Hi all,

I have tried to save up something for a special occasion. Would it be an update on the supercharger project? Was it going to be something irrelevant? Was it going to be something new and interesting?

I don't think I have seen any mention of folk doing this with their RX-8's, and a quick search seemed to back that up. So here goes...

One of the things I missed on the RX-8 compared to my previous ride was the trip computer, in particular the ability to display average and instantaneous fuel consumption. When I started researching stuff for the sCANalyser scan tool, I was slightly puzzled to note that the ODBII specs don't mandate a "fuel flow rate" type of measurement. So that was going to make things hard.

But we are fortunate with the RX-8. It has some good sensors that provide valuable information. And by combining some of the readings of these very accurate sensors, additional information can be derived.

So as a small test, I added a temporary display into the sCANalyser live-update screen, just to display some fuel consumption info...



I guess it is time some of you got used to metric . See down the bottom, there is a text box with three figures. The first is the instantaenous fuel flow, as used by the engine in grams/second. The 2nd is the same, but displayed as litres/hour. The third is instantaneous fuel economy, expressed as litres / 100 km. The same things could be done in imperial units.

I didn't do any averaging of these figures - it was updating at about 10 times / second.

How is this derived? Well - all the information you need is already displayed on the gauges. The MAF sensor tells us accurately how much air is being drawn into the engine. The WB02 (Lambda) sensor tell us the amount of O2 in the exhaust. And the lamba value lets us derive the air:fuel ratio based on the stoichiometric value of the fuel (Gasoline is about 14.7:1). So knowing how much air is going in, and what the air/fuel ratio is, we can calculate the mass of fuel that went in as well, i.e grams/second. If we use the specific gravity for the fuel, we can get a volume for the mass. I used 0.76 for this experiment, I'm not sure how close that is to what was in my tank - I didn't have my digital hygrometer on hand, nor does ODBII provide one .

So once we know how much fuel we are consuming per second, we can include the speed we are traveling at and calculate the useage per kilometre. Cool, yeah?

Someone might like to rip the readings from the above gauges and do some math to see if mine was right. (NB - I used 14.643 for stoich).


I have not tested this against a known quantity - such as a fuel flow meter. But the values I get on the road look about right. In the case above, I was diving on an ever so slightly downwards incline. Hence 7.5L/100km seems to be probably about right.

Imagine the other possibilities for a trip computer... Qty fuel used, distance remaining till empty etc. etc. etc. It would also be interesting to correlate with the fuel level sensor, also available in sCANalyser via the ODBII protocol.

So I guess sco will probably be cranky I was doing this instead of doing "real" stuff on the software for our beta release. Oh well, I guess that is why I am the boss, and we do it his way. Hehehehe. Just kidding, sco.

Note This is not being promised in the entry level version of sCANalyse Live! It depends who wins the argument

So in creating a new thread for this topic, I guess I have also given a spot where we can discuss the best and worst of my 3000 posts over the last 18 months or so. Hopefully the majority of them were in the "better" category!

Cheers,
Hymee 3000.

 

Nice, this is one of the things I wanted in the RX-8. I understand why Mazda left it out, though :p .

Congrats on the 3000 posts. The most memorable post for me included the pics of the Leyland .

Cheers,
Gomez.

 

Good one, Hymee. I made a few enquiries about a trip computer for my car a few months ago but the answers were too expensive. It's a bit comforting when you're outback of Goondiwindi and the fuel gauge is on quarter full to have a better estimate of how many clicks are still in the tank. Hope you win the discussion with Sco.

That bloody gorilla of yours is more efficient at catching locusts than it is at catching rocks. Better in the grill than in the condenser, I suppose.

Strength to your arm (or keyboard fingers) for the next 3000, mate.

 

Hymee, do you lay awake at night thinking this stuff up.
Me thinks you thinks to much
Pay more attention to your wife at night it will be better for your domestic life.

Congragulatonsyabastard

 

Awsome, Hymee. My friend has similar functionality on his Jeep and I had it back in the day on a New Yorker I drove. There is TONS of statistical information that could be logged with this, and I reckon you guys will be the one to do it. Any thought on polling the forum on what kinds of features we'd like or, (at least) having a decent plugin system for it...

Thanks for your hard work on this and the FI project.

