gemellocattivo.com

Which means "Evil Twin". Lets see your projects where you change boring into fun or create the fun from scratch.
It is currently Thu Mar 28, 2024 10:11 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 29 posts ]  Go to page Previous  1, 2, 3
Author Message
 Post subject: Re: Common Rail Diesel
PostPosted: Wed Apr 13, 2016 2:30 pm 
Offline

Joined: Tue Apr 12, 2016 5:48 pm
Posts: 9
Quote:
*Converting injection volume to pulse width is a table from volume and rail pressure. In GDI the rail pressure is sampled every TDC but I've seen some Diesel focused prototype controllers which sample it immediately before each injection. Not sure if it's really necessary to sample more frequently.


This is actually one of the Achilles heals of the diesel engine. You can not imagine the amount of work that goes into this subject. Usually the pressure waves are modeled and compensated for. With only one injection everything is simpler.

Quote:
*Fuel temperature sensor is used to convert fuel mass to injection volume with a table of fuel density


Add a wide band lambda sensor and try to get one of theese or similar http://beru.federalmogul.com/en-gb/node/262 they are in production and will be the by far cheapest way to get pressure curves. (You need cylinder pressure (way quicker application work) for peak pressure and MBF50) Google sensata cpos get some samples and give me one :).

Quote:
*With only one injection, the rail pressure, start of injection, and injection mass are all determined by torque vs RPM surfaces. Rail pressure is primarily varied by load, SOI primarily varied by RPM, and injection mass by load, but all are surfaces which can vary by both speed and load.


Skip the torque modeling. 1mg/stroke =~5-6Nm. Do all tables with mg/stroke and rpm. Do a curve with accelerator pedal position to mg/stroke. The peak cylinder pressure is greatly affected by SOI. To much advance on high load is about the only way to quickly kill a diesel.

Quote:
*Soot limit surface used to limit immediate torque request with respect to air mass but torque scheduling is still done with respect to max boost torque


Do constant lambda do not enrich further as fuel consumption will suffer if this is for a fuel economy competition.


Quote:
The project is for an engineering school fuel economy/emissions competition, so emissions can't be too bad (EPA tier 4 off-road is our benchmark, but with a higher soot limit). We are trying to get by with only a DOC catalyst, but can also add a partial-flow DPF (which can't clog and doesn't require active regeneration). Fuel economy is more important than emissions, so a NOx increase is okay.


EGR does wonders with NOx.. High enough NOx/conservative smoke limiter no DPF needed. Which engine?


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Wed Apr 13, 2016 2:32 pm 
Offline

Joined: Tue Apr 12, 2016 5:48 pm
Posts: 9
With cylinder pressure and a dyno with something like this https://www.avl.com/documents/10138/269 ... el+Balance
the application work would take a matter of days.


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Wed Apr 13, 2016 3:38 pm 
Offline

Joined: Sun Oct 11, 2015 2:07 pm
Posts: 134
The engine we are looking at is a 'smart car' 800cc 3 cylinder, 2005-2007 generation. With an I3, I'm not sure if the pressure waves will be as terrible, since the injections are spread out so far.

For dyno/instrumentation, I use a water brake dyno with our current gas engine. The team is switching from the spark ignition to the diesel class next year, so I already access to a test cell for the gas engine (water brake, 5-gas analyzer and fuel flow/density meter). I also have many wideband sensors and controllers.

I don't have cylinder pressure sensors yet, but I expect to need them.

For gas controls I calculate air mass into the engine as one of the critical calculations. For diesel I was thinking of soot limit based on this air mass calculation, which would require a lambda sensor for cal only.

For gas we currently do a sort of torque control, so it makes sense to continue doing it. Switching from a pedal->throttle table to an intermediate normalized calculation was a great improvement, although with diesel the fuel mass is much closer to torque than throttle position is so an intermediate calculation might not be necessary.


With pressure rise rate limits at high load, would a pilot injection help to reduce the pressure rise rate/is a pilot injection normally used to extend high load operation?


For emissions, the engine already has basic EGR. Since the engine can pass EU4 levels already, and we don't need to be that good, we can probably get away without EGR, or only EGR at very low loads/idle. From what I've seen, it looks like EGR control for diesels is based on a minimum lambda vs speed/load table, so it allows as much EGR as possible using either lambda calculated from MAF or a lambda sensor.

