretron's system does in real world tests like this.
  At least the issue of calculating fuel flow info with boat speed to get the all important miles-per-gallon figures has been solved nicely using NMEA 2000, as I saw years ago (1-Garmin2-FloScan3-Raymarine4-Maretron5-SmartCraft, and 6-Lowrance), and is even truer today. And let's note that FloScan led the way in bringing mechanical flow measurement to N2K, and has now had a lot of experience with it. I also noted today that the company has introduced new DataLog software for deep fuel management analysis. I wouldn't be surprised to see screens like the ones below in Maretron's N2K View, and I wonder what FloScan thinks of positive displacement sensors.

FloScan_DataLog_screens.jpg

Comments

Incidentally, it would be great to hear from anyone who has any sort of diesel fuel flow monitor and has been able to check its accuracy by comparing flow totals to actual fuel put in tank (or some other way).

Posted by: Ben at August 25, 2011 1:56 PM | Reply

The Valkyrie is a 55 Sea Ray with dual 900hp MAN engines. The MAN controls have a flow meter. While I would be the first to state that I am not an expert on how it works or what may be better, based upon fuel burn, speed, distance my calculations are usually pretty close to what fuel remains in the tank, especially for long range cruising.

The Valkyrie can sip fuel at 8kts and suck fuel at 30kts.

For me, knowing the fuel burn is a function of calculating distance. The burn rate is actually not important, but distance is.

Anything between 12 and 30 knots gives me very close to the same maximum distance that I can travel (about 250 miles). If I reduce speed to 10 kts this range will almost double and at 8-8.5 it is close to 800 miles!

I do feel my fuel meter is pretty accurate based upon experience.

Posted by: ValkyrieYachts at August 25, 2011 2:06 PM | Reply

Interesting that they are using GPI flow meters

http://catalog.gpi.net/Asset/gm001-gm002-gm003-manual.pdf


Posted by: Jaime at August 25, 2011 3:59 PM | Reply

Thanks, Jamie. The drawings in that manual do illustrate how positive displacement flow sensing is done with geared rotators, but I'm told that those are not the sensors Maretron is using. Apparently they are related because GPI and Maretron use the same manufacturer, but GPI's sensors do not have the built-in temperature sensors that Maretron says are important to diesel fuel flow accuracy. Another difference is that Maretron's sensors can measure reverse flow. (I'm not sure that's important to many fuel installs, but the FFM100 can also be used to measure other fluid flows.)

Posted by: Ben in reply to Jaime at August 25, 2011 5:39 PM | Reply

I've found that having the ability to see in real time the consumption in distance per volume (NM/gal, NM/liter, etc.) is beneficial on a planing or semi-planing hull. This allows one to see the effect of trim and throttle changes almost instantly.

I have older Navman chart plotters connected to a Navman fuel flow sensor. This combination has taught me a lot about the relationship between trim, throttle setting, speed, water conditions, and fuel economy. Even if the flow meter is not perfectly calibrated the readings are good enough for me to extract the most from my fuel.

Raul

Posted by: Raul at August 25, 2011 6:11 PM | Reply

The specifications for all three sensors listed show 1/4" npt ports - but many larger diesels are using 1/2" fuel lines (at least for draw). Not to mention that rotors cause restriction as well.

FloScan has 1/2" ports on their larger sensors, and they are optical with very little restriction. I'm curious if adding the Maretron sensors inline with a 300+ hp diesel will cause fuel starvation?

If not, then a game changer for sure, as we're looking at less than half the cost compared to FloScan!!!

Posted by: AaronH at August 26, 2011 8:40 AM | Reply

Aaron, I've sent your question along to Maretron, but suspect the answer lies in the "Accuracy and Pressure Loss" diagrams in the FFM100 manual. What I see is that the M2RSP sensor -- rated for 200-1000 hp, or 4 to 132 GPH of flow -- generates a pressure loss of 10 kPa at 100% flow, and about 3 kPa at 50% flow. 10 kPa equals only 1.5 PSI, which I think is a minor disturbance to a fuel pump, though I'd like to have a real mechanic confirm ;-)

Posted by: Ben in reply to AaronH at August 26, 2011 9:20 AM | Reply

Aaron, Maretron confirmed my interpretation of the manual diagrams and specs:

"The sensors do introduce a restriction to fuel flow, causing a pressure drop. At maximum flow, the M2RSP causes about a 1.5 PSI drop in pressure. This compares favorably with some turbine sensors, which have pressure drops of 3-7 PSI at their maximum rated flow. In general, it is advised to keep the fuel system’s total pressure drop under 5 PSI."

