Fuel system data monitoring question

[SCChris]

I have a question relating to what can and is being monitored on the GX fuel system.

Is the internal pressure and temp of the CNG available to the GX system computer? If not, how is the fuel gage at all accurate? If the temp is being monitored and the pressure is also known then the GGE amount of the fill should be available to be calculated, displayed and recorded at the time of the fill. This method would be independent of the fueling appliance, or station. With a bit of coding on Honda’s part we should be able to retain things like overall miles per GGE, and miles per GGE since the last fill. Unlike a liquid fuel tank this would seem to be pretty accurate, Right!?

If the Pressure and Temp of the tank are known, even MPH and true MPG could be correlated in small increments given memory to dump the data to.

Chris

[Green GX]

Re: Fuel system data monitoring question

I wondered much of the same but, I e-mailed the ScanGuage people with the question. They never e-mailed me back.

Bill

[Curtis]

Re: Fuel system data monitoring question

The fuel guage is a temperature compensating pressure guage. I don't know that the computers would have enough of the information it needs without additional sensors to monitor all of the parameters you outlined. Until the Civic GX receives a fuel computer... I don't think you'll be able to get this type of information without a scanguage, or similar add on device. Even with a fuel/trip computer, it would be different from calculating liquid fuels for the reasons you pointed out.

[SCChris]

Re: Fuel system data monitoring question

I don't know the detail of the temperature and pressure measurements passed to the GX CPU, but if the detail is there and access to it is there, then it wouldn’t take long to ferret out a load of interesting info.
What does the super air filter really do for you? Does this Cat-Back really pen out or is it just sexy? Did the tune-up just screw my mileage?
You could know if they have enough fuel to get 2 full commuting runs and still have a bit spare for safety.
I’d like to know if the measurement detail and access it is there.
Who would I, should I contact?

Thanks
Chris

[Jay98]

Re: Fuel system data monitoring question

Remember PV=nRT?

The pressure and the temperature are measured at the fuel block, and the volume of the tank is known, so you can solve for the amount of fuel in the tank (number of moles). This is proportional to mass and of fuel in the tank which is proportional to GGE.

[freedml]

Re: Fuel system data monitoring question

PV=nRT may be very useful in a lab where you have controlled conditions and precision, calibrated instrumentation, but it isn't much use when the measurement and repeatability errors involved are large.

[Adrian]

Re: Fuel system data monitoring question

Remember PV=nRT?

The pressure and the temperature are measured at the fuel block, and the volume of the tank is known, so you can solve for the amount of fuel in the tank (number of moles). This is proportional to mass and of fuel in the tank which is proportional to GGE.

I've thought about somehow tapping into the temperature and pressure data and using the known tank volume to solve for GGE. I think that by the time one measures the temperature at the fuel block (under the car), it might be slightly lower than at the tank due to cooling. I've added a second tank to my car with a manual pressure gauge and it's interesting to compare that to the number of bars shown. Last week I had a 4,000 psi fill while it was pretty hot outside. The Honda fuel gauge lit up all the way, but then within 2 miles, the first bar disappeared. I pulled over, checked the manual pressure gauge, and it was already at 3500 psi. I think cooling can occur fairly fast, and if one was trying to get an accurate GGE calculation based on a temperature, pressure and known volume, the temperature should be measured at the tank if at all possible. This is easier if you add a second tank, but not so easy with the stock tank. I would also use a different correlation, other than pv=nrt , to account for the slightly non-ideal gas behavior of gases at higher pressures (see equations of state on Wikipedia). Having said all that, the composition of natural gas varies based on the supply. Assuming a fixed composition will trow off the result slightly depending on how different the composition of the fuel used is from the composition assumed.

