Jeep 4.0 Volumetric Efficiency

[Dr. Dyno]

On the topic of timing advance, do our Jeeps automatically retard timing if they sense knock? I heard something about that the older Renix will do this but im not so sure about the OBDII 4.0's. In the case that the timing is automatically adjusted, say if i was to put in 93, would it adjust it for the max potential advance, say a couple more HP?

The HO 4.0's (OBD I and OBD II) don't have knock sensors so the timing isn't retarded if the engine detonates. If the engine does detonate, higher octane fuel might solve the problem.
A higher octane fuel can allow you to have more timing advance programmed into the ECU to gain some performance, especially if your engine doesn't detonate on 87 octane and you have some room for manoeuvre.
This explains why a device such as the Hypertech Power Programmer III is tuned for higher and lower grades of fuel. For the lower grade of fuel, the timing advance is less.

 

[BBeach]

I know, I was looking more along the lines of a way for us to manually change it, because i dont have any problems with 87 and wouldnt mind cranking it up a few degrees. I'll try to see how much I can get done this weekend as far as how much force there is acting against my jeep vs how much force my jeep is putting against the elements. It should be a simple optimization problem. As far as something like resistance from tires....thats gonna be a PITA cause ive got 3 different types of tires on my jeep . Anyone know the cross sectional area and the drag coefficient with a factory roof rack? http://en.wikipedia.org/wiki/Drag_equation

 

[MaXJohnson]

Hey, I'm working on a standalone ECU, and have been messing around with ignition timing table. It had no reference point, so its been a pain. Do you think you could capture several data points for me. Like, what rpm do you get max advance, what is max advance, idle advance. Basically, it would be cool if you have several points of (map value, rpm, timing advance) that I could use to adjust my table. I don't have obd-II, so i couldn't get that info anyway. Thought I would ask, since this info would be very helpful to me. Thanks.

I just recalibrated the Scangauge and plugged it into a '97 XJ. Idling in the driveway at ~750RPM the timing was bouncing around 12-15 degrees; mostly closer to 15. Stabbing the throttle briefly jumped it up to 42 degrees. Not much to go on, but I may trade cars with my wife and drive the XJ for a while with the Scangauge plugged in.

As a side note, fuel use was at .8 gph at the same 750RPM idle. This compares to .2 to .3 gph on the Focus 2.0

 

[MaXJohnson]

... an engine designed in the 70's ... with little thought to thermal, frictional, or pumping efficiency...

those silly engineers, how would they know about these advanced concepts

 

[MaXJohnson]

The intent in a production vehicle is not to get max fuel efficiency...it is a compromise....the engine does not run at peak torque rpm at normal cruise speeds...the customer demands sparkling performance and decent fuel economy.

I'm no engineer and am certainly no expert on modern control systems, but with modern engines having variable cam timing, port volume, etc, etc. the "max efficiency" number can also be variable...something a 4.0 lacks.

surely there's a few customers out there looking for sparkling fuel economy and decent performance. Maybe the compromise depends on the target market.

 

[BBeach]

Try it out on the highway and tell me what you get for the best mpg speed. It sucks going to and from school and dishing out all this money. My guess is maybe 65-70mph.

 

[scottbky]

Going on a trip for the weekend, so I'll collect some data. Got a couple hundred miles of freeway so I'll see what comes up. Won't be best mileage 'cuz the whole family will be packed in, but will get some relative values. I've only been getting 1 data point per day on speed/rpm/mileage, using my whole trip to average the numbers. Haven't really looked at other parameters just because I haven't.

The SG2 speed is very slightly lower than the speedo. As in, SG2 says 65, speedo says 66. Total miles is within 2 out of about 250 when I refill the tank. Mileage by odo was right on 19, SG2 reported 18.9. On OBD2 vehicles, aren't all those sensors an electronic signal? Versus the old days when there was a cable with a gear to the gear in the meter cluster. I think the difference (~1-2%?) is as close as can be expected.

I have seen mileage isn't much different at 70 than 60, within experimental error to me. Will see this weekend...

ScottB

P.S. I saw a bunch of pc software for engine tuning through the obd2 port when I was looking for the SG2. Didn't pursue it, but that might be the easiest way to modify ECU controlled parameters like ignition timing. No idea if the stock ECU accepts reprogramming. If not, seems like someone would build a reflashable chip.

 

[5-90]

Going on a trip for the weekend, so I'll collect some data. Got a couple hundred miles of freeway so I'll see what comes up. Won't be best mileage 'cuz the whole family will be packed in, but will get some relative values. I've only been getting 1 data point per day on speed/rpm/mileage, using my whole trip to average the numbers. Haven't really looked at other parameters just because I haven't.

