Liquid propane injection 1

If you don't also increase the airflow, from a source bypassing the gas carby, you will go way to rich.

The system can be roughly regulated by the size of the discharge holes, provided they are the most restrictive part of the system, but the pressure will vary depending on the temperature of the tank.

A thermostatically controlled heater band on the tank will help to reduce pressure fluctions.

From your question, it seems you have a limited knowledge of how fuel systems and engines really work. A little knowledge can be dangerous. DO YOU REALLY WANT TO EXPERIMENT with exceptionally volatile, high pressure fuels combined with heat and electricity

Regards
pat

1asa,

I'm with Pat on this one. Dumping LPG into your engine won't accomplish anything without an increased supply of oxygen. N2O injection works because nitrous acts as an oxidant. Intake charge cooling is largely dependent on the quantity of LPG injected and the length of the intake runner.

Bryan Carter

I'm also with Pat and Retracnic.

The whole point of the system that "carburetes" the LPG on your vehicle is to mix the (cold) LPG with air correctly, and raise its temperature for clean vaporization. A fairly 'cheap' way to get intake-charge cooling on acceleration might be to chill an intercooler with 'cold' LPG (I'd use a water-to-air intercooler and a separate LPG-to-water heat exchanger, with appropriate antifreezing protections). But fogging extra LPG into the engine would be more or less like putting extra carburetor jets (NOTE that I don't say 'fuel injectors' -- there's a reason!) in the runners of your intake manifold. A lot more fuel (at cold temperatures as it expands, too) without more air at the same time is going to produce lower, not higher, performance, and will simultaneously increase your fuel usage substantially. (Which is more or less exactly what was more succinctly meant by "way rich mixture!"

As Retracnic said, nitrous systems are completely different. Think for a moment about what 'nitrous' is. Does it contain carbon or hydrogen for oxygenation? Does it produce nitrogen oxides (the analog of 'combustion' for that element)?

Answers to questions like these should indicate to you that nitrous is indeed providing additional combustion oxygen rather than additional 'fuel'. The "fogger" ensures complete mixing with the intake air, and helps eliminate 'cold spots' in that air, due to the expansion of the N20, that might cause problems in combustion. The reason for the cutoff solenoids (attached to the famous 'button') is not what you probably think: it's that fuel delivery on carbureted engines is tied to fuel-pump speed, and adding oxidizer will cause the engine to accelerate, and hence overspeed (much like what happens when a wastegate on a carbureted turbo setup sticks).

The way engineers have best coped with these problems in the age of microprocessor assistance is to go to port or direct fuel injection, which shares the happy characteristic (with diesels) that you can't overspeed the engine by increasing the boost if you have primary control over the injected fuel charge. Then you can 'overfuel' all you want provided you can arrange for adequate oxygen and other combustion conditions, and you've got mechanical parts that can handle the increased power, torque, etc.

I'm not going to comment on the 'common sense' of attempts to wring high performance from LPG vehicles (this fuel has inherently lower heat content, and somewhat lousy high-speed combustion characteristics, compared to alternatives for the purpose). I don't see an engineering reason why nitrous itself couldn't be adapted to work with LPG engines, and produce your desired 'chilling' effect in doing so. But be VERY careful how you introduce the nitrous! I recommend that you research proportional-nitrous systems to determine exactly how much 'fogging' you can do without grenading your engine, or alternatively, how far you'd cut back the LPG regulated flow to your 'carb' (and what lag to use at a given powerband setting following the cutback) before admitting N2O to an undercarb fogger.

Two potentially useful historical sources:

1. The early Mercedes 300SLs, in the '50s, used an auxiliary fuel injection system in their carbureted engines. This consisted of rather small 'makeup' supplies that operated only under conditions of peak anticipated engine demand, and were actuated only when turbulent air intake conditions and high mass flow ensured the injected fuel would be properly carbureted by the time it entered the chambers. Making a system that did this with LPG is theoretically simple, if available digital fuel injectors are convertible to work properly with LPG (remember lubrication, cooling, etc. that might be assumed for liquid fuel might not apply to you). Of course, if you made this system large enough, you might be able to dispense with the regulator and carb...

2. An outfit in Texas built a Ferrari Testarossa for John Carmody (one of the writers of the computer game 'Doom') a few years ago. This had a very well thought out proportional-nitrous system to reduce turbo lag and cool the intake charge without large and heavy intercoolers. (The magazine which printed the initial review of the car indicated you could go from zero to 127mph and brake to a terrified stop on a freeway entrance ramp! It also indicated that things like axle shafts were beginning to break from the massively increased transient torque this engine was capable of producing).

These sorts of things are where you should direct your attention. If you don't understand them yet, read and study until you do. And remember that everything from pistons on down will be seeing increased loads -- both pressure loads and jerk loads -- that may require you to strengthen things or cool them better.

RME

Putting a fixed feed liquid LPG jet in would be like pulling a fuel enrichment 'choke' on (like an SU). Not usually so good for power, acceleration or economy.