 

That's really awesome Hymee, thanks for the update on the scan tool and great 3000th post I just hit 300 the other day :p

 

Stop posting and get back to work on screwing the twins , and tuning them to the upmost :D

cheers
michael

 

^^^^^ Assigns more bugs to Hymee to fix, as he obviously has too much spare time

 

I'm keen to have a pluggable system for derived data and will be designing a mechanism for this after initial release. The only hassle is having time to do it - It'd be great to sell so many that I could work on this stuff full time.... but we'd probably have to sell two to every forum member to make that happen

We've had a lot of feedback about features on this thread - https://www.rx8club.com/series-i-tech-garage-22/sneek-peak-new-hymee-enhanced-product-39583/ - so feel free to drop some comments in there - we'd love to hear any suggestions. We're enthusiasts developing software for enthusiasts so we're keen to hear what you all want.

 

Apparently he has too much time thinking... frustate all the beta testers

Check your email Hymee.

Congrats on 3000th.

 

Congratulations on your 3000th post.

You have come a long way in understanding this vehicle since you started working on "real world" improvements and have enriched and baffled us all with your knowledge.

I guess your supercharger thread is the most exciting and anticipated development.

skc

 

Nice to see Hymee you have finally decided to add some substance to your posts.

Good going Tiger.

 

Congrats Hymee!

That looks great, I think I know why Mazda left the trip computer off though !

Every time you looked at it you'd

 

Congratulations Hymee as one of your many satisfied past (and future) customers and follower of way too many of your 3000 posts. Wow, awesome!

Yes, I too vote for the P76 posts causing me to ROFL & LOL, even on revisits It shows how far you have come from there through V8's to the wonderful world of rotories with the RX8!

These new ideas for the the sCANalyser are a great feature addition - maybe test them out in the PPC beta, hey?

Cheers,

Tonym.

 

Congrats Hymee on your 3000th post... It'll probably take a while for some of us to get there...

Looking forward to seeing the final product and complimenting it with a Hymee Enhanced exhaust .

P.s. If you need any more resources for testing, just ask.

 

That is a MAJOR fine idea!!!!
:D

 

Is it me, or is that gas mileage extremely good ?!?!?
7.5 L = 1.98 Gallons.
100km = 62 miles
Our gas tank is about 15 gallons.
So thats about 31 MPG

It's late and maybe I'm looking at the wrong numbers or something but it seems odd to me.

 

It is an instantaneous measurement. The fact that Hymee was testing downhill - that will give that figure - about right.

It is like in VWs and Audis that have those trip meter. When you put your foot down. It will jump to 30L/100km but when you are cruising in a constant speed - then it will show the average.

I remember driving my ex's dad's 30 year old Merc 500SEL. There is an analog switch like that

 

It was only very slightly downhill. Like Taka correctly said, it is instantaneous. And it is updated 10 times per second.

On the flat it sits around 8.5 - 9. Now remember - this is a derivation based on some 3 dynamic inputs, and 2 constants. I only estimated the stoiciometric value of the fuel in my car, and the specific gravity as well, and I haven't checked what % error to the reading a small error in those constants would cause. Heck - I only coded it up yesterday lunchtime!

Cheers,
Hymee.

 

Wow.... very very cool indeed Hymee. You make this ScAnalyser more and more compelling...! :D

Oh, and congrats on your 3000'th post too!

 

whats pppc?

 

Wouldn't it be simpler to base the instantaneous fuel economy figure on injected fuel? If injector pulsewidth is available via OBDII, then you can base your calculation on PW, RPM and injector size. The only assumption then would be that the injectors do indeed flow what Mazda puports they do.

This would mitigate the inaccuracy caused by varying specific gravity due to regional / seasonal / manufacturer's blends, right? Or am I missing something?

-Brendan

 

Only the fact that injector "pulse width" (duty cycle) is not an ODBII mandated parameter.

The MAF signal is a very accurate measure of the amount of air going in. I've collected some more logs, and I will like to correlate them to actual fuel used.

Cheers,
Hymee.

 

I think you are asking about "PPC"? Pocket PC, i.e. a PDA such as an iPAQ

Cheers,
Hymee.

 

I was hoping it meant PPC as in Apple uses PPC processors.