_________________
"Sometimes, the elegant implementation is a function. Not a method. Not a class. Not a framework. Just a function." ~ John Carmack

"Any sufficiently advanced technology is indistinguishable from magic" ~Arthur C. Clarke


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 1:01 pm 
Offline

Joined: Tue Apr 12, 2016 5:48 pm
Posts: 9
Quote:
The engine we are looking at is a 'smart car' 800cc 3 cylinder, 2005-2007 generation. With an I3, I'm not sure if the pressure waves will be as terrible, since the injections are spread out so far.


I would agree especially with only one injection

Quote:
I don't have cylinder pressure sensors yet, but I expect to need them.


It would be a great help.

Quote:
For gas controls I calculate air mass into the engine as one of the critical calculations. For diesel I was thinking of soot limit based on this air mass calculation, which would require a lambda sensor for cal only.


If you dont use egr you can just do fuel/boost pressure. For egr you need to have a maf sensor.

Quote:
For gas we currently do a sort of torque control, so it makes sense to continue doing it.


Frankly, for your needs I do not see the point.

Quote:
With pressure rise rate limits at high load, would a pilot injection help to reduce the pressure rise rate


Yes of course.

Quote:
/is a pilot injection normally used to extend high load operation?


The opposite actually.

Quote:
For emissions, the engine already has basic EGR. Since the engine can pass EU4 levels already, and we don't need to be that good, we can probably get away without EGR, or only EGR at very low loads/idle.


I do not know what limits you have but a bit of egr will decrease nox by ~50% and not raise bsfc.

Quote:
From what I've seen, it looks like EGR control for diesels is based on a minimum lambda vs speed/load table,


Not from what I have seen.

Quote:
so it allows as much EGR as possible using either lambda calculated from MAF or a lambda sensor.


What if you think like this: Do a mapping without egr. Measure the air flow with the maf sensor. Open up the egr valve gradually (Need to check the differential pressure, make a model.). The difference in air flow is the amount of egr. Increase egr until your emission levels are reached, bsfc increasing and or soot is increasing (You need a smokemeter or a opacimeter).

So you need a maps with speed and fuel weight/stroke on the axis with:

Boost pressure.
Rail pressure.
SOI
Air mass (This is egr).

The lambda sensor is not needed, only for calibration. It really is that simple.

You will probably come to the conclusion that you want high rail pressure, advanced timing, low boost pressure and low levels of egr. This way of calibrating is great for bsfc not so for NVH and emissions..


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 2:22 pm 
Offline

Joined: Sun Oct 11, 2015 2:07 pm
Posts: 134
Quote:
What if you think like this: Do a mapping without egr. Measure the air flow with the maf sensor. Open up the egr valve gradually (Need to check the differential pressure, make a model.). The difference in air flow is the amount of egr. Increase egr until your emission levels are reached, bsfc increasing and or soot is increasing (You need a smokemeter or a opacimeter).


This is similar to the air model we use with gas. With gas engines, we map the engines airflow vs speed and pressure and use this as the engine load. Comparing this to MAF would show the % of gas that is EGR, but not the % of gas into the engine that is burnable (since EGR is partially burnable with lean combustion).

Quote:
You will probably come to the conclusion that you want high rail pressure, advanced timing, low boost pressure and low levels of egr. This way of calibrating is great for bsfc not so for NVH and emissions.


NVH is not really a concern. The competition's noise component measures pass-by sound pressure at 35mph. Emissions are a concern, but because of the scoring formulas not nearly as important as fuel economy and the minimum performance in emissions is very low.

_________________
"Sometimes, the elegant implementation is a function. Not a method. Not a class. Not a framework. Just a function." ~ John Carmack

"Any sufficiently advanced technology is indistinguishable from magic" ~Arthur C. Clarke


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 3:02 pm 
Offline

Joined: Tue Apr 12, 2016 5:48 pm
Posts: 9
Quote:
but not the % of gas into the engine that is burnable (since EGR is partially burnable with lean combustion).


If you like to add complexity where it is not needed, go for it!

Quote:
Emissions are a concern, but because of the scoring formulas not nearly as important as fuel economy and the minimum performance in emissions is very low.


I do not know anything about the comp but low diesel emissions is kind of in the news recently..

Getting a decade old smart car engine to run is not a big feat. Doing the same WITH low emissons is one!


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 3:23 pm 
Offline

Joined: Sun Oct 11, 2015 2:07 pm
Posts: 134
glen999 wrote:
If you like to add complexity where it is not needed, go for it!

