Diesel / LPG - The first tank results are in!

[KiwiBacon]

TD42
was naturally aspirated.
now turbo.
running about 15psi through a 50thou jet.
NFI what the flow figures are, no one would tell me, i got in trouble for opening the unit, which voided my warranty and he got a bit grumpy when i pointed out how simplistic the system is...
i dont know how to measure how much gas im flowing, but if its simple i will do so...
and its D-GAS, not dieselgas australia, just to be clear.
what setup is it you ahve? or was that just a bbq bottle attached?

i was told my motor would do what yours is doing if i was to add gas at idle but it doesnt.

it MUST have to do with injection timing, as it doesnt burn until the diesel burns so if your injection timing is a little late for diesel, its gonna be right for the increased burn rate with the gas.
which i would guess is your issue with the direct injection, probably easier for them to knock, seeings as the gas is in more direct contact with the flame front...
idunno, over my head but it seems to make sense.

my thoughts on the matter are, if you had a common rail motor, so could monitor exactly how much diesel is being injected, you could inject the exact right amount of gas to keep it at 25 or 30%. whatever ratio you want.

the D-GAS system works by adding X litres at Y pounds of boost.
irrespective of throttle position or RPM, so their claims that it injects the right amount is rubbish in my opinion.
how can it inject the right amount based on boost only, its going to be rich down low and lean up high.
all in my opinion. i am happy with the D-GAS unit, its just not as sophisticated as i would like.
this is the older system, but i think the newer system isnt too much more advanced...

cheers, andy

Yes those are my thoughts regarding direction injection versus indirect. The indirect gas charge gets cooled by the head more and combustion has a more tortured path to follow. In a DI engine everything is in one bowl just waiting to go boom. Not being shuffled to and from a swirl chamber.

But common rail engines are direct injection. Most (maybe all) use sequenced injection with up to four pulses. How lpg reacts with that will be interesting, I'd expect it to all go up in the pilot.

Regarding timing. I first ran this experiment about a year ago when I was running very advanced timing (experimenting with the full run of timing adjustment at the tim). Right now I'm on the retarded side of the factory marks but the behaviour on the video is the same as I experienced a year ago.

I don't have a real fumigation setup, this is just a simple test using a bbq bottle. If I can score some better gas measurement gear I'll repeat it at a later stage.

[JK]

The Sequent II system isfar more advanced than this.

I have been in cab during a calibration run holing the laptop showing the real-time readout from the Sequent II ECU. it is incredible.

My system is running abouu 25% LPG to diesel ratio.

LPG is still being injected at idle but at a very low rate - even if it was 25% diesel ratio and my vehicle sucked 1-2L per hour at idle means LPG is in at 0.25 to 0.5L per hr! - no wonder the rover was chugging at 4.5L per minute...

[KiwiBacon]

The Sequent II system isfar more advanced than this.

I have been in cab during a calibration run holing the laptop showing the real-time readout from the Sequent II ECU. it is incredible.

My system is running abouu 25% LPG to diesel ratio.

LPG is still being injected at idle but at a very low rate - even if it was 25% diesel ratio and my vehicle sucked 1-2L per hour at idle means LPG is in at 0.25 to 0.5L per hr! - no wonder the rover was chugging at 4.5L per minute...

So how does the sequent know what the diesel volume injected is? Does it hook into EFI diesel systems, does it have a flow-meter in the supply/return line or is the 25% just a long term average (100L diesel to 25L gas)?

Does it display lpg:air rates?

[JK]

The Sequent II system is far more advanced than this.

I have been in cab during a calibration run holding the laptop showing the real-time readout from the Sequent II ECU. it is incredible.

My system is running about 25% LPG to diesel ratio.

LPG is still being injected at idle but at a very low rate - even if it was 25% diesel ratio and my vehicle sucked 1-2L per hour at idle means LPG is in at 0.25 to 0.5L per hr! - no wonder the rover was chugging at 4.5L per minute...

So how does the sequent know what the diesel volume injected is? Does it hook into EFI diesel systems, does it have a flow-meter in the supply/return line or is the 25% just a long term average (100L diesel to 25L gas)?

Does it display lpg:air rates?

On EFI diesels I believe it hooks into the main ECU although I can’t confirm that.

On mech injected like mine they install an electronic boost gauge, electronic EGT, and take tacho readings off the alternator and I think there might be another sensor as well but can’t remember.

From this the D/G Tech ECU 'learns' how the car had been driven and maps the amount of LPG required depending on the engine load, revs etc. This is done in real time and adjusted many times per second.

LPG is metered thru an infinite adjustment needle valve injector – i.e. The Sequent II unit.

Yes 25% is calc at refuelling time. The % varies all the time depending on what the ECU tells it to do.

From what I can gather, the system developers have managed to work out a correlation between the boost, EGT and tacho etc and been able to accurately map the diesel fuel i.e LPG fuel delivery without needing flow rates, meters etc.

