Tuning Standard Triumphs Vizard

8/14/2019 Tuning Standard Triumphs Vizard

1/159

1


2/159

2


3/159

3

TUNING STANDARD TRIUMPHSOVER 1300 cc

MOTOBOOKS

PORTPEEDS

David Vizard


4/159


5/159

5

INTRODUCTION7

chapter 1

OVERHAULING THE SIX CYLINDER ENGINE 11

chapter 2

IMPROVING THE LUBRICATION 21

chapter 3

BOTTOM END PREPARATION 25chapter 4

MODIFYING THE EARLY 6 CYLINDER HEAD 33

chapter 5

BIGGER VALVES FOR EARLY HEADS 47

chapter 6

MODIFYING THE LATER 2000 AND 2500 cc HEADS 53chapter 7

MODIFYING THE TR4 HEAD 63

chapter 8

THE ULTIMATE HEAD 69

chapter 9PORT MATCHING AND DOWELLING 75chapter 10

SPRINGS ROCKERS PUSHRODS AND TAPPETS 77

chapter 11

ENGINE INTERCHANGEABILITY 81

chapter 12

STRENGTHENING THE BOTTOM END FOR COMPETITION 85

chapter 13LIGHTENING ENGINE COMPONENTS 89

chapter 14

HIGH PERFORMANCE CAMS 95

chapter 15

CARBURATION FUEL INJECTION AND MANIFOLDING 105chapter 16

ASSESSING THE RESULTS 109

chapter 17THE TRANSMISSIONS 111

chapter 18

SUSPENSIONS AND BRAKES 137chapter 19

BODYWORK MODIFICATIONS 153

contents

page


6/159

6


7/159

7

IT would appear that over the lastdecade, Triumphs policy has been tointroduce a car of a basically modern

design-and, over a period, develop itto suit various markets. This policyhas been adopted in preference to themore normal method of introducingan entirely new car for a given mar-ket.

Up to a point, this type of thinkingmust be commended. Since the ma-

jority of Triumphs are endowed witha respectable level of roadholding and

braking power, the only items whichbecome vastly changeable are thepower units and transmissions. Sincethe non-sporting driver is unlikely tocomplain if the car has a far higherlevel of roadholding than he is everlikely to use, the manufacturers have,in effect, killed two birds with onestone. The two birds in question are,of course, the sporting and the

nonsporting type of driver.There is inevitably a disadvantagewhen going about things in this man-ner, and this can be easily summedup as a weight penalty. If we design acar in which a range of engines are to

INTRODUCTIONbe used, ones, that is, which vary con-siderably in size, weight and poweroutput, we must build the rolling chas-

sis to cater for the most powerful en-gine it is likely to have fitted. Thismeans the car can be a little over-weight for the smaller engines, thusresulting in a more sluggish car thanis really desirable.

For an example, to take the Her-ald/Vitesse range of Triumph cars.The early Herald with its 948 c.c. en-gine and an all up weight of around

17 cwts. could hardly be consideredthe spriteliest of machines, rather itshould be considered to be the heavy-weight amongst 1 -litre vehicles. Withthe introduction of the 1200 the weightto capacity situation improved but wasstill not altogether remedied. Whenthe 1600 Vitesse came along thingstook a turn for the better. Here we hada car which was not unduly heavy for

its 1 1/2litre power plant. Not only didthe car have-a good weight to capac-ity ratio, but also an engine whoseoutput could be substantially raisedto produce a relatively quick machine.But for competition purposes the 1600


8/159

8

Vitesse still lacked the edge neededto make it a successful race machine.When the capacity of the Vitesse wasincreased to 2-litres and the G.T.6

was introduced, one would have ex-pected the competition scene tochange substantially. It would nothave been unreasonable to supposethat these two cars would have domi-nated the competition in their respec-tive classes, but this has not been thecase. Half the reason for this couldbe that old ideas die hard. The thought

of one of Triumphs products sweep-ing the field and making hay with therest of the competition may not yethave caught on. The idea that it couldbe possible and that they should doso is not altogether unreasonable.

Let us look at the situation in ananalytical fashion and compare, say,the 2-litre Vitesse with some of its

successful competition rivals pro-duced by other manufacturers. Forour purposes a good comparisonwould be to see if we can achieve amachine with a similar performanceto an F.V.A. engined 2-litre Escort. Ifthis can be done, then without ashadow of doubt we can say a Vitessecan be competitive.

