The only place for a coil spring is up Zebedee's arse.
994 posts
Nov 16 2016
10 December 2016 - 19:31

Sometimes when plugs have been cut off it's because they corroded, so someone might cut the plug off to just screw wires together.

Agreed with Gilbert it could be a pre-N system, there is no N designation on the ECU sticker... but if it is pre-N it will be a rare example. Most pre-N systems have pressure sensor built into the ECU (so pipes attached directly to the ECU), most of the rest will have (at least originally) been fitted with the (external) AEB013 pressure sensor, even early N systems used the AEB013 sensor. Most pre-N internal and AEB013 external sensors will have failed by now, not easy to swap ECU internal pressure sensors but easy enough to swap an AEB013 for the later AEB025 type that is still used today (means swapping connecting plugs and changing settings in software though). Most pre N systems used ECUs with only 4 injector driving channels, for vehicles with more than 4 cylinders an external injector driver was (supposedly) added to the system (not always fitted, some installers bodged by just wiring 2 injectors in parallel to one driver channel).

Though V6 software will work on most N systems, some early N systems need software within a certain version range.

Pre N system ECU's work in a very different way to N systems, need different software altogether and are calibrated very differently. N type work as a slave to the petrol system, read petrol injector pulse duration and add fiddle factors as per user mapping to arrive at gas injector duration, so gas injector duration is calculated based on input from the petrol system. Pre N type work standalone, output specific gas injector pulse duration described by the installer's map which is based on engine rpm and engine vacuum, automatic closed loop operation is afforded by connection to lambda input (necessity).


11 December 2016 - 13:31

The BMW 4.4 engine has a couple of wide diameter breather pipes that run across the top of the back of the engine. It seems common for these to perish, some people put the perishing down to the material being incompatible with ethanol that is in modern day petrol, others put perishing down to material not being compatible with oil/HC's that inevitably make there way into breather pipes. At the time of design there probably wasn't ethanol in petrol, I tend to agree with the ethanol idea because before ethanol was in petrol we LPG installers could use LPG pipe for petrol but modern petrol causes LPG pipe to similarly perish, similarly the BMW4.4 breather pipes didn't seem to have a problem before ethanol. Since the pipes feed into the back end of the inlet manifold, leaks in them amount to a vacuum leak. I've changed them on a number of L322's that I've converted to LPG - even if there isn't a leak to start with, moving perished pipes around during other work can cause them to break up. By perishing I mean the pipes tend to become gummy, almost like they're melting, become more flexible and easily tearable. There comes a point in the process where in a cross section of pipe it would be difficult to make out where the pipe starts and the gum/oil ends. The pipes are fairly easy for a home mechanic to change and, as I remember, cost a total of around £35 from BMW dealers who seem quite used to supplying these specific pipes (common problem).

Swapping tranny fluid is something I do early on all cars I buy, particularly in autos if it doesn't look or feel like I think it should, I'll also swap coolant, even brake fluid if that looks old. But there is the possibility of causing an old box to whine or affect the shift by swapping in new fluids even of the correct spec. And getting correct spec can involve a bit of research, especially on boxes that were supposedly filled for life.


12 December 2016 - 18:40

That's an AEB025 pressure sensor, so is the later (and latest) type. Difficult to see the attaching wiring, if the wiring doesn't look tampered with and there is no adaptor (to adapt from old type AEB013 connector on old type wiring loom to AEB025 type connector, i.e. a short bit of wire with a plug on each end) then I would suspect you have an N type ECU (so is slave / much more common type similar to even the latest type systems). If that's the case then, as explained above, the V6 software is very likely to work with it. Even systems fitted 11 years ago are unlikely to be pre-N type. One way to confirm would be to check your petrol fuel trims while running on gas - if fuel trims (when running on gas) don't drift to one extreme or the other then it will almost certainly be N type, if your fuel trims do drift to one extreme (rich or lean) when running on gas then this doesn't prove that you have pre-N type because this could also be explained by there being a problem with your LPG system or with how it is calibrated. Yet another way to check is by looking on the back of the LPG ECU, if it is N type it will be stamped AEB2568, I suspect yours will have a B or C suffix at the end of that board code too (still implying N type).


