Overheating 3 litre

[Robert Clifford-Wing]

Having recently had the magnetos retimed on our 3 litre and carburettors jetted and needled, can members comment on the following please.

Both magnetos were set at between 8 & 12 degrees with the manual advance/retard set at fully retarded, does this seem correct?

Fuel mixtures were indicating around 4 to 5% accept at very high revs when it reached around 6%.

I now appear to have starting issues. Oddly the engine won't start at first but, after leaving for a few minutes (to recover my sense of humour), it starts easily.

My last journey of 30 odd miles found the radiator spewing out coolant from its overflow pipe at the end of the journey, without any stops but slower speeds to negotiate parked cars etc near the end of the run. Temperature gauge showed 80c and once the fan was running normal temperature resumed. After letting the block cool I replaced 3/4 gallon of coolant and managed to replicate the same easily in the garage after 5 minutes of running at 1200 revs. There appears to be no thermostat housing and the significant turbulent visible in the header tank is caused by the water pump I presume?

Up until now the engine has run between 50/80C, how best should I proceed?
Rob

[Steve Welton]

I am slightly hesitant to comment on this .................
On my car the magnetos are set at 45° before TDC as the maximum permissible amount of advance and I usually run the car at just under this, say about 40° but that's about the way it feels as you drive.

I don't understand the fuel mixture settings because of my car I know it runs slightly rich and needs different needles.

I have had many veteran and vintage cars it is essential to be secure in the valve timing settings, the ignition timing settings and it helps for the mixture to be about right . If the car continues to overheat then obviously you are looking at cooling either the efficacy of the water pump or indeed the capacity of the radiator to cool.

[John Murch]

I find that I have to leave at least 2inches of space between the water level when cold and the overflow pipe, otherwise it just spits it out!
John

[Robert Clifford-Wing]

Sadly I had left a good 3 to 4" of space to allow for expansion so nothing so simple

[John Murch]

Thinking back some years, I had a similar incident and convinced myself that I had a crack between the valve seats and hence gases were getting into the cooling water. This was soon after the block had been lifted and some valve work done. It was suggested to me that something might be contaminating the cooling water and that the water was foaming - I put in a dose of 'water wetter' and the problem vanished!
I think it was this:
http://www.redlineoil.com/Products.aspx?pcid=10
John

[Christopher Carnley]

Retarded ignition causes serious overheating, to boiling point, as so much unused energy goes into the water.

[Dan Suskin]

I don't have my reference really handy, but I believe the ignition timing on the 3 liter is set with the ignition advanced, and I think that 45 degree sounds right - but please check the book (which I would do if I could get to it now).

[Chris Card]

Dan - that is so, but many people now set it at 42.5 btc when fully advanced, something to do with modern fuels.

Chris

[Stephen Blakey]

I've not seen AFR represented like that. If those are by mass (and how else could AFR be quantified?), then they are lean which would cause overheating. 14.7:1 is 6.8% and you should be running richer than that under load. Maybe those are CO readings, in which case I can't comment as they are off the scale compared with anything I have experience of.

I agree with Chris that retarded ignition will also cause overheating. The thermal efficiency drops and all that wasted energy has to go somewhere.

The notion of retarding the ignition for modern fuels seems counter-intuitive. The octane rating is now so much higher than when the engines were built, I can't see the need. My pal's son wrote a PHD thesis on burn rates (I made the mistake of trying to read part of it). Maybe it's something to do with those? I'll ask him.

[John Murch]

I would agree with Chris that 45° advance is usually agreed as the maximum and that better running is achieved at a bit less.
I also agree with Stephen, re Air Fuel Mixture ratios, I was advised that while a figure of 14.7 is generally regarded as ideal when cruising, vintage engines would be better a little richer.
This could account for your poor starting and lean running can cause overheating and it doesn't help your valve seats.
Is the water temperature getting up to 100° overall, or do you think that you are getting localised boiling (or as I mentioned before foaming) around the valve area - do try the water wetter, its inexpensive and may fix the problem.
The needles generally used in the 3ltr ( from memory) are CV or K, I have a pair of CV should you want to try them.
John

[Chris Card]

The spark plug electrodes can give you a good indication of whether, or not, your mixture is about right.

Get the car up to normal running temperature and then give it a good run under load, such as up a long hill, pull off the road and switch off immediately. (If you let the engine idle, it will start to soot up the plugs).

Taking care not to burn yourself on the manifold (gloves and a long extension on the plug spanner will help), remove nos 1 and 4 plugs on the exhaust (hot) side and check the colour of the electrodes.

If white, you are much too weak, if black, too rich, but a nice mid/dark brown is about right.

