MGB V8 - Troubleshooting Engine Problems
I've never been this baffled by an engine problem before. The engine started off with some occasional missing, then progressed to the point where it wouldn't start at all. I've written about the troubleshooting and included relevant specs on this page.
The problem turned out to be a number of things, but mostly plug fouling caused by a weak spark. Changing the Mallory distributor for a standard electrical SD1 part has transformed the car, although I couldn't see any problems with the Mallory - just the spark was too weak.
The problem was a complete pain to diagnose - this page covers the first week, but I've since spent another week on the damned thing. Getting rid of the Mallory has cured all of the other running issues I had initially put down to mixture.
The engine is second hand. It came out of a Dakar Range Rover special after the owner decided that only a turbo charged 5.7 Chevy engine would do. The Rover V8 is in quite a nice spec - it's a 3.9 built by RPI. It has a 270 Piper cam and runs a Mallory distributor and Edelbrock Weber 500 4 port carburetor. I've put a new set of 4.2 heads on, and made some custom exhaust headers.
The engine worked really well. The car was completed in November so I've only really driven on damp roads. It's a bit of a handful on throttle.
Day 1 - The Symptoms
I test drove the car on a 12 mile trip to work. It was pretty good, although the engine would occasionally miss, and became increasingly tricky to drive at low rpm. I examined the car and found the distributor vacuum advance has failed. The plugs seemed a little lean, but I thought I'd leave that one for further evaluation later on.
Having fitted the new vacuum advance the car ran really well from cold. I took it out for test drive, and after 2 miles it started missing again. The missing got progressively worse, and the engine was definitely a few cylinders short by the time I got back to the workshop.
Had to be a weak spark - the spark kept missing while I was checking the timing. Tried a different coil and coil electrical supply. Fitted a pair of new points.
I had a look around. I found a vacuum blanking plug on one of the spare vac supplies on the carb had perished. Fitted a new one. Checked the plugs - black and sooty. Hmmmm.
Day 2 - The Breakdown
Took the car for another test drive. After 3 miles I went around a corner in 4th gear instead of second. The engine bogged down and showed no signs of recovery. I dove into a convenient lay by just as the engine conked out.
It had previously worked well from cold so I let the car cool down. I checked the obvious things while I was waiting. Nothing. After an hour or so I tried again. No luck - she wouldn't start. Looked around again...... Didn't start again... I walked home.
Had a cup of coffee at home, then loaded my very reliable and actually working Renault with spare parts, and drove back to the MG. I tried a different coil, different leads, checked for fuel in the float chamber (present). Tried a different supply for the coil. Battery went flat.
This would be much easier in my workshop. Drove home and called the recovery (included in the insurance). They insist on sending some bloke along to try and fix the car (what do they think I'd been doing for the last few hours - it just can't be fixed). The bloke was to turn up at some unspecified time within the next hour or so. Have another coffee, and drove back and sat in the car listening to Radio 4. It was the afternoon play - rather good.
The recovery bloke turned up. He was actually pretty switched on, although he did connect up a second battery and attempt to burn out my starter motor. The end result was "it's got fuel, it's got spark - what else does it want?". He called the tow truck.
Day 3 - Investigation
Back in the workshop. Plugs are really black. It's running rich! Hey, maybe it's like when you forget to put the choke in - car runs until it gets warm then starts bogging down. Keep on going and the plugs will get so covered in carbon that a spark just won't be possible.
Cleaned the plugs up - hey it started. Then ran increasingly badly and stopped. Must be way too much fuel in there if can only run for a few seconds before fouling the plugs. Notice some overflow from the float chambers - the floats weren't correctly set so adjusted those. Tried starting it off the fuel in the float chambers. Nothing.
Measure the fuel pump pressure using an old tyre gauge. 3psi. The manual says it should be 5psi at isle, but the tyre pressure gauge is probably not all that accurate at that sort of pressure.
Maybe the spark is being easily overcome - and the misfiring is the cause of the soot - after all if it only fires every other time then there could be too much petrol left over. Try another coil, and dismantle and re-set up the distributor.
Day 4 - Robert turns up to help
Not such a nightmare this time as Robert has come along to help. Actually he came along to help me fit insulation to the garage roof and some extra lighting, but I'm now fixated with this V8.
It's so handy to have 2 people.
We decide it's a weak spark. It's got to be the coil. Try yet another - no luck. Find another distributor cap and rotor arm. Try that - nothing.
We attach a strobe light (one of those ones that fits around the spark plug lead and works by induction). It doesn't strobe. The spark isn't sparking. We take the spark plug out. It sparks and the strobe works.
Robert talks a lot about dielectric strength and suggests the plug has more difficulty in sparking when it is at high pressure and covered in petrol. Maybe there is some breakdown in the ignition system that doesn't break down when it's easy to spark but does when it's difficult to spark.
