Terrys87

iTrader: (1)

I must of missed your write up on the MSP430F5529 and even at that it sounds more complicated then most could accomplish on here.

I have seen your past experience with circuitry and I would say it is better then most, maybe even in the top 5% if I was guessing. Instead of maybe offering suggestions on how to make this system work better, you come on criticizing. Outstanding.

Unless it is some safety issue that is going to get the owner or some one else hurt, I find it disturbing to tell someone how or what to do to their truck and there is plenty of modifications on here that I disagree with. If they put the time and effort into trying to help themselves or someone else improve several different factors from how their truck performs, improves their driving habit or some other benefit, I will be the last one to criticize.

In support of more fallacy, I will be doing a wiring harness swap in a project in the near future and will add this to my list as that would be the easiest time to install it. The gauges in these trucks are very basic and the more information that I can have at a glance is something I can really appreciate.

Terrys87

iTrader: (1)

I lived in California and your driving out there will effect the mileage. With your stop and go traffic and then the next minute you are at high speeds just to keep up with the flow of traffic.

Here in the Midwest, I drive 37 miles with both lanes to myself and only get 20.5 at best and that is doing 60-65mph. If I run 75-80, I get about 17.5 mpg. Just a few miles an hour can make a noticeable difference.

I was surprised that I got 17 miles fewer to the tank on my last fill up with just a few stops in town. It doesn't take much to bring your mileage down.

jennygirl

Sometimes I wonder if EE's are upset by the rapid prototyping kits and open source community because it steals some of their thunder? I could understand that. But yes I agree, the extra $40 and accompanying 1hr electrical hookup time is worth it for me, too. Makes it easier to not spend my life on the details and focus more on the broader picture and adding more useful things, though I do occasionally like to really get into the details sometimes. Plus if it works, it works. The kitted open source things are a nice intersection of group powered think, diy customizability, and "just works" mentality. There seems to be a lot of this kind of stuff out, especially in the arduino powered realm- it really caught on.

Yeah I am noticing now how just the slightest change in throttle position = a huge difference in mpg. It's one thing to understand that's how it works... compared to actually watching the dollar signs adding up. Like I said, this completely obliterated my lead foot immediately. This device will have paid itself off in less than 2 tanks of gas! Let's see if it really can curb my appetite for tight acceleration.. Only time will tell!

RJR

Don't get too enamored of watching the instantaneous MPG and thinking that avoiding the dips in that parameter will give you great increases in overall MPG. Gasoline engines don't quite work that way. Throttled engines (non-diesels) are actually most efficient at turning gasoline into motive power at wide open throttle. The pumping losses across a partially opened throttle are quite large, and on a 3vze amount to about 3-5 horsepower at 3000 rpm. That's about 1/6 to 1/4 of the total HP required to move the truck down the road at 60mph. What you really want to do is keep careful records of your average MPG over identical trips with different driving styles.

It's very true that your instantaneous MPG will plummet when you mash the throttle during acceleration. But, you'll accelerate faster and get to the point where you can release the throttle and get back to maximum MPG's sooner, and that will offset the short period of low MPG's.

After I wrote the program last spring to read the OBD1 diag port data (which includes rpm, speed, and fuel injector pulse width), I did a couple of runs on my daily commute. The first run I mashed the throttle to max when pulling away from stoplights and shifted at 4000+ rpm every time. The second run I drove like a real grandpa (of which I am one:-), shifting at <3000 rpm, very mild throttle, just barely staying out of the way of traffic. The results -- < 1% difference in fuel consumed over the same route. The route was 10 miles roundtrip, 5 lights each way, identical (dry) weather conditions and very similar (fairly light) traffic. If I get a chance in the next couple of days, I'll share the graphs from the log files on those runs.