On gas engines, it's important to know the burnable air to calculate fuel, so I'm coming from that. I guess it isn't really required if burnable air doesn't need to be known.

glen999 wrote:
Getting a decade old smart car engine to run is not a big feat. Doing the same WITH low emissons is one!


This is a 'clean snowmobile' competition. Most snowmobiles have 2-stroke engines with terrible CO and HC emissions and terrible fuel economy. The EPA limits for them are complicated, but the NPS requirement (national parks service - to ride in a national park) is set at 120 g/kWh CO and 15 g/kWh HC with no NOx requirement. For comparison the EPA Tier 4 off-road limit is 5 g/kWh CO and 4.7 g/kWh HC+NOx. The competition has SI, CI, and EV classes. In the CI class, most teams start with a ~20-40kw mechanically injected Diesel already certified to Tier 4 and improve from there, usually adding a DOC at minimum. There are so few engines small enough to fit in a snowmobile that make enough power, hence the smart car engine (which is the only engine which packages in a very tiny space and makes 40hp).

The EPA formula for emissions calculates a 'score' from 0 to 210. The competition adds NOx to HC in the EPA formula. Minimum performance in the competition requires an EPA score of 176 (equivalent to the NPS requirement), but EPA tier 4 would score 205.6. So, improving emissions will go from 206 to 207 which is worth almost nothing. It's still a MASSIVE improvement compared to snowmobiles, but not so good compared to on-road engines.

_________________
"Sometimes, the elegant implementation is a function. Not a method. Not a class. Not a framework. Just a function." ~ John Carmack

"Any sufficiently advanced technology is indistinguishable from magic" ~Arthur C. Clarke


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 3:44 pm 
Offline

Joined: Tue Apr 12, 2016 5:48 pm
Posts: 9
Quote:
I guess it isn't really required if burnable air doesn't need to be known.


I did not say that. But the oxygen content in the egr gasses will be roughly the same so not much of a point calculating it. Main thing with a diesel is that its the fuel that sets the torque and not the air as on a otto. If you go from lambda 2 to 4 the torque would not differ that much.

Quote:
This is a 'clean snowmobile' competition. Most snowmobiles have 2-stroke engines with terrible CO and HC emissions and terrible fuel economy. The EPA limits for them are complicated, but the NPS requirement (national parks service - to ride in a national park) is set at 120 g/kWh CO and 15 g/kWh HC with no NOx requirement. For comparison the EPA Tier 4 off-road limit is 5 g/kWh CO and 4.7 g/kWh HC+NOx. The competition has SI, CI, and EV classes. In the CI class, most teams start with a ~20-40kw mechanically injected Diesel already certified to Tier 4 and improve from there, usually adding a DOC at minimum. There are so few engines small enough to fit in a snowmobile that make enough power, hence the smart car engine (which is the only engine which packages in a very tiny space and makes 40hp).


I was guessing some sort of eco marathon vehicle. So performance is a issue.. You need a smoke limiter that reduces fuel during transients. Try to run on a lambda level that you feel comfortable with (Please dont run below 1, thats for trucks with mullet equipped owners.). Use a doc. If you run with a lot of NOx, HC/CO will be very low.
You problably want the glow plugs functional.

Quote:
The EPA formula for emissions calculates a 'score' from 0 to 210. The competition adds NOx to HC in the EPA formula. Minimum performance in the competition requires an EPA score of 176 (equivalent to the NPS requirement), but EPA tier 4 would score 205.6. So, improving emissions will go from 206 to 207 which is worth almost nothing. It's still a MASSIVE improvement compared to snowmobiles, but not so good compared to on-road engines.


Point taken.


Top
 Profile Send private message  
 
 Post subject: Re: Common Rail Diesel
PostPosted: Thu Apr 14, 2016 4:13 pm 
Offline

Joined: Sun Oct 11, 2015 2:07 pm
Posts: 134
We don't try to run past 1 lambda with SI. Usually we run 1.15-1.35.

Smoke limiting is the biggest issue with the mechanical injected engines now, and a big reason to use a common rail engine.

Planning on using a doc.

_________________
"Sometimes, the elegant implementation is a function. Not a method. Not a class. Not a framework. Just a function." ~ John Carmack

"Any sufficiently advanced technology is indistinguishable from magic" ~Arthur C. Clarke


Top
 Profile Send private message  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 29 posts ]  Go to page Previous  1, 2, 3

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group