Running the Rocket Three with an SU carb.

Look at a group of BSA/Triumph triples and you will not find many which are totally standard. It never ceases to amaze me how many triple owners change their bikes in different ways. I suppose it reflects motorcyclists in general in that we are all individuals who have our own preferences. To this end the triple lends itself perfectly. We all love our bikes but if not all, then most of us, realise their limitations and shortcomings and because of this most of us modify and so improve our bikes in our own particular way. This is how I did it with one of my Threes.

This article is to show what I did when I fitted a single SU carb to my Rocket Three. I will not be showing you in detail how the SU works or how to tune one except where applicable to the changes I made. An excellent book on this subject is available published by Speedsport MotoBooks ISBN 85113-072-0 and info on the web at

http://www.jetlink.net/~okayfine/sutech.html

A few years ago I finally put my Rocket Three on the road.

After building the engine with almost totally new parts, including Boyer ignition it ran well and did what I expected it to, until I had done about 5000 miles, when it would not tick over properly or pick up cleanly on acceleration and gave at best 32 MPG. The tickover problem was cured by fitting adaptors to the carb intake stubs, and balancing with a manometer. This however, only lasted for a few hundred miles and then needed setting up once more. This was the old problem of Amals wearing out again. As the bike was used regularly and a trip to Germany was coming up, I looked for alternatives. A friend had had an SU fitted to his triple, I recalled, a few years back so I asked him what had become of the setup. It turned out that the bike had been sold, but the new owner had put the carb and manifold to one side. I contacted Tom Kerr the new owner and borrowed the setup to see what it was all about. The manifold was made from ¾ inch pipe fittings interconnected with rubber hose, with brazed on flanges for the carb and cylinder head. A 1¼ inch SU was used taken straight from a Morris Minor with no changes except for cable pull for the throttle rather than a rod linkage.

My curiosity got the better of me and after cleaning and checking the carb I fitted the assembly to my Rocket Three. To my surprise, it started first kick. Normally the bike was a difficult starter. Things looked good. I took it out for a ride and it ran well but was weak at the top end. At this time I knew nothing of this type of carb and decided to get some knowledge. Amongst other things, I found that the fuel ratio was metered by a needle, which was tapered toward the tip, that went into a fixed jet, through which the fuel was sucked by a depression of air over it. That is about as technical as we get so don’t worry. When the needle is raised out of the jet there is more room round the thinner parts of the needle so more fuel can rise up. The needle is raised by a piston which also allows in more air as the engine needs it, so giving a large amount of air for a large amount of fuel at full piston lift and a small amount of air to a small amount of fuel at low lift. Simple hey? Well not quite but here is where it gets interesting for our purpose. There are about 350 needles available, with differing tapers giving varying amounts of fuel at each part of the needle. This gives different ratios of fuel at every part of the needle, which can be matched to the needs of each individual type of engine. The jet can also be raised and lowered using a nut under the carb, which richens or weakens the mixture at all points of carburation.

As the bike ran weak at the top end, I set the carb jet with the adjusting nut so that the mixture was right and it ran well at about 80 MPH, by plug chopping as that is our normal cruising speed.

Chris, the wife, on the back gives me a tap on the shoulder at anything above this. How she senses that speed I don’t know, but I digress. This setting meant that the bike was rich at low RPM although it did start easily without any choke of course. Being unable to compare needles, and therefore chose the right one for our purpose at this point, I rode the bike around for a few hundred miles, with only the odd fouling of plugs at low speed being a minor problem. Even then the bike ticked over more evenly than when the Amal carbs were fitted. I decided to go to the BSAOC Summer Camp in Germany which was near Wolfsburg with the SU carb fitted. The bike was loaded up with camping gear and panniers with enough clothes for two weeks and we both set off. As I wanted to gauge the fuel consumption from the bike I noted all mileages ridden and fuel used for the trip. Cruising at 80 MPH on the motorways and Autobahns we did about 800 miles to the camp, from there we went to the Black Forest for a week and on the way home did two back to back days of 450 miles to get home via an overnight stay with friends near Paris. We did over 2500 miles in the two weeks. Most of the travel was on motorway but a good deal was around towns and through Black Forest roads. I checked the fuel used against the mileage and found that overall we got 56 MPG. This now fulfilled the original objectives, of easy starting, good tickover, good acceleration and a top speed over 100MPH. All this, without even tuning the carb to the engines requirements. Unfortunately I had to return the carburettor and manifold before I could do any more with it. A forced return to the Amals soon made me want to build my own version of this setup.