The F.V.A. Escort has a 2-litre en-gine producing a power output in theorder of 220 b.h.p. The overall weightready to race is about 15% cwt. andsuspension is independent at the frontbut not so at the rear. The rear endon these cars is served by an ex-tremely well located live axle. Howwould our hypothetical race preparedVitesse compare with this ? First off,

we should find it very difficult to matchor even approach the power outputof the F.V.A. engine without resortingto very expensive twin overhead camheads. Tuning Std Triumphs over1300c.c. This inevitably limits the

number of us who could entertain thefeasability of such a project down toa very small minority. Our best plan isto ignore such a scheme and stick

with the more conventional pushrodlayout. If we really went to town withdown draft heads, fuel injection, etc.,etc., we could achieve a possible 190-195 b.h.p. from the engine, although,might I say, this figure has yet to berealized. Straight away we find we aredown on the power of the F.V.A. en-gine by 25 - 30 b.h.p. To a certain

extent this deficiency can be offset.By an intensive lightening programmeinvolving the use of as manyfibreglass components as possible,the weight of a Vitesse can be re-duced to just under 15 cwt. so a littleof the lost power is made up for byhaving slightly less weight. Even so,our power to weight ratio is still down

by 8-1/2%. There is a possibility thatwe can offset some of the disadvan-tages of this lower power/ weight ra-tio in the suspension department.

Unlike the F.V.A. Escort, the Vitessehas an independent rear suspension.This means we have less unsprungweight to contend with which theoreti-cally should result in better road hold-ing. Against this must be added thefact that the Vitesse is one of the fewcars which still utilizes a chassis whichin all probability is not as stiff as a unitconstruction body. The result of thisis that our carefully planned suspen-sion geometry may not follow the ex-act motion it was intended to. In point-ing this out I may be leading you tobelieve that even a fully developed

suspension will be inadequate for ourpurposes. This is not the case. TheTriumph rear end will give better re-sults than a conventional solid axlebut its superior


9/159


10/159

10

ity might not be quite as much as onewould expect.

Having made our comparison, itcan be seen that a really competitive

Triumph is not by any stretch of theimagination impossible. It may be abold assumption, but I would predictthat we are likely to see far more ac-tion in the next few years providingno great restraint is placed on the situ-ation.

In an effort to encourage Triumphowners or would-be Triumph ownerson to greater deeds in the field of com-

petitive motoring, this book has beenprepared. In the main it will give triedand proved methods of obtainingmore performance. Here and there

ideas will be put forward which arealthough untried based on sound en-gineering practice.

These ideas will show a gain in

power but the amount gained can onlybe estimated. It has also been recog-nized that not everyone is bent on pro-ducing an all-out racer, so all modsdescribed will start from square oneand progress from there. For thoseof you who are not fully acquaintedwith your car, it is a good idea to pur-chase a workshop manual for the rel-evant machine. This book is not in-

tended in any way to replace anowners manual but to supplement itfor the purposes of achieving a higherperformance.


11/159

11

SINCE the conditions of the bottom

end of an engine can have a majorinfluence on power output, we shalldiscuss, thoroughly and in detail, howto go about building up the bottom endwith a view to extracting the maximumhorsepower for a given state of tune.

If the car you are working on is sec-ond-hand and has seen a normalmileage, then it is more than likelybelow par. It is also noteworthy, since

we are on the subject, that 99 out of100 new engines can benefit as faras power is concerned by a carefulprecision rebuild. In this chapter weshall deal with the 1600, 2000, and2500 c.c. six-cylinder engines. The2litre and 2.2-litre engines will be dealtwith in the following chapter.

Degreasing

The first move is to take the engineout of the car and degrease it with asuitable cleaning agent. Then, givingyourself plenty of working space, stripthe engine down and where neces-sary, number components for replace-

ment in their original position. Check

the cylinder bores for wear. The stan-dard sizes of the bores are as follows:1600