The SG2 speed is very slightly lower than the speedo. As in, SG2 says 65, speedo says 66. Total miles is within 2 out of about 250 when I refill the tank. Mileage by odo was right on 19, SG2 reported 18.9. On OBD2 vehicles, aren't all those sensors an electronic signal? Versus the old days when there was a cable with a gear to the gear in the meter cluster. I think the difference (~1-2%?) is as close as can be expected.

I have seen mileage isn't much different at 70 than 60, within experimental error to me. Will see this weekend...

ScottB

P.S. I saw a bunch of pc software for engine tuning through the obd2 port when I was looking for the SG2. Didn't pursue it, but that might be the easiest way to modify ECU controlled parameters like ignition timing. No idea if the stock ECU accepts reprogramming. If not, seems like someone would build a reflashable chip.

If you've still got the links to hand, I'd like them - as a jumping-off point for further research, if nothing else.

Most OBD-I/OBD-II ECUs use EPROM, EEPROM or EAROM chips - which is why they can be "popped" or "flashed." The key difference is in software support for setting up a new chip - the chip itself is a standard part, available at most electronic stores that cater to builders (which leaves Radio Shack out...)

Differences?
The EPROM (Electronically Programmable Read-Only Memory) can be programmed from scratch, but require erasing (typically by a minimum exposure to UV light - you can buy "EPROM Erasers" from places that carry EPROMs, usually)

The EEPROM (Electronically Eraseable Programmable Read-Only Memory) doesn't require exposure to erase, but it still needs to be pulled to be reprogrammed. The EEPROM programmer has an option to send the electronic signal to "wipe" the chip before programming - it can't be reprogrammed "live."

The EAROM (Electronically Alterable Read-Only Memory) can be reprogrammed "live" (as I recall - I'd have to check to be sure...) with the right signals.

The use of PROM chips allows automakers to design a few "standard" ECUs (standard set of inputs and outputs, standard chassis, &c) and use them with a large number of vehicles, simply by changing PROMs. The ECUs are probably built by contract, with the automaker "flashing" (programming - also "burn" or "pop") PROMS as required. This also allows the use of a "standard" ECU over several model years - again, just burn PROMs as needed.

GM and Ford will both sell copies of the software needed to "flash" their PROMS - ChryCo "licenses" their software for an inordinate fee (I just found this out last week from a guy who flashed PROMs for a living.) That's why "flashed" ChryCo PROMs aren't common - most shops don't want to pay a large recurring license fee, and have ChryCo "flashes" at three or four times the cost of GM or Ford - just to break even (while they do make some money on the GM/Ford flashes.) That's why I'd like to see about adapting the GM V6/DIS to the AMC242 - that would allow for easier reprogramming...

 

[scottbky]

Only saw the hits when I was looking for OBD2 gadgets. Didn't track them as it wasn't what I was looking for. I'm sure a google search would find something worth pursuing.

ScottB

 

[5-90]

Only saw the hits when I was looking for OBD2 gadgets. Didn't track them as it wasn't what I was looking for. I'm sure a google search would find something worth pursuing.

ScottB

I can do that - I just wasn't sure if you still had them, and there's no point in reinventing the wheel, y'know?

 

[BBeach]

Using some info on a wikipedia page...http://en.wikipedia.org/wiki/Drag_coefficient
And....
Using the CdA value of 11.7 - 1993 Jeep Grand Cherokee which is similar in height by .9 inches and width in .1, and generally similar shape (ZJ) so I think it was a could constant. In any case I converted that to metric, because we all know thats much better and when you sub it into the equation I got 513.7N at 27m/s (~60.4mph). Seems like a reasonable number to me. If anything is fishy about those whether the wikipedia info or my conversion then let me know. Does anyone know how much force an engine will have if you were to not have your foot on the throttle at all (basically engine braking?)

 

[BBeach]

I looked around some more and found the Cd to be .52 for XJ's and the frontal area to be 2.58m^2, making the CdA=~1.3 and therefore the total force to be 608N. Anything Im missing?

 

[BBeach]

Quite true - especially the angle on VVT. However, the 4.0 does have a broad, flat torque curve - which partially makes up for the lack of VVT (and does so without additional moving parts.)

Part of the reason cruise RPM tends to be so low, tho, is because that's what people have just come to believe. I don't recall all the theory behind it (although I have verified this "fact" experimentally - on the road, with friction and air resistance, parasitic drag, and even in various parts of the country with different fuel formulations!) but I think the "best cruise at best torque" works because the engine has the ability to produce more torque - so a lower percentage of that output is used, and the engine doesn't have to work as hard to maintain forward speed.

