[dhanushs]

With increasing number of diesel vehicles in India, and the majority of them being crdi technology, I thought of starting a thread, to shed some insight over the topic. I bet most of us would love to know whats happening under the hood: Common Rail Diesel Injection System.

Recently, I was discussing with my friend at Bosch regarding the fuel consumed while coasting in gear. It was then he mentioned that, there are 7 types of injection in a Common rail diesel injection system.

Broadly classified into three.

• Pre
• Main
• Post

Pre: Mostly, there are 3 types of pre injection. One of it is pilot injection. The pilot injection (as I was told) was to create a perfect burning environment for the main injection. And to avoid any hot spots and make the combustion even. Another type of pre injection is used to (As I was said) control the NVH of the engine (I'm unclear of this).

Main: As termed, the main injection, which causes the combustion.

Post: Again, mostly, there are 3 types of post injection. I'm not so clear on this regard, but then, the post injections are used to regulate dpf, and also said to be used to control the temperature. And also, while coasting, if there is any kind of injection, then its post injection.


I have shared my tiny bit of knowledge on the subject. Now waiting for the guru's to share the major part.


P.S - There might be a risk, that this thread may turn extremely technical, and like most of fellow members, I would be a good listener.

P.P.S - The info I have gained may not be concrete. Please feel free to correct me.

@Mods: Didn't find an appropriate thread on search, hence starting one.

[vina]

Originally Posted by dhanushs

...

Post: Again, mostly, there are 3 types of post injection. I'm not so clear on this regard, but then, the post injections are used to regulate dpf, and also said to be used to control the temperature. And also, while coasting, if there is any kind of injection, then its post injection.

....

That may be to provide some fuel to keep the catcon warm enough to be useful when coasting ends.

By the way, I hope you don't mind but I'll ask another question regarding CRDI technique not related to injection per se but to fuel pump:

First some background (to show where I'm coming from):

In non CRDI engines, the fuel pumps generate pressure just at the time of injection. the CRDI differs in that there is one high-pressure fuel pump that feeds an "accumulator" (i.e. the common rail - acts as both an accumulation and distribution mechanism for fuel) and the accumulator keeps the diesel at high pressure till the "injectors" (which are complex and rightly glorified faucets) let it into the cyliders.


My question is - diesel being practically incompressible, the moment even a tiny bit is removed from the "accumulator" the pressure should drop drastically. Then how does this accumulation thing work? Is it that the fuel-pump for CRDI also generates pressure right before and during injection? Or is it that the common-rail expands a little, and the extra volume keeps the diesel at high pressure? Or is there a third explanation (e.g. there really is a very high pressure accumulator somewhere)?

[Mr.Boss]

Post deleted by the Team-BHP Support : Typing a message which is not a value-addition to the topic being discussed, or is just entirely irrelevant is considered as “Spam” on this community. Spamming is STRICTLY prohibited on Team-BHP.

We advise you to read the Announcements and Board Rules section before proceeding any further.

[vina]

A good video on CRDI

YouTube - ‪Diesel Common Rail Injection Facts 1‬‏

YouTube - ‪Diesel Common Rail Injection Facts 2‬‏

there are other videos in suggested links on the youtube pages

[DerAlte]

Originally Posted by vina

... diesel being practically incompressible, the moment even a tiny bit is removed from the "accumulator" the pressure should drop drastically. Then how does this accumulation thing work? Is it that the fuel-pump for CRDI also generates pressure right before and during injection? ...

@Dr.Vina, why should 'a tiny bit' of volume reduction (fuel taken away due to injection) cause a 'drastic' change in pressure?

The video you put up should have answered your question - this is a continuous flow system, not linked with injection timing. Pressurizing the 'accumulator' (rail) is asynchronous with injection. And the 'post' injection is not for keeping the catcon warm - the catcon is anyway 'hot' due to the accumulated waste heat from the previous cycles!

@dhanushs, I think the 'post' injection is for cleaning / burning up some of the by-products of combustion (emission reduction). Could you please check with your friend? Also about how many times does the nozzle spray (it is definitely more than 3 times) for every cycle? I have a feeling this is proprietary Bosch information, so discretion is advised for him.

[vina]

Originally Posted by vina

My question is - diesel being practically incompressible, the moment even a tiny bit is removed from the "accumulator" the pressure should drop drastically. Then how does this accumulation thing work?

...

I may not have explained what I mean.

Let's take the example of a 1ltr jerry can and let's say we want to fill it with air till pressure reaches 1MPa (roughly 10atmos)

Assumeing a rigid jerry can, while we fill in air, it is getting compressed, so more mass of air is getting in as pressure increases to 1MPa. Once the pressure reaches 1MPa let us say the can is sealed.

Now imagine the can develops a leak, and equivalent of 10mL of air @ 1MPa pressure moves out - this will lead to a slight drop in pressure of the air. But drop in pressure will increase the volume of the remaining air.

A crude calculation will show that pressure will reduce by about 1% as long as temperature doesn't change much (which it wouldn't)


Now consider the same experiment but this time with water. One you have filled the can fully with water upto 1atmos (i.e, 0.1MPa) to increase the pressure hardly any more water needs to be injected - can is rigid and water is incompressible.

