12H-T performance Build (2 Viewers)

[gbentink]

You appear to be not giving due consideration to the impact of mixture ratios on the combustion event. Unless the engine is running in HCCI mode, which requires a different cam profile on the exhaust or a restricted exhaust port flow, the engine will detonate with rich mixtures. By rich I mean greater than ~ 60:1.........

When the mixture is leaner than 60:1 at std 18:1 compression, the fumigated mix needs a higher actication energy than the heat of compression alone to ignite the mix.

This is why (taken from a forum you suggested in an earlier post) the Engineering Gentleman Franz or something like that quoted for Sparki Ignition (SI) LPG engine something very interesting:

"One item that seems to be overlooked is that when a gaseous fueled engine goes lean, the exhaust temps also LOWER, and when the mixture goes rich, the temps RAISE. There is no evaporative effect with the gaseous fuel. I ran one 6.0 liter engine to L=1.9 and the exhaust temps were around 800 deg F! This was at 3000 rpm, 30 psi turbo boost, and about 290 bhp. Bringing the L ratio closer to 1.0 raised the power levels, but also the exhaust temps, and the NOx went through the roof!"

This is due to the same principles that he was able to run the SI engine at 30psi boost without detonation.

I don't know about your obviously scientific background, but mine includes chemistry with covered kinetcis of chemical reactions. What is observed in the above quoted based on practice is clearly what should be expected in theory.

Also note, the above referenced SI engine is likely as efficient as an equivalent Diesel, possibly moreso and ran at ~ 30:1 based on my rudimentary back calc……

Anyway, back to your last post, I propose that if combustion were to occur the way you state, detonation would always the inevitable result, and that is clearly not experienced in practice - what I see in practice is what I would expect based on my understanding of theory.

Note that I say "my understanding", this is a caveat......

Cheers

 

[gbentink]

getting the head back tomorrow from the head shop. Going to cost a fortune I think, because I used shorter valves, it was hard to escape coil bind. Anyway, they found another spring after a couple of hours of searching -thats when I got scared about the potential price. Time = Money...ouch....

 

[gbentink]

Hi Dougal, knocjk yourself out. Here are the figures for the lower power 2 valve head. The 4 valve makes 230kw... and a little more torque but holds it to 3000rpm.....

 

[gbentink]

Dougal, thanks for the links to those threads. Both read now.

I am interested in Franz comments about completely different combustion flame fronts etc. I maintain that the diesel ignites the lean propane mix below a "certain" propane AF ratio.

I am not certain what that number is, but expect less than 30% and likely around the 15-20% range dependant on compression pressure and temp. Impossible for me to model.....

There are alot of factors to be considered in order to make the engine safe with fumigation. Since I am intending to run high boost on thye 12H-T, the complexity to benefit ratio becomes an issue. For me to proceed, it will ONLY be after I max out the std pump and only after that I will setup LPG and Water injection concurrently.

I like simplicity under the hood. As it is there will be a huge amount of sensors etc......

Will revert next week hopefully with photo of head. Will do a dummy assembly after receipt of head and check piston to valve clearance.

A friend who installed a 1HZ cam in his 1HDT for performance (on my reccomendation :-( ) had to machine valve reliefs, but instead of small pockets where the valves are, he machined a flat circle over the whole piston. It blows white smoke on startup now but is OK hot. I expect this is due to a possibly lower cranking pressure with the cam in addition to the low compression....

In a nutshell, I do not want the white smoke on startup in mine......so I need to be very careful regarding the cam/piston etc etc....

Ahhh if only a vernier cam drive......

 

[Dougal]

You appear to be not giving due consideration to the impact of mixture ratios on the combustion event. Unless the engine is running in HCCI mode, which requires a different cam profile on the exhaust or a restricted exhaust port flow, the engine will detonate with rich mixtures. By rich I mean greater than ~ 60:1.........

When the mixture is leaner than 60:1 at std 18:1 compression, the fumigated mix needs a higher actication energy than the heat of compression alone to ignite the mix.

