MSefi.com read only forum (Jan.15,2018)

Some time ago there was a long thread about this topic. (which for some reason don't seem to be available any more)

Anyway, I want to share my experience on this topic with the forum FWIW.

I've been running 2squirts/alternating on both of my cars with great success, with good fuel consumption (Eg: Jag: 8.6Km/L = 11.6l/100km = 20.3mpg (US) = 24.3mpg(UK)
*Distance measured with GPS, fuel as per pump reading upon re-fill, freeway only) and what seemed like good power output.

Recently I decided to experiment with a more frequent (but shorter) injection scheme having read somewhere on the forum that this might improve engine response, so I started on the V8 Land Rover by changing to 4 squirts/Simultaneous (which doubles the frequency of squirts with PWs being halved from 2/alt).

The following immediately became evident:
1) Engine response to throttle changes was much snappier;
2) "Much more" torque became available at low rpms (< 1500). "Much more" is defined as: so much that I'm convinced that I'm not imagining things;
3) Much later I realised that at idle (when the engine is hot, IAT is hot and without any additional loads like A/C and lights), my PWs had now become so short that some injectors were no longer actually opening such that some cylinders were no longer firing and this made the engine idle unstable. To fix this would mean either matching my injectors so they all have the same opening time (they vary between 1.2 and 1.55ms), OR lowering the fuel pressure at idle so that I can increase the PWs during idle OR making the car idle at VERY rich mixtures (Around 12 AFR). I've already bought an adjustable fuel pressure regulator to see if it will solve the issue (not major in my book).
4) a complete re-tune of the VE-map, AE, WUE was required.

Then I decided to try the same on the Jag (this time with a new MS2/Extra):
1) Again I found MUCH snappier engine response to throttle input;
2) Again found more torque at low RPMs. It does also feel stronger towards the upper limit but I concede that I might be imaginening things there;
3) Engine note changed at higher RPMs now sounding more like a V12 rather than 2 straight 6's - go figure...;
4) No hassle with idle PWs since PWs are still quite a bit longer than opening time (about 1.25ms)
5) This could be my imagination but it certainly feels even more smooth than the turbine-like power delivery I was used to with 2/alt.
6) Even without ANY AE, the car was quite driveable. Only rather sharp throttle inputs require AE. I put this down to the 12-cyl having so many firing strokes per rev. My point is that with 2/alt, the engine still relied heavily on good AE settings whereas now it is much less sensitive to AE settings.

In both cases I tuned the engines to the same AFR table and in both cases duty cycles did not become a problem.

My conclusion is that both engines "like" having shorter squirts more often.

Have you figured out your after fuel consumption yet? I did the same change to my car and like the 4 squirts better. I had a horrible spark table, so my gas milage was bad around 22mpg US and when I switched it was about the same after getting my ve table fixed. I finaly got my spark fixed and now im up to 25-26mpg.

Lance wrote:So I think people should start with 2 squirt/alternating as the manual recommends for port injection, and tune it as well as they can.

I fully support that view Lance. As it was, I achieved very good results with 2/alt. See, it works to play dumb! That way we get to learn more from those with the knowledge.

If I, as a MS newby, had started my MS implementation with 4/sim, I would have run into that "pw less than opening time" problem and THAT WOULD HAVE messed me around BIG TIME as then I would have been totally unequipped to find the problem as I did not nearly know what I do now ( and still learning).

Your comments do explain many things I did not understand such as the tremendous reduction in AE settings that had to be made and why small changes in the VE table now has a much more drastic effect on AFRs than with 2/alt.

I have not quite figured out the "quartering" but your comment has reminded me that one should remove the opening time from both the 2/alt and 4/sim PWs and then compare.

One aspect I really WOULD like to understand though, is why both engines undoubtedly deliver more torque (more noticeable at low rpms) on 4/sim than 2/alt, given they were both tuned to the same AFR table before and after? Better evaporation of fuel?

Carl Gianonatti wrote:Have you figured out your after fuel consumption yet?

The consumption figures I quoted for the V12 was based on 2/alt and was measured over about 260km (162 miles).

