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 Midget and Sprite Technical - Camshaft choice

I'm planning to build a new engine in the next few months so I'm just beginning to think about some of the choices I might make.

My current engine has a Kent 276 which works very well though in a perfect world the torque would come in 500rpm or so lower.

I had been considering building the new engine with a SW5 camshaft but a bit of googling suggests it can lead to more tappet noise, does anyone have any experience of the SW5 and can comment on all aspects of it but particularly tappet noise please.

Any other comments on camshaft choice more than welcome.

Thanks.

Jeremy Cogman

I haven't used the SW5 but I do have the Kent 276 in my 998. I'm interested in your comment about tappet noise, because that's what I had with this camshaft. I followed the received wisdom to equalise the rocker movement, which means shimming the pedestals to get the rocker angle at valve closure the same as at valve open. This maximises the lift. However I got quite a lot of tappet noise. I now have a different engine which is exactly the same except for a slightly higher deck height, and the rocker angle is more for the open valve. This is not quite ideal for breathing, but the noise is much less.

The reason for this rigmarole is that I wonder if the reputation of the SW5 for noise is because of this valve geometry issue. I should mention that I did reface the rockers by setting up a jig on the lathe, and that did reduce the noise. I also re bushed and reamed the rockers. So there is quite a lot of factors involved in this noise problem.

The Kent 276 is about as far as I could go for a roadgoing 998, as there is not much torque below 2000, but it comes on like a banshee about 4000. I'm guessing though that your engine is a big bore one and you will have a much wider choice of camshaft for that.
Les Rose

I looked at an SW5 but after chatting to Swiftune settled on a 246 from Kent. Its got a bit more bite than the SW5 in timing terms but has slightly more lift. I was looking for more torque lower down to pull a higher ratio axle. It will bimble around 1500 rpm quite happily in top but will pull strongly when needed.
SW5: intake 16/48 Exhaust 52/12 244 duration with 0.281 lift
246: intake 18/54 Exhaust 58/18 252 duration with 0.285 lift
engine is 1310cc with Slark fast road head HS4 carbs with maniflow inlet and exhaust 10.5 CR
Bob Beaumont

Hi Les
I would be very interested in your method and details for refacing the rocker arms as I have a few sets that require attention.
S G KEIL

Not used the SW5 but I’m happy with the Kent 256 in my 1293 Mk1 midget. I built a similar spec engine a few years ago with a 266 in it but I wasn’t overly impressed. It was quite lumpy on tickover and didn’t seem to give much performance.
The 256 is nice and smooth on tickover and gives decent power and torque from low revs.
Engine is 1293, 10:1 cr, PB Econotune head and standard carbs and filters.
John Payne

Mine has the MG Metro cam (and pistons) as decades ago a mate was on the Rover sponsorship scheme and could get them at works price. I've never rolling roaded it but it seems to pull well on the road and survived a few laps of Castle Combe once.

Google tells me it is similar to, but not the same as the Kent 266.
AdrianR

The MG Metro cam is the old Special Tuning '500' profile.

We used them in the MGOC Race Championship, back in the '90s, and they used to cost £27 from Roversport Parts. They were also used in the Mini 7 Championship.

Worked very well and I would certainly use one again if I was building a fast-road engine.
Dave O'Neill 2

A couple of acquaintances were unhappy with the road behaviour of their 1275 engines so with my assistance we fitted cometic head gaskets to optimise the Cr at 9.5:1 and fitted A.C.Dodd RT camshafts. I also built a 1/2 engine for someone to be fitted with a stock head again fitting an RT camshaft from Adrian Dodd. All have been pleased with the performance / torque. They are the fittings that I can currently recall there are probably more.
The camshafts are ground in Kent by Newman Cams.
A.C.Dodd can be found on Facebook and Youtube. He is always free with his advice and he has other spec’ cams to suit various needs / wants.
Alan Anstead

I use a SW5 cam in my 1293 with 1.3:1 ratio zero tolerance rockers supplied by AC Dodd and have no noticable tappet noise. Tappets set as instructions from Swiftune (14 thou if memory serves). You can see the work that goes into AC's rockers on his Youtube channel.
I am happy with the performance of the SW5 in my car.
R Farrar

Swiftune. I was very fortunate that Nick Swift allowed me to take a group of Kent Masc members on a visit of the Swiftune works. He personally gave us an extremely informative tour followed by a Q & A session.
I have also, in the past, arranged a tour at Newman Cams whom I mentioned previously with regard to the A.C.Dodd camshafts.
Alan Anstead

My engine was rebuilt by Southam Engineering in Warwickshire and it is an absolute jewel. It pulls strongly and is smooth and quiet. They did a lot of work including fitting a Swiftune SW5/07 billet cam shaft and a vernier duplex timing kit.The latter was set to 107 dgrees full lift on number 1 inlet. I also have a Maniflow fabricated inlet and exhaust manifold. The tappets are set at 15 thou and there is noticeable extra noise. I also have Swiftune valves, bronze guides and Hi-lift double springs and all work perfectly.