The fact of the matter (for the fourth/fifth gear comparison) is established experimentally - try it. This was in a 1988XJ - both with stock (237/75-15?) and oversize (30"-9.5-15) tyres. The gain was ~3mpg. Considering I was working with getting 15-16mpg before and 18-19mpg after, and I took notes (which I cannot now find...) for a month under each condition, I'd consider that "experimental verification."

I look forward to seeing what I can get when I finish renewing my engine management sensors - since they've got 280K on the clock, they're getting a little worn and slow...

So would i get best mileage with225/75, 235/75 or 30" tires with half miles on highway ~70-57mph and the other half in the city.

 

[5-90]

So would i get best mileage with225/75, 235/75 or 30" tires with half miles on highway ~70-57mph and the other half in the city.

You should get best mileage when you cruise at a forward speed that is consistent with optimal crankshaft speed for cruising - in our case, that's a pretty big window (~2000-4000rpm, as I recall.) The closer to the middle of that (3000rpm) you are, the better - since that's right about where the torque curve peaks.

However, since our torque curve is so broad and flat, we have a good deal of latitude in cruising. Not so with a lot of these little four-cylinder econoboxes - you've got a fairly pronounced torque peak, and it's relatively high (4000-4500rpm.) V8s have a slightly flatter torque "peak," but it also tends to be fairly high (say, 3500-4000rpm or so.) That's speaking very generally, of course - it varies depending on intake/exhaust combinations, camshaft selection, and even from engine to engine (due to variations in the castings.)

However, the cruise cranksahft speed I had in fifth gear (~1800rpm) was entirely too low to sustain fuel mileage - I was actually using most of the torque my engine was generating to stay moving. I stopped using fifth gear - bring my cruise RPM up to about 2600rpm - and noted a significant increase in fuel mileage. While I'm not fully clear on the theory involved, I think it's due to using a relatively smaller percentage of torque generated - so the engine didn't have to work as hard.

Picture yourself having to dead-lift 250#. Then imagine, say, Lou Ferrigno doing the same task. He could do it easier and more often than you probably could, simply because he's using a smaller portion of his physical strength each time he does the same work you do. I'm not sure, but I think it's much the same way with engines at cruising speeds.

 

[BBeach]

I agree with the last statement completely, however I was always under the impression from when I saw it on the jeep website tht 90% of peak torque was generated at idle or some very low rpm <2000rpm. Im pretty sure that within that range the power from the engine overwhelms the power of drag.

 

[j99xj]

I think it has more to do with vehicle speed than rpms. But I'm not sure. My engine seems to love 2000-2200 rpms while cruising.

Assuming a static speed, the only forces the engine has to overcome are frictional losses in the drivetrain and the form drag force.

Frictional losses and drag are both minimized when vehicle speed is minimized.

So in theory the slower you go the better the gas mileage.

I've gotten mid teens for gas mileage driving really fast at 85+ mph, and I've got 24 mpg going 55 mph with the cruise control on.

 

[BBeach]

I think it has more to do with vehicle speed than rpms. But I'm not sure. My engine seems to love 2000-2200 rpms while cruising.

Assuming a static speed, the only forces the engine has to overcome are frictional losses in the drivetrain and the form drag force.

Frictional losses and drag are both minimized when vehicle speed is minimized.

So in theory the slower you go the better the gas mileage.

I've gotten mid teens for gas mileage driving really fast at 85+ mph, and I've got 24 mpg going 55 mph with the cruise control on.

Well vehicle speed is directly related to engine rpms when the torque converter is locked. Ive been calculating drag forces, somewhere around 600N at 60mph or so but idk how much drivetrain parasitic losses are. Im basically trying to figure out how force is required to be overcome vs how much force the engine can produce at that rpm in a ratio to found out when the least load and best AFR for fuel economy is there. Then again, this is all theoretical and itd be much easier for me to go and get a scangauge OO.

 

[MaXJohnson]

Hey, I'm working on a standalone ECU, and have been messing around with ignition timing table. It had no reference point, so its been a pain. Do you think you could capture several data points for me. Like, what rpm do you get max advance, what is max advance, idle advance. Basically, it would be cool if you have several points of (map value, rpm, timing advance) that I could use to adjust my table. I don't have obd-II, so i couldn't get that info anyway. Thought I would ask, since this info would be very helpful to me. Thanks.

mapping out an accurate matrix of timing values at various rpm's and load values would take a considerable amount of time and effort; more than I have right now. Including my previous post, here's a few observations that may help you:

idle timing typically at ~15 degrees +- 2 degrees
stab the throttle at no load: timing jumped to 42 degrees
typical timing at steady cruise or light acceleration: 33 to 36 degrees
transition from steady state cruise to acceleration: timing briefly drops to ~ 15 - 17 degrees, then rises to mid 30's or greater depending on load and %throttle.

