Relooking at warm-up

Ok,

I've been working on the problem of using the physical metrics to calculate the BSFC (g/J). Not a trivial problem because the data is also 'noisy' but here we go:

• ~70C - threshold between high efficiency, fuel trim and enriched mixture
• ~225-235 g/J - operational, peek engine efficiency

I need to take more cold temperature metrics and warm-up rates. With the 'cold temperature' efficiency and warm-up rates, we can model different warm-up scenarios ranging from:

• cold-start -> drive off
• cold-start -> noload warm-up to 40C -> drive off
• cold-start -> noload warm-up to 70C -> drive off

Understand this is going to be a series of equations but once we have them, we can begin to model different scenarios. Once the models are created, we need to validate the results. Understand, I don't know what these equations will be BUT they should be reasonably accurate and reproducible.

Bob Wilson

ps. Frank we can use the injector volume and rpm to calculate the fuel burn in grams. It is not a trivial relationship but appears to be reproducible. We should also be able to use injector timing with suitable constants. Also, I'm getting 'noise' due to the slow scan from the current miniVCI capture. I may be able to increase the sampling rate AFTER

 

The NHW11 and NHW20 have hundred watt, 12V, electric heating elements in the cabin as well as the fans. We should be able to hack a 12V plug-in solution that powers the heating elements and fan to pre-warm the cabin. It can also put a pre-charge on the 12V battery to hopefully extend its life.

The alternative is to put say a 400-600 W AC heater in the car and run it off of an extension cord.

Bob Wilson

 

It is an important metric. But it is premature to worry about it for now.

We will develop a metric:

• rate of fuel burn to 40C

This will let us know how much and how long it takes from any given starting temperature. What will be harder is the rate of cooling as this has a non-linear factor.

Bob Wilson

 

Former valley and river bottom land, there are hardly any elevation changes.

After warm-up to engine stop on the driveway, I drove to the cleaners using neighborhood streets and cutting through or around shopping centers. This allowed me to stop before the miniVCI finished each segment; restart, and; proceed.

Bob Wilson

 

We finally had a cold snap, 28 F (-2C), outside. So just as the Eastern sky was starting the lighten, pre-dawn, I connected the miniVCI and recorded the warm-up:

• Time starts from when the 12V bus voltage changes indicating the car goes to "READY".
• Each fuel metric is for the duration of the samples, not a cumulative but the grams burned.
• The heater was off.
• The engine auto-stops when coolant reaches 40 C.
• After a pause of 20 seconds, turning on the heater started the engine.
• At seconds 780, the engine auto-stops.
• After 5 minutes, the engine starts and the cooled radiator coolant hits the block.

• At seconds 22, the engine starts and fuel consumption is at the highest.
• At seconds 120 and 25 C, the limit for high-fuel consumption occurs.
• At seconds 150 and 30 C, we see the first low-fuel consumption occurs.
• There after, we see fuel consumption alternate between high-fuel and low-fuel.

• To start the engine, turn on the heater.
• The car continues to warm up until 55 C when the engine auto-stops.

• With the heater still on, the engine remains off until the coolant drops to 46 C.
• When the engine start, the radiator coolant comes in and the coolant holds at 46 C.
• Engine fuel rate goes from high-fuel and soon resumes oscillation with low-fuel rate.

CONCLUSIONS

1. Warm-up on the driveway should continue until 30 C when we see fuel trimming begins.
   1. Need to check O{2} sensors to make sure it is coolant that triggers conversion.
2. From here on out, the engine will work at efficient BSFC
   1. The heater could be turned on.

I need another cold-soak test but that may take a while for a cold front to reach Huntsville. But we've got an initial set of data that gives a starting point on driveway warm-up.

Bob Wilson

ps. Let me know if you'd like a copy of the recorded data.

 

Let me insert some thoughts:
* * * Quoted material * * *
About coolant 30° reminds me of the end temp of S1a that i'm used to see, the 55 sec approx from ICE ON. Which usually corresponds to 220-230° CAT temp.