The second run over the same route and more or less same distance did seem to use less fuel (4L, about one gallon US?) but I won't make that claim as I recon its quite possible that this amount of fuel could have been added to the tank after the run with a bit more patience. Too many other factors could have contributed (like the cooler weather we had on the second run, camber of the forecourt, etc)

For me the major score in changing to 4/sim was on the "torque at low rpm front". It still (and can never) pull like american iron at low rpms, but any and all gains are welcomed!

Philip,

did you measure the injector opening time ?
Seems that everyone else with the injectors for the 3.9l RV8 engine uses 1.0ms rather than 1.3ms

Yep!! I used MS' injector test mode when I lifted the rail to have them cleaned, gradually increasing the PW time until I saw dribbles of fuel forming at the tips of the injectors which I took as "opening" time.

But here's the thing!!! Not all injectors had the same opening time!!!!

One came in at 1.2ms and two came in at 1.55ms and the rest came in at 1.4ms. This leads to grossly different AFRs on the different cylinders at small PWs (during idle and at lean trailing throttle situations).

It also leads to an unstable WBO gauge, jumping around as the pockets of gas comes through at different AFRs. I'm now at the point where I want to bench test and hand pick my injectors such they all have the same opening time - not an issue on sequential of course if you can set the opening time of each injector individually. This should also contribute towards a smooth running batch fired engine.

My "measurement" was re-confirmed when I activated 4/simultaneous. At idle without A/C and lights and fans (ie as small a load as possible) the PWs became VERY close to 1.4ms. At this point, the AFR needle started jumping around like crazy (as those two 1.55ms injectors were no longer opening), and rpms became unstable. As soon as I switched on the A/C causing the PW to be > 1.5ms, the engine settled down again.

So yes, I firmly believe the opening time to be in the region of 1.4ms on the Rover injectors, BUT I'm all ears !!

Are you running larger cam(s) in your motors Philip? As Lance mentioned, the injection timing delay can make a difference to getting the squirt between the overlap period. I noticed this on my pontiac engine. Cam is in the range of 236-242 @ 0.050 on 110LSA.

No Chris this is a fairly mild "broad band" cam that was meant to produce lots of torque low down.

I'm running 2.862 fw. Does that support this "injection timing delay"? Can't remember seeing it anywhere.

Mild so around the 210-220 @ 0.050 on a wide LSA?... I wouldn't worry too much about timing delay as you would pretty much have near no overlap then depending on the seat timing but it's worth a play.

Under fuel setup, general, you'll see Injection Timing Delay (%). I set my timing delay to 60%, how this works is since there are 90° between sparks on a V8, The equation is - (90 * 60% - trigger offset) = x°. My trigger offset is 10°BTDC, so my injection events wont happen until 44°ATDC. I had found a smoother running low/midrange engine by tweaking the delay but yeah, my cam is a decent size for a 400" engine.

While it´s easy to understand that making 4 squirts instead of two halves to PW, I struggle to see why going from alternating to simultaneous does the same. I would have said that the PW would not change - only the time when the PW´s start (injectors opening) varies.

very well explained, now I´ve got it !


At the moment I used 4 alternating squirts but to my understanding it makes sense to reduce to 2 alternating s., if there are possibly that big variations in the inj.opening time (which I cannot exclude even though the injectors were cleaned and flow tested professionally) then that effect would be reduced. At high revs it might "pamper" the injectors to open less often but compensatory a little longer on each pulse. After all pressure fluctuation in the fuel rail is minimized.

Can´t see any reason to use simultaneous (thinking about a standard engine like mine) ..

(is there any issue with wiring and the number of injectors - well, it´s just that I think I started with wires that were too thin because thicker wires were hard to solder to the DB37 connector. Then, out of bad conscience, I uprated the injector wires and the ground wires (added wires to the "unused" pins)(After all - didn´t change anything for me. Problems had and have to be solved by tuning). At least if there is a "bottleneck" for the electrons to flow now it will be in the pins of the DB37 connector itself - whose size is fixed. I could imagine that this is no issue in a 4banger, but might be with 3 times the amount of injectors ???)

landybehr wrote:very well explained, now I´ve got it !

(is there any issue with wiring and the number of injectors - well, it´s just that I think I started with wires that were too thin because thicker wires were hard to solder to the DB37 connector.

I agree, very good explanation Lance, would be nice to have that in the manual.

