Rover V8 3.9 EFI - some running issues!

[mosesthemonk]

Hi everyone,
I am new here and have no mechanical training other than owning a haynes manual and plenty of trial and error, so be gentle!
I have a 1992 EFI 3.9 Range rover. She's converted to run on LPG via a vapouriser system, not injection. A few months ago I had an issue where one of the hoses leaked and peed out some coolant. I thought it might be due to pressure, but found a small area of wear and tear on the hose (near another jubilee clip, so probably someone caught it with a tool). I topped back up with reverse osmosised water (as filtered as you can get!), got the car home and replaced the pipe. The temperature gauge never got above 1/4 as the weather was quite cool. The engine didn't smell or seem to be burned (a smell I HAVE previously encountered in my first RR, a £300 toy)
I have always had a little white steam from the exhaust (for the last year or so, as long as I've owned it) but in the last month this has increased somewhat.
Well anyway I was on a fairly long journey and the coolant light came on so I stopped, let the engine cool and checked it, sure enough the level had dropped an inch. I topped up and drove a further 200 miles, and the level had dropped another 2 inches in the expansion tank.

I did a compression check cold as I didn't have time to get to it when hot, and all the readings were about 12bar ±.8 bar. Well while this isn't informative as it was cold... I did notice something odd!
The spark plugs in the left bank were a light cream colour indicative of lean running, and those on the right bank were blackened, indicative of rich running.

I run the car on LPG 2/3 of the time, and petrol 1/3, and on hope most of the time. The latter tends to be most expensive indeed.

At no point during these issues has the temp gauge climbed above 1/2. When I first got the car I drove it up a mountain on a warmish day and it got a bit above this and exhibited no issues so I know the gauge works well.

Also, when first started and for the first few minutes, if you let it idle for a minute, and then press the throttle very slightly, the revs drop from 500 to about 200 before it revs. (not always but sometimes).

So I have a few questions:

Does anyone know if this is likely to be a head gasket, cracked head, cracked block or LPG fault? And how to diagnose each

If I put a set of rebuilt heads and new gaskets on it and it is still steaming (block damage) can this damage the new heads? And thus will I lose my money:?:

If when running on petrol it still uses coolant and steams is it still possible the LPG vapouriser coul be leaking coolant into the LPG pipes
Yours Sincerely,
Mosesthemonk

[kiwicar]

Hi welcome
I would think the next step is to do a hydrocarbon test on the coolent to see if you have combustion products escaping to the water (local garage should be able to do it, or get a kit). That will tell you if it is a gasket issue. I am afraid I suspect it may be as you have light cream plugs on one bank. This can be because you have water in the cylinders during combustion, lean mixture I would associate with light brown and the electrode with white deposits, though very lean mixture and water in the combustion arnt easy to tell apart from the plugs, it is obvious when the heads come off. All 4 plugs on one bank is very confusing The other thing that messes up plug checking is the LPG, so really I would get it hydrocarbon tested.
Sorry cant give you a definate answer yet, let us know how the coolent tests out?
Best regards
Mike

[mosesthemonk]

Hi Mike,
thanks for the advice. I haven't done a hydrocarbon test on the coolant yet, I might well still do so, but I suspect this will be positive as there is a slight whiff of exhaust gas from the header/expansion tank when opened up. If I do the test and it is positive does that mean it is definitely head gasket failure, or could it still be block or head cracks? Not nearly as worried about it being a head gasket problem as I am about the block cracking I have heard so much about.
I'll keep you posted if I test the coolant,
thanks again,
Mosesthemonk.

[kiwicar]

Hi Mr Monk
If you can smell exhaust in the header tank then I wouldn't bother with the hydrocarbon test after all, I think that will do as confirmation!
I would pull the head with the light coloured plugs off, if it is 4 blugs then my bet is it is a gasket, if it was 1 cylinder alone It would be more likley a cracked head or dropped liner, however it is a 3.9 and they are much less prone to dropping liners.
My bet is a failed head gasket due to warped head/deck surface due to a big overheating some time in its past and I don't mean getting a bit hot going up a hill, the damage will have been before that you just stressed it again and it failed a second time.
When you have the head off take a photo and post it up, and of the block and bores. To check the heads you need a straight edge, a window and a set of feeler gauges, I think it fairley obvious, but check the heads with the straight edge side to side, end to end and corner to corner, first by placing the straight edge on the head and use the light from the window to look for a gap, then measure it with the feeler gauges. If you are more engineeringly inclined then put a sheet of glass on a flat surface (3/4" mdf on a kitchen work top spray engineering blue onto the head surface and let it dry, put brasso on the glass and move the head back and forth on the brasso until you make the high spots on the heads obvious (it can be fasinating on an old head!
Best regards
Mike
Ps Get back to you later when I get home.