[SCChris]

Re: Fuel system data monitoring question

It occurs to me that an instrumented nipple could always gather pressure and the temp during the fill. I'll bet it's a bear to replace the nipple. The other place to measure would be at the tank / fuel line connection.. Again this could be a bear. Drain, remove, insert sensor package, test sensors, replace fuel line, pressure test, test sensors.. I have also wondered if I could check the incoming CNG temp passively by measuring at the fuel nozzle. If I were off a degree or two Kelvin, how much would it affect the accuracy of the calculation? If I use the Trillium station in Anaheim, I have the pressure, although the accuracy is unknown. I understand that this passive sensor calc is only good at the fill moment, but fun food for thought nonetheless.
Chris

[Jay98]

Re: Fuel system data monitoring question

PV=nRT may be very useful in a lab where you have controlled conditions and precision, calibrated instrumentation, but it isn't much use when the measurement and repeatability errors involved are large.

Probably good enough to use for assessing fuel consumption trends and changes due to tune ups, mods, etc...

[kierandill]

Re: Fuel system data monitoring question

- the pressure relief on the tank might be a good place to measure the pressure.

- The Hondata people, while they have no GX stuff now, certainly have the know-how to read whatever ECU info is there.

- Adrian - how much did your aux tank cost and how big? Is it in the trunK?

[CraziFuzzy]

Re: Fuel system data monitoring question

Probably good enough to use for assessing fuel consumption trends and changes due to tune ups, mods, etc...

Agreed, it would be enough. Add filtering to smooth the irregularities, and the average trend will tend towards ideal gas law. I do know I've used a basic add-on scanguage2 on my crown vic, and even though it reads wrong number by default (as far as GPM fuel rate), after it is calibrated through a few fill-ups, it reads quite accurately.

[Adrian]

Re: Fuel system data monitoring question

- the pressure relief on the tank might be a good place to measure the pressure.

- The Hondata people, while they have no GX stuff now, certainly have the know-how to read whatever ECU info is there.

- Adrian - how much did your aux tank cost and how big? Is it in the trunK?

I think using the pressure from the OEM pressure sending unit would be sufficiently accurate. I think the OEM temperature sensor however, due to its location, might not be sufficiently accurate; it likely indicates a temperature lower than the actual tank temperature. It's the temperature that could benefit from being measured on the tank. My extra tank is a 6 GGE Lincoln Composites type IV (used), in my trunk. I have no trunk left. I would have to find all my receipts to figure out the exact cost. It took me over 6 months buying the parts one by one and then installing it with the help of some very knowledgeable friends to get this done properly. It took this long because of the overly protectionist attitude of almost everyone selling parts in this industry. Luckily, not everyone is a hard liner, and I eventually was able to purchase all my parts. Either way, the total was less than $2,000 and it gives me a range of 400+ miles. I generally drive at least 80 MPH, so if I drove 65 MPH, my range would be 500+ MPH.

[larrycng]

Re: Fuel system data monitoring question

Hey Guys, I don't think I would try to put a pressure gauge/transducer in the PRD line. If the gauge fails you loose all pressure in the tank and there is no way to isolate the PRD for gauge installation. With electronic valve the PRD is usually in the valve (I have a picture if you need it). If the is a way to isolate a PRD line to install a gauge it, then the installation does not conform to NFPA 52

On the GX, since it has electronic valve, you might place the gauge/transducer in the fill lline close to the tank or in the HP line close to the tank.

Just suggestion

Larrycng

[SCChris]

Re: Fuel system data monitoring question

I haven't looked at how the shutoff valve is setup, but I wonder if the motor side of this valve could be where instruments could be located. It would certainly be easier to use the shutoff, remove the existing line, insert an in line instrument package, and connect it all back. As long as this area is close enough to the tank and the flow is sufficient then the temp should be close to the real tank temp. And as long as the package is between the tank and the PRD then pressure might be ok. I don't know how much the line pressure would vary. I could see that the flow might not be steady. I'm envisioning something like analog vs. digital where the fuel system caches a bit of fuel and as this cache drops to a low pressure threshold more fuel at higher pressure is allowed to replenish the system, so what the pressure over time graph looks like is a square wave rather than a clean slope from 3600 to 500 psi. If the existing Honda pressure sensing is good then why recreate the data, temp on the other hand...

Chris