Chris

[Robert Craven]

Rob,
In your post under another heading you said you'd recently had the carbs adjusted on a rolling road with good results. If it was done properly you shouldn't be having the problems you now describe unless there's something faulty with the carbs.
And as others have said, the figure you gave for the magneto timing is not the usual way it's measured.

[Robert Clifford-Wing]

[John Murch]

Rob, have you got the 'Technical Facts of the Vintage Bentley'? If not, it is well worth buying (available from the club) and covers the vast majority of jobs and problems that one is likely to get.
John

PS, you also need to be a member of the vintage spares scheme to buy it.

[Christopher Carnley]

When you think about it for a moment, 4 to5% is an air fuel ratio of 25 to 20:1 by weight, and very lean which makes for overheating in itself,the ideal at stoichiometry being around 7%.

As I say to people, these rolling roads are alright but the technicians try to impose modern car conditions on car engines designed around lower grade petrol probably no more combustible than 55 octane, and not run in everyday conditions.

In the "Old Days" of the Sun and Crypton tuners, I had the carburettors set at 3.5% to 4% CO (carbon monoxide) on the advice of the tuner, and the engines always ran very well.

[Robert Craven]

Rob,
Have you tried the simple test of raising the piston very slightly with a small screwdriver as the engine is idling to see if the engine speed changes? If the speed (revs) increases and stays increased the mixture is too rich and if it decreases the mixture is too weak.

[Stephen Blakey]

At WOT the engine will want to be somewhere near Lambda 0.85 which is 12.5 to one or 8% by mass.

I can't see much point running an engine as valuable as this above Lambda one as any benefit, from slightly improved fuel economy and a reduced pong of partially burnt fuel in the exhaust, will be more than offset by the worry caused by the higher temperatures.

Hence I reckon a good starting point would be to aim for 7% at light load and 8% at high load and to err on the rich side. The engine should then tell you what makes it happy.

It's worth bearing in mind that there are big AFR variations between cylinders so this is as much an art as a science.

Robert makes a good pragmatic point about testing by raising the pistons.

[Chris Card]

I have a fairly basic air flow meter that i use for balncing the Slopers on my 4.5l - you hold it against the front of the carb. And it reads the suction against a scale. You can then adjust the butterflies in the throttles to give an even suck at fast tickover.

Having got the two carbs more or less balanced you can then work on the mixture by individually lifting the pistons slightly to see what effect that has on the engine speed. If it stays more or less constant, you are about right.

In my limited experience, you never seem to get a costant result, but it gives you a good guide point. You can then see how the engine performs on the road, check the colour of the plug electrodes and adjust according to what you find.

The consensus of opinion suggests that richer is safer, and ultimately less expensive, than too lean.

Chris

[Dan Suskin]

With SU carbs, you do better to put a dial indicator on each carb via the dampner dashpot and simultaneously measure piston lift. Just need to make a small adaptor to screw into the top of the carb.

[Chris Card]

Dan - wouldn't work with slopers.

Chris

[Stephen Blakey]

A good idea, though, and something for those of us with later carbs (on later Bentleys, in my case) to bear in mind. Did you have to screw-cut the adaptor or is a die available?

[Dan Suskin]

Stephen - I cut the thread on the lathe.

True that it doesn't work with slopers, but it was the method I wanted to show. I've used these on many other cars.

[Chris Card]

Dan - I didn't mean to sound dismissive, it's just that the thread started with slopers. For the later carburettors, it's a great idea. Obviously. a dial gauge is very sensitive - in real life, how close would you expect to match two (or more) units?

Chris

[Robert Clifford-Wing]

It would appear I have successfully adjusted the carburettors to run on a slightly richer mixture. Having driven six or seven miles, then checking the plugs in cylinders numbers one and four, the electrodes were brick red and there were signs of carbon on the outer ring of the spark plugs.

I am pleased to report the engine did not overheat, although there were signs of some foaming within the top of the radiator, that dispersed quickly, and very slight weeping marks on the water jacket and signs of coolant on the sump adjacent to the oil gauge frame. I suspect some of the bolts securing the water jacket on the exhaust side of the engine will need helicoils fitting as some are not locking into their threads.

Unfortunately our local mechanic failed to bring his electronic timing equipment to his garage today, so I was unable to adjust the timing as per BDC forum suggestions of between 42 and 45° advanced.

In spite of this I feel the engine ran very well and delivered and even power curve.

[Robert Craven]

I'm glad your engine is well, but spark plugs having brick red electrodes sounds odd to me. I hope it's just my misunderstanding.
As to the water jacket screws being loose, see Technical Facts 5.13.