If 1asa has an O2 sensor and closed loop control (stepper motor mixture control or gas injection) it may lean the vapour feed to compensate for whatever liquid flow there is. But some controls switch to a fixed map and go rich for max power. If it does, it may be possible to reduce the 'max power' vapour feed by getting the ECU re-mapped but as Pat pointed out this would give variations due to changes in tank pressure as the compensation would be fixed. The only way to avoid this pressure variation is to do what petrol EFI does and fit a fuel pump with regulator. On LPG these can suffer from early vaporisation on the suction side, so the tank has to be cooled. Cummings have been working on an 'assist' pump that only boosts delivery pressure when it below the liquid injection system regulator pressure ~ below 0 degree C tank temp.

Regarding Nitrous oxide injection. 'Wet' NO2 always has an additional fuel feed system in addition to the NO2 jets. Both the fuel and NO2 are delivered though precisely matched jets. The fuel is fed into the NO2 fogger stream so the NO2 atomises it. Add "50bhp" of NO2 and you must add an extra "50bhp" of fuel. Add either without the other and the motor will most probably be so far rich or lean that it won't run (Murphy's law says it will run when weak and do lots of damage). There are some so called 'Dry' NO2 systems, these are dependant on the OEM fuel injection ECU running in closed loop at max power. The O2 sensor detects the mixture is lean and the ECU adds more fuel though the standard injectors. This doesn't work on ECU's that switch to closed loop and rich operation for acceleration as they have a fixed fuel map for max power. It will run weak if the OEM injection system maxes out on fuel feed, as will a 'wet' system if the fuel demand exceeds pump feed - fit an up-rated pump.

Ok folks, had to jump in here. Since I specialize in Gaseous Fuel System Development and LPG, CNG, safety training, I thought I should chime in.

For the record, LPG is not especially more dangerous or more volatile than gasoline under similar conditions. LPG is however, a pressurized fuel and when released to atmospheric pressure does drop to –44 deg F. As with any compressed or pressurized product (freon, cutting fuels, steam) it must be handled with respect. In the US, the NFPA publishes pamphlets on the approved handling processes for almost any flammable product, LPG, CNG, LNG H2, etc.

Propane also has an expansion ratio of 270 times (the liquid at atmospheric temp but at pressure) will expand 270 times to atmospheric temp and pressure vapor. One cubic liter of liquid LPG will form 270 cubic liters of vapor at the same temperature.

Nitrous is not a fuel, it is an oxidizer, and performs differently when used in an engine as a power enhancer. The ONLY similarity is that when Nitrous is released, it rapidly drops the surrounding medium’s temperature.

Now, the concept of liquid injection has been discussed before, but its been a while. Although the concept has been done internationally, here in the US, we use primarily HD-5 propane, which averages about 95% propane, and the trace being propylene, ethane, butanes, and other trace gasses. The rest of the world tends to use a blend of butane and propane, around 40-60, 50-50, or 60-40. What this accomplishes is to reduce the pressures. Liquid injection requires propane to remain a liquid under almost any temperature, and its not uncommon to see propane reach 400 psi when it absorbs engine heat. This pressure transfers to the fuel tank where it must be boosted to over vapor saturation phase to retain its liquid form.

Developing electronic fuel injectors that can handle these high pressures, and perform under those pressures accurately while metering mg of fuel is quite a feat. Not too many have been successful.

Developing a liquid propane injection system is not for the faint hearted person, nor is it a project for one with shallow pockets. The development of one platform can easily cost over $250,000 USD.

Franz

Thank you for all the informations.
Mine concept was based on a system like the one sold from this company, a quite simple system used for turbocharged cars: take a look here and tell me what do you think about it..
my only doubt is about the use of a "VAPOR ONLY" propane tank: it's the propane still at -44 degrees in that conditions?
And my thoughts was to add this propane shot when my OHG vapour system leans out at higher rpms, to richen and cool down the mixture under full throttle conditions..
Riccardo De Cal

As I cautioned earlier, throw the concept of liquid fuel combustion out the window when working with vapor fuels. They DO NOT perform the same either when mixing with air or during the combustion process.

Adding more fuel will not necessarily cool down the fuel mixture. When adding propane and gasoline together, some fuel will not burn due to the overly rich condition. Whether it is the propane or the gasoline, you will never know. Propane will mix more completly with the incoming air than will the gasoline, so it may well be the gasoline that doesn’t burn completely. It MAY then cool down the combustion charge.

As for your question about the DOT tank in vapor service, when you draw vapor from the tank, the phase change is performed in the tank and the resulting vapor temp may well be near ambient. At no time is the tank at –44 degrees unless the vapor draw is so drastic as to reduce the vapor pressure due to evaporation.

As for the OHG leaning down at higher loads, the vacuum operated power piston should help alleviate that. This carb should provide fairly even air/fuel ratios at most engine speeds, providing it is sized properly. Making sure the three metering pins are set at the proper height and that the air valve moves freely will help too. This is one of the most efficient mechanical vapor carbs around, but requires considerable maintenance. Don’t overlook your vaporizer supply and pressures, plus filter flow.

The website you refer to uses a small amount of propane under boost to enhance the octane. It is not the primary (or what we call the “major fraction”) fuel. I would guess the total amount supplied would be in the 5% range. The gasoline oxygen sensor would trim the gasoline leaner to make the average air-fuel ratio somewhat closer to Stoichiometric.

Franz