Wet manifold questions / Batch / Port Injection.
Turbo Mini Forums

small update, back onto carb for last weekends tour, oddly found the fuel pressure had shot up with no adjustment from me, partly why it was running so rich, also confirmed the richness on 1 and 4 is a quirk of the manifold, its biased that way. Now converted back onto sequential injection with new ecu, adapted and additional wiring, cam sensor from dizzy. It runs and sounds mighty fine. road tuning next.

eta to add - put the injection period to end 60 deg btdc as a starting point.

Edited by Richspec on 3rd Apr, 2017.

Here for the craic..

Supercharged Arden powered

First dyno sesh came to an abrupt halt when the fuel pump failed. So I've fitted a Walbro 255 to replace the original mpi pump, also fitted a new regulator with gauge so i can monitor it easily.
Then back to JDM dyno this weekend gone, I also was paranoid about the injectors being too small so refitted the big 630's, this meant a lot of time redoing all the fuel mapping as the figures were now all to cock.
At the end of the afternoon the figures were 117bhp and 107 lbft.

https://www.youtube.com/watch?v=WYiNOAEn96Q

I've plotted the torque figures onto the minisport graph (I know different dynos!) to see how they compare, I'm up mid range but down at the lower end but both graphs follow the same shape.
Its a good start but i feel theres more to come with manipulating the injection points later in the year.
All plugs are showing a nice even mixture, perhaps looking a tiny bit rich but better than lean!

I'll try and get the graphs on here shortly.

Attachments:

91.78kb

280.92kb

Edited by Richspec on 24th Apr, 2017.

Here for the craic..

Supercharged Arden powered

fixed some exhaust issues and added a lambda in branch of the lcb and yesterday did a run down the motorway to a show, sitting at 4000 rpm I'm seeing a distinct 4 point difference between inner and outer cylinders, the outers sitting around 10 and the inners high 14's and even getting into the 16's at times on he way home, at the (lumpy) idle the figures are closer, maybe a difference of 2 or so.

Here for the craic..

Supercharged Arden powered

That's odd, usually the outers run leaner than the inners with untimed port injection.

Check that you have the sensors the right way round, inner/outer and that you are not getting an air leak on the inner leg of the LCB.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Paul, it has to be an oddity of the inlet manifold design as it was the same on a carb, you could see it by looking at the plugs, now i have actual figures though! I suppose the next thing will be to look at moving the injection timing around. I need to read up on the DTA systems of doing this to get an idea on what to do, theres an rev/tps table involved in this i know but its blank as yet.

Here for the craic..

Supercharged Arden powered

update time, still fighting against the inbalance between the paired cylinders. At the moment I've got 2 and 3 at 6% extra fuel and 1 and 4 at -26% (i think). Once the AFR is matched across all four then of course the 6 will drop to 0, its just to stop those two from getting too weak as the fuelling is slowly fine tuned. I did look at refitting the original alloy manifold that didnt have noticeable cylinder imbalance (or not as bad), sadly its too long to fit. I did see inside the current manifold that you can see where the fuel is hitting the far side of the radius, as mentioned earlier its significant wall wetting.. I think turning the ida flange round 180 deg would fix that entirely. Current mpg is approx 28 on a run round the scottish borders last weekend. So certainly not giving up and constantly revising things, despite calls to 'put the weber on' or 'put the arden on it'..

Here for the craic..

Supercharged Arden powered

chopped and rewelded the inlet manifold last weekend so the injectors are now firing down the inlet manifold, not at the back wall of it. had a drive around today and no matter what i tried the outer cylinders are still black rich, using the injector trim option i had the max allowed -50% and it made sweet fa difference, discovered a slight air leak at the manifold and fixed that, reset the idle fuelling, more testing tomorrow.

Here for the craic..

