LPG Forum


All times are UTC



Post new topic Reply to topic  [ 58 posts ]  Go to page Previous  1, 2, 3  Next
Author Message
PostPosted: Wed Apr 13, 2016 3:35 pm 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
Hi all,

OK, I had a few hours to hook up nearly all of the wiring.

All wires are all now connected and being read correctly by the LPG ECU except the LPG injectors. Reason for this is that my old loom doesn't have the red band on it to determine whether is it bank A or bank B, and annoyingly I didn't think to take note when I removed it from the old car. I can't find a pin-out diagram/table anywhere online for the Romano ECU. Does anyone have one please? :)

I've noted down the following:

Bank A (Left side of engine whilst looking from the front)
Cyl 1
8 - Original petrol signal
9 - Petrol signal from LPG ECU
Cyl 2
10 - Original petrol signal
11 - Petrol signal from LPG ECU
Cyl 3
12 - Original petrol signal
13 - Petrol signal from LPG ECU
Cyl 4
14 - Original petrol signal
42 - Petrol signal from LPG ECU

Bank B (Right side of engine whilst looking from the front)
Cyl 1
29 - Original petrol signal
1 - Petrol signal from LPG ECU
Cyl 2
2 - Original petrol signal
3 - Petrol signal from LPG ECU
Cyl 3
4 - Original petrol signal
5 - Petrol signal from LPG ECU
Cyl 4
6 - Original petrol signal
7 - Petrol signal from LPG ECU

LPG Injectors
56 - Cyl 1 - (Yellow)
55 - Cyl 2 - (Orange)
28 - Cyl 3 - (Red)
27 - Cyl 4 - (Brown)
21/49 - Common

26 - Cyl 1 - (Yellow)
25 - Cyl 2 - (Orange)
24 - Cyl 3 - (Red)
23 - Cyl 4 - (Brown)
21/49 - Common

But I don't know which set of LPG injector goes to which side of the engine. Grrr. If no pin-out table is available, how do I found out (other than trail and error!)?

Otherwise, I am really happy with the petrol injector loom, there are no connectors, it's all spliced into the original loom and you can't tell there are extra wires in there :) Still runs perfectly on petrol. Pics to follow once the LPG injector loom is also installed.

EDIT 6:45pm: Ignore the above, it works!
Trial and error technique utilised. All 8 cylinders now running on LPG. Hurrah! I need to tidy up the wiring now and a few other jobs, then can start testing... More tomorrow... :D


Top
 Profile  
Reply with quote  
PostPosted: Thu Apr 14, 2016 10:51 am 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
Only 2 options regards which injectors go on which bank anyway :lol: If you'd connected them wrong would see one side of the engine go rich while the other went lean.

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Thu Apr 14, 2016 6:02 pm 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
LPGC wrote:
Only 2 options regards which injectors go on which bank anyway :lol: If you'd connected them wrong would see one side of the engine go rich while the other went lean.


Yup, and surprise surprise, I got it wrong first time! :D

Sooo... It works!

Mostly. :)

I didn't want those huge LPG loom connectors everwhere, and well, there wasn't space for them anyway. I spliced the loom into the existing one, although it was a little scary cutting the first wire on an injector. Still, it worked out well, and I think looks rather neat. When all back together, you have to look hard to see any differences:

Left
Image
Right
Image
Before air filters/MAF's/plastics
Image
All back together
Image
Switch/gauge hidden in coin tray
Image

First startup after working out what side was what, then completing the wiring/loom...
Petrol - all 8 cylinders. OK, good start! The LPG ECU still had the map from the e55 so I figured this was a good base. I've got slightly larger nozzles (3.2mm rather than 2.8 last time), regulator pressure is 1.25 bar - up slightly (half a turn on each regulator). It should at least idle. And it did!

This is where my lack of understanding kicks in - no laughing at the back there! On petrol, AFR was 14.7, so figured I need around this figure on LPG Torque suggested the the AFR(measured) was around 18, so I upped all map values by 10%, then 20, and finally 30%. It settled around 14.7, and idled well. Is this actually a good way of noting the mixture, or am I being silly here?

Here are some screen shots of the numbers that I currently have. Any feedback, as ever, would be most welcome!

Before adding 30%
Image

Image

No idea what these numbers do...
Image

After upping by 30%
Image

Image

First drive. Surprisingly good! AFR showed it running a little lean to start with, but it seemed to sort itself out after 5 mins of driving. It pulls well, and I genuinely can't notice any power drop, but if I move my foot quickly on the pedal, it hesitates quite badly before picking up. It will take full throttle though all the way to the rev limiter. Oh, one thing I should note is that I still only have a 5mm balancing pipe between the gaseous pipes between each bank. I forgot to order some T-pieces, so these will arrive in a few days.

I am rather happy with the project so far, and certainly enjoying it. I hope with a little tweaking it will OK to give it an extended test on Sunday when I drive back to Germany :D


Top
 Profile  
Reply with quote  
PostPosted: Thu Apr 14, 2016 6:23 pm 
Advanced Member

Joined: Wed Apr 09, 2008 9:00 am
Posts: 7882
Images: 0
Location: Peterborough
You've set the AFR wrong. Stoichiometry on petrol is 14.7:1 but on Propane is 15.5:1 so you've set it too lean which is probably why it initially ran rough until it had chance to alter the fuel trims.