The only issue I have had with the engine was a VERY leaky rear lip seal conversion. This was fitted on the inststence of the previous owner. Southam still had the original scroll seal from my engine several years after they had done the work and that is now refitted to my car with minimal oil leak.

Jan T

J Targosz

Refacing rockers

I used the unworn section of an old rocker shaft, mounted exactly vertically on the tool carriage of my ancient lathe. I dropped a rocker on the shaft with the valve pad facing the chuck. The rocker was lightly sprung to keep it towards the chuck, which had mounted in it a small grinding disc. With the lathe running, I moved the tool carriage towards the chuck, emulating as closely as I could the normal operating arc of the rocker. The idea was that whatever the angle of the rocker the pad would be tangent to the grinding disc.

I'm sure that the professionals among us will have problems with this method, but the result is that the rocker adjustment is now more accurate and the valvetrain is quieter. I'm aware of a number of dangers. I am using forged rockers and I'm not sure if they are case hardened. If so I may have gone through the hard surface. Time will tell. The arc on the surface of the valve end pad was achieved with a large amount of guesswork. But at least the pad was ground accurately side to side.

The main reason for doing this job is that old rockers that have been used on standard camshafts for decades have pads that are worn deeply, creating a ridge around them. If you use these with a higher lift camshaft the ridge catches on the valve stem and increases side load. Also it will be impossible to set the clearance accurately. Many many years ago I realised this and manually removed the ridge on a grindstone. This worked when the wear was not too bad, and if I was lucky. It didn't work when I tried it more recently and could not get the pad ground accurately by hand. Hence cobbling up a jig on the lathe.
Les Rose

I've got a Kent Cams 286 in my 1275 plus 60 with MG Metro head, HIF44 Maniflow LCB and silencer. When I bought the engine it had been fully balanced with centre main strap but I have no idea about what head mods have been done nor what CR it is. Many have said a 286 would not be very good but I really enjoy it. The idle is a bit lumpy (about 1000rpm) and it doesn't really come good till 2500-3000 rpm but when it does it seems to have no end of power. I used to have a 3.7 diff but found 30mph traffic weary with lots of 3rd/4th changes according to urban traffic conditions.I've since fitted a 3.9 and in that respect it is much better.I've got a 4.2 somewhere I might try sometime this summer as I have a 5 speed gearbox.
Rob
MG Moneypit

There is no magic in cam profile design, at the end of the day you cant fool old Isaac Newton. There will be other compromises the factory couldn't make if you specify long period aftermarket cams. You will suffer from backflow during the large overlap at low speed thus adversely effecting torque, idle stability and emissions. The momentum of the incoming charge can overcome the backflow at high engine speed. Personally, I wouldn't go above 264 deg for a 2 valve engine used on the road. You get lift for free when you extend the period as well as increasing valve float speed and lowering contact stress as the nose radius increases (providing you don't increase spring forces too much) It depends very much on what character you want your engine to exhibit. There comes a point where the curtain area around the valve is bigger than the throat area, so extra lift does nothing to increase flow, but the manner in which you get to this point does. Then there is another can of worms because there is no industry standard way of measuring period. The cam profile designs for A and B series are the patented 3 harmonic Multi Sign Wave method devised by JLH (Jack) Bishop. The problem was if handed an impossible task by the development engineers in terms of lift on a given period, he used to fib about the period meaning its difficult without the profile drawings to compare them. They were often longer period than people thought.
On the rocker geometry, its designed to give pure rolling at 1/3 lift. Either side of that there is sliding imparting a friction force to valve head.
Paul Hollingworth

Great thread, thanks. I try to save these recommendations and experiences for ideas for potential future engine builds for good torque and easy to drive from low revs (tractable).

There was a similar thread a year or two ago that started off with a query about engine spec and cam choice for high altitude from a member in the US. I’d be keen to locate and re-read that.

It is a pity that a Vizard like tome is not written to capture what has happened in A Series tuning since the publication of the blue and yellow books and earlier references like Mike Garton’s ‘Tuning BMC Sports Cars’ and the BMC/BLMC/BL Special Tuning leaflets that cover building good road engines, or ones for roadgoing hillclimb/sprints rather than all out maximum race or rally engines. Info on modification specification and tuning for small bore A Series engines (1098/998/948/850) is even thinner on the ground - thank goodness for the wise heads on this BBS and the Morris Minor Owners Club Forum.

Cheers
Mike

M Wood

Yes, great thread with lots of different perspectives. Because of the trade-offs between idle, low speed torque and top end power, camshaft choice is very subjective: what A loves, B might not - because either it is too lumpy or it lacks top end. The range of preferences above (from Kent 246 to 286) illustrate that well ... there is no absolute right or wrong, it is all about what each of us prefers.