If you are serious about learning more I suggest you contact the guys who developed it – they are easily contacted on the net or by phone.

I’m just some dummy that builds stuff for a living.

Cheers, JK

[ISUZUROVER]

The Sequent II system is far more advanced than this.

I have been in cab during a calibration run holding the laptop showing the real-time readout from the Sequent II ECU. it is incredible.

My system is running about 25% LPG to diesel ratio.

LPG is still being injected at idle but at a very low rate - even if it was 25% diesel ratio and my vehicle sucked 1-2L per hour at idle means LPG is in at 0.25 to 0.5L per hr! - no wonder the rover was chugging at 4.5L per minute...

So how does the sequent know what the diesel volume injected is? Does it hook into EFI diesel systems, does it have a flow-meter in the supply/return line or is the 25% just a long term average (100L diesel to 25L gas)?

Does it display lpg:air rates?

On EFI diesels I believe it hooks into the main ECU although I can’t confirm that.

On mech injected like mine they install an electronic boost gauge, electronic EGT, and take tacho readings off the alternator and I think there might be another sensor as well but can’t remember.

From this the D/G Tech ECU 'learns' how the car had been driven and maps the amount of LPG required depending on the engine load, revs etc. This is done in real time and adjusted many times per second.

LPG is metered thru an infinite adjustment needle valve injector – i.e. The Sequent II unit.

Yes 25% is calc at refuelling time. The % varies all the time depending on what the ECU tells it to do.

From what I can gather, the system developers have managed to work out a correlation between the boost, EGT and tacho etc and been able to accurately map the diesel fuel i.e LPG fuel delivery without needing flow rates, meters etc.

If you are serious about learning more I suggest you contact the guys who developed it – they are easily contacted on the net or by phone.

I’m just some dummy that builds stuff for a living.

Cheers, JK

SO then on a mech inj. diesel it works exactly how loveke70 described it. They inject LPG based on boost pressure (and possibly EGT).

[JK]

The Sequent II system is far more advanced than this.

I have been in cab during a calibration run holding the laptop showing the real-time readout from the Sequent II ECU. it is incredible.

My system is running about 25% LPG to diesel ratio.

LPG is still being injected at idle but at a very low rate - even if it was 25% diesel ratio and my vehicle sucked 1-2L per hour at idle means LPG is in at 0.25 to 0.5L per hr! - no wonder the rover was chugging at 4.5L per minute...

So how does the sequent know what the diesel volume injected is? Does it hook into EFI diesel systems, does it have a flow-meter in the supply/return line or is the 25% just a long term average (100L diesel to 25L gas)?

Does it display lpg:air rates?

On EFI diesels I believe it hooks into the main ECU although I can’t confirm that.

On mech injected like mine they install an electronic boost gauge, electronic EGT, and take tacho readings off the alternator and I think there might be another sensor as well but can’t remember.

From this the D/G Tech ECU 'learns' how the car had been driven and maps the amount of LPG required depending on the engine load, revs etc. This is done in real time and adjusted many times per second.

LPG is metered thru an infinite adjustment needle valve injector – i.e. The Sequent II unit.

Yes 25% is calc at refuelling time. The % varies all the time depending on what the ECU tells it to do.

From what I can gather, the system developers have managed to work out a correlation between the boost, EGT and tacho etc and been able to accurately map the diesel fuel i.e LPG fuel delivery without needing flow rates, meters etc.

If you are serious about learning more I suggest you contact the guys who developed it – they are easily contacted on the net or by phone.

I’m just some dummy that builds stuff for a living.

Cheers, JK

SO then on a mech inj. diesel it works exactly how loveke70 described it. They inject LPG based on boost pressure (and possibly EGT).

Sounds like the same broad concept but I'm guessing the devil would be in the detail i.e. how the ECU brain works.

Also the needle valve injector compared to the stepper injector.

Would be interesting to see a side-by side comparo.

Cheers, JK

[ISUZUROVER]

Sounds like the same broad concept but I'm guessing the devil would be in the detail i.e. how the ECU brain works.

Also the needle valve injector compared to the stepper injector.

Would be interesting to see a side-by side comparo.

Cheers, JK

Regardless of how sophisticated the microcontroller (ECU) is, it has no information on the mass flow of air, how much diesel is being injected, or if detonation (or worse) is occurring.

[JK]

Sounds like the same broad concept but I'm guessing the devil would be in the detail i.e. how the ECU brain works.

Also the needle valve injector compared to the stepper injector.

Would be interesting to see a side-by side comparo.

Cheers, JK

Regardless of how sophisticated the microcontroller (ECU) is, it has no information on the mass flow of air, how much diesel is being injected, or if detonation (or worse) is occurring.