_________________
'96 Saab 900XS, AEB Leo, sold
'93 Range Rover 4.2 LSE, Lovato LovEco, sold
'98 Ex-Police Range Rover 4.0, Singlepoint AEB Leo, my daily motor
'97 Range Rover 4.0SE, eGas multipoint, a project.....

Proud to be a member of the YCHJCYA2PDTHFH club.


Top
 Profile  
Reply with quote  
PostPosted: Thu Apr 14, 2016 10:27 pm 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
On the subject of AFR though, 14.7:1 is richer than 15.4:1.. Also, I believe meters work on the principle of reading oxygen levels in the exhaust and therefore need to be calibrated for the fuel in use... Correct mixture (detected by the proportion of remaining oxygen in exhaust) running on one fuel will dictate a particular AFR (14.7:1 for petrol) while running on another fuel the same proportion of remaining oxygen in the exhaust will dictate a different AFR (15.4:1 for LPG). If we want the proportion of remaining oxygen in the exhaust to remain the same when running on petrol or LPG and we use a meter calibrated for petrol, then even running on LPG we want that meter to show 14.7:1. In which case using such meter to set AFR at a measured 15.5:1 while running on LPG will see the engine running lean..

Best way to calibrate a technically quite demanding engine is by monitoring fuel trims using an OBD live data scanner - Aim for the same trims on LPG as on petrol under all closed loop driving conditions, and then check high load open loop mixture using the same tool (will either show actual lambda value, lambda equivalent value or calculated AFR).. Without an OBD live data scanner you can do it the old fashioned way - see if petrol duration goes up or down when you switch to LPG (if it goes up the LPG calibration is too lean and vice-versa). You cannot really use an exhaust gas analyser to set up calibration for any driving condition where the engine is running closed loop (because calibration might be out by over 20% which is the range to which the petrol closed loop system can adapt for incorrect mixture, so the EGA will show correct mixture within maybe a 40% range!) but you can use an EGA to set up open loop operation (usually this means foot at least close to the floor on accelerator), though again you would need the code reader to know when the engine is running in open loop mode... and if you have the code reader you don't need the EGA! Edit 15/4 - If you had a newer ECU (and added a couple of wires to the loom) you could run those wires to the vehicle's OBD socket and could see fuel trims on LPG software screen (use instead of a live data scanner), though the LPG software doesn't sample the fuel trim data as often as an OBD live data scanner. .

If your calibration is near correct (or on the lean side) then you will need to increase pressure or nozzle size.

Your bank 2 petrol duration figures are a little higher than bank 1 trims, if this is not the same running on petrol then there may be need (later, no rush!) to add a few % positive bank 2 trim (maybe after increasing balance pipe diameter!).. First calibration needs to be done and checked (1 of the 2 methods above), and then it's a good idea to see if switching any cylinder back to petrol effects calibration and bank to bank trims more than switching any other cylinders (an injector flow comparison).

Those 1000-6000 and 2ms to 18ms ranges define what figures appear as row and column headers on your map screen, so you can re-define those figures. For your engine, since it idles much lower than 1000 rpm, might be best to re-define at least the rpm column (this won't be totally necessary since you fitted MJ injectors but can make for finer control and better results nonetheless. I wouldn't expect this to make any difference at the moment but if you increase pressure or nozzle size it may help make a positive difference to control). On my car I re-defined rpm headers to 500, 750, 1000, 1250, 1500, etc.. because if/where you need rpm adjustment it is more likely you will need that adjustment between small changes of rpm at the bottom end of the scale than you would need adjustment even between big rpm changes at the top end of the scale... More critical on mine than it should be as it runs too much pressure at the moment (I fitted my reducer underneath and have been too busy working on customer cars to mess with my own!), so injectors pulse for less duration than they really should at idle.. with lower injector duration pushing towards the minimum at which the injectors will predictably perform, fine rpm adjustment can becomes more of a factor.

Hesitation when putting foot down may be due to any combination of pressure/nozzle size, mapping, extra injection filter settings, enrichment in acceleration settings. For the moment, concentrate on achieving same fuel trims on LPG as on petrol (or using the old method, achieving same petrol injector duration when you switch between fuels).. Then we get around to dialling out hesitation etc. Also while you're booting it with calibration as it is, either watch the gas injection figures in software (do they go to zero) or watch the switch (does an LED start flashing?).. If so, then the system switches back to petrol temporarily due to excessive duty cycle of LPG injectors - This would be expected at high rpms given the high numbers (170's etc) in the cells at high rpm and high loads... These figures would see the LPG injectors pulsing for around 1.7 times the petrol injector pulse duration and at 6000 rpm this would mean the injectors wouldn't have time to close before they would need to open again for the next cycle (depending on which injectors you selected in software...?).

Without re-reading the full thread, remind me what nozzles and pressure currently are?

Great looking job!

