Cruise Control (with MPG sensitive logic)
 

I was wondering if there was a way to set up a cruise control but instead of speed, You use mpg. of course you would need a min and max speed. Since I got my scangauge I have gotten great MPG by just tring to keep a certain mpg. Sorry i cant use other driving techniques as i have an auto and crowded highways. but a 36 to 41 boost is good for a start.

A smart cruise control would be a fantastic feature for many, and would save a lot of fuel. RH77 modified his so that it could only pull a certain amount of extra throttle when encountering hills, but nobody has done exactly what you're talking about.

FYI: What features on your ULTIMATE fuel economy instrumentation?

I bet there are lots of other small things you could do to improve MPG at either end of your drive on the crowded highway.

Watch for a new feature on the site later this week: 100+ efficient driving tips for all kinds of situations.

As you have stated, the biggest improvement u can do is driving mods. if you are not going to do any P&G driving then a mpg cruize control would be like the ultimate invention. Then any car mods you do will either make you go faster or allow an increase in ur mpg setting. This seems so amazing and also easy that why isnt anybody doing it.

You have given me an idea! This may be possible to do in a typical ECM, as the software already governs cruise control...

I'm going to see if I can modify this parameter in the ECM...

thats what i wanted to hear, it seems simple easy and highly productive.

What happens when you hit a hill and can't maintain 40mpg? Your engine shuts down?

How about cutting the power to the cruise control motor once it's engaged? That way you can freeze a certain throttle position.

If you hack the cruise for MPG it may be easier on the input end to use TPS voltage instead of VSS volts.
This would allow constant throttle instead of constant speed.
Let me know how you do.
I am not a good hack but I can follow instructions.
Thanx,
S.

I'll look into this for different cruise control systems when I get a chance. Right now I'm looking for software solutions.

Theoretically: Vacuum-controlled cruise, set cruise to specific throttle position and then isolate the vacuum source solenoid and electrically cut off the dump solenoid to maintain specific TP. Tap the brake going over a hill to release cruise, de-isolate cruise vacuum valves, then hit reset at bottom of hill, re-isolate...

If you have to do a total system you would have a minimum and maximum speed. if not i would supose you would know you are going to slow. But that is the whole purpose IS to slow up going up hill. Do you have a scangauge. if not you wouldnt understand what maintaining a mpg setting details.

I don't think you'll be able to maintain 40mpg going up any long, steeply graded hill...going slow works to a point.

If it is constant throttle you seek, why not just hook the throttle servo to a potentiometer.

I'd imagine using a cruise control based on MPG might pose some dangerous situations. A car will shift its speed greatly as it is loaded differently throughout a drive. I can't imagine that driving 65mph, slowing to 40mph, and speeding up again to 60mph would be particularly safe in any form of traffic. If you are completely isolated, then I suppose it might be useful, but not as a daily use item.

I'm interested in watching any developments, though. :)

- Lostcause

I've been in cars with badly running cruise that did the pulse and glide automatically! I imagine any of the old mechanical systems could be tweaked to do this. I don't think that there would be any advantage though at a manual trans car would still have the engine braking on the glide. Maybe if you had an auto trans that pulsed in a lower gear and then glided in top gear it might be worth while.
Ian

Yes, I do have a Scanguage, and in PA, we also have some fairly steep grades that are miles long. While I can easily hit 38 MPG on the entire trip, it is absolutely impossible to maintain even 30 MPG in any shape or form for a certain 1.5 mile stretch. Trust me, if you try to maintain cruising MPGs up extended hills, you will have to slow down to a stop. Your way only works if your car's momentum is enough to put you over that hill.

With hills like these, I stay in 5th and give it about 80% throttle to maintain speed. A greater percentage of the fuel I burn gets turned into motion at higher loads, so I take advantage of it.

Also realize that if I were to maintain 38 MPG instead of the infinite MPG I get going down that hill by laying off the throttle, I'd soon be at 110 MPH.

Maintaining a certain MPG is pointless. You'll be able to average even higher MPGs by using moderate throttle up hills and zero fuel down hills. There's nothing wrong with gassing it up hills if you know you'll be able to coast down the other side without riding the brakes. Take a look at a BSFC map sometime.