this is on a '97 4.0/auto/231, completely stock

 

[czb83]

Thanks for getting that info. Know that it cruises about 33-36 degrees is helpful, because typically that will be max advance, as your usually running lean during this period.

mapping out an accurate matrix of timing values at various rpm's and load values would take a considerable amount of time and effort; more than I have right now. Including my previous post, here's a few observations that may help you:

idle timing typically at ~15 degrees +- 2 degrees
stab the throttle at no load: timing jumped to 42 degrees
typical timing at steady cruise or light acceleration: 33 to 36 degrees
transition from steady state cruise to acceleration: timing briefly drops to ~ 15 - 17 degrees, then rises to mid 30's or greater depending on load and %throttle.

this is on a '97 4.0/auto/231, completely stock

 

[MaXJohnson]

You should get best mileage when you cruise at a forward speed that is consistent with optimal crankshaft speed for cruising - in our case, that's a pretty big window (~2000-4000rpm, as I recall.) The closer to the middle of that (3000rpm) you are, the better - since that's right about where the torque curve peaks.

This is not true. For steady state cruise, best fuel mileage will be below 2500 RPM. Within a practical range, lower is better; fewer gulps of air per minute = less gas per minute. Lower RPM = less friction(piston rings, valve train, bearings). Keeping RPM at the front edge of the peak torque plateau will minimize friction at close to peak torque efficiency. Your statement that vehicles are geared high(low numerically) because people expect that is wrong as well. Higher gearing equates to better fuel mileage on the EPA test cycle, less engine wear and less noise and vibration.

However, since our torque curve is so broad and flat, we have a good deal of latitude in cruising. Not so with a lot of these little four-cylinder econoboxes - you've got a fairly pronounced torque peak, and it's relatively high (4000-4500rpm.) V8s have a slightly flatter torque "peak," but it also tends to be fairly high (say, 3500-4000rpm or so.) That's speaking very generally, of course - it varies depending on intake/exhaust combinations, camshaft selection, and even from engine to engine (due to variations in the castings.)

"little four-cylinder econoboxes" don't get best mileage at their torque peaks either. My Focus has a torque peak at 4200 RPM. For steady state cruise in 5th gear (OD) on level highway:

75MPH - 29.7MPG
70MPH - 32.9MPG
65MPH - 36.5MPG
60MPH - 38.8MPG
55MPH - 45.7MPG
50MPH - 50.0MPG

In 4th gear (1:1)
60MPH - 32.7MPH, RPM's increase from 2500 in 5th to 3300 in 4th
55MPH - 34.1MPH, RPM's increase from 2250 in 5th to 3000 in 4th

*These figures were taken on level ground in western Kansas. Each figure represents an average mileage reading for a 5 mile run. All in the same direction, all at ~61 deg F with ~5 MPH cross wind.

The negative impact on fuel mileage at the higher RPM is obvious as is the impact of aero drag as speed increases.

However, the cruise cranksahft speed I had in fifth gear (~1800rpm) was entirely too low to sustain fuel mileage - I was actually using most of the torque my engine was generating to stay moving. I stopped using fifth gear - bring my cruise RPM up to about 2600rpm - and noted a significant increase in fuel mileage. While I'm not fully clear on the theory involved, I think it's due to using a relatively smaller percentage of torque generated - so the engine didn't have to work as hard.

"using most of the torque my engine was generating to stay moving" Isn't that what you would expect?

Picture yourself having to dead-lift 250#. Then imagine, say, Lou Ferrigno doing the same task. He could do it easier and more often than you probably could, simply because he's using a smaller portion of his physical strength each time he does the same work you do. I'm not sure, but I think it's much the same way with engines at cruising speeds.

Picture a little engine vs a "Lou Ferrigno" engine keeping you XJ at a cruising speed of 65MPH. Both are pushing the same .52Cd Box through the air and both are overcoming the same drivetrain losses and rolling friction, but the bigger "Lou Ferrigno" engine has more internal drag (piston rings, valve train, bearings). 50% more parts = 50% more friction & 50% more rotating/reciprocating mass to accelerate/deccelerate for each revolution of the crankshaft.

The real reason people are seeing mixed results with lower gears is due to the variation in their driving habits and the variation is their percentage of highway cruise vs city driving. Low gears (numerically higher) can make stop and go driving (lot's of acceleration cycles) more efficient. High gears (numerically lower) aid in steady state cruise. YMMV