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Wed May 04, 2016 9:24 am 
New member

Joined: Wed Feb 05, 2014 12:56 pm
Posts: 68
LPGC wrote:
On the subject of AFR though, 14.7:1 is richer than 15.4:1.. Also, I believe meters work on the principle of reading oxygen levels in the exhaust and therefore need to be calibrated for the fuel in use... Correct mixture (detected by the proportion of remaining oxygen in exhaust) running on one fuel will dictate a particular AFR (14.7:1 for petrol) while running on another fuel the same proportion of remaining oxygen in the exhaust will dictate a different AFR (15.4:1 for LPG). If we want the proportion of remaining oxygen in the exhaust to remain the same when running on petrol or LPG and we use a meter calibrated for petrol, then even running on LPG we want that meter to show 14.7:1. In which case using such meter to set AFR at a measured 15.5:1 while running on LPG will see the engine running lean..


Makes sense about the metering, I've thought about this before.

Correct me if I'm wrong..........

14.7:1 is richer than 15.4:1 ON THE SAME FUEL.

Runing propane at 15.4:1 as opposed to 14.7:1 looks like we have more air = leaner mixture but you have to be careful about what you mean by "lean" and "rich".

Air/fuel ratios are quoted by mass. The molecular mass of propane is lower than that of petrol and in fact you're always burning very many more molecules of propane (as opposed to petrol) for the same amount of air. So by confusing the terminology you might even say that propane always runs "richer".

[strictly speaking there is only an average molecular mass for petrol since it's a blend of different molecules, whereas uncontaminated propane isn't]

Ah.........but the calorific value of propane is lower than petrol, so the same mass of petrol should be using up more air, so therefore running propane at 15.4:1 (on a mass:mass basis) is still running weak, in fact more so.

Well no actually, because that's another terminology confusion. The calorific value of petrol PER LITRE is higher than propane, but the value of propane PER KG is actually higher than petrol. That's why the ideal A/F ratio (mass:mass) for propane is higher.


Top
 Profile  
Reply with quote  
PostPosted: Wed May 04, 2016 6:45 pm 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
Oldskool wrote:
Correct me if I'm wrong..........

14.7:1 is richer than 15.4:1 ON THE SAME FUEL.
Correct, because the figures denote a ratio, where the first figure in the ratio is the proportion of air, the second figure is the proportion of fuel.
Oldskool wrote:
Runing propane at 15.4:1 as opposed to 14.7:1 looks like we have more air = leaner mixture but you have to be careful about what you mean by "lean" and "rich".
Got to be careful with wording though! 15.4:1 is the correct (stochiometric) ratio for propane, 14.7:1 for petrol, so burning propane at 15.4:1 or petrol at 14.7:1 will leave a similar proportion of oxygen in exhaust gases in either case, even though more air is used to burn a kg of LPG than is used to burn a kg of petrol. It is the proportion of oxygen in exhaust gases that exhaust gas analysers use to calculate afr, so if we use an afr meter calibrated for petrol then for stochiometric burning of LPG it should still read 14.7:1 as opposed to 15.4:1. I.e. If an exhaust gas analyser afr meter calibrated for petrol reads 14.7:1 running on propane, then the actual afr running on propane will be 15.4:1.

Oldskool wrote:
Air/fuel ratios are quoted by mass. The molecular mass of propane is lower than that of petrol and in fact you're always burning very many more molecules of propane (as opposed to petrol) for the same amount of air. So by confusing the terminology you might even say that propane always runs "richer".
I don't know much about chemistry, so if you say so! At school I quickly took dislike to chemistry because I like to know things from the base up and in chemistry you don't get to know the route of much because the real roots of chemistry are in the domain of quantum physics.. So it didn't really wash with me when the chemistry teacher attempted to explain the cause of chemical reactions, all to do with numbers of electrons in specific orbital plains etc, when he couldn't then explain why certain electron orbital plains could hold only a certain number of electrons... but this did serve to make me more interested in physics, though I'm not exactly up to Brian Cox' standard in physics either! Nevertheless a quick search on the net gives me the impression you're essentially right.. If molecular weight of propane is much lower than molecular weight of petrol but by mass (not molecular mass) a very similar amount of oxygen is needed to burn either fuel, then less air must be needed to burn each molecule of propane than each molecule of petrol, and in this context we might consider propane burns richer than petrol.. But the AFR's we commonly refer to are based on mass not molecular mass... If we have a kg of two fuels and one of those fuels has half the molecular mass than the other fuel, then we still have a kg of both fuels but we have twice the number of molecules of the lightest molecular mass fuels.. I know there is a way of calculating the number of molecules of either fuel that will react with a certain number of molecules of oxygen but then to arrive back at an afr useful for our purposes we would need to factor in how mow many molecules of fuel weighed a kg and how many molecules of oxygen weighed a kg.. If you were giving someone a 1/4 pound of raisens and a half pound of apples because the person needed these ingredients for cooking and their recipe said to use them in that ratio (and the person didn't want to have any raisens or apples left over), one approach would be to calculate how many raisens and apples to give them based on the weight of one raisen and one apple, or a far simpler approach would be just to weigh the ingredients... With ratios based on mass, the number of raisens in a pound and the number of apples in a pound are pretty irrelevant, as are the number of molecules of fuel and number of molecules of oxygen. In cooking the writer of the recipe might have some formula for the perfect ratio of number of raisens versus number of apples but when writing the recipe book will simplify the ratio to weight of raisens versus weight of apples - It wouldn't make much sense for the cook to count out raisens but that would be more sensible than us trying to count out molecules.
Oldskool wrote:
strictly speaking there is only an average molecular mass for petrol since it's a blend of different molecules, whereas uncontaminated propane isn't
yes but would think since the smallest drop of petrol will contain millions of molecules, the different molecules in each tiny drop will average out so comparing the properties of one tiny drop to another tiny drop would reveal no discernible difference.[/quote]
Oldskool wrote:
Ah.........but the calorific value of propane is lower than petrol, so the same mass of petrol should be using up more air, so therefore running propane at 15.4:1 (on a mass:mass basis) is still running weak, in fact more so.
No, calorific value tells of how much heat will be produced by completely burning a set quantity of fuel, only related to afr because there must be enough oxygen to burn that quantity of fuel. We already know (from the AFRs based on mass) that burning 1kg of petrol should use 14.7kg of air and that burning 1kg of propane should use up 15.4kg of air. If calorific value of petrol and propane were the same, this in itself would tell us we would need more air to generate the same heat burning propane compared to burning petrol. We already know (from the calorific values) that more heat will be generated by burning 1kg of petrol than burning 1kg of propane. If AFR for petrol and propane were the same this in itself would tell us we would need more air to generate the same heat burning propane compared to burning petrol. The effect is double whammy/compounding... we need to burn more weight of propane than weight of petrol to make the same heat, while by weight propane also needs more air to burn.

