Development history of cdi?

[BertTrack]

I'll fantasize here about how cdi's came to be and hope to unravel some of it's quirks.

Common diesel knowledge: Injected fuel evaporates and ignites because the temperature of the air inside the cylinder is high enough to do both jobs.

Good ignition of the fuel requires an even spread of fuel and no droplets. Requiring high pressure at the injector for a good spray that evaporates easily and evenly and can ignite the same. Let's assume the fuel is diesel. And that has standards. Like ignition temp/ ignition delay time etc.

First a step back. Direct injected big diesel engines.
Large volume. Much compressed air. A load of energy in that. A small cooled surface and a high volume of hot air. Easy to evaporate the fuel injected because there is plenty of heat to go around. (Evaporate then burn) These engines are generally kept hot. Or at least warm. A cold start situation hardly ever occurs. And if it does it may take a while before it gets ignition.

Then Direct injected smaller engines. Taking old vw tdi as an example, this has a much smaller volume and heat in it's compressed air, and the cooled surface is relatively high. This could cause a cold engine to not have enough heat in the air to maintain ignition temperatures if all the diesel is injected in one go (and has to evaporate too). Running glow plugs could help out a little heating the air more but it would be perfect if that could be the ignition source of the partly evaporated diesel which hasn't been able to ignite due to low air temperature.
But what if you inject at first shot some small amount of Diesel and allow that to evaporate and ignite before you inject the main fuel load? And that's what they did. Using a hydraulically controlled 2 way injector which opened first time at a low pressure (as the fuel pressure builds up) and then again the main opening at high pressure.
The fuel from the main injection will land in a warm bed of turbulent air which has possibly a burn going on and if not that at least it's hot and toasty.
One of the drawbacks of the hydraulically controlled injector was it's fuel pressure. At low rpm this isn't exactly as high as you'd want. Thus the pre-injection, already at lower pressure still needed the prolonged help of the glow plug.

I know VW came up with the TDI because Bosch wasn't sure if it would be able to deliver enough CDI units. But it's a nice step inbetween.

And i end up with the Merc 800cc 3 cylinder tiny diesel. With just 266cc per cylinder. Even smaller volume of heat (cold start anyone?) And even larger relative cooled surface.
This is where the cdi shines. Being able to deliver a high pressure at the injector even at low rpm's it allows the spray of fuel in the pre-injection to evaporate easily (big surface area) so that it doesn't eat too much heat through evaporation (and cooling of the air through the metal surrounding it). Thus it ignites. Heats up the air and increases the pressure enough to let the main fuel load do it's business. One of the side effects is that the uncontrolled burn period is shortened. Thus smoothing the already sharp knock from the CDI (which is normal).
The electronically controlled injector with ample fuel pressure, even at low rpm, allows the engine management to even inject at a time later than normal for those moments when you're idling with a cold engine and the pressure drop(heat loss) is very big. Which allows you to inject less fuel in the main injection and that way limit the knock of the cold cdi engine as well. (at higher rpm the knock isn't as prominent due to higher piston speeds. And the later injection isn't needed anymore anyway at higher speeds)

I'd be happy to hear extra information/corrections to update this basis.

[Sphere]

I don't know much in detail about diesel engines, but I'm pretty sure modern CDI injectors spread the volume of fuel over multiple discrete injection moments for the work stroke.

[tappy]

Interesting post.
What's the reason that the common rail systems don't like veg oil?
It strikes me that ordinary injection systems are volume-based, whereas common rail are pressure & time based. In the latter situation, the fuel viscosity would have a large effect on delivered quantity of fuel, so without viscosity measurement & compensation, the engine wouldn't fuel in the same way.
I guess if the common rail systems also use multiple injections to give even burn & reduced knock, then the timing of these would also be wrong for veg oil fuel rather than standardised diesel.
Other than that is there anything about the sealing & pumping systems on the common rail set-ups? What kind of rail pressure do they run at?

[pietenpol2002]

Thanks for furthering our understanding of that process. Very helpful. But it then raises for me a related question.

As the potential of a diesel engine is dependent in part on the amount of air you can force into the cylinder, we resort to turbochargers and superchargers to achieve that end. We also add intercoolers to force the air particles to shrink in size and huddle together so that we can squeeze more of them into the cylinder. If the diesel benefits from a heated charge it would seem that the undesirable heat (at least for petrol engines) typically generated by a supercharger would be to your advantage in a diesel.

Now, it makes sense that the intercooler is upstream of being compressed by the supercharger in that the air particles are all huddled together in the chilly air prior to being compressed. Thus, the positive displacement pump delivers the same volume of air to the engine (irrespective of density), thereby forcing a more densely charged volume into the cylinder. Am I understanding this correctly?

And finally, how can an air-to-air intercooler cool the air to anything below ambient temperatures, which is exactly the temperature at which the air is entering the intake, if the intercooler is upstream of the compressor? And now I'm wondering if this is better addressed over in Exhaust, Turbo, Cooling & Induction Related Topics.

[BertTrack]

Thank you all for replying! I'll try to build upon that.

I don't know much in detail about diesel engines, but I'm pretty sure modern CDI injectors spread the volume of fuel over multiple discrete injection moments for the work stroke.

Sphere. I think you're right. The only time I've seen that information though is in an Audi tdi film. And the odd text here and there. Up to 5 minute injections during the main burn phase i believe.

What's the reason that the common rail systems don't like veg oil? Other than that is there anything about the sealing & pumping systems on the common rail set-ups? What kind of rail pressure do they run at?