The major driving habits that affect fuel economy are:
- Use of the brakes. Brakes turn gasoline into heat. The less you use them, the better. Note: Using heavy engine braking instead of the foot brakes doesn't help - you're still turning motion into heat out the tailpipe. What does help is anticipating stops and coasting down gradually, and avoiding full stops when possible.
- High speeds. Wind resistance goes up as the square of the speed (i.e., double your speed and the wind resistance goes up by 4x), and really starts dominating the drag on our boxy trucks above about 50mph. My truck will hit 20+mpg at 45-50mph, but at 75 I'm lucky to get 17.
- Hi rpm at part throttle. This maximizes the pumping losses I mentioned. Note that high rpm at full throttle is not nearly as bad for mileage, because the air has a free path into the engine.

I'm by no means downplaying the usefulness of the device. Take careful records and you'll be able to learn what helps mpg and performance, and what doesn't in terms of driving habits. It's also a great tool for getting fairly quick feedback on mods you make to the engine (timing, higher octane gas, etc.) It's much quicker and more accurate than keeping records on multiple tanks of fuel. Have fun!

jennygirl

Wow RJR, some really great tips here. Thank you! That all makes complete sense. I will absolutely be trying it out.. And it means I can still accelerate fast as long as I am maintaining that speed afterwards, without braking. Really appreciate the input

RJR

If you're interested in this sort of stuff, here's a good article that discusses specific fuel consumption (a measure of engine efficiency) in terms of various operating conditions. Some interesting charts and graphs.
http://www.autospeed.com/cms/A_112611/article.html

wireguy

OK, so lets say you have run through a couple of tanks of gas in a perfect blend of city/highway driving, and have established an accurate average. Now your unit has a standard to use that allows it to measure pulse width and extrapolate from that reasonably accurate MPG.
You need to drive north to Bakersfield to pick up a bargain part for your 'yota. You drive north up over the grapevine and practically coast down the other side. You cruise up I-5 into Bakersfield, pick up your part and climb south up over the grapevine and again practically coast down the south side. Nearly this entire trip has been spent far outside the parameters that you used when you established your standard. Is the unit still going to be accurate for the fuel used on this trip?
I can't figure out whether the radically different conditions would throw off extrapolated MPG or whether the unit's standard would give it the information it needs to extrapolate accurately.

RJR

To a first order, it should be pretty accurate regardless of the driving conditions. Basically, all the arduino is doing is continuously adding up the total time that the injectors are open, and multiplying that by the "gallons/second" constant for your particular injectors to get total fuel used. The injector constant doesn't change for different driving styles, so the arduino can always work that equation and get pretty good answers.

When you calibrate it, you are simply telling the arduino how much gas you used for a particular trip. It already knows the total injector time for that trip, so it just works the math backwards to get the constant, which it then stores for future use.

The volume/second constant for injectors is pretty stable. It primarily depends on orifice size and fuel pressure, which is held constant by the fuel pressure regulator. The ECU needs the injector constant to be stable in order to accurately calculate the mixture when running open loop.

BTW, this technique is exactly how most new cars calculate instantaneous MPG, average MPG, miles until empty, etc. It's all based off of fuel injector timing.

wireguy

It is amazing what can be done with electronics and programming. That price is very attractive for such an upgrade.

scope103

I, too, believe that's correct.

But a year ago Circuit Cellar published an article where the contributor based his MPG meter on the VAF signal, not injector time. http://circuitcellar.com/cc-blog/a-r...ption-monitor/ If I recall the article, he didn't discuss why to use one over the other, but the VAF signal would require less conditioning than measuring pulse length.

Also, he wasn't as concerned about numeric accuracy as I am (and I think JennyGirl and RJR are), concentrating more on "relative" efficiency. Like the unlabeled analog MPG gauges on 2014 Subarus.

Terrys87

iTrader: (1)

jennygirl..I am glad to see some post have not deterred you from your project. I too am interested in what you are doing as gas cost is one of my larger expenses that I would like to cut down on. Several posters on your thread, including abe, I have learned a lot from. Anytime these guys have posted I always look to see what they have to say and usually learn something from them. Electric is still a weak area of mine and I am further then I ever thought I would be. RJR and scope103 have posted some interesting info and somethings I had not thought of.