Oldskool wrote:
Well no actually, because that's another terminology confusion. The calorific value of petrol PER LITRE is higher than propane, but the value of propane PER KG is actually higher than petrol. That's why the ideal A/F ratio (mass:mass) for propane is higher.
The AFR is due to a point you made earlier (how many molecules of oxygen are needed to burn one molecule of fuel, then upscale those figures to account for how many molecules of fuel make a kg and how many molecules of air make a kg).. then we can arrive at a mass based AFR (one of my points above) so the 14.7 and 15.4 are handily related to mass not volume (handy because volumes go up at different rates due to heat expansion, mass is not effected by heat expansion - if we used AFR's according to volume instead of mass the information would only be true for a certain density/pressure of both fuel and air, so only true at a specific temperature). We buy fuels by volume, not by mass/weight, and a litre of propane weighs less than a litre of petrol, or to put it another way a kg of propane takes up more space than a kg of petrol. The last word is that a 50L tank of petrol contains a bit more heat energy than a 50L tank of LPG.

As said before, using an afr meter to calibrate a slave type LPG system is not necessary or particularly helpful. With a slave type system the calibration could be incorrect by a good margin but the petrol system will compensate for what it can see is incorrect afr, so LPG calibration would need to be way out before a separate afr would see any change to the afr that the petrol system wants to see. The only conditions an afr meter become useful in are setting LPG calibration while the petrol ECU is in open loop mode but the petrol system might run an open loop mode during only near full throttle operation.

Much has been said on this forum and elsewhere about comparing mpg and peak engine power between petrol and propane (LPG). The power engines generate is related to the amount of air that they can suck in (which is related to engine size, engine volumetric efficiency, any turbo/supercharging), which in turn tells us how much fuel the engine can burn (according to AFR for the fuel used), which in turn tells us how much heat will be produced (according to calorific value of the fuel) - It is the heat produced which expands gasses in the cylinders causing pressure to push on pistons, so there is a direct link between heat produced and engine power. It is a common misconception that engines run hotter on LPG than on petrol... If more heat was produced burning LPG than petrol that would be an advantage because the engine would then also make more power, while engines make varying amounts of heat in any case depending on throttle position etc! Though it can be true that heat is distributed differently in an engine running on different fuels - If, say, one fuel burns more slowly in the cylinder than another, then a greater amount of heat might be in the cylinder when the exhaust valve opens, which could mean that the exhaust valve runs hotter. The amount of air an engine can suck in will be very similar regardless of what fuel is being burned, so without need to read very far between the above lines this implies that running on LPG the engine should make quite a bit less power. But the above does not take into account any of the great many other factors.. Examples - If we were to use a fuel with twice the calorific value of petrol with the same afr as petrol, the above would imply that the engine should make twice as much power running on this other fuel compared to petrol but that would not quite be true because the law of diminishing gains would work against us. Petrol has a slightly higher calorific value than LPG but the law of diminishing gains in this case is in reverse so works in our favour to lessen the effect. Engine pumping losses (partly from sucking in air through the partly closed throttle valve) are detrimental to engine efficiency (and hence engine power) but at part throttle where intake air pumping losses have most effect we can expect an engine running on LPG to suffer less from the effects of pumping losses. Already said that engines make power by increase of pressure in cylinders - so if the cylinders contained no air (and the fuel used required no air to burn it), then the engine would not even run never mind make any sort of power.. so given engines need air in cylinders in order to expand and push on pistons, in many cases a bit more air, as needed with a fuel with a higher ideal AFR, can also increase efficiency. There are flip sides on the one hand and double whammies on the other hand for all so far mentioned aspects of engine efficiency and there are plenty more aspects to engine efficiency besides those mentioned. Just for a bit of a taster of some of those aspects and how they inter-relate to make things very complicated... Running on LPG, as said, pumping losses on intake strokes at part throttle might be lower, but on the flip side the engine then has to compress that extra air leading to compression losses, but mitigating compression losses the air itself gets hotter and expands so at high enough rpm acts like a spring to press down on the piston on the power stroke thus reclaiming some of the losses during compression. Now we're talking engine compression, higher compression is generally a good thing as the same explosion in a smaller space will lead to higher pressures to push on the piston, while the decreased space in the cylinder at top dead centre will mean more of the exhaust gas can be expelled and then intake charge can begin flowing into the cylinder at an earlier point on the induction stroke, so more intake charge in total might come in, but on the flip side the exhaust gas remaining in a lower compression engine adds to the volume of gas in the cylinder that will be expanded along with the incoming fresh charge without affecting the fresh charge afr... a bit like a high volume default exhaust gas recirculation system, lowering the peak temp of the burn but increasing the burn duration (whether that is that a good thing or not will depend on other aspects of the engine).. Compression goes on to link to flame speed and other factors that will all have an effect.