TUNING

I obtained some pipe fittings and copied the design with one change which was to angle the carb flange so the carb had a downdraught of 5 degrees which is about the angle they are fitted to most cars. This also gave better clearance for the dashpot under the frame top rail.

The carb used was a standard SU HS2 carburettor from a 1968 Morris Minor. Specification:- AUD13, 1.25 inch bore, 0.09in. jet AUD 9098, Red spring, Damper AUC8103 and AN needle. Carb was fitted with a 5 degree downdraft.

Round type Amal filter was adapted to fit from a D14/4 Bantam as this was easily available and allowed the engine breather to run into the carb inlet as per standard with a small modification to the rear plate of the filter. Another plus was that the filter was tucked under the tank so not exposed to the rain as on a standard setup. After fitting and setting the carb up so that it ticked over well I then plug chopped the bike at 500 RPM intervals up to 7000 RPM to find where the bike ran weak or rich. I checked out at a list of needles and decided to draw graphs showing one needle’s diameters against another starting with the AN needle fitted. The following are some of the graphs I drew, you will need to rotate the page so the diameter is at the top to read them. The needle is graduated at eighth of an inch landings which all have their own dimension according to each needle’s chart. An example is shown below left.

The original needle which came with the carb was a AN which was used with various settings of the jet. IE. The jet nut was turned up to weaken the mixture at all levels of the needle and down to enrich the settings.

From experimenting with this the E31 needle was tried but this was too rich at tickover due to first diameter being 0.088. Also, very rich after 4000RPM.

Conclusion:- For even tickover the first two graduations should be at 0.089 and 0.0855 (or 0.085) with the needle shoulder flush with the underside of the piston and the jet turned out 6 flats.

M needle fitted and a lot of running done on this. Tickover 600 RPM and top speed at about 100MPH but it pinked at about 2500 to 4000 RPM. This was found to be because the needle was weak at the 3^rd and 4^th angle on the needle. Power came in at 4000 RPM up to 5500 RPM then slowly dropped off.

This showed that the needle was too rich above 5500 RPM.

GR needle fitted with jet still 6 flats out.

Tickover at 500 RPM and starts well cold and hot. No more pinking at 3000 RPM. Picks up cleanly but struggles to run at over 90 MPH. Turned jet in 3 flats and it runs better at 90 and runs to over 100.

Conclusion:- Good at low end and now correct at 2000 to 3000 RPM but too Rich after 4500 RPM.

A V3 needle was fitted with adjusting nut turned out 6 flats which gave good tickover at 600 RPM which revved to over 7000 RPM. Fast enough for what I wanted. A slight weakness was found at acceleration so the needle was put in the lathe and some 1000 grade wet and dry touched on the needle at positions 3,4 and 5 to richen the mixture for pickup. This was much better for pickup.

I decided to fit a small bicycle computer/speedo as I did not think the Smiths clocks were very accurate. According to the Bicycle speedo I was right, as it was up to 15% wrong at some speeds. At this point they bike was taken to Australia for the BSAOC rally and rode around for some time after to see some country. It ran well with no problems and returned 57 MPG. I was recently given a piece of software which allows you to choose a needle by comparing it with the one you are using and showing where it will be richer or weaker. A graph from this programme is shown below comparing the V3 needle to a A5, M2 and 2 needle. Using this programme allied with a rolling road with CO2 analyser or even by plug chopping, it is possible to tune the carb for whatever you wish.

Whether like me for cruising or for speed if you like. The options are limitless.

I am still improving the setup, and have recently found that if I put thicker oil in the dashpot, it slows the piston when rising, meaning there is a greater depression on the jet, so more fuel is sucked up, making the fuel richer for a short while which, means better acceleration. Will be trying this next.

Update to article.

Since writing this I have now done nearly 50,000 miles on the Rocket Three, taking in most of Europe, plus an extended trip round Victoria in Australia. Sometimes riding for miles and miles, without seeing another vehicle. A two week trip to New England in the USA, showed how reliable the setup is. A friend of mine borrowed the bike for a ride, and it convinced him to modify his Rocket Three. However he has a small foundry in his shed, in which he made some blank manifolds in aluminium. I machined the one he gave to me and this is now fitted.

If anybody would any more info I can be contacted by email on or Lightning-spares@lightning-spares.co.uk

ing you

Brian Pollitt