[mosesthemonk]

"Engineeringly inclined" would be an overstatement of my abilities and an understatement of my ambition!
I will have a go at dismantling the engine this weekend and see what I find (I'll post pics of the carnage!). I have some old heads I will check for warping as you've described. If they are ok, I will get the valve seats cut out for larger valves in naive preparation for the rebuild (maybe even have a crack at porting the inlets and playing with the exhaust manifold a bit!)
Thanks very much indeed for the help! This is why I love forums!

[kiwicar]

Hi
Good luck with the strip, tell me if I am telling my granny to suck eggs, but when you strip the engine remember things like keeping the push rods in order (a piece of cardboard with 8 holes in is the easy way) remember when draining the engine of water that the block drain plugs will be blocked with crud and no water comming out of them doesn't mean the block is drained, take the hoses off and blow in the top one to get all the water out before you take the inlet manifold off, put the engine at TDC firing stroke no1 cylinder before you take the distributor out, it saves alot of time later when rebuilding. Clean the outside of the engine with fairy power spray and a stiff paint brush/ nail brush before you strip off the heads, and manifolds, a clean engine is much easier to work with. Slacken head bolts and rocker shaft bolts in order and a little at a time and take photos as you go if it is your first rover engine, you then know how it should look later.
Best of luck and enjoy
Best regards
Mike

[mosesthemonk]

No not at all! Speaking to me at a very basic level is best to be honest! As I have learned everything I know from trial and error, but have occasionally got by with dumb luck!
When you sat to put cyl.1 at TDC firing is it sufficient to get this vaguely correct by looking at the rotor arm and turning the engine by hand and then to get it exact with the old stick in the spark plug hole trick?
The strip down has been put off for a week so I can get new (slightly larger) valves and rebuild these spare heads I have.
I will keep you all posted and I'm sure I'll be back with more questions.
THANKS!!!!!

[kiwicar]

Hi
Use the rotor arm and cap to get it on the right stroke for No1 cylinder, then if you look down at the pully on the end of the crank shaft there is a scale in degrees and a pointer that will get you spot on, set it to zero on the pointer, just rotate everything by hand (well a socket on the crank nut and a long handle wrench )
Good luck with the rebuild, come back to us with any questions as they come up.
Best regards
Mike

[ramon alban]

the coolant light came on so I stopped, let the engine cool and checked it, sure enough the level had dropped an inch. I topped up and drove a further 200 miles, and the level had dropped another 2 inches in the expansion tank.

I did a compression check cold as I didn't have time to get to it when hot, and all the readings were about 12bar ±.8 bar. Well while this isn't informative as it was cold... I did notice something odd!
The spark plugs in the left bank were a light cream colour indicative of lean running, and those on the right bank were blackened, indicative of rich running.

Hello Mose, First off let me declare I have no knowledge of LPG symptoms, but based on my limited knowledge, I'm not convinced you have a head gasket or cracked block problem.

Your compression tests showed all cylinders the same, the prior leaked coolant problem and subsequent refilling would indicate that you did not get all the air from the cooling system

Particularly as you also had a subsequent low coolant indication and are still losing coolant due to pressure in the system induced by an incomplete fill.

Air in the system allows water to boil locally and generate steam and pressure that forces water from the expansion tank. If the situation is not corrected it will never self resolve.

Then there are the uneven plug colours on each cylinder bank.

The light coloured plugs are clearly OK, but the black ones indicate a rich mixture and therefore a possible ECU failure.

I would certainly be looking to investigate further before stripping the engine.

[ChrisJC]

I have had an internal vapouriser leak, which leaked coolant into the LPG stream. A most problematic symptom!

Chris.

[DEVONMAN]

Hi Mosesthemonk.
You certainly got the old grey matter working on this problem.
I would say that you possibly have two separate issues here.

1. Different mixture to left and right banks.
2. Possible coolant loss.

On item 1. With all 4 plugs on opposite banks showing different colours it's got to be ECU related. The system alternatively fires all the injectors on one bank and then all the injectors on the other and so on.
As there are 2 lambda sensors (1 left and 1 right) I would assume that the ECU corrects both sides independently, (someone please correct me if that is wrong).
Based on this, I suggest you check out the lambda on the rich side as that side may be running in open loop mode.