:flamesuiton:

Sounds like you guys are quite the experts on this - perhaps you should offer your consultant services to the guys at D-Gas / Diesel Gas Technologies etc who have spent the last few years developing systems that pass stringent regulatory requirements and Australian Standards…

But really, is this pre-detonation an actual problem or just some theoretical issue that could under the right conditions happen in a crude and totally illegal homebrew system?

Clearly the proof that these systems work is that there are hundreds, if not thousands of cars, trucks and stationary engines now on D/G systems and this detonation issue does not seem to exist on proprietary systems.

All I know is that I have been running my system for a few months now and it’s great. Smooth, reliable, economical and the truck hauls!

:flamesuitoff:

Cheers, JK

[MightyMouse]

be interesting to know the effects on cylinder pressure rise caused by what I assume is the initial combustion of the gas.

State of the art direct injection systems rely on the multiple injections ( its up to 7 currently.... BOSCH ) per combustion cycle to progressively increase combustion pressure thus avoiding the traditional "high stress" bang and therefore the need for significantly heavier and stronger engine components.

I'd certainly be very wary of fitting it to a current, high speed; light weight DI engine.

[love ke70]

ISUZUROVER, yes, similar, only D-GAS system has no computer and doesnt even base flow rate on RPM
new one does i think, i dunno, same broad hit and miss scenario as the older model i have as far as i can tell.

JK, i tried, he wouldnt listen.
the prohibiting factor is the simplicity of a mech injection of a diesel, and the cost it would take to make the system accurate.
detonation does occur, i can put a 50 or 60 thou jet in mine and listen to it knock in the low RPM all day, and this is super bad for cylinder pressures, and mighty mouse is mentioning.

to be honest, the setup is best for stationary fixed RPM motors, where they can set it to the optimum ratio, which for most applications is over 25% im sure, as mine runs sweet on the highway with a 60 thou jet, just makes a prick to drive around town, spend half the time with it switched off to stop the knocking.

mightymouse, there were studies into such things by cummins i think it was, very expensive studies, with pressure thingies in the bores. cant remember the results, but it wasnt all bad. i would say its a different kettle of fish with modern small diesels though..

you could however run LPG injection, rather than fumigation, and match the injections or stagger them to the diesel ones, thus working the same as before, with a smooth bang.
but that would not be cheap, and i dont think that technology is even developed yet, with standard LPG injection only now getting developed after 30 years of pissing about...

[MightyMouse]

With older heavily constructed engines you would guess that there might not be issues - and I can't think of a better example of "heavily" than truck engines, Love ke70 are you certain Cummins was testing multiple injections /cycle ?

However given the generational change happening with Diesels I have a feeling that things will get very much more complex for the "squirt in some gas" systems in the very near future.

State of the art Diesel systems are every bit as complex as petrol injection systems with the same level of closed loop monitoring that makes modification a job for experts and even then sometimes nigh on impossible.

The ol' diesel tractor engine is sure a changin.......

[love ke70]

i dont think it was multiple injections, just they were testing LPG fumigation on some of their big motors, and the combustion pressures were not too bad...
im not even sure it was cummins, but i read something of it, probably in one of the million threads here or on patrol4x4. probably the latter

was thinking about this today, and my TD42 is multipoint injection accoding to the book, is this what you were talking about, on a very basic scale, or is that something else?

i appreciate the comments you are making about the modern diesels, gas would likely be less needed in a wizzbang diesel though as their efficiency is much better isnt it?
that was the point i was making, that for these systems to be competitive they need to be much more sophistacated than they are.
really need a standalone LPG ECU that knows what the diesel ECU is doing.

or a full standalone unit that does gas and diesel would be the best, but highly unlikely. not enough demand or money in the subject...

[KiwiBacon]

But really, is this pre-detonation an actual problem or just some theoretical issue that could under the right conditions happen in a crude and totally illegal homebrew system?

It's just another of those "purely theoretical situations" which happens to come true in a way that makes a 400kg engine sound like it contains 1000 angry elves with hammers.
It also makes it jump about like it's walking on scorpions.

So do you think it's worth worrying about?

[MightyMouse]

Multipoint is different from multiple injections / cycle.

As KiwiBacon is sort of elluding to those angry elves have a more sinister side - they are imposing significant mechanical stress on the engine internals.

Old engines were designed for this single "big bang" - but the down side was significant weight.

Weight savings ( and all the good things that come from this.... response, eceonomy etc... ) mean that stress must be controlled and obviously injecting the fuel incrementally per combustion cycle allows the cylinded pressure rise to be controlled.

Same or usually better outcomes without the penalities of tractor engine design. The down side to this is that the electronics involved in one of these engines is COMPLEX...... if your into petrol EFI then its all "ho hum" but if you've been playing with cams and plungers ( mech injecton ) then its all ging to come as a bit of a surprise.

In fact DI petrol and DI diesel are sort of comming together - a strange sort of twist !