MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Now I'm baffled .... Two cylinders not trying

The manifold has been skimmed and reseated. No change. The valve clearances have been checked and re-set. No change. The plug leads have been checked for firing order, there is a good spark at each plug, the rocker arms move up and down gracefully, compression is good, ignition timing has been checked statically and dynamically,and yet when I remove the plug leads from 1 and 4, even at 2000rpm, there is no change in the engine note. Is camshaft wear the only possiblity and is there a way of checking it without dismantling? Seems like it should be one of those "staring you in the face" type solutions but I just can't see it!

Thanks,

Steve


Steve, Oxford, UK, scp@ukmax.com

I took my '74 GT to be set up on a rolling road today and confirmed that it is way down on power. Checks showed that cylinders and 3 are doing all the work, to the extent that the plug leads can be taken from the other two without changing the idling note.

There is a strong spark at each plug and the car car drives reasonably. Valve clearances were checked and each cylinder compression tested, with only No2 showing slightly low. The plugs looked evenly fouled, although checking the hydrocarbons suggested that 1 and 4 may not be getting fuel? The car has Weber carb on an Oselli inlet manifold.

I can't understand what could affect just 1 and 4 so dramatically, especially if the problem is fuel related as surely 1 and 2, or 3 and 4 would be the obvious candidates.

Any thoughts would be much appreciated.

Thanks,

Steve


Posted 14 March 2002 at 01:53:32 UK time
Chris Betson, Essex, England, chris@octarine-services.fsnet.co.uk; http://www.octarine-services.fsnet.co.uk

1 & 4 do tend to run weaker than 2 & 3 - a quirk of fuel distribution in the ports I always set up for the right mixture in 1 & 4, leaving 2 & 3 a bit rich.

It may be a cam wear problem, if the lobes on cyls 1 & 4 are worn - the clearances may be OK but the lift is less thhis would have the effect of the engine actually running OK but down on ultimate power.


Posted 14 March 2002 at 13:40:59 UK time
Steve, Oxford, UK, scp@ukmax.com

Thanks for the comments Chris. The engine is an Oselli built 1995cc which is meant to be good for 128bhp and has been strengthened and balanced. It has about 30k on the engine so I am not so sure about cam wear. Could this occur so soon?

I am wondering whether I have an inlet manifold leak as I've noticed taht the flange on the Oselli inlet is proud of the exhaust flange, meaning that the nut/washer is trying to tighten down at an angle. I've asked a garage to skim the inlet flange today so everything is even, although spraying Easy Start around the join last night didn't change the idling note. Would this sort of leak only show at higher rev's?

I'll update on the manifold fix tonight but meantime any thoughts would be well appreciated.

Steve
Steve Postins

Just a WAG, but could you have swapped the one and four plug wires? This would cause the symptoms you describe.
George B.

Couldn't you remove the rocker arm and use a dial meter to measure the cam lift by measuring the lift of each pushrod? You might even be able to do it without removing the rocker arm assemble. Just measure the rise and fall of each rocker arm at the adjuster.

This may not be terribly precise, but I suspect that, if the cam lobes are worn to the point that the cylinder isn't even firing, the difference between the good cylinders and the bad ones should be pretty significant. If all appear to be about the same, that's not your problem.

Could an extremely worn pushrod end or a bent pushrod cause this? Either one might shorten the pushrod's effective length, which would result in less lift to the valve, even though the cam lobe might be lifting enough.
Paul Noble

I like the bent pushrod theory. It certainly fits the symptoms if I am thinking them through correctly and I presume it will show in the measurement of full open to closed, the same as cam lobe wear. I'll get my ruler out tomorrow evening and will report back.

Thanks,

Steve
Steve Postins

A bent pushrod would only open the valve clearance - it would not affect the lift.