On item 2. I don't know how the vapouriser is set up but to eliminate it as the culprit, is it possible to temporarily disconnect or bypass the coolant feed and run on petrol. If the steam dissapears from the exhaust and water loss improves then you will be nearer to solving the problem.

Given the smell from the header tank, You may well have a head gasket or liner problem buy it is very unlikely that all four cylinders on one bank have the problem and therefore the spark plug colour is probably unrelated to the water loss. The gasket water ways on the RV8 are adjacent to the front and rear cylinders only and therefore problems with the middle cylinders gasket are unlikely to show in the coolant.

Hope this helps
Regards Denis

[Ian Anderson]

I'm confused here
Are you running on LPG of petrol

If LPG there should be no petrol entering at all
If on Petrol then you need to diagnose a petrol problem - perhaps leaky injectors

Ian

[mosesthemonk]

Hi everyone!
Thanks for all of your replies! I am still a bit confused but I will have a go at investigating these problems.

Firstly I run the car on LPG about 2/3 of the time, and petrol 1/3 of the time.

Secondly whether running on petrol OR gas I still get a bit of white 'steam' out the exhaust.

Thirdly this is worst when the engine has sat for some time unused, and then first startup, so I presumed this could be due to residual pressure in the coolant system forcing coolant into the cylinders.

My next step will be to do a hot compression test, as I have been told elsewhere that a cold one is useless, I'll post the results when I get a chance to do this test.

So my questions are these:
1. If I just loop the coolant feed from the LPG vaporiser to isolate the vaporiser from the system, and remove the LPG gas feed from the plenum and temporarily blank this then I have effectively removed the LPG system. Given the amount of steam I was seeing, how long do I have to run the car to remove risidual condensation from the exhaust to see whether the white steam goes away?

2. I am completely uneducated and a total novice to injection and O2 sensors in the exhaust, but would I be right in saying that I should check the apparently rich bank for rust and blowing around the lambda sensor as that would raise the ppO2?
If this is not how I do so, can anyone help explain how I can check these sensors or the ECU without visiting the nearest LR dealership?

3. the smell in the header tank is not strong (and could be the smell of LPG OR exhaust I guess given my limited experience ), should I watch the header tank for bubbles with the car running?

4. By internal inspection (I poke about inside the vaporiser a bit) can I determine if the vaporiser is OK visually? Will the leak be obvious as it would be if you see a head gasket failure etc.?

5. How do make absolutely sure I have bled the system of ALL bubbles?

6. If these things do not show faults is it still worth popping the heads off and having a look for what I find or should I run away and hide?


Thanks again guys and sorry for my very very basic approach!

[kiwicar]

Hi
I seem to be "going against the general wisdom" at the moment but I would suggest taking the heads off, I think your problem is water in the combustion on the light coloured bank and that it is either due to the head gasket or that you have severe damage to the heads around the exhaust port due to the LPG.
A hot compression check may help determin this without pulling the heads, but I don't think you will have a clear answer until you pull the heads.
Bypassing the vaporiser, running the engine for 10 minutes then allowing it to cool and looking for white smoke on start up may isolate a Vap problem, but I cant see how the vaporiser leaking water could result in one bank clean/normal lean and the other looking rich.
I am biased here but I thing LPG is a bad thing on an engine intended to also run on pertol, I don't think you can build an engine to run properly on LPG that can also run on petrol. Further LPG engine development just hasn't been done, none of the big car manufacturers sell LPG cars with the same warrenty as a petrol version, they don't sell them in any quantity and I think the reason for this is that they do not believe it is a good fuel to put in an engine designed to run on petrol. Sorry but I think just like my second range rover that had an LPG conversion on it, you have a problem with the top end of the engine as a result of running LPG.
Ill now go off and crawl under my rock while I get shot at, it is my day for it
Best regards
Mike

[ChrisJC]

Hey, guess what, I don't entirely agree with Mike.

Well, I do mostly. LPG is OK on the RV8 in my experience. I have run an engine for 60,000 miles on LPG exclusively (apart from starting), with no problems.

I'm presently piling the miles on another one, so my opinion may change.

But he is right, the engine was never tested by the manufacturer to run on LPG, so it is certainly outside its design envelope.

But to the point, I can offer advice on testing the LPG vapouriser for leaks. I reverse pressure tested it, by which I mean I pressurised the water jacket, and dunked it in a bucket of water. Bubbles coming out of the gas port told me I had a problem.

Chris.