14 December 2016 - 19:10

Good going with the mates P38 Gilbert. Out of curiosity (no need at the moment), does anyone have the scan tool for setting / initialising suspension on P38s and Disco's? I once changed bags and height sensors on a Disco but then had to take it to a local Landrover specialist (whom I do all the LPG work for) for them to do the setting up with the tool. Did a bit of research before going there, apparently such tool for P38s is cheap and easily available (Ebay etc) but the different tool for a Disco is a lot more expensive to the extent wouldn't buy one unless using it regularly for paying customers?


13 December 2016 - 09:06

Hmm, well without a stamp the jury is still out on whether this is a preN or N system. There is yet another way to tell... If there is a connection to TPS (likely a blue and yellow wire) it will be a preN system.

I've converted a few P38's myself years ago using a sandwich plate.. but I made up heat shields for the injectors, fitted a reducer that is more on top of the job for the V8, used another gasket with the sandwich plate and mapped it properly! ;-) With manifold nozzles in the correct place near petrol injectors, short pipes will run to injectors sat at either side of the engine, OK except this can mean they're above exhaust manifolds so pick up both infrared and convection heat and with rails widely apart at each side of the engine it isn't a complete no-brainer keeping injector rail feed pipe lengths to the reducer both the same length (not a biggie really!).

It is possible to convert these engines in the same way without even using a sandwich plate, can simply machine the ribs on the rocker covers and the underside of the upper manifold a bit to make room for 5mm internal diameter pipes to run to injectors at either side of the engine. 4mm pipe will fit even without the machining but would mean using different injectors (Matrix have 6mm outlets, 6mm or 5mm pipe will fit but 4mm won't). I've converted some P38s with both of these methods (the machining / the 4mm pipe) too.

These days I machine metal from between the finger spans of the banana manifold, so injectors can sit atop the engine with pipes running to nozzles in the manifold in the correct place through the machined holes in the upper manifold, doing away with the need for heat shields and keeping pipe lengths (including between reducer and each injector rail) all similar..


15 December 2016 - 10:39

I think when I looked into it, like you said, I found cheap gear for the P38, equipment for the Disco started at about £1k.. I let 'Simons Services' use their tool, same name as myself, unrelated except I do their LPG work!

Thanks Gilbert, Yes I've replied on the Etagas thread.


15 December 2016 - 17:07

Not useful but on the same subject - One of the easiest cars ever to get into without a key was the last shape Vauxhall Senator. Just pull up the plastic strip that runs the width across the outside of the door at the bottom of the window (thing with the outer felt window seal attached) to reveal the door lock and latch levers! Pull on either of those (can even get your finger in if not too chubby) and the door will open.. No damage, just slot the plastic trim back in place. I remember reading about Senators being unbelievably easy to get into in a newspaper, obviously the paper didn't explain how or why they were easy to get into, I found out a few years later when I bought my first Senator and had some work to do inside the door. Major design issue!

29 December 2016 - 14:51

I don't want to speak out of turn, especially not knowing the nature of the problem, but I've known P38s come up with gearbox warnings and shift gear incorrectly when there's been a problem with the throttle position sensor, a problem with the TPS could also cause the engine to cut out...

I appreciate people and firms that know their stuff (including Gilbert, wouldn't want this to read otherwise) and I've just had a look at the Ashcroft website. I have gearbox problems of my own on another type of vehicle, wondered if they might work on my type, they don't but I'm still impressed and found their site interesting.


03 January 2017 - 10:30

The negatives of long pipe lengths are not nearly as pronounced on a group injected engine as on a sequential injected engine (because group injection fires injectors twice as often, and in advance of inlet valves opening), so pipe lengths associates with fitting 2 x rails of 4 injectors on group injection Gems engines are not as critical as on sequential Bosch engines. When I convert a group injected Gems I mount a rail of 4 (down-flow) injectors to the UK driver side middle plenum allen bolt and the other rail to the passenger side front plenum allen bolt (throttle body prevents mounting rail in the middle on this side). Don't worry about keeping all pipes same length, the only 2 pipes that will be a bit on the long side anyway are those that run to the rear 2 cylinders on the passenger side, makes negligible difference on Gems particularly if you use 4mm ID pipe (low internal pipe volume compared to wider diameter pipes). The bosch/sequential is a different story, I mount side outlet injectors atop the bananas on these but remove metal from between the banana runners to allow pipes to fit through to the correct place (for outlets) in the lower part of the manifold beside petrol injectors.