This is why (taken from a forum you suggested in an earlier post) the Engineering Gentleman Franz or something like that quoted for Sparki Ignition (SI) LPG engine something very interesting:

"One item that seems to be overlooked is that when a gaseous fueled engine goes lean, the exhaust temps also LOWER, and when the mixture goes rich, the temps RAISE. There is no evaporative effect with the gaseous fuel. I ran one 6.0 liter engine to L=1.9 and the exhaust temps were around 800 deg F! This was at 3000 rpm, 30 psi turbo boost, and about 290 bhp. Bringing the L ratio closer to 1.0 raised the power levels, but also the exhaust temps, and the NOx went through the roof!"

This is due to the same principles that he was able to run the SI engine at 30psi boost without detonation.

I don't know about your obviously scientific background, but mine includes chemistry with covered kinetcis of chemical reactions. What is observed in the above quoted based on practice is clearly what should be expected in theory.

Also note, the above referenced SI engine is likely as efficient as an equivalent Diesel, possibly moreso and ran at ~ 30:1 based on my rudimentary back calc……

Anyway, back to your last post, I propose that if combustion were to occur the way you state, detonation would always the inevitable result, and that is clearly not experienced in practice - what I see in practice is what I would expect based on my understanding of theory.

Note that I say "my understanding", this is a caveat......

Cheers

The observations of EGT's on gaseous engines are exactly the same as those experienced by diesel engines which run on th lean side of stoich. Pump excess air through a diesel and it cools down.

There certainly is detonation happening in practise, did you see catserveng's comments about higher metal counts in oil for fumigated engines?
Why do you think the compression ratio is reduced to such levels in dedicated LPG engines? Higher compression gives higher efficiency so a commercial power plant will always run as high as the fuel allows.
Some of the marketing literature alludes to the gas forming a plasma which causes earlier and more complete burn of the diesel. They carefully ignore any reference to preignition.

How can a gas be above it's autoignition temperature yet not have the activation energy required to ignite? It seems to me the energy required and the temperature required are one and the same.
A fumigated diesel engine is running in HCCI mode. The exhaust treatment and cam timing don't matter, only the charge that is trapped in the cylinder and how much it is compressed.

Do some calcs on adiabatic compression and compare those to the autoignition point of propane (published in supplier MSDS's). The adiabatic temps are right there without factoring in any charge heating, heat soak from the head etc.

On second thoughts regarding an engine detonating at idle, I'm convinced the "governor fuel cutoff" argument is bogus. As fuel will be cut equally from all cylinders and can't cause the noise or shaking experienced. Essentially under governor fuel cut it's behaving the same as the slow down from free-revving.
Next time I'll cut the fuel completely at idle and see if it still runs. Need more hands for that.

My background is I'm an engineer, plenty of thermodynamics but only a few years chemistry background. My chemistry reference books were stolen 10 years ago, I should replace them.

One other thing you need to consider is the safety aspect.
You're running maybe 50psi of air and propane mixed. If that vents either through hose rupture or back through the turbos (sudden throttle let-off) then that's a big volume of flammable gas you've just released.

Why do you consider the 1HZ a performance cam? Sure it'll have more duration, but it has to because the higher compression ratio limits valve lift.
I'd also be concerned about valve overlap and performance at lower revs under boost.

 

[gbentink]

Hi Dougal,

I read the comments on the extra metal and those results speak for themselves. I do not recall what the propane % was. If people are munching gearboxes and the like, (the truckies etc), they must be dumping in a heap of propane....

If there is detonation happening, I am not interested in running with fumigation. I drove in the 1HDT 1 hour ago; nothing audible happening there.

You only need to turn the key off to check if it runs without diesel. As for the argument being bogus, I look forward to hearing your results. Idle circuits are slow to react, this I expect is the reason for the shudder. Your test would be valuable I think. If you do check by cutting the fuel, I will state the obvious that with a steady state propane supply (as typical for a test like this), if the engine does detonate AND drop in revs, the AF ratio will be richer and there is big risk of damage. As I said, the obvious - I don't know how you planned to do the test, probably with a level of sophistication that I did not employ.

By the way, I also held revs steady at 2000rpm and could easily rev the engine with the propane, but I di not like the sound at all. Didnt sound like typical detonation, but may well have been. This is a really bad way to test, and driving a properly setup engine (ie: the 1HDT) displayed none of what I just mentioned.