Measure the valve lift by rotating the engine to compress each valve in turn and then measure the height of the compressed valve spring.
Chris Betson

This has happened to me and I almost pulled my hair out. Quick story, took my BGT out for a ride, it was running great, decided to fill the fuel tank with premium fuel, car ran great all the way home. Wanted to take the car out to a meeting a few days later.fired it up and it had dropped a cylinder.Thought it was a fouled plug.could not find the problem. Eight hours later my buddy says what did you change. I say I put fuel in it. I drained the fuel from the carbs and disconnected the fuel pump. Made a gravity feed system with different fuel. BGT ran like a champ. The fuel was the problem.drained the tank and put in a different brand and the GT runs super.
Jim

take the Weber off & put the original carbs back on & I'll bet it smoothes out & operates as it should--on all 4 cylinders....problem is between your weber setup & your distributor & coil...coil & distributor don't like one another...you changed the distributor when you installed the weber, didn't you? &, you still have the original coil, right?
Tony Barnhill

Steve,
Just a quick thought, is the distributer cap OK and not cracked between 2 and 3 lead positions, I once had a crack on one lead position that caused that cylinder not to fire.
Graham
Graham

Thanks guys for all the tips and here's where I'm up to:

George, thanks for the lead idea. Just for thoroughness I've watched each rotor arm and valve alignment and they are all in synch.

Jim, fresh fuel has sadly not helped but I will keep to top brands in future.

Tony, I'm unable to test your theory as I bought the car fully modified. The car is exactly as described on the Oselli site at http://www.oselli.com/grad.html with the Oselli mapped distributor and coil, so I can only assume they have mtached the components carefully.

Graham, nice to know there is another enthusiast locally. When the car was dyno tested they checked for a good spark at each plug and annoyingly under the circumstances concluded that they had rarely seen such a steady and powerful spark on a B, probably thanks to the Lumenition and Lucas sports coil.

The dyno mechanic rang earlier to say he'd been dwelling on the problem and now wonders if the cam is timed incorrectly, or wrongly ground, or worn as Chris suggested. I've booked it in for Monday for him to look over but will measure the valve spring compressions myself in the morning.

Happy motoring!

Steve
Steve Postins

Steve. I wonder if the test equipment was functioning properly. I have no experience with the 1950 engine. Perhaps, it is much more powerful than the stock 1798 engine and will run on two cylinders. But, when my daughter's 77 cracked an exhaust valve, and the car ran on only three cylinders, it was immediately obvious that something was seriously wrong. You could hear it and feel it when driven. If you really have two cylinders that are not firing, I would suspect, based on the experience, that the car would not be drivable. Since you were able to drive the car to the rolling road, this suggests that something is being over looked. Les
Les Bengtson

Steve, listen to Graham.This is a very common occurance. You can still get a good spark from each individual wire even though the cap is cracked. Change the cap and check for sideways play in the distributor shaft while you are in there. Sideways play can cause your points to be closed at peak point opening because of the related pressure from the point spring pressure. Dann
Dann Wade

I have a few observations...

'Compression is good' Now that is like saying how long is a piece of string!

What were the figures and what was the difference between dry and wet tests? Was compression on 1 and 4 actually down or up over 2 and 3, or were they all over the place. Was there a variation of more than 20psi between the highest and lowest reading?

Les makes a very pertinent observation in that when driving, the loss of two cylinders on any B series, whatever the tune, would leave you in difficulty out accelerating an electric milk float! The engine would also vibrate seriously and shake the whole car in the sub 3000rpm area, above this the vibration frequency rises to a point where you notice it less, but notice it you do.

Compression testing is done at cranking speeds only and electronic cylinder drop out tests are done at fast idle speed. Both relatively low rpms where if there is leakage from a cylinder the maximum time is available for maximum leakage. As revs rise so the time available for leakage is reduced and so the engine smooths out and produces more power.

A worn camshaft profile reduces the degree of overlap and this tends to see increased cranking compressions. So if the pressures read during the compression tests were higher on 1 and 4 then this would point to worn cam lobes. However this also tends to boost low speed engine power and torque at the expense of power at higher revs which is contrary to the circumstances described.

No the key here is that there is sufficient power for the car to be driven and then the testing shows number 1 and 4 are lame ducks at idle and fast idle. This points towards cylinder leakage from these two cylinders that is at the maximum effect at lower rpm and less so at higher rpm, although it will still lose power. This can be through a number of possible causes or a combination of them. Such as badly seating valves or bent valves. It can also be caused by valves sticking in the guides and through having seriously mal adjusted valve clearances. (I note this has been checked though)

Gasket leaking is less likley on unconnected cylinders, but remains a possibility. Related to this is a further possibility re the seating of the head on the block and the correct torque loads. Is your mention of skimmed manifold actually relating to a skimmed head face? I have clearly missed some earlier posts. If so then consider that the original head will almost certainly have been previously skimmed and as such has anyone checked the thickness of the head casting, and taken steps to ensure that there is sufficient stud thread to allow correct tightening of the nuts whilst transferring the torque onto the head and not tightening against the end of the stud threaded?