04 January 2017 - 16:59

I never did get around to reading Rob's build thread Richard, seen the video though. Equally as clever as bonkers hehe.

05 January 2017 - 14:45

I've heard that parking near certain types of transmission towers can lead to battery drain on P38s and on other vehicles...?

06 January 2017 - 12:46

Yes it's an AEB2568B, which is an 'N' system, later type, the B suffix means it's 2nd generation N system (like I implied above!), it is the sequential slave type.

This is for intents and purposes fully interchangeable with even the most recent AEB2568 systems that are still made today.

If V6 software doesn't work, an earlier version such as v4.x or v3.x should. V5 software usually works with C/D suffix boards. As memory serves, B added sequential changeover, C added more advanced on board diagnostics, D is around when they became truly sequential and added connectivity to vehicle OBD2 to allow the LPG ECU to read petrol fuel trims.

Although it is termed sequential, on a B suffix it probably only reads petrol injector pulse duration from the front cylinder in each bank and bases gas pulse duration for all cylinders on the bank on only this reading, in which case it doesn't match the modern definition of truly sequential (wait for petrol pulse on each cylinder before beginning LPG injection pulse on that cylinder), but this isn't an issue on a P38.


08 January 2017 - 19:30

Don't quite follow, Marty. If you want to keep the back of the engine clear of LPG bits in order to allow better access to coil packs, splitting LPG pipe at the back of the engine (presumably to run pipes to Hana rails mounted under the manifold with gas pipes feeding to beneath the manifold from the rear) might not be a great way to go?

Given the limited options for injectors available with a Zavoli ECU would go with the Matrix setting.

No need for the balance pipe and won't make any difference.


08 January 2017 - 15:50

They are that interchangeable but it is unlikely you'd see any difference updating to a more recent ECU and in some cases swapping to a more recent ECU can even prove a negative move.

200Gb worth of music on the hard drive in this laptop, backed up from another laptop from when I used to be a DJ. And this doesn't include the hundreds of CDs full of MP3s (maybe 100 tunes per CD) that I never put on a hard drive!


11 January 2017 - 18:31

Sequential system reducers can be mounted in any orientation, though there may in theory be slight advantages for longevity if they are installed in a certain orientation, would depend on reducer type and where inside it heavy ends are likely to build up. But since any slight advantage here would be completely overshadowed by the negative effect on reducer longevity of allowing the reducer to become a block of ice, it is better on vehicles with high risk of airlocks to mount the reducer in orientation that will best prevent airlocks.

I use loads of different types of filters on new installs, the handiest types are usually the smaller types, if I get too many of the big types in stock I'll fit them on vehicles I'm converting that have loads of space! The big type probably contains a gauzy type filter, reckon the small paper type do the best job of filtering but will need changing more often. Probably no advantage to being anorak about type of any filter, biggest job they do is prevent big bits of crispy heavy ends clogging injectors (vapour) or bits of foreign stuff from clogging reducer (liquid).

A new engine wouldn't put me off running it on LPG, but I'd be 100% the LPG system was working properly.

What to do with a new engine with a new cam... bottom ends like to be run in gently at first, cams are often better run in at high rpm. But nobody is going to fit an old cam in a new engine and swap the cam for a new one when then bottom end is run in, not in a pushrod engine anyway...


13 January 2017 - 13:46

No expert on plugs or chemistry but can say what I've noticed, my point of view and what the wider consensus of opinion seems to be.

If you're not sure what make of plugs to fit on any given vehicle, the NGK equivalent to whatever is standard is generally a good bet, not really relevant if NGK is standard.

Single pronged plugs are usually the best bet on LPG, unless standard plug has more prongs. General take on this seems to be that the extra prongs shield the mixture from the spark.