I am wondering if the "phenomena" for lack of a better description of the combustion is actually closer to HCCI. Typically the HCCI event occurs near TDC and is over in ~ 1deg crank rotation and occurs simultaneously over the entire combustion chamber area.

The lift is greater in the 1HZ according to the manual. The duration is longer and the lobe centres allow a small amount of overlap.

Why do you say the higher compression limits the valve lift? The pistons share the same compression height and the the 1HZ has valve reliefs in the piston. The 1HDT would not turn over with the 1HZ cam.

I would expect that as you get leaner and use a gaseous fuel, the propensity to knock would be reduced....

By the way, I wasnt meaning to question your training, even though it may have sounded like that. Apologies for sounding arrogant. Your engineering prowess is evident in your posts, no question.

 

[Dougal]

I didn't think through the valve timing properly. With the combustion bowls in the HD-T pistons the squish will be similar.
I was thinking of two similar pistons with different compression ratios, clearly not the case.

The interesting part about pre-ignition in a diesel is the amount that is is rpm dependent. Anything that happens after TDC will be a whole lot smoother than the same event happening before TDC. At 700rpm idle it could take 3 times longer to get from ignition to TDC than it could at 2100rpm.
Of course diesel injection advances with revs so the timing difference will not be 3x as much.

At the time of my experiment I was running quite advanced injection timing, I don't know how many degrees crank angle but at the limit of manual pump rotation. Combusting lpg causing the diesel to burn even faster may have been exacerbating the knock.

But back to your 12HT.
What is the intake manifold like? Can you see any gains there?
Is a plenum and runner style manifold desirable for a boosted diesel?

 

[gbentink]

I would also like to say that my automotive experience is as a hobby, wheras your appears more as an occupation. So kudos to you for all your information and knowledge, I am happy to be proved wrong - that means I learnt something. I have many ideas, but limted experience.

For the most part people associate lean on a SI engine with higher EGT's. What they fail to consider is "leaner on which side of stoichiometric"??

My interest in the engined was simply that Franz managed to get the engine to spark with 30psi of boost and at 30:1 AF ratio - no mean feat! On the basis of actually being able to achieve it, then naturally the low temps would follow. Quite some time ago I wanted to do this in a daily driver but with limited operating range and complex control requirements I decided that I would only be able to achieve it in a stationary type situation - something I have no immediate requirement for

 

[gbentink]

Ahh…. The 12HT…..

The inlet manifold is a log type, tapered at the ends to encourage equal VE accorss the cylinders.

As such, there is no inlet runners to speak of except what is found in the head itself, and this accounts for ~ 8cm at a guess.

In the photos you can see I bell mouthed the inlets to reduce turbulent flow with the idea that the reduction of turbulent flow at the inlet should allow for a higher percentage of laminar flow and raise the CFM at which choke flow occurs. It is all theory, since I was not interested in taking measurements on this engine. I have to also make some improvements to the log manifold where the air enteres and remove the cold temp air heater (no glow plugs in this engine)

I have wonederd about tuned inlets many times. Since the tuning frequency is density dependant, it would be near impossible to find the "optimum" runner length. As a result, when looking in racing circles, log type manifolds with bell mouthed inlets are what is typically found. I think an ecu controlled continuously variable runner could realise gains but would be an expensive option; even just from the tuning hours…..

 

[gbentink]

All comments were related to turbocharging by the way - regarding runners, manifolds etc.

 

[Dougal]

Tuned inlet runners are only effective on NA engines. The speed of sound in air (speed of pressure waves) varies with temprature so it's impossible to tune for an engine with varying boost and inlet temps.

So I think a plenum which provides enough volume to fill each cylinder with short runners is called for. But I have no examples to base this on.

I don't do this for a living, I just do it as a hobby to keep my brain alive. My work is providing solutions to industry, the fluids and thermodynamics I encounter in a usual job don't get more complicated than pneumatic circuits.

 

[Dougal]

Hi Dougal, knocjk yourself out. Here are the figures for the lower power 2 valve head. The 4 valve makes 230kw... and a little more torque but holds it to 3000rpm.....