(Tappy's full post can be found further back in the thread)

Tappy as far as i've heared, read and seen in films, the high pressure pump of the cdi delivering 2000 bar easily has such tight clearings that even the acid of your skin will damage the pump surface enough to scrap it. Speculating on that: Veg oil draws in a lot of moisture and that could cause damage on the pump surfaces through corrosion.
I think you're right that the old injector system is more rugged. The common rail system being minutely controlled by a computer checking the r.p.m. of the engine every 2 milliseconds or so (speculating) will allow it to adjust it's injection time and possibly moment for the variable fuel viscosity. It should be able to pick up on that change in fuel quality and regulate the amount of fuel injected based on the effect it's having. (I'm thinking sub zero behaviour of diesel and an cold engine and still idling automatically at the desired r.p.m.)
My 1.2crti VW diesel has rail pressures between 1700's at idle and over 2500 bar when running higher loads. I can't really imagine that they wouldn't use other seals for common rail engines than before. (tdi). But the main problem is corrosion in the pump.

As the potential of a diesel engine is dependent in part on the amount of air you can force into the cylinder, we resort to turbochargers and superchargers to achieve that end.

(Pietenpol's full post can be found further back in the thread)

Pietenpol, setting up few steady state variables in/around the engine will help explore this. So let's assume that the engine is warm. And it's not running in an extreme environment. Then what remains is the compression ratio to come to the ideal inlet temperature which is around 15c in a diesel engine. The compression ratio is a given so to end up at 350c temperature 15c is in the middle of the range before compression. Of course this is only for a warm engine, where the heat loss through the cylinder/piston area is much less than with a cold engine. Requiring pre-injection shots with a cold engine e.t.c. With a cold engine it may well help to have a hotter than normal charge. (Big rig pre heaters in the US heat up the charge air usually) When you want power however you will want to keep the inlet temperature around 15c. (Compression ratio is set-up to operate in that area) because as you've said already this will allow more air to be pushed into the cylinder. And so you can also burn more fuel.

Your supercharger assumption is I think correct. The amount of air delivered by the supercharger can be changed only by changing the gearing between engine and supercharger. This compressed air will be heated by it's compression and would ideally be cooled. Except for losses in supercharger and resistance increasing with pressure the amount delivered by the supercharger should be the "same" per engine r.p.m.
It will also be charging the air at idle (when you don't need it). But at any time when you give it more fuel the extra amount of air is already available. Trading efficiency for instant response. The ideal inlet temperature stays the same since the compression ratio doesn't change.

The inter cooler(air to air) can not cool the air below ambient. But since it's placed behind a compressor of some sorts it already has a lot of temperature to dump into the ambient air as it is. I'm seeing 50c temperatures on my inlet piece of my car when I'm riding in 35c conditions at normal speeds.(after inter cooler but inside an insulated engine compartment)


I've come across this very good information on the topic of diesels and charged air online http://www.thermoguard.com.au/tech.html I wish he had continued the series.

[Blunt Eversmoke]

As for the heat of the super/turbocharger being useful for diesel operation: When starting, of course, supercharger heat (if any available) is useful. However, after starting it is harmful, lowering charge density.
There is an exception to that rule, though: Once you go to really low compression ratio, it becomes different. I mean low like 16:1 on atmospheric diesels of some old Harvest International tractors, 14:1 for the Mazda Skyactiv D, 13:1 on Charomski two-stroke diesels used on Soviet TB-7 long range bomber of WWII era, 12:1 on the diesel Lanz Bulldogs, and, of course, extremes like 7:1 in hyperbar diesels of the french Leclerc tank or 9:1 for some US hyperbar aero diesel project from the 1970s. In these conditions, it may make sense not to cool the charge air until sufficient boost figures are reached so the fuel can evaporate and burn more completely and fast.

For the aero diesels, this moment of "sufficient" may never come because of the colder air, especially at high altiitudes; the TB-7, for instance, was doomed to fail because the one engine powering one supercharger to boost all engines was not mounted on some of the planes due to some short-sighted decision, (like "We aint got enough engines for one bomber, let's take the supercharger engines of the six that we've got!") causing the other engines to fail to ignite because they never reached ignition temperature.
For the Leclerc, intercooling becomes a must at the latest when its engine reaches full boost of 7:1 (yikes!).
For the Skyactiv D, Mazda went a different way, using ceramic glow plugs which operate constantly, and re-inhaling some hot exhaust through the exhaust valve when at low boost. The plugs are the reason the Skyactiv D injects relatively late (around TDC) in most RPM conditions.
The non-charged HI tractors used an intake manifold that shared a rather long wall with the exhaust, which made for reliable operation especially in winter, but made the charge less dense and thus cost them some power.
The Lanz Bulldog runs at pretty low RPMs even as a diesel and has an uncooled cylinder head (not glowing like the hot-bulb ones, tho), which also makes it digest crudely-filtered WVO and WMO without problems (it's to burn the really heavy oils you need the older hot-bulb engine designs for).


And now for something (not so) completely different: The early 20th century four-stroke Monovalve diesel engine from the US used, as the name suggests, only one valve per cylinder for gas exchange. Exhaust gas went into a straight wide pipe, its inertia pulling fresh air charge through a narrower passage to the air filter for the engine to suck in through the valve. The air warmed up to some degree by the heat of the exhaust pipe, but was still cool enough to cool the valve, prolonging valve life.
Power was 75 HP at 800 RPM and 100 HP at 1000 RPM. One can imagine it could be higher if they made a resonator exhaust pipe for it...
Oh, did I mention it was a side-valve diesel?