Supercharged Arden powered

Keep it up Rich, following this

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

thanks Paul, I can try that, i've currently got the injection period starting at 113 btdc, the reason being that I sat down and worked out the transit time from injector to valve, then worked that out in degress of rotation, the thinking being to have the pulse of fuel arriving on the opening valve. I've also just downsized the injectors to 480/490 cc and reset the injector trims to 0, just to see what happens. Apart from resetting the fuelling round idle and some sites in the static revving area its untested as yet, but the afr's were comparable. But thats the beauty of injection - I can alter stuff so rapidly, i just need time to drive around and monotor the laptop / make alterations.

Here for the craic..

Supercharged Arden powered

In hindsight my suggestion of 0 deg btdc was wrong as it was based on injectors near the head, whereas yours are much further away.

The problem with injecting at 113 btdc is that the air in the port destined for the inner cylinder is not moving. I'm guessing that by the time the fuel gets near to the cylinder the valve has closed. The injection event for the outer cylinder is into moving air and is probably hitting the open outer valve.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

that makes sense, I'm of the opinion i've got a setting fundamentally wrong somewhere as altering that timing from 113 to 0 and trying various offsets both plus and minus on the injectors made no difference to the cylinder imbalance, and even turning sequential off made not enough difference for me to think its working correctly. Back to the dyno and the dta expert asap!

Here for the craic..

Supercharged Arden powered

so I read this last night:

http://www.starchak.ca/efi/siamese.htm

and I'm starting to get my head around it, I'm still far from 100% like others on here but its getting clearer!

That got me thinking about what Paul said about the inner valve being shut as the fuel arrives and as detailed above the fuel should be sent in much earlier on the inner valves.
I then checked the 'injector offset' as listed under the 'unequal injection angles', by default its:
inj 1 0deg (cant be altered)
inj 2 180
inj 3 360
inj 4 540

2-4 can be anything from 0-719deg. in ascending order.

i knocked the inners (inj 2 and 4) back and afr's are better.
I tried to set the fueling to the outer cylinders but had to compromise for the moment as it was way weak on the inners.
Repeated laps and tweaking on my test route have got rid of the biggest flat spots in the fuel map and its driving quite well, the gauge is still showing a fluctuation between inners and outers of 2-5 afr points depending on revs and load but pulling the plugs when i got home its not looking bad at all, if anything 3 and 4 are richer than 1 and 2 so tomorrow I'll check the balance between the two throttle bodies and do some more tweeks to see what effect it has.
Still hoping to get on the rollers before i travel down to mitp though.!

Here for the craic..

Supercharged Arden powered

So how have you wired the injectors? I would suggest:

inj1 -> cyl1
inj2 -> cyl3
inj3 -> cyl4
inj4 -> cyl2

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Wow, top marks for sticking in there and trying to get things working for you!

I'd be wary of bore wash with running a pair of cylinders as rich as you are, as well as fuel dilution of the oil.

Change the oil often and perhaps mix a small percentage of oil with your fuel to counter excessive ring / bore wear. The latter reduces knock resistance so only use it whilst testing at part throttle until you get the cilinders to run even.

Aye I'm nothing if it not persistent, 'pig headed and stubburn' are also appropriate.!!
So today more tuning up and down the road in readiness for the trip to croft tomorrow.
Injection corrections are all 0.
Injection period is set to end at. 0, 170, 360, 530.
Injection angle is still at -113 degrees btdc
Injection angle map, which corrects against that with rev/tps etc is unused. There's no way I can keep a steady load and rev on the road to see what effect that has.
Afrs are still generally weaker on 3 and 4 but at times they are even, and I think it's as good as I can get it road tuning, watching 3 lots of lambda, revs and the road is tricky :)
Yes Paul, that's exactly how I wired, barring a wiring mistake somewhere of course!!

Here for the craic..

Supercharged Arden powered

Good effort.

How do you know that it's weaker on 3 and 4?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

sorry typo, weaker on 2 and 3

Here for the craic..