Just ran those figures, it's okay but I expected better.

The litres/hour they quote is not at the full engine power. So at 2500rpm (peak torque should be close to peak efficiency) the graphs read 18 litres/hour and about 66 kw.

BSFC comes out to 232 g/kwh.
I expected better but it's not under full load, only delivering approx 250Nm.

At 3200rpm it's delivering 365Nm for 34 litres per hour, that's almost the same at 237 g/kwh.
Only requires 10lb boost to deliver that with an AF ratio of 20.

Compare that to the Yanmar L100V (single cylinder 10hp industrial) which does better than 230 g/kwh from 1500-2500rpm.

*edit*
Well maybe comparing them to industrial isn't fair, I've just done the same graph conversion on the Isuzu 4BG1TCX marine engine, at 2600rpm it produces 110kw for 30 litres/hour. BSFC comes out at 232 g/kwh also. This engine in an industrial application does about 210 g/kwh.

 

[gbentink]

The 1.9 VW Tdi Does 235g/kwh @ 3300rpm, full load.

I am a little suprised it is worse at very light load - I calculate 257g/kwh @ 1900rpm.......

I designed and built extractors for an old Honda 1300 Coupe 9 (1300cc 2 valve OHC) Car. I used to collect them. Power went from 87hp to 106 Hp @ wheels. Torque went up a massive 30%. It was night and day driving experience. With decent valve overlap and a tuned exhaust, you can get the VE to be quite above the theoretical 100%.....

 

[Dougal]

The 1.9 VW Tdi Does 235g/kwh @ 3300rpm, full load.

I am a little suprised it is worse at very light load - I calculate 257g/kwh @ 1900rpm.......

I designed and built extractors for an old Honda 1300 Coupe 9 (1300cc 2 valve OHC) Car. I used to collect them. Power went from 87hp to 106 Hp @ wheels. Torque went up a massive 30%. It was night and day driving experience. With decent valve overlap and a tuned exhaust, you can get the VE to be quite above the theoretical 100%.....

That's an excellent result, especially in a car that light.
I've heard the S2000 does 109% for parts of it's rev range. Makes me laugh when ricers take off a factory tuned intake system and replace it with a pod filter.

 

[gbentink]

By "It" I meant the Marine 1HDT.

Not great economy!! I wouldnt have thought the Landcruiser varient would be any better either.....

Oh well, all the better the challenge!!

 

[Dougal]

By "It" I meant the Marine 1HDT.

Not great economy!! I wouldnt have thought the Landcruiser varient would be any better either.....

Oh well, all the better the challenge!!

As the load drops, your parasitic loads (friction, accessories etc) become a bigger proportion of the output so BSFC starts to really suck.
An idling engine has a BSFC of infinity.

 

[gbentink]

Actually the factory carburation consisted of 4 x 40mm Keihin CV type carbs with ~ 3--35cm runners. I didnt touch those, the difference came from the exhaust. I did a 4 -1 with 900mm long runners. All on the std cam too.

They are rated 116hp, 7300rpm @ flywheel without the alternator connected, so I thought I was doing well. peak torque went up massively in the 4500-6000rpm range and peak power which measured at 7300rpm was now at 8200rpm, so a really fat power curve.

I might get back into them one day - very impressive 1971 vehicle (air cooled, dry sump inline 4)

Dead stock origninal motor too!!!

 

[gbentink]

30-35cm runners, boy I really need to proof my tardy posts.

 

[gbentink]

Yeah, it shows how important it is to reduce parasitic losses. It would be interesting to know idle fuel consumption on various vehicles.

If I can get lowish EGT's and huge torque at ~ 1600rpm, I will be very happy since that should equate to a good bsfc result. I should aim for the 1.9 TDi EGT curve.

I did consider destroking my engine at one stage..... I just cant make up my mind. Maximize the potential capacity or make it less. I wont know what I want to do for sure until I get the twin turbo running!

You can get a 7psi back pulse in the exhaust, so in effect you can supercharge it in the resonance range - to good effect!

 

[gbentink]

I have been interested to try the "pressure wave supercharger"