Summing up... All points here, plus many others not mentioned, plus no doubt many I am not aware of, will all play a part in determining how a certain engine design's power and economy will be affected running on LPG instead of petrol. What they generally (mode average) boil down to is that the majority of engines running LPG will use 10% more fuel than they would running petrol during the same driving conditions, even if doing the maths on calorific value might suggest an engine running on LPG would use more than 10% more fuel than the engine running on petrol. There are some engine designs that use more than 10% more fuel running on LPG (worst case scenario might be maybe 27%) and some engines that actually use less LPG than they would petrol, but the mode average would be in the range of 10% more LPG used than petrol. In all cases including the worst case scenario, large savings are made running on LPG compared to running on petrol. In most cases LPG users will save nearly 50% in money running on LPG instead of petrol.

A marginally suped-up Vauxhall Monaro that I converted a few years ago has just been in for it's LPG service.. The owner reported it was still running great, no problems, so the service was a safety check, filter change and check of calibration (I made only slight adjustment). The owner also reported seeing no difference in mpg and feeling no difference in performance between running on LPG or running on petrol, and said he recently had it on a rolling road where it made 380bhp on petrol / 360bhp on LPG. He's more than happy at that and I think most people would be too, considering the un-discernible difference in performance combined with saving over 50% on fuel bills (petrol in area maybe £1.06, LPG 49p or as low as 43p at the Calor depot). Even if he fills up at local motorway services at 53p he will save over 50%.

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Thu May 05, 2016 10:02 am 
New member

Joined: Wed Feb 05, 2014 12:56 pm
Posts: 68
Agreed :)


Top
 Profile  
Reply with quote  
PostPosted: Tue May 17, 2016 1:02 pm 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
Firstly, apologies for the lack of communication the last few weeks.

Life.
Work.
Family.
Work.

I've had a little time here and there to play with the car, but certainly not a half hour to sit and write this... However, the car has done around 1500 miles on LPG since my last posting.

Interesting read on the 14.7/15.4 subject. Whatever the correct value for LPG, the car's ECU will always adjust the trims to try to achieve what it perceives to be 14.7. I can't decide whether that means the car is running slightly rich on LPG, or whether the 14.7 is actually a false reading on the Lambda as LPG stoichiometry is met and all is well? I am now more confused after your in-depth discussion! I thought I followed Olskools post, but then I read Simon's reply and my head went pop! The most important take out in this piece is that it used to cost me €130+ to drive to Germany - it now costs a little over €40! LPG is currently €0.38/litre (under 30p) in Belgium. 98 unleaded is €1.46.

So...

Short and long term fuel trims are now in line with petrol figures - or as close as possible - petrol trims tend to vary by +/- 7%, on LPG it swings a little wider +/- 15%, but both usually centering around 0, and neither ever going off the scale (as they were when I first started!)

Screenshot from 'Torque' on my phone:
Image

Also the petrol duration is within a few tenths of a second on LPG compared to petrol - Certainly at least on the lower half of the rev range. Although one bank is quite different to the other on either fuel, as per the screen shots on previous posts. The biggest problem with this diagnostic is that I reach silly speeds far too quickly. It's a brutal car in a straight line. You can't just put your foot to the floor in 2nd as it simply spins the tyres, and it will still sometimes cut loose in 3rd. The upper revs of 3rd gear put you well into 3 figures, and I'm too busy looking ahead to see what the numbers are on the computer. A fun 1st world problem! I guess I need to get someone sat next to me.

On saying that, I'm now pretty sure I have the mixture more or less right throughout the rev range (mostly based on seeing the STFT numbers). But in some places the LPG injector map is as high as 230. If this is a percentage of petrol injector timing, then how is this working so well?! As you say, the injector would just be open most/all of the time! (It is set to "Romano FAST" btw if that makes a difference) The car genuinely runs well on LPG apart from this change in throttle response problem described hopefully in more detail below. Regulators are currently running around 1.4 bar - (a 12mm balancer pipe is now in place) - is it safe to go higher pressure and reduce the map numbers? The magic injectors can apparently handle 4.5 bar. Nozzles are 3.2mm.

Interestingly, the car does not go open loop on full throttle - even at 7000rpm in 4th - foot hard on the floor at 130mph (it has 7 gears ;) ). Although Torque says I only ever get 92% throttle when it is on the floor. It goes open loop only when decelerating.
Also on that topic, I have petrol enhancement turned on above 4000rpm on the Romano CPU, but the lights never change on the gauge. It properly pulls hard all through the rev range, as demonstrated in the Instagram video below that a friend took on Sunday (car running on LPG):

Video of gentle acceleration :)

I need to get that speed limiter removed!