One of the better aftermarket models (Rosta?) has adjustable gain... Set the gain low and there you go.

Different cruises are different.

On our Accord stick the cruise is really course and aggressive compared to the Santa Fe. In normal slower traffic situations engaging the SF's CC always seems to do better according to the ScanGauge iMPG than my right foot for some odd reason - unlike the Accord. That's okay, because I'd rather tap buttons than press a pedal.

wow i didnt know there would be so many people who would put in their two cents as to why it wont work. I guess there is some people who cant take over once the cruize is set, even if it had a minimum and maximum speed. i am sorry some people have mountains, But i drive 100 miles a day with just small grades. I just try to keep my auto metro on 52 mpgs. so far my milage is skyrocketing with this system. But it is har to constantly monitor the gauge. a system to set and forget is what we want, keep the negativity down.

Adjustable gain is cool, my Dad's last van (92' E-350 with 460) had an after market cruise + factory cruise. The After market cruise had the Gain CRANKED as the previous owner was pulling a 42 foot (yes that's correct) camping trailer with it. When he used the Aftermarket he'd get 9.5~10 mpg @ 70 (loaded) and with the Factory cruise he would get 11.5~12 @70 (loaded). Pretty significant gain.

The van he had before that (84 Dodge 250 360 engine) had a cruise that would best be categorized as weak, it would essentially DWL, until you dropped to 8 or 10 mph below set speed, then go 90% throttle. In retrospect that was a good setup, but at the time we thought it just sucked.

Hi Terry,
I am curious what you are doing with your cruise now? Is it normal cruise control you are using? Is it different than pulse and glide in gear?
I just got my scanguage and it is an eye opener, the rate of fuel used accelerating and climbing any sort of grade. So now I do watch and try to keep it below 12L/100km climbing hills when I can and EOC down the other side.
Ian

I am using the old foot cruize. But as you can see by using the scangauge is your mpg goes up and up as you get use to holding a constant gas useage, i set my gauge to mpg and try to maintain 50 mpg. as you can see from my fuel log it is working great. I notice that you can be driving say 55 mph and getting 45 mpg, with just a minute backing off of the throtle, the mpg's go up to 50, but speed stays the same. its like free mpg's. but if i was doing a mpg cruize control or a constant throttle. i wouldnt get the same results.

I do agree, current cruise controls are not as effective as they could be...

The worst part is when the car starts to go uphill first it loses some speed then the cruise OVER reacts, every single unit I have ever used does this, just some are worse than others.

But I'm not really sure how we'd address this issue...
How is the CC supposed to know about an approaching climb?
I think, however, possibly some type of digital level gauge could send 'angle of inclines' information to the CC... This of course would be affected by acceleration and deceleration but then if the CC is active this could be compensated for since our foot isn't doing it.

What I am getting at, is if the car is climbing the CC should at this point merely attempt to maintain speed (most CC's try to get back to the set speed even while still climbing)... Even if it has lost a few mph's it shouldn't try to regain this until things level out some at which point the CC could THEN regain the set speed.

A little smoother on the throttle, basically.

jpgmtech
How about if the driver can adjust the delta, and the gain. We need to incorporate the driver's ability to fore see what is going to be needed up ahead. Perhaps a switch to select vacuum controlled or speed controlled. A constant 5 inches of vacuum would put you pretty far up the hill until you needed to intervene with your foot. I would like to limit the throttle position so as to not downshift the transmission or unlock the converter unless i intervene with my foot.

I see what you are talking about now, I have noticed the same phenomenon in my tracker but I have basically no flat roads on my commute so its hard to see what is going on as my instantaneous mileage moves around alot even with constant throttle postion. Also the areo drag is quite high so backing off a little results in lower speed almost right away.

I can see though having cruise that tries to optimize mpg's would help abit, I'd think it would be possible to add some "intelligence" to help, sounds like a good college electronics project!
Ian

I add a step or two of ACCEL at the bottom of the hill then remove those as I feel the load. That tends to keep the rapid throttle additions (adjusting to instant decels) down to a minimum. I've been using that in the RAM for years to eliminate that overcompensation.