A lot of modern vehicles use platinum plugs as standard, platinum is supposedly consumed during the burn process of LPG because it acts as a catalyst (in the LPG burn process). While I'm sure iridium will last longer than platinum (it's a harder metal), I don't see any problems with platinum plugs in LPG engines, though they do wear out a bit quicker than running on petrol. At school I was taught something that acts as a catalyst aids a process without being consumed itself... and there is platinum in catalytic converters anyway.

LPG needs a hotter ignition source than petrol but I would think any electrical spark should be hot enough, so I don't tend to go with theories that refer just to spark temperature. Where coil voltage is marginal for making a spark jump across a plug gap, a spark may be much less likely to form running on LPG which is a dry fuel compared to running on petrol, so closing the gap is likely to help make for a good consistent spark. For any given condition in an engine cylinder, coil voltage needs to rise to the point at which a spark will jump the gap, so closing gap can also slightly advance ignition timing, I would think a negligible effect. No harm on most engines from using a slightly narrower gap than standard, gap will increase as the plug wears anyway (used to be common to gap points just a bit narrow knowing that soon gap would be to spec). Probably no big gains to be made from different plug types and gaps, as long as there are no misfires.

Used to be fairly common to have to close plug gaps, these days seems much less common, probably due to the vehicles we convert becoming later year models with stronger coils but never had misfire problems with P38s except in cases of worn ignition system components. If I had a P38 I'd fit the NGK's and gap them down a bit.

Having said all the above, I recently did a favour for a mate, changed the plugs on his petrol only Mondeo ST220 with Duratec V6 just for the cost of the bits. He'd had the plugs changed 6 months back and gave me the impression it had run OK since then until recently, so I had to wonder if the garage had changed the 3 rear cylinder bank plugs which is a manifold off job (front bank easy access without manifold off). I drove it in high gear at low speeds high loads to make it misfire and bring on the flashing MIL light, then OBD pointed to misfire cylinder 3, which is the right side rear plug. Getting the plugs out I noticed that 5 were new NGKs but the one with the misfire was an old original type Motorcraft plug. I changed them all for new NGKs, same spec as the 5 NGKs that came out, NGKs supposedly the equivalent to the Ford spec plug. It seemed fairly obvious that I'd found the problem (the plug that hadn't been changed) and I nearly didn't even bother testing it, but I did test it and it still misfired! Then my mate told me he'd had no problems with the car until having the plugs changed last time, and he'd been back to the garage about the misfire twice but they gave up when they found no OBD codes. Now I re-gapped all the plugs (closing) to spec (not pre-gapped to suit the Mondeo) and fitted new leads (his idea). The car now drives well under nearly all conditions but I could still force an occasional misfire at very low rpm near full throttle (few people would drive like that normally), and I reckon this will be due to the NGK plugs not being as good in this engine as the Ford recommended plugs, putting the Mondeo V6 in the same bracket as the Vauxhall engines Gilbert mentioned... but I only learned this while messing with plugs on this car, and since I have converted loads of the same engine in other V6 Mondeos and V6 Jags without having to even consider plugs, can read into this that standard plugs are the best bet on this engine on either fuel and NGK equivalents problematic even on petrol (on this engine).


18 January 2017 - 20:58

You haven't just swapped the two black plugs over that carry the petrol injector signals to the LPG ECU have you...
I can send versions of Zavoli software.


19 January 2017 - 14:41

Orangebean wrote:

Mine's the 2568B Tony, but I imagine the pinouts are the same if the Vogels is a clone....

Ferryman wrote:

Is the pinout of the AEB 2568D 50-pin socket the same as installed by Vogels? I do have a hardwaremanual with pinout if you wish.

If reference is to Vogas it's a completely different ECU, different pinout. I will have a pinout somewhere but it'll be one I made myself and it's finding it... Might as well do as Gilbert suggested though and pull the cover off the back of the ECU plug, there are earth connections between wires in there that can corode. As long as ECU gets battery live and earth and battery live on red/white the switch should light momentarily anyway, regardless of rpm signal, petrol injector signals, etc etc.