OK, so the only remaining problem, this stuttering...

It runs fine all the time, except this flat spot/glitch/stutter. It usually only happens when the engine is below 3000rpm, and when the throttle is pressed between anywhere between 0 and about 30%. If you are above 30% already, then floor it (as per video above) it is absolutley fine, but during normal cruising it's bloody annoying! The glitch/stutter is so quick, that it doesn't register on anything that I am metering, but it feels like a full loss of power just for a moment. This also puts the engine check light on (hard code) with a very generic and unhelpful "powertrain" error in the log. Where do I need to start looking?


Top
 Profile  
Reply with quote  
PostPosted: Wed May 18, 2016 11:19 am 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
1. With 14.7:1 afr reading from car (stoch for petrol), LPG mixture will actually be 15.4:1 (which is stoch for LPG) but torque will still say 14:7:1 because the car afr reading is calibrated for petrol... Your car isn't running rich on LPG, it is running correct mixture. This is the gist of what I was saying regards afr readings... :lol:

2. Your Romano ECU may not even be capable of petrol addition, if it isn't and you try to enable petrol addition in software this may cause it to simply switch fully back to petrol at the point where you set petrol addition or do nothing at all. Switching fully back to petrol or a substantial addition would of course be an easy way of getting the car to drive properly when you put your foot down but does rather defeat the object (maybe not so much on a mega power car as on most vehicles)!

3. The map numbers are similar to % but not quite the same. Nevertheless, if actual gas pressure is the same as the reference pressure you entered in software would expect a map figure of 230 to mean LPG injectors pulse for over twice as long as petrol injectors at that point on the map, which isn't a good situation... Not least because the available window for injectors to open and close decreases as rpm rises and at 7000rpm the window is only 17ms, attempting to pulse injectors for longer (or even approaching) 17ms at 7000rpm will mean they exceed the window.. Just re-read your post and realised there's no need to describe windows/duty cycle etc here again (I've done so on many threads in the past!) as you already seem aware. But you asking 'Are map figures %', makes me wonder if you have compared gas injector duration to petrol injector duration (with 200 in map would expect gas injector duration to be around double petrol duration).. Both are shown at the bottom of screen.

4. The jerking could be due to: Switching back to petrol due to petrol addition settings (2) / Switching back to petrol due to exceeding duty cycle (3) / Gas injector duration well in excess of petrol injector duration can cause drive-ability problems such as stuttering (3) / Extra injection filtering settings / Enrichment in acceleration settings / Maybe a combination of those points. Then there are less likely points which could be due to you using an older ECU such as gas injector pulse not starting at same moment as petrol injector pulse per cylinder, LPG injector firing order not the same as petrol injector firing order, Romano Fast injector setting not very compatible with your injectors.

You already said trims are about the same on petrol/gas. At idle what are Pinj/Ginj (petrol/gas durations)? What is gas pressure? What is reference pressure in software? With the high numbers in your map (3), maybe you need bigger nozzles and/or more pressure.

Next time you go for a drive, watch the LPG software screen. During the glitch - Does the map ball bounce to top (or bottom) of screen, do petrol injector / lpg injector durations flutter to low or high figures (e.g. 1ms / 30ms), do gas/manifold pressure readings flutter, in gas/petrol screen does the light come on to signify switching to petrol or petrol addition, does the switch on the dash start to flash the amber gas light ? Any correlation between below/above 30% throttle and gearbox kickdown/engine rpm/fuel trim swing etc?

Just one other point on the above.. If the engine runs closed loop nearly all the time it will have 'command mixture', where at high loads fuel trims will steer mixture toward a rich mixture (below 14.7:1) anyway. Most code readers don't show command mixture in terms of afr, instead they show it as lambda equivalent ratio. In lambda equivalent ratio a value of 1 is stochiometric, below 1 is rich and vice/versa. Might expect an equivalent ratio of 0.9 to be around 10% rich (13.2:1 if calibrated for petrol). If you aim for 14.7:1 but command mixture is 13.2:1 then problems will occur, most likely fuel trim errors. Don't worry about afr, aim to get fuel trims correct.

Any chance of LPG map screen shots / screen video while you're driving?

Over the next few weeks there's a 50/50 chance I'll be converting a Merc CL65AMG (I think that's the model) with 6.5V12 twin turbo motor... For the same owner of another 2 AMG Mercs I already converted (his supercharged 5.5 and wife's n/a 5.5 as I remember).. But I don't take vids of calibration as that could prove unfortunate if pulled over!

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Wed May 18, 2016 3:21 pm 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
Hi Simon,

As ever, a thoughtful and informative reply. Thank you so much.

1. Yes, good summary. That was what I was getting at :)

2. Oh really? Interesting. I will switch the petrol enrichment off and see if it make any difference. It's just always been set to on with my previous Mercs, so I left it on... That does kinda make sense though, as I've been booting it for sustained periods this weekend, and it's used virtually no petrol at all.