[John Murch]

Regarding your plugs, the odd colour could be caused by very old petrol - or perhaps old deposits in the fuel tank now getting mixed with new fuel.
As the car is new to you, it may be a good idea to drain the tank and inspect what comes out.
John

[Julian West]

If your car tends to overheat when on tickover, for example after stopping at traffic lights, then the engine waterways could be partially blocked by scale, particularly at the back of the block.

This can be removed by one or more applicatons of Fernox DS-3 (sulphamic acid) limescale remover. As removing limescale from the radiator can cause unwanted leaks, the engine should be isolated by disconnecting the top hose and blocking the inlet from the water pump with a copper gasket. After filling the engine with the sulphamic acid (a yellow paste) dissolved in water, it can be run until the temperature reaches 70 C. The top hose needs to be loosely blocked with a rag to prevent splashing as carbon dioxide gas is vented. This procedure should be repeated until the water remains a pale yellow colour and/or still smells of sulphamic acid. Raising the front of the car when descaling prevents the formation of airlocks in the waterways.

After each application the engine should be flushed via the top hose until the water is debris free. If a fine mesh tights sock is fitted at the radiator inlet, any remaining sludge will be filtered out. This filter needs to be cleaned, initially every hour or so, as otherwise the engine will start to overheat again. When the cooling water finally runs crystal clear, the sock can be removed, and antifreeze or water wetter can then be added.

This procedure worked a treat on my 20/25 after recoring the radiator and replacing the water pump impeller failed to make any improvement.

[Stephen Blakey]

Steve Lovatt told me that Ristes used to send their mechanics out to buy ladies tights so they could use the feet in the radiator inlets of their customers' cars, however it seems this policy met with some resistance from the mechanics. They now sell purpose-made radiator inlet socks.

[Julian West]

Stephen,

Yes indeed they do, and I've got one in my toolbox. The mesh is very coarse, so that it doesn't get blocked and can be left in the system, but it only traps relatively coarse debris.

[Robert Clifford-Wing]

Hi All, having recently rebuilt our 3litre, I am still experiencing overheating issues. The water jacket was perfectly clean before gasketing and plating. There is a new radiator core (2 to3 years old) and mixtures and timings appear to be correct. There appears to be no leakes within the system, I suspect coolant is being lost via the overflow as, over a 100 mile journey 10 litres + of wager were lost. Using the manually switched fan allowed a safe return to base but there is amps issue somewhere.

On the basis the block is clear, hoses are not leaking and radiator is new I have a question. What type of water movement should one expect to see within the header tank from a serviceable water pump and, is there a method of testing its output as there is little else obvious left to inspect?

[Dan Suskin]

I see pretty turbulent flow of coolant entering the header on my 3L.

I used to fret about loss from he overflow - but someone else told me to just let the level find itself - and it does. It looks low, but is not - I now rarely have to put coolant in and it says at that level. If I top it up, it simply overflows out back to the same level while driving.

I'm not sure what constitutes overheating. Is it an absolute temperature? I don't think so. Unless it boils while running, I am unconcerned. I'm sure if the gauge had 100 in the center, rather than all the way at the right end, I'd be less worried at the same actual temperature.

My early Silver Ghosts have no temp gauge - and I never give it a second thought!

[John Murch]

Rob, like Dan I have very turbulant water entering the header tank. (Normal)
On my car the overflow pipe comes right up the filler neck to about 1/2 inch from the top. If I fill the rad to the top, it will push out about 3 ltrs of water before finding its level and staying there.
Normal running temp on open roads is about 70° with ambient temp of circa 18°, it then rises / falls in direct relation to the outside temperature.
Did you try 'water wetter'? If you are getting foaming of the coolant it will push out far more water through the overflow.
It seems to me that provided you are not boiling, there is no reason to be loosing water unless the overflow pipe is too low, could a piece have broken off?
Has a second independent 'expert' checked the mixture and timing?
All most frustrating!
John

Edited
Presumably the overheating is caused by the drop in water level in the rad, so if there is no external leak, and the coolant is not foaming, the water is going elsewhere. Not into the sump (that volume would be very obvious!) could it be that you have a crack between valves allowing gasses to get into the coolant and thus blow the water out of the overflow?

[Chris Card]

Rob - can we take it that the water manifold inside the inlet side water jacket is in place - it disperses the inlet water evenly among the cylinders.

Also, do you have a thermoststat? If so, is it openning fully? What can happen is that the "open" flow is restricted, which then directs some hot water down to the radiator outflow and therfore back into the block, bypassing the radiator.

Even with the bellows functioning, I was having this problem, so I removed the bellows and blocked off the pipe to the bottom of the radator and all is now well. Visually it looks correct, but I don't have a functioning thermostat.

