Fitting a Roller Cam to a 1250cc TF

Some time ago I fitted a Roller Cam from Len Fanelli to my TD after it had eaten 3 camshafts and two sets of tappets, admittedly over a period of 40 years. After having discovered my TF had eaten a cam lobe after only 3000 miles, I decided to fit it with a roller cam. Following a lot of consideration, I went with the same performance cam as in the TD, even though the TF is more of a concours car, it keeps everything the same.

I decided to go the whole hog with a new set of roller rockers, 6 blade water pump and plastic fan while I was at it.

While communicating with John on his steel tappet gaskets he asked me to do some photos and write the installation up for TTT 2.

The kit ready to go.

I know some people out there manage to change springs using special tools and rope or air in the cylinders, but because these are 350lb open pressure springs, I went old school and removed the head.

This is the first time I have had to do any engine work since the car was restored so I am now finding out what an absolute pain it is to work on a TF compared to a TD.

Why MG didn’t leave the bonnet sides attached to the top as Morgan did, I don’t know. Off with front bumper, radiator and shell, giving access to the timing cover. Drop the sump. remove the distributor and generator, and off with its head.

Swapping out valve springs was not too difficult on the bench taking care not to damage seals, which had to be removed to get bottom steel cup on.

Removing the timing gears was not hard, but would have been easier if (1) chain had split link, or (2) I had two pullers. Constantly moving between the two gears, pulling a bit at a time on each one was very time consuming. there is not a lot of room to get leavers in.

Next, out with the old and in with the new camshaft. It was the usual pain to get the centre bearing in, fortunately I have a 4-post hoist which made life easier. I would not recommend doing it on axle stands but it could be done. Remember it has to go in one particular way, so it gets its oil feed.

Now for the cam followers, there are two ways to fit these. Link outwards or link inwards. To fit link inwards, you need to grind a bit of metal off the block to get them in, this I did with the TD.

If you fit them outwards there is a slim chance that, as the cork gasket gets saggy, they will cut a hole in it. I ordered a John James Metal Tappet Cover gasket, so it has the plenty of clearance and of course will not sag.

Now head is back on and followers in, so I can measure the length of the push rods and have them machined to size. When I did the TD a friend had a lathe, but he has sold that, so now I have to get a machine shop to do it, so more running around. Job keeps getting delayed, and finally took 3 weeks.

First mistake here – more later!

Warning – length of pushrods is quite critical so as not to get interference with rocker on full open. You will need 2 turns below the body of the rocker, in my case that was 4-5 turns from bottom of recess.

I already had a set of roller rockers but if you buy a set, bear in mind there are 4 different ones, just as there are 4 different original rockers. The only thing with the rollers is they are a bit harder to tell the difference, but you must get them right.

Refitted the side plate, timing gear cover, distributor, generator sump, put in fresh oil.

What to do with the exhaust manifold was another problem. For years I have had a 4 branch manifold, which I paid a lot of money for, but it will not fit with a Laystall Head due to it being 12.5mm wider. This pushes the manifold out and interferes with the steering shaft. It was no problem on the TD as that is left hand drive. I am trying to modify it, but in the mean time I have sent the original manifold and down pipe off for ceramic coating, which will be another month.

I managed to get the 4 branch modified so it would fit, but am having problems with pin hole leaks on my welding. I re-fitted the original, now it is back from coating.

Engine all back together, fresh fuel in the tank, set the CSI distributor and selected curve 11 and fired it up, ran OK, will get to tuning when I have connected up the radiator and everything else.

Heard a crack, didn’t think too much at the time so went ahead and refitted radiator and surround along with front bumper. Fired her up again and heard another crack. Pulled off the rocker cover and saw two rocker pillars had cracked between the bolts in line with the shaft. Between puling the valve gear apart and measuring and refitting several times, I managed to break another 6 pillars before I noticed the marks on the bottom of the rockers. It looked like the cup of the push rod had been making contact with the rocker when the valve was fully open.

The instructions that came from Harland Sharp (the manufacturer of the roller rockers) said to set the adjusting nut two threads down, I set it 2-3, but that was forgetting about the recess; it should have been 2 turns from the surface of the rocker, not from within the recess. This is to give clearance at full open for the pushrod cup. Until I get a new set of pillars I cannot prove this, but it looks the likely suspect. I have had the same set-up in the TD for about 8 years and never had a problem with pillar breakage.

I don’t have the tools to remove the cups from the push rods. They were quite a firm fit and required a hydraulic press to get them in. So, as I did not wish to damage them, I took the alternate route.

I drilled/ground out the recess until I could get full angle without interference at 3 turns. Probably a bit over the top but I only wanted to do it once

Now the wait for the new pillars.

John had wanted some information on driving with this set up, but after what happened I put him off, forgetting about the TD. So, after downloading an app I did some quick runs. Unfortunately, the only quiet 100km/h road near me is not level or straight, also I was starting on a gravel hard shoulder, so I could probably do better. But I think the times shown (see Note 1) are good for a road car in this day and age.

I have always liked the roller cam in my TD, but I will say, it feels better with the CSI distributor. Thanks to the supplier* and one of his clients who put a similar spec motor on the Dyno and found the optimum curve (see Note 2). No pinking or missing and even the gear stick rattle at 3000rpm seems to have gone.

Bernard W Wood (New Zealand)

Ed’s note:      *Roller/Lifter camshaft kits are supplied by Len Fanelli of Abingdon Performance: laf48′ ‘at’

I am aware that there has been much debate about these kits on T Series sites and there are some who say they would not entertain fitting them.

The last thing I want is to get caught up in any controversy regarding this. I am not advocating fitting the kit; my sole objective in publishing Bernard’s article is to provide a technical article which readers might find interesting. Bernard had previously mentioned to me when I sent him the steel tappet chest gasket kit, that he was just about to fit one of Len Fanelli’s roller/lifter camshaft kits to his TF. Always keen to gather in articles, I asked him if he would write one for me. Although he would have preferred to have waited until he had overcome the difficulty mentioned in the article, he agreed to my request because I was short of copy at the time.

The performance figures from his TD with the Fanelli roller/lifter camshaft kit are set out below, after Note 1 (the engine spec). Note 2 shows an advance curve using different CSI distributor settings. The optimum curve is setting 11.

These should be possible with Bernard’s TF, once he has sorted it (to be reported in a future issue).

Note 1 Bernard’s TD’s XPAG is 1250cc +80 thou Laystall Aluminium head with 32cc chambers – about 9.8:1 CR, Fanelli performance roller cam and roller rockers, 1½” Carbs, 4 branch manifold, CSI electronic distributor curve 11. Uses 95 octane fuel and has a 4.55 diff.

Performance figures (100km/h road not level or straight, and starting on a gravel hard shoulder

0-60 km/hr                               5.0 sec

0-100 km/hr                             15.0 sec

0-60 mph                                 14.0 sec

¼                                             14.0 sec

1/8 mile                                   7.0 sec

Note 2 Bernard says that curve 11 is very close to the original curve if you set the static timing at 10 deg. BTDC instead of 0 deg.