3. Gas pressure and reference are set similar - 1.5 reference, about 1.3-1.4 actual. I'll bring them closer together. I confess, other than a quick glance down to the switch/gauge, I've never actually checked the software screen when I floor it to see if it momentarily switches back to petrol. I'm too busy watching the stars fly by as the car hits warp factor snot, but the gauge definitely always says it is on gas! I am driving back to the UK tomorrow, so will borrow a friend hopefully over the weekend and do some testing. I understand duty cycle and rise/fall rates etc which is why I was querying how it can even work when the map it set to 220 or more... I have compared petrol injector duration whilst running on petrol to petrol injector duration whilst running on gas. Is that right, or have I mis-read your previous posts?

4. I don't think the system is reverting back to petrol when it stutters - it actually feels like no fuel at all, and it's only for a fraction of a second. But it would make sense if it is something to do with the prolonged duty cycle - or at least I need to bring these down to sensible levels anyway, if nothing else to eliminate that from the diagnostic. Important question! ... What is a safe/sensible pressure to turn the regulators up to? This is obviously the first choice rather than having to strip the injectors to replace the nozzles.

Edit: I've just read the magic3 spec, and working pressure goes up to 1.6 bar. Possibly not enough?

Running on gas, Pinj is around 2.6-2.8 and Ginj is around 5.1-5.4, which makes sense as I'm getting these numbers from a photo on my phone and the green dot is hovering over a map setting of 189 at idle (590 rpm). As I type this I can see how silly my numbers are!
I know the map ball goes down when the glitch occurs, but i have not seen what the numbers go to. I will have to get some better results to post up here - hopefully a video of the computer screen. Torque actually has variables for measured and commanded afr, so I will compare those at various loads if both are readable (can't tell when sat in the office). But on saying that, my fuel trims are pretty good on LPG all through the rev range and at different loads.

CL65 AMG would be the M275 motor I guess - 600bhp and over 700 ft/lb. Beast! I bet there is even less room in that engine bay than in mine - be interested to see some pics of the install if it goes ahead.


Top
 Profile  
Reply with quote  
PostPosted: Wed May 18, 2016 8:50 pm 
New member

Joined: Wed May 11, 2016 6:47 pm
Posts: 9
Interested to see what solves the stuttering at rev's below 3k and light throttle opening, as I have the same problem with two jags both using the brc 56 system. one is a 2.1 that does not need the fuel return as its factory fitted and the other a 2.5 with a fuel return fitted.


Top
 Profile  
Reply with quote  
PostPosted: Thu May 19, 2016 4:30 pm 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
2. If that's the case then most likely your ECU does support petrol addition, as said if the ECU doesn't support it and you try to enable it the usual outcome is switching fully to petrol.

3. If you adjust the reference pressure (and/or reducer pressure for that matter) then you will need to recalibrate - No big loss if you have 200+ numbers in the map anyway! Your reducers won't go to much higher pressure than 1.4 bar, so will need to increase nozzle size, best done before recalibration.

4. Watch the switch for flashing lights when you boot it... During changeover no fuel at all (or double dosing), at least some sort of mis-fuelling can occur. As said above your reducers probably won't adjust to much above 1.4 bar, though injectors would be fine to maybe 1.8. If you have 200+ numbers on your map (and those figures provide correct fuelling) then an increase in pressure of only 28% or so won't be enough, you'll need to increase nozzle size.

You have the standard MJ reducer, spec says will go to 1.6 bar but I find in practice most don't. The taller HP MJ reducers are adjustable between around 1 and 2 bar.

Even if your trims seem OK running on gas, don't forget there must be some issue at higher rpm given the necessary duty cycle of injectors with figures like 200 in the map. Also I would expect problems with temperature compensation with such high figures in the map, maybe not now but in winter...

No news on the CL65AMG as yet!

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Thu May 19, 2016 4:39 pm 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
Xtype_LPG,

Though the symptoms with the Merc seem similar to your Jag symptoms, it is likely they are due to different issues. I would start by checking the fuel return is working correctly, best done by comparing petrol pressure in OBD live data when running on LPG / when running on petrol.. If that's not at fault then check the usual (plugs etc). If your system has the BRC lube which injects lube into the gas pipe between reducer and injectors I would suspect the LPG injectors are gummed up with lube fluid. Then it's onto other points such as LPG system components (reducer / injector problem with one or more injectors / calibration / etc).

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Thu May 19, 2016 6:34 pm 
New member

Joined: Wed May 11, 2016 6:47 pm
Posts: 9
fuel return is working correctly and fuel pressure is the same when running on petrol and lpg. I have a dash display showing fuel trims and petrol rail pressure.

The lube is fed in at the throttle body on both systems and the injectors have been changed for brand new ones on one car. plugs have been changed on both cars. vapour filter has been changed and reducer also changed

the calibration has been run on both cars.

still get missing when lite throttle opening, stops if switched to petrol.

hence my interest in this symptom if it gets solved


Top
 Profile  
Reply with quote  
PostPosted: Thu May 19, 2016 9:41 pm 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
xtype_lpg wrote:
hence my interest in this symptom if it gets solved
It doesn't quite work like that, your problems though similar in symptoms are unlikely to be related to anything to do with the Merc... But I could sort your problems ;-)

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Fri May 20, 2016 10:08 am 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
I looked on my past invoices, and I am actually running 2.8mm injectors, not 3.2.