They also discovered that the link to the literature for largest sensor, the M4ARP-2-E8, is wrong. That sensor does in fact have 1/2" NPT ports. And finally in terms of real world testing they wrote:

"Prior to the product release, we’ve done field testing with our distributors both in the US and Europe, so far with engines ranging from 26 HP to 480 HP. Each sensor is individually calibrated prior to shipment. K-factors for each sensor appear on the label for each product. We’re also starting to get extremely positive feedback from customers who have installed the FFM100."

Posted by: Ben in reply to AaronH at August 26, 2011 4:27 PM | Reply

If I understand this technology correctly, this Maretron system solves the fuel flow part of the equation admirably. But the other side of the equation is speed. I'll be interested to find out if Maretron allows the option of toggling between SOG and SOW. The only accurate measurement of nmpg is to use speed over water (instead of ground from gps) to compensate for current (including tide flow). Of course SOW still won't compensate for windage but I'm assuming that head-on wind is minimal. There are times when SOG is desirable such as the prediction of fuel available until the next fuel stop, but only SOW can give an accurate measurement of drivetrain efficiency to eliminate the affects of moving water.

Posted by: Jim W. at August 26, 2011 5:41 PM | Reply

I agree, Jim, but only in theory. The problem is that many STW sensors aren't very accurate and can't be calibrated. I have two running on Gizmo's N2K networks and my displays are set up to show STW right beside SOG. Plus I have lobster pots everywhere that let me know what the current is actually doing. Sometimes the speed differences makes sense, but not that often.

In fact, when I collected that feul flow data seen in the graph above, I noted STW as well as SOG. We also did two runs, in opposite directions to normal tide current flow, and averaged them. I'm pretty confident that the SOG numbers are more accurate than the STW numbers.

Fortunately, most modern MFDs let you choose between SOG or STW for calculating values like NMpG, "True" wind, etc.

Posted by: Ben in reply to Jim W. at August 26, 2011 5:59 PM | Reply

This is a very interesting development. I personally have a high regard for Maretron and have quite a few of their bits of kit, and in fact used their great "bomb" proof cabling throughout my boat when I did a complete electronics refit winter before last.

I will be interested in live experiences with these units.

I have had floscan in a previous boat but found them not as accurate as I had hoped and fitted the Navman units when they came out about 4 years ago in my current boat. Price of the floscans now for N2K is beyond belief!

When the Navman units worked these were very accurate, however the sensors were aluminium with plastic fin wheels inside. The aluminium used to fracture if you exerted the smallest amount of undue stress when tightening the pipe connector, and the plastic fins were suseptable to the high return fuel temperatures and any fouling of the fuel. The display heads failed twice in 4 years. So last year I removed the whole system.

So as soon as we have some live data that supports their accuracy and reliability I will be queueing to buy the maretron units. Maretron as I say are a great company however their ultrasonic tank transducers are, like all of this type, unable to read the top 2 inches of the tank so it is these sort of issues I would like highlighted in the the flow sensors operation before I make the decision to jump.

I watch avidly!

Posted by: Michael S at August 30, 2011 4:27 AM | Reply

I have the Floscan, FloNET installed for the past two seasons. FloNET suffers from a serious RFI emission problem that makes using the single sideband radio impossible while FloNET is connected to the N2K network. Disconnecting the FloNET drop cable makes the problem go away. Floscan sales and tech support investigated this and in the end said there was no problem. Unfortunately, I have contacted other FloNet users with a single sideband radio who have the same loud, pulsating RFI on their radio. This season I started seeing the FloNET freeze up after a long day of motoring, (50 Hp diesel on sailboat). Disconnecting the FloNET drop cable and reconnecting it temporarily solves the problem. In my opinion, this solution is not acceptable and inconvenient. For the high cost of FloNet I’m disappointed that it came with a series of issues.

Maretron’s FFM100 is something I am very interested in, but this time I will be cautious. I hope Mareton has done the required FCC testing to insure there is no RFI emitted from the control box. As far as displaying the data I assume when Maretron says “compatible NMEA 2000 display” they mean the FFM100 data can be displayed using a Garmin 4000/5000 chartplotter or GMI-10 display. If I can be assured of this and an RFI free product I’m willing to abandon FloNET system and try the Maretron FFM100.

Posted by: Richard C at August 31, 2011 9:39 AM | Reply

Leave a comment