So somehow you reach 1MPa. Now let's say can has a leak and 10mL leaks out - given that after adding the first 1L of water you probably didn't even have to add a full 1mL to the can to increase the pressure (water is practically incompressible - so any forced introduction of water will increase pressure very fast), a 10mL leak, if it happens, will lead to the can deforming or the pressure reducing drastically.



Apply the same principle to diesel (instead of water) and common-rail (instead of rigid jerry can) and if the common rail is truly rigid, then unless the pump is pumping right at the time of injection pressure should fall in the midst of injection.

[figo_mba]

Originally Posted by vina

I may not have explained what I mean.

Let's take the example of a 1ltr jerry can and let's say we want to fill it with air till pressure reaches 1MPa (roughly 10atmos)

Assumeing a rigid jerry can, while we fill in air, it is getting compressed, so more mass of air is getting in as pressure increases to 1MPa. Once the pressure reaches 1MPa let us say the can is sealed.

Now imagine the can develops a leak, and equivalent of 10mL of air @ 1MPa pressure moves out - this will lead to a slight drop in pressure of the air. But drop in pressure will increase the volume of the remaining air.

A crude calculation will show that pressure will reduce by about 1% as long as temperature doesn't change much (which it wouldn't)


Now consider the same experiment but this time with water. One you have filled the can fully with water upto 1atmos (i.e, 0.1MPa) to increase the pressure hardly any more water needs to be injected - can is rigid and water is incompressible.

So somehow you reach 1MPa. Now let's say can has a leak and 10mL leaks out - given that after adding the first 1L of water you probably didn't even have to add a full 1mL to the can to increase the pressure (water is practically incompressible - so any forced introduction of water will increase pressure very fast), a 10mL leak, if it happens, will lead to the can deforming or the pressure reducing drastically.



Apply the same principle to diesel (instead of water) and common-rail (instead of rigid jerry can) and if the common rail is truly rigid, then unless the pump is pumping right at the time of injection pressure should fall in the midst of injection.

I am just thinking aloud here.
Instead of having a hole in the jerrycan after you have reached 1MPa.

If we have a jerry can with a 1mm dia hole and we try to push in air with a high pressure pump. Will the pressure inside the jerry can will remain constant ?

[DerAlte]

Originally Posted by vina

I may not have explained what I mean.
...
Apply the same principle to diesel (instead of water) and common-rail (instead of rigid jerry can) and if the common rail is truly rigid, then unless the pump is pumping right at the time of injection pressure should fall in the midst of injection.

@vina, your method of going reverse from engineering to scientific principles behind it (that's exactly what you are trying to do) is rather flawed:
1. You can't / mustn't use approximations and shortcuts - this will cause more misunderstanding than understanding. If one were to do the same while going from scientific principles to technology to engineering, it will result in a product like Frankenstein!
2. Disbelief gives you much less understanding than belief
3. Trying to disprove a current working technology principle (since it does not fit the paradigm you chose from the science world) takes you further away from understanding. If at all, the onus of proving it is on you (go ahead, you can do it - write an appropriate equation without ignoring any parameter )

* 'Practically incompressible' and 'absolutely incompressible' are not the same
* Visualize a flowing fluid dynamic (not static) system at 1200bar, independent of injection sipping from it
* Take another look at the videos you put up - and consider what the pressure regulator and fuel return sub-systems do

BTW, anything above 800bar is good enough for finely controllable injection at the injector. So, if the rail pressure is 1200bar (or even varies 1100-1200bar), there is still a margin of 300-400bar to play around with. Now please work out with your equation how much pressure drop you will get with the injector opening for a few microseconds at a time.

[vina]

Originally Posted by figo_mba

I am just thinking aloud here.
Instead of having a hole in the jerrycan after you have reached 1MPa.

If we have a jerry can with a 1mm dia hole and we try to push in air with a high pressure pump. Will the pressure inside the jerry can will remain constant ?

If by high pressure pump you mean a pump that generates pressure more than what is already in the can - then yes the pressure will increase further.

PV=nRT, T is roughly constant, R is a constant, V is constant provided jerry can is rigid, and n will increase if the pump injects more air.

So P must increase.

[dhanushs]

Originally Posted by vina

So P must increase.

Vina, I think Figo mba has your answer.

Infact, while the engine is running, the pressure constantly 'leaks' through the four injectors, and the high pressure pump remains active all through to provide pressure. Further, the pressure 'leakage' through injectors is a known value and is predictable, hence, IMO, by using the pressure regulator, rail pressure can be kept as a constant. Diesel is continuously flowing through the system.

Originally Posted by DerAlte

@dhanushs, I think the 'post' injection is for cleaning / burning up some of the by-products of combustion (emission reduction).

Yes, the post injection is mainly used for DPF (Diesel Particulate Filter) regeneration.

Originally Posted by DerAlte

Also about how many times does the nozzle spray (it is definitely more than 3 times) for every cycle?