I've ordered a set of 3.2mm nozzles - I know I could drill them out, but for €0.25 each, why bother...

I've also ordered a pair of the bigger Magic reducers. They may not technically be needed, but I don't like that my system is running on the upper limits of it's ability. Also, the car will be off to the tuners soon to get rid of that pesky speed limiter, and that also pushes the peak power to about 560bhp apparently. Woo. Man maths justification for buying the bigger reducers!

For sale: two used but perfectly working regular Magic reducers. They worked brilliantly on a 350bhp v8, but I think they are pushing their limits at over 500bhp. They are about 50 quid each new, so £30 each inc. postage...

New bits should arrive mid next week, so will update soon... :)

Regards C.


Top
 Profile  
Reply with quote  
PostPosted: Thu Jun 16, 2016 8:02 pm 
New member

Joined: Wed Dec 05, 2012 2:36 am
Posts: 29
Ooo, found some time!

I fitted the 3.2mm jets as well as the taller reducers. It looks much neater in there now, as I used the second output from each reducer as the balancer between the two, so no T pieces, and presumably a slightly smoother flow.

I've upped the pressure to about 1.8 bar. The LPG riser inputs are about 3.1mm, so they are now more restrictive than the actual jets behind them.

Was able to reduce the map to around the 150 level, so it has made a difference. Except it hasn't! The hesitation is still exactly the same if I put my foot down in the lower half of the rev range. Slightly disappointing, but at least it is one more thing ruled out - it's not gas starvation. I gave the car a flat map of 150 across the board this evening. It didn't seem to make any difference, but the trim level were exaggerated on hard acceleration. I'll put it back tomorrow, but rain stopped play.

Otherwise the car drives fine - actually now done a good few thousand miles on LPG. However, I am just not sure where to go from here to resolve this hesitation. What do I try next?

Shiny bits make me smile :)
Image

Fitted
Image

Oh, and cheapest LPG found to date - that's about 26pence per litre! :D :D :D
Image









In other news, a mate has bought a heap of a Toyota Previa. Horrendous thing, but it came with a very badly running and badly fitted (or am I just a perfectionist?) Dream X-something or other system. It sounds like it runs on only a couple of cylinders on LPG. I managed to get a software connection to the ECU, and it is pretty much the same as mine, software wise.

Any quick fix idea what would cause this massive petrol injection timing number on only one cylinder?

Image

Image

Image


Top
 Profile  
Reply with quote  
PostPosted: Fri Jun 17, 2016 8:34 am 
Installer

Joined: Wed Sep 29, 2010 5:01 pm
Posts: 2983
Location: Yorkshire
3.2mm nozzles and 1.8 bar... Would expect the combination to flow way too much gas.

Without re-reading the whole thread, can you see petrol rail pressure in OBD live data? If so: When running on petrol what is frp? / when running on gas what is frp? If there is a big difference in figures (e.g. frp rises by more than say 20%) then you will need to fit a fuel return if you haven't already done so (and before you'll get anything like good results).

I am currently converting a Merc C65AMG 6.5 litre V12 twin turbo, that has been tuned to maybe 800bhp with a wider stainless exhaust, increased boost, etc. But I don't expect to run as high flowing a combination of pressure and nozzles. This model has a return-less petrol system and does read frp, so while I was running Faro between the tanks and the reducers I also ran a petrol pipe from the engine bay back to the petrol tank.. Will see if I need to actually plumb it in but it seems very likely.

I just re-read and noticed you said spuds have only 3.1mm hole - Remove those spuds and fit spuds with bigger hole! As you know, the spuds are the restriction now, but this means your sequential system will become a continuous injection system at higher loads, which will really mess things up.

I expect the OMVL system has been wired incorrectly to petrol injector 1. Do the figures for inj 1 fluctuate as we'd expect (in fact do any/all of them fluctuate)? Do cyl 1 figures read properly when running on petrol? Does the car misfire on cyl 1 on petrol / on gas?

Run a pipeline from that pump over here!

Simon

_________________
Full time LPG installer
Servicing / Diagnostics / Repairs to all systems / DIY conversion kits supplied with thorough tech support
Mid Yorkshire
2 miles A1, 8 miles M62,
http://www.Lpgc.co.uk
Twitter https://twitter.com/AutogasSimon
07816237240


Top
 Profile  
Reply with quote  
PostPosted: Sun Jun 26, 2016 9:50 am 
New member

Joined: Fri Jun 24, 2016 4:41 pm
Posts: 2
Hey Piffle,

nice install on the AMG! I registered on this site so I could post this....

Ive done something similar in my BMW 550i, but wanted to increase the range per fill from what I had on petrol. Managed to get a 120 Litre 4 hole tank in the boot after quite a bit of faffing about.

I have a couple/3 of questions as I think that the only improvement over what I have may be with an AMG.lol. there are a lot of other alternatives that would be better for me, but not viable as the engines are direct injection. natural progression for me would be an M5 as I now have deacent knowlege around car.

is your car the M156 engine?

also, what range do you get on average per l of LPG?

what width and depth is availible in the boot for a large cylinder 4 hole tank? dont know how you put up with a donut tank, as your mpg has got to put you permanantly planning your next fuel stop.

cheers

Brad


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 58 posts ]  Go to page Previous  1, 2, 3  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group