For that reason, and the amount of work involved in changing a camshaft should the result be disappointing, I always suggest trying to score a test drive in a car fitted with the camshaft being considered.

Paul H touches on a very important aspect of comparimg camshafts: the point at which duration is measured. It is almost never the actual duration from the valve leaving the seat to when it returns. Apart from the aspect Paul mentions, there is the very practical difficulty of identifying the actual commencement or conclusion of lift correctly when trying to measure it because at clearance take-up the valve train velocity is very low compared to crank movement in order avoid impact loadings that accelerate wear (and make for lots of tappet noise). As a result, measurement resolution is low.

To address this, the valve timing/duration is specified at a timing clearance, or a specified lift off the base circle, which makes measurment more accurate. But things are never that simple, as different manufacturers use different figures for lift off base circle - for a variety of reasons no doubt, one of which will be to make comparisons harder (that's my conspiracy theory for today!).

BMC specified a timing clearance (from memory) of 22thou B series and 21 thou A series, which calculates to 16-17 thou lift off the base circle. I don't know what Swift, Kent, Piper use, but it makes on-paper comparison between different manufacturers' cams tentative at best. (Within a given manufacturer's range, the spec point will be likely the same, making it much easier).

It is for this confusion that the 50 thou lift figure is a popular reference point for timing. It also has the advantage of being closer to when max lift velocity occurs, so giving even better measurement resolution. (Which is also why degreeing the cam using lift at TDC is more accurate).

One final observation: while Paul is correct about remembering to compare valve curtain area with throat area, dont forget to consider what part of the valve curtain is masked by chamber walls, and also bear in mind that the valve is at peak lift only momentarily so comparing at 90% lift can be useful.
Paul Walbran

Quoting the period a some arbitrary lift doesn't really work because by then the follower is accelerating like billy-oh. Some consultants use 1mm which is very confusing. If the follower is accelerating then that is part of the flank of the cam and should be counted in the period. On the factory drawings the open and closing ramp lift and eccentricity are tabulated separately from the cam proper (flank and nose). Its obvious that Jack cheated because the eccentricity which is proportional to velocity isn't constant at the ends of the ramps. In about 1998 I was tasked with re-designing one of Jacks A series cams to make more tolerant of the valve clearance not being spot on. People were going back to the dealers complaining of tappet noise before they had got to 1000 miles. This was on the MPI Mini. So I had to go through all of this to come up with an equivalent which would still produce similar performance (72 Hp). I cant recall what lift and period I cam up with now and I don't have access to the records of that time. They might have some at BMM Gaydon or possibly Kimber House but we can be sure everything left at Longbridge is in the skip. All drawings were on microfiche, I wonder what happened to that.
Paul Hollingworth

Peter Burgess gave me a Piper 255: Very hi torque, nice idle, much power till 6000+ rpm, no need for other rockers, 40 mpg . And yes indeed, it worked out as he promised. (1300 fast road head, two HS 2 carburetors, maniflow medium bore LCB).
Flip Brühl

Paul, that is very interesting about the MPI Mini. Prior to designing my own cam profile (laborious with lots of maths, not having access to the design tools you had) I did a detailed analysis of several BMC profiles and noted the constant velocity section you refered to above, and understood the function of it.

What interested me was that B series tappet clearance was specified pretty much at the end of that section, after which the valve train accelerates hard. On the other hand the A series clearance is almost a couple of thou less after adjusting for rocker ratio. I concluded this was a likely origin of B seties seeming so tappety, as a small error setting or small amount of wear would result in much higher velocity at the clearance take-up. So ever since I have used the A series clearance adjusted for rocker ratio. This approach gave the desired results with cars set this way over the last 40+ years - reduced valve train noise and wear with no downsides in terms if idle quality or valve life.

I did another cam profile design recently when an E-type customer asked for the engine to be tickled during rebuild, and produced a magazine article extolling the gains of higher valve lift in those engines (which figures given the valve size). As he wanted to retain nice idle, none of the off the shelf options met both requirements, so it was back to the design exercise. In the intervening 40+ years, new computer lools made that so much easier, especially as my computer-savvy son produced a nice bit of software that would draw the lobe shape so it could be zoomed in on enormously to detect any issues.

Ironically, the biggest hassle with the whole exercise was that new camfollowers from all sources all came with a champher at the outer edge, so reducing the diameter of the working face. As I had pitched peak velocity to optimum for the std follower (so gaining a very useful 10% increase) this presented a major problem. Eventually resolved, job finished and the result what was aimed for.

As for specifying duration at a random lift (50 thou), it isnt something I use myself but its main purpose is as a benchmark for comparing different profiles between different manufacturers, which can be difficult if using timing from the valve seat due to variations in specifying where on the ramps the clearance should take up.
Paul Walbran

This thread was discussed between 27/03/2023 and 19/04/2023

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.