Supercharged Arden powered

well that was dissapointing :( run to croft on sunday and it was running richer than i would have liked and still with the imbalance at times. wtf!!
Reading back through the previous helpful replies above and something Paul said was pricking at my mind.
'I don't understand why you are getting too much fuel in the outers unless the injectors are only firing once per cycle. '

So tonight for shits and giggles I ticked a larl wee box entitled ' injectors in two stroke mode' (which Stevieturbo mentioned) to see what would happen, the answer is the afr's are way better on the outers (11+ versus 8's and 9's) and the middles were a happyish 13 gentle cruising, however on a big bootful the imbalance went way up 9 v 16-17.

So I think either all or some of these are happening
1. The manifold is biasing towards the outer cylinders, and even with the sequential working and swinging injection timing and % adjustments about between inner and outer cylinders it cant be overcome.
2. The injection timing is so far out that I;ve not got near the sweet spot
3. Theres a wiring issue and the injectors arent working as intended
4. The injectors are the wrong size for alternating between batch/untimed and seq port, 630cc for seq, 226 or 480 for the others.

I'll start with no 3 tomorrow night and see what develops

Here for the craic..

Supercharged Arden powered

Can you borrow an oscilloscope so that you can actually see what the injectors are doing instead of guessing and trying random settings?

http://www.jbperf.com/

Very doubtful mate, although i do wonder if there's some diagnostics in the software (i'll have to look), there is for a lot of bits plus there's the data logging, but that doesn't read the exhaust branch lambda's just the combined exhaust one.

Edited by Richspec on 8th Aug, 2017.

Here for the craic..

Supercharged Arden powered

I've just read through this from the beginning and have a few comments but first, what ECU do you actually have ???
It starts as an Omex600 but later you say "Now converted back onto sequential injection with new ecu, adapted and additional wiring, cam sensor from dizzy".

It would be helpful to read the manual to see exactly how it does some of the things you mention like different injection timings.

For something like a siamese ported engine I tend to agree with Jean, you really need to be able to look at what is actually happening if the settings don't make sense or don't do what you expect.

Have a look at this
https://www.poscope.com/product/poscope-mega1/
It's what Graham T and I use and compared to the cost of the injection hardware it is peanuts.

Schrödinger's cat - so which one am I ???

Rod, must have missed that rather useful bit of info! It's a DTA S60pro


http://www.dtafast.co.uk/download_files/Ma...es%20Manual.pdf

Here for the craic..

Supercharged Arden powered

That looks a lot better (and matches what you said in later posts in the thread).

I'm still not clear how you have wired your injector(s) to make it sequential on the siamese inlet ports but the one thing I was going to comment on was your earlier statement "the reason being that I sat down and worked out the transit time from injector to valve, then worked that out in degress of rotation, the thinking being to have the pulse of fuel arriving on the opening valve"

You are absolutely right about wanting the fuel arriving on the openning valve (esp. for the outer cylinders, it can be early for the inners) but it doesn't really matter whether you consider the transit time in mS or degrees of rotation, it is simply not a fixed value. With any form of port injection, whether the injectors are really close to the valves or a bit further away, the transit time will vary with RPM and especially engine load. The further away the injectors are, the harder it is to deal with.
Now I was going to say that those of us who use an MS2 deal with it using the siamese specific code on an MS2 which has tables (because Jean wrote the code that way) but on page 66 of your manual it appears you do too,

"(section 10) USE 20 X 14 INJECTION ANGLE MAP
Allows the injection angle, discussed in 1 above, to vary with speed and load of the engine.
See Essential Map Settings/Injection Angle Map."

From personal experience with the MS2-E code, I think you will need to implement this if your setup is more port injection than wet manifold.

And I do strongly recommend you get one of the USB scopes I mentioned to see what's happening.

Schrödinger's cat - so which one am I ???

Plus it would be nice to see a photo of your inlet manifold to see how far it falls between the definition of port injection and wet manifold....

Schrödinger's cat - so which one am I ???