Landybehr, I have 6 LOW IMPEDANCE injectors on one output on the Jag with no issues so you need have no worries about your setup.

Hohoho ;)

just a BTW:

have been looking through a Bosch book.

1) The Rover V8 uses 2 squirts per revolution of the camshaft.

2) And, for the "vintage" L-Jetronic (which is equivalent to the Rover Flapper-type ECU of the 3.5l V8´s) they had problems to drive more than 4 injectors per driver. So an 8cylinder was the most (with 2 drivers) that ECU could cope with. UNLESS they did some trick with the injectors = a high voltage was used to open the injectors and then a lower current used to keep them open. This relieves the drivers and then 2x6cylinders were possible.

I ended up with 6 squirts alt for inline 6 cylinder and the driveability is fantastic. Tuning was MUCH different than 2 squirt batch

I need to test my opening times (I built the little opening time thingy that Bruce provided plans and circuit boards for at March meeting) but I'm not really sure how to use it yet.

What a interesting thread.. I have gone from 4/Alt to 4/Sim on my Rover V8.. with such an improvement.

Currently using Lucas Green injectors..is there any impact on injector wear when doubling the firing?

Linzerl wrote:What a interesting thread.. I have gone from 4/Alt to 4/Sim on my Rover V8.. with such an improvement.

Currently using Lucas Green injectors..is there any impact on injector wear when doubling the firing?

less than the impact of different fuels (with different lubricating additives) or using significantly higher fuel pressures I would think

a lot of cars come with a form of 4/sim firing injectors, witrh no ill effects.

BTW, Diesel Common Rail Injectors run upto 2000 bar's pressure, with reasonable life expectancy...

Does anyone know what most OEMs currently do as far as injector firing? Sequential, 2 or 4 simultaneous, 2 or 4 alternating, etc? Does it differ by country of origin? Do the Japanese do one thing while the Germans do something entirely different, for example?

Thanks,
Robert

PS I am running 4 simultaneous and am very happy with the results. It pushes the high-RPM duty cycle a bit, but I'm still under 90%.

73Inka2002 wrote:Does anyone know what most OEMs currently do as far as injector firing? Sequential, 2 or 4 simultaneous, 2 or 4 alternating, etc? Does it differ by country of origin? Do the Japanese do one thing while the Germans do something entirely different, for example?

Thanks,
Robert

PS I am running 4 simultaneous and am very happy with the results. It pushes the high-RPM duty cycle a bit, but I'm still under 90%.

I think all new released engines or magements use sequential injector firing, but then a lot are direct injected engines so no choice there...

Lance wrote:landyvehr,

Suppose you have a 4 cylinder 4-stroke engine. Then in 720° (two revolutions) you have 4 spark events (at 0°, 180°, 360° and 540°, then starting over at 720°=0°). MegaSquirt can only inject on an ignition event, so 1, 2, or 4 squirts/cycle for this engine.

If you have 2 squirts/simultaneous, you have:

degrees 0 ---- 360 ---- 720 -->
| bank1: x-------x------>
| bank2: x-------x------>

but with 2 squirts /alternating, you have:

degrees 0 ---- 360 ---- 720 -->
| bank1: x-------------->
| bank2: ---------x----->

Both are two squirts per cycle, but alternating only has 1/2 as many total squirts per injector (i.e., since only one channel is active per squirt), so the pulse width must be doubled.

Philip's 4 squirts/simultaneous would look like this:

degrees 0 ---- 360 ---- 720 -->
| bank1: x---x---x---x-->
| bank2: x---x---x---x-->

There are 4x as many injection events with 4sqt/sim compared to 2sqt/alt.

The req_fuel calculator will show this - watch the top and bottom numbers as you make changes to the alt/sim and number of squirts. The top number stays the same (it is the 'unadjusted number' for 1 squirt simultaneous), while the bottom 'adjusted' number used for the pulse width calculations changes with the number of squirts and simult./alternating.

Lance.

But why couldn't it inject more often in Alternating mode? Especially with a trigger wheel you know position pretty well. You would just do the injections on bank 2 based on a calculated time between the spark events. I don't know if it would offer any advantages, unless you have two fuel rails and split each rail between both banks (for less variation in fuel pressure).