STD 2,6276-2.62872000

2.9405-2.94162500

2.9405-2.9416In the standard bores the pistons

are graded in three different sizes and

the bore clearance on a new enginewill be between 3.5 thou and 4.2 thou(0.00350.0042). If wear has takenplace we must determine the amountof clearance which exists before wecan say whether or not a re-bore isrequired. These engines were fittedfrom the factory with either solid skirtor split skirt pistons. Ideally the boreclearances which we should aim for

are 0.005" for the 1600 motor, whensolid skirt pistons are used, or 0.0055"when split skirt pistons are used. Ifthe engine is to be used solely forcompetition work, the bore clearanceshould be 0.0055" with solid skirt pis-tons. The use of split skirt pistons

chapter I

OVERHAULING THE SIX CYLINDER ENGINE


12/159

12

checked for size. If they are too tightwe are likely to encounter stickingrings and if too loose, the rate of ringwear will increase and cause gas

blow-by.The ring gap can best be measuredby using a ring and a feeler gauge.As new, the normal ring clearancevaries for the compression rings from0.0019" to 0.0035". For our purposesthe ideal clearance is 0.002". It maybe difficult to achieve this figure with-out sorting through a number of pis-tons which can make life a little diffi-

cult. Whatever the case may be, thering gap should not be allowed to ex-ceed 0.0035' which is the top limit ofclearance as new from the factory.The oil control ring should be a slightlycloser fit in its slot. The normal toler-ances for the clearance are 0.0007'

in a competition motor is not to be rec-ommended. Also one should avoidthe use of pistons with a ring belowthe gudgeon pin as this creates extra

bore drag On the 2000 and 2500 c.c.engines we should have a piston tobore clearance of 0.0055" for fast roadwork and 0.006" for competition pur-poses. Here again we must avoid theuse of split skirt pistons or pistonswhich have an extra oil control ringbelow the gudgeon pin.

When determining the bore clear-

ance which exists, be sure to mea-sure the piston diameter at rightangles to the gudgeon pin and at apoint as shown in Fig. 1. The reasonfor this is that the pistons are both ta-pered and oval. Should the bore clear-ance prove excessive then ideally arebore is called for. The best pistonsto use are those supplied by Tri-

Grade F G H1600 2.6264/2.6267 2.0968/2.6271 2.6272/2.62752000 2.9384/2.9388 2.9388/2.9392 2.9392/2.93962500 2.9380/2.9384 2.9384/2.9388 2.9388/2.9392

umphs. In the event of the bores hav-ing less clearance than required,some honing should be done. Thebores, whether rebored or not, must

be honed to a 20 micro inch finish witha cross hatch patter at 450 Do not at-tempt to use any other type of finishother than this or it will cost you a dropin power in the long run.

Piston fits

Having prepared the bores one

should next ascertain whether the pis-tons are fit for re-use. The diameterof the skirt will have already beenchecked in connection with borepreparation. For reference purposesthe following chart gives the gradedsizes of the pistons.

The ring slot widths should be

to 0.0027'. For our purposes weshould try to achieve a clearance of0.0015'.

Before fitting the rings to the pis-

tons, they should be gapped. The1600 motor should have the gaps onall the rings set at 0.010". The cor-ners of the gap when viewed fromabove should have a 0.005" - 0.010"radius stoned on to them. The gap onthe 2000 and 2500 c.c. engines is bestset at 0.012' and again a radius of 5to 10 thou. should be stoned on to thecorners of the gap.

Finally, before fitting any rings,check them for flatness on a surfaceplate or a piece of plate glass. If thereis any detectable sign of warpage re-

ject them. Check the fit of the gud-geon pin in its bore. If there is theslightest evidence of sloppiness


13/159

13

then either piston, gudgeon pin orboth must be replaced.

Con-rods

We will now focus our atten-tion on the con rods as these itemscan cause a profound drop in power

if they are not true. Since con rodsare amongst the most heavilystressed items in an engine, it is awise move, before using them, tohave them crack tested. Once theyhave passed any crack detectiontests, checks for truth should be car-ried out. First off measure the bore ofthe big end (without shells). This mustbe 2.0210'-2.0215dia. for all engines.

The little end should be asnug slide fit on the gudgeon pin, i.e.the gudgeon pin will not quite fall outunder its own weight but can easilybe pushed out with thumb pressure.The rods can now be checked forstraightness. For our purposes the

maximum error in bend whenchecked across the gudgeon pin mustnot exceed 0.0005" and the maximumallowable twist should not be morethan 0. 001 .

To make things absolutelyclear as to what we are checking, re-fer to Fig. 2. This shows the effect ofbend and twist greatly exaggerated for

the sake of clarity. If you do not botherto check the rods, then you will surelypay for it in increased oil consump-tion and reduced power output. Thoseof you who have a lathe and a dialindicator, can check the rods your-selves. For those who have not, mostmotor machine shops possess suit-able equipment to do the job. Shouldany of the rods prove to be out bymore than 0.007", then it is a goodidea to replace them with new rods.Even if you have had to resort to newrods, do not accept the fact that be-cause they are new


14/159

14

they must be right. Check them to besure they are right.