Hello everyone, I'm new at this forum. I have a 1998 Suzuki Swift with a 1.3 litter engine and manual. I recently installed an Audiovox Cruise Control CCS-100 and also a Scangauge II. I was thinking, maybe we can fool the cruise control, instead of connecting the Speed Input wire of the CC to the ECM or VSS, we can connect it to a MAP sensor. I know, everyone is going to tell me the MAP generates a 0-5 volts analog signal, but I found that Ford's MAP Sensor generate a pulsating signal depending on the vacuum on the intake manifold. So if we add a Ford MAP sensor to any vacuum line, we take a reference voltage of 5 volts, lets say from TPS or Original MAP sensor, and the output terminal we connect it to the speed input wire of the CC, maybe we can set and lock it by load instead of speed, and this should maintain MPG constant, if the hill is not to long.

Any ideas, please let me know, I've been working the math on the pulse per miles vs. the pulse per vacuum and I think it is possible. If any one is interested, I made calculations and tables in Excel.

Hi jomelmaldonado -

I would think the open source MPGguino platform would be ideally suited to being modified to do exactly what you describe: monitoring fuel consumption and generating a signal to feed the AudioVox cruise unit.

The Ford MAP sensor would be easiest, if it's "plug and play", of course.

I am going to try at least to test with the Ford Map Sensor first. If I have any luck, I will be posting the results here.

Sounds good. Please keep us updated - even if it doesn't work out!

Hello again. I continue to work on the CC idea, I already found a used Ford Map Sensor, took it out of an Aerostar Van from a Junker. I bench tested it and it is producing an AC output signal. I powered it up with a 5 VDC Fixed Voltage Regulator from Radio Shack, it's the IC 7805 or Radio Shack Cat. # 276-1770. I have to use this because the sensor, like almost all sensors, works with a reference of 5 volts. I didn't like the idea of taking 5 volts from the car wire harness from an existing sensors like the TPS or MAP sensor, so to not strain the internal voltage regulator of my Suzuki Swift's ECM, I am using this external one. It's less than $2.00 and we assure not to put additional loads on the ECM. Tomorrow I am taking this setup to my University to bench test it again, but hooked up to a frequency counter or an oscilloscope to measure the exact frequency it is producing. I will also apply different vacuum levels to make a chart of Vacuum vs. Hz
I will be posting this information and photos probably on Saturday.

Any suggestions or ideas, please let me know.

Hello again, as I mentioned, I bench tested the MAP Sensor with a Vacuum Pump and an Oscilloscope. It is working just as described, the nominal frequency and the measured frequency is almost identical. I am including some photos of the measurements with this link of photos. The parameters for the sensor is:

0 inHG = 159 Hz
30 inHG = 80 Hz

Also I included an Excel file with more detailed information on Vacuum Vs. Hz
and MPH Vs. Hz


Any question comments or suggestions, please let me know. I will be mounting it in my car on the next week probably.

Do you know at this point what frequencies the cruise control can work with? Does it care, or is it only monitoring for increase/decrease?

Well, at least the cruise I have, an Audiovox Universal CCS-100 can handle a wide range of frequencies since it's supposed to support different voltage levels, Pulse Per Miles and Wave Patterns because different cars have different ways to report the speed. The Auidiovox CCS-100 has an option to choose from 2000, 4000,5000 and 8000 PPM's If you see the Excel file, there is a table I made that converts from MPH to Hz based on the 4 options of PPM's. And in the 8000 PPM setting, the output frequency in Hz from the MAP Sensor falls in between the 8000 PPM setting, actually from about 36-72 MPH.

Remember that with this modification, it will think it's monitoring MPH from the VSS, but actually it will be monitoring Vacuum from the Ford's MAP Sensor. It will only checkl, like always, for any increase or decrease in frequency on that gray wire and it will compensate adjusting the throttle.

I also have the CCS-100 in my car. (Hope yours is more reliable than mine...)

FYI, mine is set up to read only the frequency from the ignition coil, rather than VSS.