Have you recorded CAT temp. in your testing? Yes but I only have one set of data from a serious cold-soak of less than 10C. I belong to the school of engineering that says 'one is interesting,' 'two is curious', and 'three is a pattern.'

Can you affirm that is the value of 30° that makes the change to an efficient BSFC and not simply the change of status (so the whole settings) from S1a into S1b? Remember the terms "S1a" and "S1b" were provided by Ken@Japan as a way to help explain the different modes. I'm 'Relooking at warm-up' because I want to see what the vehicle metrics are telling us. I'm after quantitative numbers that might provide better insights to the boundaries and practical impacts on cold-weather, Prius driving.

For sure, timing advance changes from S1a to S1b, so the efficiency in burning fuel. Adjusting the timing is an effective way to reduce the thermodynamic efficiency and thus increase the temperature and heat of the exhaust gas. When the car is stationary, it removes spark advance as something also impacted by the engine load.

Anyway, i see that the simple suggestion to make S1a standing still and S1b in motion, should be confirmed, is'nt it? For the single set of data most recently recorded, yes. We really need more records and Frank might have something handy.

If i'm getting it right, apart the battery usage, you seems to demonstrate that S1a is affected also by a low fuel burning efficiency. Right? That is exactly what I'm seeing but we can look more closely in the data, the engine torque, to see if that hypothesis bears out.

* * * End quoted material * * *

When I get home this evening, I'll go ahead and upload the excel data in a zip file. I have no problem with additional eyes looking at the same thing I'm seeing. <grins>

Bob Wilson

 

The miniVCI data also shows a binary signal, "Request Warm-up" that is associated with faster coolant warm-up:

• 2.61 C/min - with "Request Warm-up" OFF
• 9.62 C/min - with "Request Warm-up" ON

Bob Wilson

 

Background:

With a recent cold snap, new data:

This time, I started the car with the heater set to low fan. This means a smooth transition after the coolant reaches 40C. The engine continued to run to 55C, just as the earlier test showed.

What was surprising is the fuel consumed decreased until 2:20 seconds after the start and then the engine torque went to zero. The engine continued to run, spark retarded, until 20 seconds later when the torque came back. In effect the first loss of engine torque corresponds to the optimum fuel consumption.


So it looks like we have several clues as to when the engine is trimming to an optimum fuel efficiency:

• engine torque goes to 0 - difficult to detect unless one has a Prius-aware scanner
• ~325C O{2} B1C1 temperature - also coincides with best engine fuel trim but also requires a good scanner that can find this emissions related field (maybe).
• 25C coolant temperature - due to the rates of warm-up, this very likely tracks closely with the O{2} sensor WHEN starting from ~0C

So I checked the previous data:

It appears the fuel trim occurs:

• engine torque to 0 (Nm)
• ~25C - coolant temp
• ~325C - B1C1 catalytic O{2} sensor temp

The coolant and catalytic converter, O{2}, temperatures occur at about the same time when the starting temperature is close to 0C. So I need another run at a higher, initial temperature so the engine coolant has a 'head start' on the catalytic converter. Meanwhile, I'll check my previous records to see if I might already have some.

I found some:

OK, this was a warm day and sure enough, fuel efficiency starts at a coolant temperature of ~25C. The engine torque increases (i.e., more power) while the fuel flow remains the same. It is independent of the catalytic converter, O{2} sensor B1C1.

Bob Wilson

 

Conclusion:

• Wait until the coolant temperature reaches 25C when fuel trim starts.
• Drive modestly until coolant reaches 40C, without heater, or 55C with heater because the hybrid, engine-off, functions are inhibited.
• After reaching 55C, full hybrid mode works.

As I have time, I'll investigate the remaining questions ine new threads:

1. What is the minimum fuel burn to reach 25C as a function of starting temperature?
2. Compare and contrast the <25C driving effects versus >25C driving?
3. Impacts of bumper inlet blocking on transaxle cooling and engine warm-up?

Bob Wilson