Warmup time over the NEDC & FTP-75 drive cycles 1

[murpia]

I'm putting together a simulation of the energy budget for a car driving the NEDC and FTP-75 emission drive cycles. I need to incorporate the effect of warmup enrichment for gasoline and diesel powertrains.

As a first approximation, I plan to simulate a simple heat capacity model of the engine thermal mass.

Can anyone help with data or references regarding how long (in seconds) a typical powertrain takes to warm up from cold to operating temperature on the two drive cycles?

Thanks in advance, Ian

 

[SomptingGuy]

Oh c'mon, they're all different. Modern diesels have a heater to supplement the heat-rejection heating at startup. How long is a not generally an output, it's an input to the system.

- Steve

 

[mattsooty]

murpia

how would you define a 'typical powertrain' ???

MS

 

[BrianPetersen]

Aside from the enormous scaling differences between the powertrain between a smart and a Ferrari that preclude "typical" from being of any use whatsoever ... diesels don't do warm-up enrichment. They'll use a little more after a cold start because of greater heat losses, more friction, glow plugs are often left on for a period of time after the engine starts which increases electrical demand, etc., but by no means are they running "rich".

And, modern gasoline engines with good fuel injection also have minimum enrichment. It might be a smidge rich for the first couple of revolutions to make sure the engine starts, and it won't be able to run in closed-loop until the oxygen sensor starts giving a signal (and those are electrically heated and often very close to the engine, so this takes seconds nowadays), but given that the HC and CO emissions during cold start "count", they lean out ASAP after startup. Modern fuel injection minimizes (almost eliminates) "puddling" that necessitated enrichment in the old carbureted days. Modern fuel injection is capable of taking an extremely accurate "guess" at the required fuel volume even in the absence of the oxygen sensor signal. They'll use more fuel for the first little bit for all the same reasons that diesels do (above) with the exception of the glow-plug heating bit.

 

[JSteve2]

Most modern diesels don't have "glow plugs" anymore. And they typically adjust timing rather than fueling to make up for cold conditions, except for maybe a very transient early period. As stated above, most systems cannot afford to run an extensive cold cycle and still meet emissions, so they will try to line out to nominal conditions early.

Also, extreme cold conditions may not be emission controlled (less likely) or may have another emissions tier (more likely), so results may vary if the temperature is cold enough. What is cold enough depends on the regulations, engine size and application, and negotiations with EPA (or appropriate regulatory body), so there is no specific number to throw out there where you might see the change.

 

[BrianPetersen]

Brand new 2009 VW diesels have glow plugs. So does every other automotive diesel that I know of. I realize that electric intake air heaters are also possible, but "nothing" isn't really an option for an automotive engine that has to start in -30 degrees.

They do adjust the timing on cold start, but it takes more energy to spin the oil pump and to make up for heat losses, so fuel delivery is naturally higher.

 

[70btdc]

For any vehicle you would need to know the thermal mass of the coolant system with the thermostat closed. You need to know the required power output of the engine required to drive the cycle in a particluar vehicle, then convert this to rejected heat through thermal efficiency. Assume little or no heat rejection from the coolant as radiator not in circuit during warmup. Small engines that work harder warm up quicker than large powerful engines designed with a coolant system that can dissapate lots of heat when needed.

Most modern small diesels will take between 400-600 seconds to reach 70 deg C on a 25 deg C ambient NEDC. For exapmle, 10% lower fuel consumption on same engine can result in 200 second extra warmup.

Warmup period as suggested consists of a 200 rpm higher idle and advanced injection timing. Smart alternators also have a heavy charging period in the first minute of startup to replenish battery charge when glowplugs dragging alot of current. Many things to consider, but just heating the water from rejected heat will give you a rough estimate at first.

 

[JSteve2]

Brian,
I'll admit a heavy duty bias. Not familiar with the small engine market, although if they are direct injected now then I would suspect glow plug usage is minimal at best. Virtually no trucks use them.

 

[BrianPetersen]

ok, I'll bring you up to speed on the automotive side a wee bit. They're all direct-injection and they've all gone to common-rail injection (VW just gave up on their VW-specific electronic unit injector design for 2009). They all use glow plugs and VW always has since the beginning in the 1970's. Most modern designs are 4 valve per cylinder and DOHC although there has been a phase-in (my 2006 VW is SOHC and two valves per cylinder and has electronic unit-injectors, VW calls this "pumpe-duse" = pump-nozzle in German).

The last couple generations of VW diesels (along with everyone else) use ceramic glow plugs that are extremely fast-heating - only a second or two, even at -10 C. Below that, they start taking a little longer, it's probably 5 seconds at -20 C. The VW TDI engines leave the glow plugs on for some period after a cold start (although the indicator lamp on the instrument panel doesn't show it) to improve emissions and noise after the cold start.

All VW TDI diesels since the TDI began (phase-in starting 1993) have catalytic converters and exhaust-gas recirculation. The 2009 models have oxidizing catalyst, DPF particulate filter, de-NOx catalyst.

I've seen (and smelled) the huge clouds of stinky white smoke coming from heavy-duty diesels after a cold start ... this is what the electric assistance (glow plugs or an intake air heater) is meant to avoid.

 

[TDIMeister]

Ian, I have been already doing exactly this type of simulation in GT DRIVE since 2007 (NEDC, not FTP-75, but shouldn't matter), so I feel I have some (minor) authority on the subject. I am also writing a thesis on this subject.

First of all, modern gasoline engines do not use cold-start fuel enrichening to quite the same extent as they used to for obvious emissions reasons. The key word is catalyst heating, and most engines use a combination of measures, predominately retarded ignition timing, to raise the exhaust gas temperature and thereby get the catalyst to light-off as quickly as possible.

Nonetheless, in my models I use a very simple fuel quantity multiplier by some factor (in the order of 110% -- don't misunderstand, that means 10% in addition) during the period of catalyst heating, which in the cars I have investigated are under a minute. One of my mentors' favourite sayings is, "Make a simulation model as simple as it can be, and complicated as it needs to be."

Increased friction during warm-up is done internally in GT DRIVE if I supply a coolant or engine-oil heating curve over the course of the driving cycle simulation. At least in the NEDC cycle, many cars reach 60°C oil temperature between 350-450 seconds, and 90°C (full operating temperature) by 800 seconds, corresponding with the end of the ECE urban phase and start of the EUDC extra-urban phase.

During the initial phase of warm-up, the engine operates at an increased idle that is a function of coolant temperature. Again, given this function and a coolant heating curve (typically reaching 60 and 90°C somewhat sooner than the engine oil), then I know precisely what the mapped idle speeds.

Diesel engines do not use fuel enrichment at during cold-start at all, but also features retarded injection timing, increased idle speed (again a function of coolant temperature, and there may be auxiliary coolant glow-plugs in use), and different EGR settings than a warm engine. In my simulations I never use a cold-start multiplier. The internal friction compensation as a function of oil temperature suffices.

With proper model build-up and calibration, I have found excellent correlation with road-test data.

 

[TDIMeister]

murpia, I have a relevant portion of my draft thesis that I could share with you. I'm not going to host it for public download, but if you're interested, get in contact with me and identify yourself. Since this forum frowns upon sharing personal contact info, look me up under the same username @ the TDIClub.com forums. I don't think what I'm doing here is against the rules in any way?