In addition has the block face been checked for flatness and also for stud thread pull out. (This is where the thread of the studs lifts slightly from the block face and creates a raised area around each stud and which can prevent the head being correctly torqued down.)

The 1995cc bore of 84.45mm or thereabouts is really out to beyond what I would ever recommend. The MG Comps Dept in 1967 were forced to develop a 2004cc engine for the then new MGCGTS which couldn't run with a 6 cylinder motor as it was prior to the official announcement of the MGC. This was not a successful engine conversion and the fact that the factory only ever then went to 1892cc (plus 80 forged pistons) tells it's own story!!

Piston design in terms of height and crown shape also dictates alternative pistons are less successful and so lose out significantly in terms of squish. (mixing within the combustion chamber during the compression stroke) This is why so many plus 060" bore engines are easily able to hold their own against 1925cc (Lotus piston at plus 20) and 1950cc (Lotus piston at plus 40) conversions. I have only got expereince of a few bigger capacity engine beyond 1950 and only two for Oselli 1995 engines. This was not deep, just with rolling road testing, and both had serious reliability problems. I therefore can't comment on the details inside the engine and I didn't investigate the reliability problems that aflicted these two cars.

It would seem logical at this stage to do a cylinder leak down test and see what happens. I suspect that the pressure loss will be significant and immediate. This may give a very good indicator to the source of the leak and it will also probably benefit from the engine not running so the bore distortion which affects these big bore engines under load will not be influencing things.

If you have the head off and simply invert it, you can pour white spirit or parafin into the combustion chambers in turn and note how quickly the fluid drains and to where. Cracks in the head will also swallow up fluid and these are also far more common now with the hotter burning more volatile unleaded fuels that impose much greater thermal loads.

There is also a reasonable presumption that the head has been modified. This on it's own should not cause any problems but it may alter the flow balance in the siamesed inlets so that natural tendency for the inner cylinders to run richer is increased. This can lead to local overheating in the end cylinders that often melts plugs in really ardous conditions. It also explains why there is often different plug grades in many of the early Works competition engines between inner and outer cylinders. This is a point to raise the question of what loads the engine has had to suffer in the 30k miles.

It also begs a simple additional question which is why have a seriously modified engine unless you intend to use that performance. If it has been used then it is very, very reasonable to expect that 30k hard miles with the expected extra perfromance is easily the equivalent of three to four times normal road mileage of a standard engine.

Hope this helps you identify the problems.

Rog
Roger Parker

Graham and Dann -advice taken, I have a cap on order.

Roger, thanks for a fantastic analysis. Every time I start to think of the B-series engine as a simple device I find it's me who needs education.

Let me add some numbers in for you. The compressions were 170, 155, 162, 172 for 1 to 4 respectively. A further compression check was performed used compressed air with the engine static (a leak down test?) which verified "good" compression in 1, 3, and 4 and "moderate" in 2. I guess this rules out head gasket, piston ring, or valve wear except for 2. In terms of the power curve, the car has bags of torque and accelerates well up to about 3500rpm after which the graph goes flat, giving a maximum output of about 90bhp.

I'm thinking this fits rather too well with your description of cam wear. Does this mean cam followers will also be suspect? I'm trying to convince myself I don't need to take the head off.

For the record it was the inlet manifold that was skimmed as it was not seating well and the head has not been off so far. The head has been gas flowed and polished but is fundamentally standard.

I take your point about the choice of a higher powered engine. My logic was to have an MGB that could take in it's stride today's road speeds, something that has "head room". Come to think about it maybe a milk float is sufficient on the Oxford ring road. The car is used every day and so far has been more relaible than the Saab or Mercedes I drove previously. Needing the car to get me to work and doing most of the work on it myself gives me a mechanical sympathy that limits my use of the performance.

Steve
Steve Postins

If it is the cam, the head does not have to come off.