Well, I was planning on only using the RPM, but I found the VSS wire, taped it near the ECM. It was a little bit confusing because I could only find the Geo Metro diagram, and mine is a Suzuki Swift, it changed the color code for the wires, but it turned out OK. You can set your CCS-100 to read VSS with the DIP Switch settings. I will be posting the progress of this experiment on the week.

While the idea of cruise control maintaining a constant MPG is a worthwhile cause, it will really create problems for us. I am sure with the great minds we have here we can develop and test a functioning bench test model, but here are my concerns.

I use CC everyday on my commute, in fact it is the only way I can maintain the high mileage averages I get. This gives me time to monitor the scangauge reading during my commute. I notice that my instant MPG fluctuate continuously throughout a range of about 7mpg due to traffic, wind conditions and the most obvious road elevation. Those small changes in MPG's would crash any computer trying to determine how to react so quickly to those changes.

Instead, I feel the CC should have a P&G mode utilizing they latest technology in order to sense distance ahead of it in order to avoid accidents while doing P&G or even simple locked CC. Including a logic for hill climbing and simulated EOC would be icing on the cake, but it would take one of us on the inside to convince auto manufacturers to pay so much attention to something percieved to only be used on long trips to Wally World.

Hello again, I mounted it in my car today and it is working fine, after ironing out a kink. I knew that it was supposed to be set to PPM = (8000), also the Speed Signal was set to (VSS or Magnets); so I did that before making the first test run. The only problem I had was that when I pressed the set button, it over reacted and opened the throttle to much, immediately after, it canceled out. I tried a few times at different speeds and loads, but it kept doing the same. Then I changed the DIP Switch setting to Sensitivity (LOW) and it worked fine.

I tested it in flat roads and grade roads and it maintains a constant engine load. In the ScanGauge II, it was displaying very close to 52 MPG, it went up and down about 3 MPG. Also I was monitoring the TPS and Load with SGII and it wasn't jumping up and down erratically, instead, it was fairly smooth. I consider this very good for what it's calculations are doing and at the speed it has to perform the calculations and react to throttle position to compensate.

I am going to further test the car and will be posting my results. Also I will be making a block diagram and taking some more photos. I will put it in my album. I also included another Excel file that describes the theory of operation.

The important thing here is that instead of setting a Speed, know we can set MPG's to maintain it constant.

Good stuff! I'm looking forward to your progress.

Hello, just wanted to comment that the Cruise Control is working fine and it is maintaining a constant engine load. I am including an additional picture of DIP Switch Setting so everyone could have a better idea of how to set up there CCS-100.

Hello again, just wanted to keep up to date with the experiment. I added a block diagram for those who want to try to modify there CCS-100 or similar. Also I would like to comment that my mixed city driving went up from about 41.8 MPG to 47.5 MPG with this modification. It's posted in my fuel log.

Hi,

That is a great result! Now, if we can get the car makers to implement this on the stock "cruise" controls...

Well, probably not all drivers will be willing to sacrifice MPH to gain MPG, but we hypermillers are willing to give the extra miles, well in this case, the extra MPG to reduce fuel consumption. Any way, it is possible to do it, so perhaps some company will think about it and put an option in the OEM cruise control for steady MPH or steady MPG mode. That's the next mod I am going to do to it.

This is cool. One could possibly even modify a stock cruise control to work like this.

Hi,

I think it would be best if the MPG could be the main criteria, but below a certain speed (say 5mph less than where it is set -- this speed drop should be programmable) then the throttle could be increased to keep going above the minimum speed.

In other words, If you set the control while getting 60mpg @ 55mph, the control would try to hold the MPG (constant throttle position?) as high as possible (or even increase it, if that starts to happen), but if the speed drops below 50mph, then the throttle would be increased just enough to keep the speed above 50mph.

If you are going downhill, and you start gaining speed, the opposite might help: let the speed go up to 60mph (the speed increase should be programmable, too), and then back off the throttle to get better MPG, and then only increase the throttle back to original setting when the speed drops back below 60mph.

So, dual (triple?) curves would make this sort of control ideal, IMO.