There are basically 7 types of injections and in some sophisticated systems, 10 types. But then, in most common cars, in a cycle only 5 types are used to the max. ie, at the max, an injector can spray 5 times/cycle.

[vina]

Originally Posted by dhanushs

...

Further, the pressure 'leakage' through injectors is a known value and is predictable, hence, IMO, by using the pressure regulator, rail pressure can be kept as a constant. Diesel is continuously flowing through the system.

...

Is there a pressure regulator in the system? I thought it was only a pressure limiter (valve opens to let fuel out if pressure increases too much) - but if the system has some small storage "enough for one or two stroke's worth of injected fuel) that probably will do the trick

[DerAlte]

Originally Posted by dhanushs

... Yes, the post injection is mainly used for DPF (Diesel Particulate Filter) regeneration. ...

By NOx generation or some other method?

Originally Posted by vina

Is there a pressure regulator in the system? I thought it was only a pressure limiter ...

What is the difference?

[vina]

Originally Posted by DerAlte

What is the difference?

Pressure (or similar quantities) can be limited in many ways, limiters - bleed out excess fluid (steam for example in boilers - but also water sometimes) or store the fuid at constant pressure. e.g inverted dome type regulators for gas-storage (one picutre attached here from Bio Gas Plants you can find it in school text books) - the weight of the moving dome and horizontal cross-section area is constant so pressure of the gas remains constant - when you fill gas dome moves up, when you take gas out, dome moves down. The second systems are more typically called regulators, though the terms are interchanged often in case of things like boilers (with good reason - see below)

Another example of a pressure regulator is a common sight - municipal water tanks (even overhead tanks in houses) - the pressure is generate by the head of water between the outlet and the storage - as water flows out, since the cross-section of outlet is much smaller than that for the tank, even when water flows out the heaf doesn't change much - you get practically constant pressure of water.


A similar principle was used in lab burettes, but to prevent liquid from reaching the mouth of the lab technician (not really a regulator ONLY - serves many functions at once) - the (small) second bulb after the main bulb in the middle - the second bulb would provide storage for short duration, while allowing the guy sucking on the tube to maintain enough pressure not to allow the fluid from draining out of the tube again. The pressure doesn't drop, in fact because as liquid is sucked into the bulb the cross section of its surface increases even fast sucking through the tube wouldn't let pressure build up (or in this case down) quickly.





The bleeding limiter is sufficient as a regulator when the fluid is highly compressible, the storage unit attached is large and fluid outflow over short periods of time is not enough to let the pressure drop very fast (for incompressible fluids - pressure will drop very fast even if miniscule amounts of fluid are taken out, and the voulme of the container can't change). This is true for gas storage in units where gas is being generated by some process usually.



The regulators, by providing variable volume (at roughly constant pressure), can handle situations where inflow-outflow will not be matched in time, but pressure must be maintained - almost all high pressure liquid storage systems use some form because most liquids are highly incompressible.




Even gas tanks use proper regulators (rather than limiters) in situations where constant pressure is needed but there is no possibility of an inflow in any reasonable time. E.g. nitrogen or oxygen cylinders (LPG cylinders maintain pressure by the fact that at high pressure the gas just turns into a liquid - a on/off valve is enough to get a pretty good pressure regulation for the purpose served - kitchen burner fuel pressure regulation. This is also similar to the situation discussed above where gas is generated inside the cylinder by boiling of the liquid at room temperature the moment pressure is released) used in industrial processes. In such situations gas is invariably filled at much higher pressures than eventually needed and the regulator at the output maintains the flow (for flow regulators) irrespective of the pressure at the output or the pressure of the output gas (more advanced regulator systems - they can maintain the head pressure of outflowing gas)

[vina]

Further to the doubt I had expressed earlier - I found the bulk modulus of Jet Fuel on Physics Forum (they say diesel should be similar).

Attached here.


Now two pieces of puzzle remain - can somebody tell me (1) how much fuel is (milligrams - typical and max) injected in an exhaust stroke into the cylinder (per ltr - if you are unsure, let us know the figure and the engine). and (2) what is roughly the volume of a typical common rail (i.e. how much diesel does it carry all by itself)



EDIT:
Here's the abstract of a paper (couldn't download - need cash for that - if someone has information that is accessible free kindly let me know - mine is a non-commercial, hobby/curiosity only usage)

The Impact of the Bulk Modulus of Diesel Fuels on Fuel Injection Timing - Energy & Fuels (ACS Publications)

EDIT2:

attaching another paper I found on basics of diesel engines. This has some info on compressibility (based on google - I haven't read it at all)

[vina]

Here's a very good presentation from the internet with lots of technical data:

http://www.akautomotivetraining.co.u...on_000.pdf.pdf


For a Fiesta engine with diesel (from tank) consumption of about 0.1mL/stroke, with a roughly 20mL storage in the rail, volume of the rail will decrease (assuming constant pressure) by less than 0.5% (0.1mL was at 1bar, 20mL is at 2000bar). With diesel bulk modulus of about 1GPa, this means a drop in pressure of 5MPa or about 50bar - not much really, even if rail pressure was as low as 500bar.


By the way, ECU seems to control the rail pressure in accordance with engine conditions.