You can change the cam once the radiator is out the way (you also need to pull the dizzy drive )and the followers MUST be changed or you will wreck the new cam - they can be reached through the side covers but you will need to pull the inlet and exhaust manifolds.
Chris Betson

Well, I've now had the cam measured for lift and duration and it's spot on, each valve within a fraction of a mm of the next, I've changed the dizzy cap, and have even run the engine with the exhaust manifold off in case of weird pulse effects that may have been occurring (yes I am clutching straws).

My attention is now focused on the Weber as we all know that carbs are slightly arcane. I think I can read on the castings that 32mm chokes are fitted, and I'm going to switch these to 36mm at Oselli's suggestion. If this gets me four cylinders working I will be amazed but next step is moving to SU's.

Thanks everyone for your advice,

Steve
Steve Postins

What weber is that? A single 45? Seems you should be using the 36 choke. No wonder it's flattening out at such a low rpm. Especially with the 1995cc displacement.

Take a look at the Weber tuning info on this web page for some additional insight. http://members.aol.com/dvandrews/webers.htm
Gerald O'Docharty

Thanks Gerald,

Sadly I had a good look at the carb again last night and it was my eyesight at fault, as a closer look showed that 36mm chokes are already fitted. The carb is a 45DCOE so thanks for the link. I'll search the site for ideas as at the moment I'm all out!

Regards,

Steve
Steve Postins

still think its somewhere between distributor & coil....there are 2 types; 1 is externally ballasted....you may need to remove the ballast resistor or the resistive cable if its there...or, if you don't have it, you may need to add one
Tony Barnhill

Tony,

I don't understand the science behind it but I will certainly give it a try. I think I have a standard Lucas coil somewhere.

Many thanks,

Steve
Steve Postins

Tony,

what about the rubber O-rings between the manifold and the DCOE?

You mentioned in your first thread that you removed the plug leads from cylinders 1 and 4. Was that one at a time, or simultaneously?

I agree with you and would put a pair of known SUs, preferably from someone's car which runs well, on to see what happens. Berlin-Oxford is a bit far for me though.

Best wishes,

Mike
Mike

Which sparkplug are you using ?
.

In answer to your questions, I can remove the leads simultaneously with little change in the running of the engine. I'm now using Champion plugs but have replaced NGK's as part of the checks.

I've got a pair of SU's arriving tomorrow so will be able to give nore info at the weekend.

Thanks again,

Steve
Steve Postins

Mike...i believe the rubber o-rings that go in the end of the carb are there to make a seal to the intake manifold...don't know how they could affect things unless a vacuum leak was occurring...plus, i just don't see how a weber is any better than good old su's...either hif-4's or hs-6's for the big engine
Tony Barnhill

Do you have a colortune? It would be interesting to see if you're getting any fire at all in 1 & 4. If you're pulling hard up to 3500 and then it all drops off then perhaps you're suffering from fuel starvation.

I know you said the timing was checked dynamically. Was that at idle or through the entire rev range? You may have issues with vac-advance or a fouled mechanical advance spring.
Lin

Ok, here's another thing to ask; have you upgraded the fuel pump or is it the standard unit? The standard MGB fuel pump may not be up to the demands of your engine. Above 90HP or so you will need a higher volume pump. Since you thought it may be a fuel problem look there. Also it may be helpful to temporarily install a fuel pressure guage for diagnostic purposes.
Gerald O'Docharty

"Above 90HP or so you will need a higher volume pump"

I doubt it, the same AUF305 was used on both 4-cylinder and factory V8! And fuel starvation would affect all cylinders, particularly with a single Weber, and not cylinders 1 and 4 in any case.

"the car has bags of torque and accelerates well up to about 3500rpm after which the graph goes flat, giving a maximum output of about 90bhp"

You say that 1 and 4 are doing nothing even when idling, so I don't see how you can be getting the above.

Still sounds like the wrong firing order to me. Go back and prove that the rotor is pointing to each cylinder on its compression stroke and not its exhaust stroke.
Paul Hunt

I've sat back on this one long enough. Because the problem is supposedly on only two cylinders, #1 & #4, which have the highest compression of the four, it can't be the camshaft being ground badly out-of-phase, or the vacuum advance mechanism, or the coil, or the fuel pump, or an intake leak, or the carburetion. However, there is one telling symptom: It pulls hard up to 3,500 RPM and puts out 90 HP on a dynometer while running on only two cylinders!? Hey, why hasn't anybody asked themselves "What's wrong with this picture? Is Steve Postins playing a practical joke with this thread?"
Steve S.