We attach the strobe to the HT lead from the coil to the distributor. Looks like it's sparking 7 times out of 8. OK so there's a lead down somewhere but it should start on 7.
I try the engine without the fuel pump - just in case it was still flooding the float chambers. No - it works less well without the fuel pump.
Try running without primary metering rod springs - that would make it very lean. No difference to running. Try it without any primary metering rods (needles) at all by accident. No difference again. This problem seems carburetor insensitive.
We check the spark strength. The spark from the coil can jump 8mm gap. The spark from the plug lead can manage at least 4mm. That should be enough.
Robert makes an excellent point. OK so we've got a spark and fuel, but it's not even trying to start. Doesn't even want to start on easy start. What if it's not fuel. What if the stuff in the tank is some odd stuff that looks like fuel but is really diesel or something. We connect the pump to Robert's spare fuel can and try it. No difference.
Day 5 - Sunday - a day of rest
Day 4 isn't a nightmare at all. We pushed the MG out of the garage and fitted the insulation to the garage roof, and installed additional strip lights. Goodness the garage does look bright now.
There seemed to be a lot of coolant where the MG had been standing. I'd noticed some coolant under the manifold. It appeared to be coming from the pipes into and out of the manifold.
What if there was a really small crack in the (heated) inlet manifold. A crack between the water passages and the intake passages. When the car was cold - no problem - no water would pass through the crack. When the car gets hot the water is pressurized and starts squirting water through the crack - which puts the fire in the cylinders out.
Day 6 - Help
I've checked the inlet manifold for cracks. No luck there. I did find that the manifold to head bolts weren't as tight as they should have been - they just seemed nipped up rather than torqued. It's just possible that this could have allowed water or even air into the cylinders. I've torqued them to 30lbft.
Maybe I'm clutching at straws now. I checked all the compressions - all around 175psi so doesn't look like head gasket or similar.
Phoned Chris Crane from RPI. He feels it's an ignition problem. I've ordered new plugs, leads, coil and condenser. He suggested that there ought to be more engine earths on the car - specifically one from the starter, one from the alternator and one from the engine to the bulkhead. That'll keep me busy until the new parts arrive.
Day 7- Success
In making the measurements for the right side of this page I thought I'd measure spark plug resistance. 9k ohm. That's not right. Turned out I'd been running suppression spark plugs for the last couple of hundred miles.
Fitted new plugs. No luck. Tried the new coil. No luck. Fitted some extremely expensive Magnecor leads. Still not there. Tried a new condenser. No luck. Tried starting some more. The car sprang reluctantly to life and ran smoothly.
The problems had indeed been weak ignition - The plugs certainly helped a lot, but I think the biggest thing was fitting the Magnecor leads. I'm really impressed with the Magnecor leads from RPI. They were cut to exactly the right length (other leads I've used in the past have been way too long), and the connectors to the distributor cap have been well thought out to avoid the top hose and bonnet. Also they made the car work. The downside is the ridiculous expense.
Took the car out for a run and returned without breaking down. Great. Tried a longer run. She's fine on heavy throttle, but still misses on part throttle. Took the plugs out - white as a sheet. Looks like she's running lean.
Anyway, I set up the timing as she's been a little jerky on starting as if she was too advanced - found the mechanical advance seems to be advancing even at idle. I don't like that - I'd noticed some evidence of previous fiddling with the advance mechanism when I stripped the distributor down. Tried retarding the ignition a little to compensate. The missing at part throttle is better.
I'm going to order some new needles ...err metering rods, and think a little about the distributor.
Day 8 - Still Working
Starting still isn't a strong point - maybe it's my starting technique. Previously I would give the throttle pump 3 squirts then crank the engine and she'd start first time. I'm probably too scared of flooding after the engine trouble.
I went out for another test drive. She does run lean when hot, and I've checked for vacuum leaks so it must be the jetting. She's fine on half choke. I've ordered an assortment of metering rods (needles) and jets from RPI. Should be here tomorrow. Unfortunately Mallory distributor rebuild kits are no longer available - RPI are going to look around for advance springs for me. I know I should get myself completely happy with timing before altering the carb.
Day 9 - Tuning
The jet and needle kit arrived from RPI. I fitted slightly richer jets and took the car out for a run. She performed very well so I kept going. I managed about 40 miles through towns, A-roads and motorways without a missfire.
I'm still not completely happy with running just off idle - the car is still a little jumpy at supermarket car park speeds, but everything else is excellent. Best of all I didn't spin the wheels once during my test run. With the part throttle performance so much better it's easier to feather in the throttle and prevent sideways moments.