RJR

Air flow is a good first order measure of engine power output, and therefore fuel flow, but I don't see how it would be as accurate as measuring the fuel injectors directly. The ECU uses multiple sensors and a fairly elaborate memory lookup table to figure the exact fuel amount for each injector pulse. Unless you have access to the other sensors and can replicate that table, air flow is only going to give you an approximation. It could be fairly accurate when the engine is in closed loop mode and you know the exact mixture, but during heavy acceleration, hill climbing, and coasting, when the engine goes open loop, it will be more of a guess.

rworegon

Woot! Another great mod, Jenny, for the pre-OBD-II bunch. I have a OBD-II ScanGauge with similar functions in the truck and find it useful.

joshtee

Jenny, any updates on this? Did you get it dialed in? Have one on order, and am trying to figure out the pulse per mile value for the 22re.

jennygirl

Hey sorry for the delay! Yeah I did get it dialed in pretty well. After my first attempt at calibration the next tank info on the mpguino read 15.2gal used and the pump filled me up at 5.12. Unfortunately it was a different pump, so I was unable to know exactly how accurate.. But it seemed accurate enough for now.





One of my better runs on a trip home from a friends house:


Overall I'm getting anywhere from 13mpg in really awful conditions and heavy foot, up to 21mpg with city and highway mixed driving. The other night on the way home from work (no highway) I came in at 20.5

This has all been with the top off as well. Eager to see how it does with the top on, which I just installed because tonight is the first rain in LA all summer!

It's definitely hard to get good mpg under crappy conditions like traffic uphill. But I am getting better at it. This device is awesome!

This past weekend I took her wheeling at Angeles National Forest OHV offroading area. Then on the way home had 80mpg heading down the hills the whole trip was an avg of 17mpg. Not bad for wheeling around in L4 for a couple of hours!

EDIT: Forgot to mention the VSS setting was spot on after the first calibration. The injector setting should probably be calibrated again, but it's close (within .1gal out of 15)

jennygirl

Another thing I forgot to mention, I set the tank size to 17gal. The gas light came on with about 15 miles left on the Distance To Empty readout (on the Car Sensors mode/page). So super handy since my gas gauge acts funny (hits E at about 60% tank). Oh well, no need for a gas gauge now

jennygirl

I am excited for you.. Here are some tips for use

"Custom" page is awesome, what I most frequently use. Shows instant mpg (momentary) "Im", current mpg "Cm", injectors on trip time (T) in hr.min, and cost of trip (set the cost in settings when you fill up.

Car Sensor screen is nice if you are wondering about your battery voltage and distance to empty.

Tank screen is nice to see average mpg for the tank (which is independent of trip time).

Left+Right = settings
Left+Center = tank reset
Right+Center = trip reset (use this a lot)

Middle button changes backlight brightness (4 settings: off-1-2-3)

I love this thing. But it does make me a little sad seeing what terrible amounts of mpg this car is capable of even when trying to hypermile a trip, haha.

joshtee

Sweet! Thanks for the update. Is the 7893 pulse per mile value something you had to enter in? If so, where did you find that?

jennygirl

Both of those settings I entered in after following calibration directions on the mpguino wiki. The only thing I didn't do was drive 200 miles during the calibration, it was more like 100, so that's probably where the accuracy of the injector pulse fell a bit.

I am going to calibrate again when I am able (you have to fill up at the same exact pump twice in a row and recently I've been doing some traveling & filling up at random pumps).

Hope that helps. These settings can definitely be improved upon! But the VSS should be pretty darn near perfect.

joshtee

Ok. Guess I should really wait for it to show up before I get to far into it, anyhow. I'll post back here when I get it set up.