yoeddynz

DIY Fuel injection thread.

Recommended Posts

Body loom wiring is about as much fun as a 3s - all yours mate, I will stick with 4ages and engine looms haha

Share this post


Link to post
Share on other sites

It's amazing how much my wiring has downsized on new loom. 

Toyota e-throttle unit for example. 

Previously 12 wires, 3 plugs, and a throttle cable arrangement. 

Bosch unit - 6 wires, 1 plug, no cable. 

Huge difference to how messy that area is.

Then just putting some effort into placing things so they're less of an eyesore (InB4 "Throw a blanket over the 3S")

  • Like 2
  • Haha 1

Share this post


Link to post
Share on other sites

Anyone tuning with lambda?  What do you sit around when cruising? The mazda seems to like it between .98 to 1.03. Any leaner and it's starts to run rough. 

Share this post


Link to post
Share on other sites

Technically everyone does! 
Then widebands assume Lambda 1 is 14.7:1 and then scales it from there.

But then the funny thing is that if your stoich value was actually say... 13:1 for a different fuel, it will still read as 14.7:1 and you'll still effectively be tuning it to Lambda 0.9 or Lambda 1.1 or whatever and it will still be fine haha. Which is why everyone insists that using Lambda as the scale makes a lot more sense. Good habit to get into.

Yeah that's not hugely lean, but keep in mind that as you go leaner you can end up needing a fair bit more ignition timing as the burn rate slows down by heaps. 
Like at cruising conditions you could probably quite happily add another 5 or more degrees over what's good at Lambda 1.

Supposedly with petrol 15.2:1 ( Lambda 1.034) is the best point to aim for lean burn power/efficiency. But obviously engine to engine that's gonna change.

 

  • Like 3

Share this post


Link to post
Share on other sites

I find no difference in performance between 1 (14.7:1) and 1.23 (18:1) at 100 km/h cruise in Starlet. But the reported fuel consumption drops almost 20% when lean.

  • Like 1

Share this post


Link to post
Share on other sites

Yeah it kind of sits around 1.034ish, jumps between  .98 and 1.04. Sitting at 14 Initial and 32 all in around 2800,  power it drops down to .78-.8.....

What about decel? I would imagine you'd want no gas being I injected at all so lambda 1.2.. 

Share this post


Link to post
Share on other sites
1 minute ago, ajg193 said:

I find no difference in performance between 1 (14.7:1) and 1.23 (18:1) at 100 km/h cruise in Starlet. But the reported fuel consumption drops almost 20% when lean.

 

Starts to jerk if I go too high.. 

 

Share this post


Link to post
Share on other sites
2 minutes ago, yetchh said:

Yeah it kind of sits around 1.034ish, jumps between  .98 and 1.04. Sitting at 14 Initial and 32 all in around 2800,  power it drops down to .78-.8.....

What about decel? I would imagine you'd want no gas being I injected at all so lambda 1.2.. 

What ECU are you using? 

You might have an option for fuel cut decel you can turn on, shuts off injectors completely.

  • Like 1

Share this post


Link to post
Share on other sites

Does your link system have auto tune?

Just aim for lambda = 1 for any cruise settings (40-70 kPa, 2000 - 3500 rpm). You can adjust it after you have everything else set.
0.85-0.90 above 90 kPa everywhere

and maybe a bit leaner for high vac everywhere, but 1 should be fine

Around idle region 0.90-0.95 is good

Share this post


Link to post
Share on other sites
6 minutes ago, yetchh said:

 

Starts to jerk if I go too high.. 

 

Yeah, you could possibly get around that issue by advancing the timing a bit so it has more time to burn. But each engine is different at cruise settings.

Do you know if your ecu lets you tune by just adjusting your desired AFR once you have your volumetric efficiency table dialed in?

Eg on the megasquirt once it is fully tuned all I have to do is say "I want 18:1 at this cruise region" and it automatically changes to that without having to check what the actual afr is and adjust stuff.

 

 

Also, my AFR values could be complete bullshit as I am running a chinese LSU 4.9. But I think they are pretty good from the way the ECU gets the predicted fuel for each AFR value pretty much right.

Share this post


Link to post
Share on other sites
1 hour ago, Roman said:

What ECU are you using? 

You might have an option for fuel cut decel you can turn on, shuts off injectors completely.

 Link g1 lemv5

Share this post


Link to post
Share on other sites

Fuel use on deceleration is pretty minimal anyway. So fuel savings by enabling it won't really add up to much, but they do make you feel superior to people running carbs.

Running stoich under engine braking will use less than 10 cc/min.

Driving at 100 km/h will use 100 cc/min or more.

Idling will use around 15-20 cc/min

  • Like 1

Share this post


Link to post
Share on other sites
3 hours ago, Roman said:

What ECU are you using? 

You might have an option for fuel cut decel you can turn on, shuts off injectors completely.

 

Actually I do have an overrun setting, just haven't set decel kpa yet.. 

 

Share this post


Link to post
Share on other sites
14 hours ago, ajg193 said:

Fuel use on deceleration is pretty minimal anyway. So fuel savings by enabling it won't really add up to much, but they do make you feel superior to people running carbs.

Every little bit you can do all adds up eh. Plus better engine braking/no fumes on downhills. Well at least around our place with many hills it certainly helped on the v6. 