Steve,

I was just asking this same question a few hours ago...glad to hear I'm not alone. Let's see, 90HP as measured on a dyno, but flattening out at 3500 rpm...pretty darn impressive, huh. Are we talking at the wheels? Wow, most modified B's don't make that on all 4, much less perform to such a level on 2 cylinders, just when most of our engines are only starting to make power (well, I guess the other two just aren't "trying", right?).

On the other hand, perhaps Steve is having a real problem with #1 and #4 and may be somewhat overstating some of his numbers (hey, I'm pretty proud of my B at times). The problem here, is it has occurred to me that we've pretty much exhausted all possibilities (every time something was suggested, the original poster ruled it out...at the moment, we're left with attempting to return to SUs, which in and of itself might help, but who knows? It's certain to change things anyhow.). I'm sort of waiting for the SU swap to fail, so we can all scratch our heads and come up with something new and interesting (e.g., the inability of the standard SU fuel pump to "keep up" [one of the suggestions already offered]). OR, perhaps the dyno operator is pulling a fast one on Steve and he's the victim here. (Steve P. you may REALLY wish to get a second opinion, should this prove to be the case.) He did afterall note that he purchased the car in its modified state. However, I'm starting to feel the same way as Steve S. on this one.

FWIW (and that's not always much),

Barry
Barry Kindig

Steve,
would you be able to post the cylinder rpm drops along with the hydocarbon increase for each cylinder. Sounds like your dyno man has the proper equipement. Also an idle balance/power balance test would confirm that 2 cylinders negatively effect the engine. When ever I get an engine with suspect mechanical problems the first thing I do is check the vacuum waveform. This will tell you DYNAMICALLY what the valves/(and pressures) are doing while the engine is running. If all four intake ports share the same manifold you should see 4 evenly spaced sine waves(or very close to it, depending on intake design). Please post this graph if you can. As Steve S. says, it would be very hard for your engine to produce enough power to run somewhat OK on 2 cylinders. It is possible to have bad intake valves so that removing the spark lead to the bad cylinder will have NO or a POSITIVE change in the engine rpm. This is caused by the vacuum dynamics of a bad valve diluting the vacuum or the mixture of another cylinder. As others have mentioned, don't rule out ignition parts. Can your dyno man provide pictures or results of the 4 spark firings at idle and under load as well a spark durations. Particularly interested in number of coil oscillations. If he has the equipement he should be able to produce the data. Often times I see customers who have spent alot of time and money on something that is fundamentally flawed as in the case of points/carb system. I hope the SU swap cures your problem as you state that 1 and 4 may not be getting fuel according to hydorcarbons. As a note, the best rpm to remove the leads is at the lowest rpm you can so that when the leads are removed the engine does not stall, usually about 900-1500 rpm for 4 cylinder. After individual leads are removed, 2 leads should be removed at a time (1 &2, 3&4) to check carbs balance.Note that if you have electric fans, they should not come on during cylinder shorting tests. Good luck.
Joaquin

Thanks for all the comments everyone.

The car is going back to the original builders after Easter for their opinion, although I will try putting a new fuel pump in if I can find the time this week JIC. Barry, I'm guessing the rolling road results were calibrated back to flywheel output as the tester's comment was "that's less than a standard B". He diagnosed the lack of effort on the two cylinders and then spent two entire days trying to isolate the cause without success so really, I'm not hallucinating, just seeing strange things.

Joaquin, I'll see if the tester can remember the actual figures for hydrocarbons and get a copy of the plot.

If anything works, I'll let you know.

Regards,

Steve
Steve Postins

'lo all...

I have to say I'm intrigued by this one. Has this engine ever run right? What colour are the plugs in the bad cylinders after the engine has just been running?

My B has a nice lumpy idle and I dare say I could take one plug lead off without changing the engine note much. I wouldn't want to be holding the lead with a bare hand though :o)

ttfn,
--
Oliver Stephenson
Oliver Stephenson

This thread was discussed between 14/03/2002 and 23/03/2002

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.