Chris

[Robert Clifford-Wing]

I am happy the mixture and timing are accurate, as checked.
We drove 100 miles yesterday, (50 there and back). The cooling system appears not to be able to cope with sustained 50/60 mph. The coolant temperature rose to 80C+, causing us to stop in a lay-by. On removing the radiator cap (I know silly thing to do, but I thought a non pressurised system would be different), I was showered with what appeared to be boiling coolant. Leave 10 minutes to settle then administer small amounts of water until radiator topped up to within 2 to 3" of the neck. More sedate driving delivered longer periods before stopping to check coolant level when temperature rose above 75C. No further boiling experienced, leading me to think that the water pump was unable to move the coolant through the radiator sufficiently quickly to cope with raised heat levels of faster driving.
Use of the electric fan as fitted kept the temperature manageable.

There is a visible movement of water in the header tank but far from violent as expressed by others owners.

[Robert Clifford-Wing]

[Chris Card]

" I was showered with what appeared to be boiling coolant."

Rob - it sounds as if your radiator is becoming pressurised somehow, and John's concern over valve seats could be a possibility. I think you ought to discuss this with a WO Bentley specialist, such as William Medcalf, John Ambler or VBE. They should be able to recognise the symptoms you describe.

If, with the engine up to temperature and at fast tickover, say 700rpm, there should be a steady flow of coolant across the header tank, with no foaming or excessive bubbling, as these could indicate exhaust gas from a cracked valve seat getting into the coolant.

Where do you live, and do keep us informed.

Chris

[John Murch]

A worn or damaged water pump will not help matters. Forgive me for repeating myself but foaming of the coolant will also give these symptoms.

[Dan Suskin]

Why did you pull over when it reached 80C?

[Chris Card]

Rob - following on from Dan's comment, where is the temperature gauge bulb fitted. Is it in the header tank or the water rail and, if in the rail, which end?

Chris

[Robert Clifford-Wing]

[Robert Clifford-Wing]

[Robert Clifford-Wing]

[Dan Suskin]

Rob,

I'd be very surprised to find many 3L (especially early ones) running at 65C in anything but cold weather. Many (maybe most?) are running 75-80 with occasional drift to 90.

Unless it's boiling, there is no harm to keeping going.

Incidentally, are you using a coolant, or straight water?

[Chris Card]

Rob - as Dan says, it's not unusual for a 3L to run hot in modern traffic conditions. I am, however, surprised at the amount of water loss that you are experiencing. Does it actually come out of the overflow pipe, and if so, how close to the radiator cap is the top of the pipe? Are you getting a syphon effect?

I don't know which radiator you have on your car, but Bentley did increase the size of the header tank on the later cars. This was, undoubtedly, due to cooling probems. The position of the temeratur gauge bulb is fine.

My pall Roy's Speed Model didn't actually have a temperature gauge and was always close to boiling, and would vent through the overflow. You may be worrying unnecesarily.

Chris

[Robert Clifford-Wing]

[Dan Suskin]

Just drive the car and let the radiator find it's own level. Don't worry about it unless it boils while running.

[John Murch]

Rob, re reading one of your earlier posts,** you mention the water rail as being flattened. Is this creating a restriction? You don't have a thermostat housing, so that cannot be causing a restriction. The block has recently been off, so one can assume that the water passages are clear and the core has been replaced, so that does seem to leave the flattened pipe or the water pump.
The photo above is I think fairly standard, showing car with the low radiator and small header tank, but as there was nothing in the thermostat housing, there is no return pipe to the pump.
In UK hot summer weather my car would run at 75° when cruising at circa 60, but would rise to over 80 if I pushed up to 70+ on a motorway.
John

** edited. You mention three flattened pipes (not the water rail) can't quite think what you are refering to. Can you post a photo?

[Chris Card]

John - The "flattened pipes" are the water distribution manifold attached to the inside of the water jacket plate on the water inlet side of the block. You can't see it without taking off the water jacket plate.

Chris

[John Murch]

Thanks Chris -

[Robert Clifford-Wing]

Having completed the flushing process using Holts two stage product, I am pleased to report a significant improvement in temperatere. The maximum being 75C and an average of <70C while airflow was passing through the core. It would seem there was sufficient sedemient within the core to effect its efficiency. Once I am comfortable the system is stable I will drain and replace the coolant with the normal anticreeze inhibitor and water.

[Chris Card]

Good news.

Chris

[John Murch]

Rob, really pleased for you. One wonders how sediment builds up like that is such a relatively short time.
John

[Robert Clifford-Wing]

[Julian West]

Rob

As a matter of interest, what colour was the sediment - white, brown or black?

[Robert Clifford-Wing]