The starting technique from cold is three squirts of the trottle pumps then turn the key. She starts instantly. When hot she'll only start on full throttle. That suggests fuel vaporisation issues to me, and the lack of insulation spacer between the carb and the manifold may well be the cause.
Day 10 More Tuning
I've improved the idle by unscrewing the mixture screws by about half a turn. Coincidentally this puts them back to the original settings. There is still occasional kangarooing off idle, but it's reasonably driveable now. Idle itself is a bit lumpy. I'll need to look at that. Later...
Days 11 to around 400
wOOt - it worked, I went to Scotland and then to Santa Pod, played a little, it was fun. Then the car went to sit in a barn
Day 401 - When will it ever end?
The V8 doesn't start. Well it does start then does all the same things as before. Same thing - weak spark. Either the carb is throwing in way too much fuel (which it does) or the distributor is poor. I'm going to change the distributor from the Mallory twin spark (which I've heard nothing good about) to a standard SD1 dizzy with lunenition rather than the OPUS (opeless?) electric ignition. More on this before day 800.
Day 500 - Finally fixed the problem
So what was the problem? A weak spark.
The cause? Something in the Mallory distributor as replacing it (without altering or replacing anything else) cured the problem. The new distributor (standard SD1) cost £20 from eBay and took a couple of hours to fit. It came with it's own inbuilt electrical ignition.
After further investigation later on I think this was down to the wonderful way that Mallory distributors are earthed to the engine (sliding plates). I measured quite a lot of resistance between the points base and the casing which was sorted by cleaning the sliding plates. I didn't try the distributor on the car again.
All of the faults described on this page reared up because of the weak spark. I returned the carb jetting to standard after fixing the distributor - that just helped work around the problem.
Since this problem I have been avoiding aftermarket "performance" parts. I can't recall a single performance part on this car that hasn't caused trouble (apart from the ones I designed - they are still good).
Much later I had a very similar weak spark problem, this time on the MGA. That made starting difficult and caused the car to pop and splutter to a halt when hot. That was caused by the rotor arm arcing to earth when hot. The test is to get the car hot, then pop in a cold rotor arm to see if it fixes the problem. If the problem reoccurrs when the new rotor arm gets hot the replacement is a bad one too - much fun to diagnose. See Helga's hot starting problems.
Set up information:
I've written all the settings down here rather than on a bit of paper which I'd lose.
Mallory dual point distributor:
Initial Points gap: 0.5mm
Dwell Angle (primary points only) 26 to 28 degrees.
Dwell angle (combined primary and secondary) 33 degrees.
Edelbrock Weber 500 carburetor
Model: 1404 (means it's 500cfm)
Primary jet: 0.086 inch
Primary metering rod (needle): 0.065inch* 0.052inch ( top and bottom of taper).
I changed the primary metering rods to 0.062inch * 0.052inch to cure a lean part throttle.
Primary metering springs: Orange (5" Hg)
Secondary jet: 0.095inch
Float min height = 7/16inch (inverted - end of horizontal part of float to gasket)
Float max height = 1 1/4 in (right way up - end of horizontal part of float to gasket)
Mine is a Facet Silver Top fuel pump. Should deliver 27 gal/hr at up to 5-6 PSI.
Edelbrock/Weber recommends a maximum pressure of 6psi at idle.
Measured it with cheap broken digital tyre pressure gauge - made it to 3psi before breaking.
I have measured a couple of 6v (ballasted) coils
Resistance between positive and negative terminals 1.5 to 1.7 ohms.
Resistance between either terminal and the HT lead: 7.6k to 8.5k ohms
The lower figures were for a new Lucas DLB 120 coil that I'm now running on the car.
I'd fitted Champion RC11PYPB4. They are early 2000s Land Rover V8 spec and are quite fancy with platinum bits. They have 9k ohms resistance between the plug lead connection and the anode. These plugs are not suitable for cars with suppression plug leads.
Chris Crane from RPI recommended NGK BP60S. They have 0.02 ohms resistance - that's more like it.
The plug gap for the NGK plugs is 0.8mm.
Resistance (note that the spark travels down a carbon sleeve around the wire conductor. Measuring the conductor doesn't tell you if the carbon has broken down).
In any case, a 700mm Mallory coil lead measured 14.5k ohms end to end.
The Magnecor coil lead (similar length) measured 4.5k ohms.
Low Tension Circuit
I run a 6V coil with ballast resistor. The idea is the resistor reduces the running voltage across the coil to 6V for normal operation, but it gets a 12V boost from the starter motor when cranking to help starting.
Cranking voltage: 12.5V
Voltage when not running or cranking: 6.2V.
The cylinder compressions for my engine ranged from 174psi to 194psi. Most were 178psi or there abouts.
All of my measuring equipment is low cost DIY stuff. None of it is calibrated, although I believe it is reasonably accurate.