We ran the Viva on the v6 up to the nats at rotorua(?) with no fuel cut. Then turned on fuel cut for trip home (were still learning things) and it gave a very noticeable drop in consumption. 

Because v6.. Not tiny 4 pot 

  • Like 2
  • Haha 1

Share this post


Link to post
Share on other sites

I'm not saying it's not worth doing (I have mine set up to cut fuel). Definitely good regarding fumes.

 

But the fact of the matter is that you can only get so much air past a closed throttle - it will use essentially the same amount of air as it would at idle. One minute of driving at 100 is about the same fuel consumption as 10 minutes of going down hills if you don't cut the fuel.

 

The reduction in fuel consumption in your case was probably mainly attributable to changes in driving style and other changes you made to the tune on the trip. Did you go on the desert road both ways? I've always found my fuel consumption to be significantly higher on the desert road (higher altitude, so lower engine power and operating in less efficient revs/load) compared to driving closer to sea level.

  • Like 1

Share this post


Link to post
Share on other sites

I'm just been lurking this thread for ages, haven't felt like I've had too much to offer but there is something that I feel maybe relevant to this "off throttle" stuff is that the "off throttle" area of the map is diabolical to get tuning accurate for using a lot of the popular tuning methods when dealing with VE tuning.  As such, most of the maps I see I feel are likely to be dumping excessive (like not just slightly too much, but massively) amounts of fuel in some of the very low load areas of the map and people may or may not be aware that their map is doing that, it kindof feels to me like that "over-run fuel cut" feature could almost be seen as a band aid to this kind of thing.   

I don't know if anyone has played with modern Ford ECUs or not, but they don't run conventional fuel maps/VE tables.  Instead, they require a slope and offset to be defined for airflow vs pressure at each rpm and essentially assume that the airflow can overall be defined a increasing as a linear function vs manifold pressure.   There are other corrections to "tidy up" things, but the basis of their mass airflow calculation centres on that.

The interesting thing about it, is the offset is defined by "Manifold pressure at 0 airmass" - which is an interesting one.  Essentially the assumption (or reality...) is that you will not get any fresh airflow through your engine below a given manifold pressure at any given rpm, and it probably goes without saying that the manifold pressure will not be the same for each rpm.   If you do a scatter plot of rpm x map after a long session of driving you can see that even with the throttle closed the MAP will consistently stop at a minimum amount, and you get a clear shape through out the rpm range - that point may not be the zero airflow point, but it still paints an interesting picture.   

Any numbers >0 in the fuel map which are at the 0 airflow manifold pressure, or below that will be infinitely surplus to requirement.   Any numbers in the area of the 0 airflow part of the map which are "close" to what the typical cruise load VE numbers are  going to be way in excess of what they need to be, as you get close to the 0 airflow pressure the VE numbers should steeply drop off. 

Here's an example calculated by Paul Yaw for one of his rants on his company blog page:  (reference : http://injectordynamics.com/articles/shelby-gt500/ - search for "just maths and physics" and btw, NSFW)

VEvMAP

Most VE maps I see don't end up looking anything like this, but that is usually not a problem thanks to overrun fuel cut - or the fact that the engine never drews enough of a vacuum to really be too much of a problem.   I do try and do my best to get this kind of thing sorted out, and have actually played with just not running overrun fuel cut and instead having the fuel table drop away and zero out at the areas I determine are likely to be "0 airflow" - which granted relies on higher resolution map spacings than you normally see <40kpaA.  Guys I've done it for have not complained about fuel consumption and I've actually had it reported back "this is better off/on throttle than stock?!?!"... like unprompted, maybe placebo stuff making to want to find ways of being nice to their tuner, but one way or another at least my fucking around with this has been interesting.   

What I don't like about the magical "injector off at x map" thing is that you have to tinker around to progressively phase timing in or out to try and alleviate any jerkiness if the injector off thing happens at an inconvenient point, and of course the pressure where shutting the injectors off at may be optimal at different pressures at different rpm - as above.   Of course there is another thing, if you are at air speeds where wall wetting is a big factor and then you shut the injectors off, you lose your puddle... so when you turn them on again you need to rebuild that as well.   Keeping this balance nice at air speeds could arguably make it nicer to drive and also save some fuel that could be wasted from the driver ending up driving around weird transient torque things.

TL;DR?  But just thoughts/observations/experiments I've played with and still ponder on, sorry if it's actually no new news - I was triggered by "fumes" and "big difference in fuel consumption"  :)  

  • Like 8
  • Thanks 1
  • Sad 1

Share this post


Link to post
Share on other sites

Just to post another graph, because, well, that's what I do.
This is a plot of my engine, RPM vs MAP and then the colour scale is MAF airflow.
So the blue line is basically zero airflow point mentioned above.
Definitely not blanketly happening at the same KPA value throughout rev range. 
It's interesting to see how it looks like some approximation of the inverse of the torque curve. 
I guess that makes sense - at the RPM where the engine is trying harder to gulp in more air, it's going to generate a stronger vacuum.


Capture.PNG.446a5bb89b9728c4144495c5dcb6fcf8.PNG

 


I know I keep banging on about it, but it's amazing how much more resolution and ease of tuning you can get with a MAF in some areas.
With a MAF as your load axis, when you are tuning that blue line its just a flat straight line across the bottom row of your table.
Elegance through simplicity.

 

  • Like 6

Share this post


Link to post
Share on other sites

Interestingly if i turn that image on its side I can see a horses head. 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.