Big end bolts can be a critical areaon the rod assembly. Repeated torqu-

ing up and prolonged use causesthem to stretch a small amount. Toavoid any unnecessary failures in thisarea, the bolts should only be usedonce. It may seem unduly expensiveto replace these bolts just becausethey have been used once, but if arod lets go, then the chances are itwrites off a whole engine.

The cam

Moving on to the cam, we find thatthe standard journal clearance variesbetween 4.6 thou. (0.0046) and 2.6thou. (0.0026). Ideally we want thisclearance at about 3 thou. (0.003).Some blocks have cam bearings forthe journals to run in and some do

not. If you have one with bearings,then it is a lot easier to achieve the 3thou. clearance since new bearingsare usually on the tight side. If youhave a block without bearings, theseare usually the earlier ones, then bear-ings should be fitted. If you intendusing a hot cam, then it is even morestrongly recommended that you use

cam bearings.When installed, the cam must ro-tate freely. To put a figure on it, thecam should, when lubricated with thinoil, rotate when a torque of 1 oz. ft. isapplied. If you have your clearanceat 3 thou., and the cam seems stiff,the usual fault is a bent camshaft.Needless to say, it is not the donething to use a bent camshaft. Inspectthe cam lobes for wear. If there is anysign of deterioration, then replace thecam. Early 1600 Vitesses had camfollowers having a diameter of 0.687".These were later replaced by camfollowers of 0.800' dia. which is the

same size as fitted to the 2000 and2500 engines. The small diametercam followers can be a weak point inthe engine, especially if a hotter cam

is to be fitted. On engines with thesmall cam followers, it is a straight-forward job to machine out the fol-lower bores to accept the larger camfollowers.

The 1600 and 2000 engines use asingle row timing chain to drive thecam. The 2500 c.c. engine uses aduplex timing chain. For the sake of

extra reliability on a tuned engine, onecan use the duplex set up of the largeengine on either of the smaller en-gines.

Crankshaft

The crankshaft of an engine is justabout the hardest working compo-nent. Bearing this in mind, it is only

fair that we give the crank every op-portunity to show good service. If thecrank has already seen a reasonableperiod of use, it will not take too kindlyto all the increased loads broughtabout by tuning and the use of extrarevs. If we hope to achieve a reason-able life from the bottom end of theengine, then we must give it a good

chance to start with. This is not to saythat the bottom end is in any way un-reliable, but if we tune the engine weare expecting the crank to withstandloads in excess of those it was origi-nally designed to cater for. Thismeans that we can accept no halfmeasures. The crank must be in A.1condition before we can consider itsuse.

The 1600 and early 2000 enginesup to about 1966 use virtually thesame cranks. The main bearing sizeis 2.0005"/2.001 dia. If the main jour-nals are more than 0.0003' below thebottom limit, or they show


15/159

15


16/159

16

The above sizes for the crank does

make the assumption that the bear-ing shells that are used are in themiddle of their size tolerance. Themain bearing clearance with thesecranks is between 1.2 thou. and 2thou. (0.0012 -0.0020-). We shouldattempt to get an initial clearance withnew shells of 1.5 thou. When theshells have run in and everything hassettled down, this clearance will have

opened up to about 2 thou., which isjust where we want it. The big endsshould also be treated in a like man-ner as far as clearances are con-cerned.

Assuming the block has been fullyprepared, we can now install the crankand in doing so make certain func-tional checks. Fit the bearing shellsinto the main bearing caps and block,place the crank in situ and oil the main

journals. Fit the caps and torque downthe bolts to 50-55 lb. ft. for 1600 and55-60 lb. ft. for the 2000 and 2500engines. Check that the crank spinsfreely i.e. it should rotate when a

torque not exceeding 1/3 lb. ft. is ap-

plied. If the torque required to turn thecrank is appreciably above this figure,then you have either a tight bearingor bearings, or a distorted crankshaft.You can determine whether the cul-prit is a tight bearing or not by loosen-ing off each bearing cap in turn to seeif this frees the crank. In nine casesout of ten, tight bearings can be curedby swopping the shells from one

housing to another. In the odd onecase out of ten, you will have to getsome more shells. If you want to num-ber your bearing clearance exactly,then I would suggest the use ofplastigauge clearance gauges. Thesegauges allow quite an accurate mea-surement of the clearances existingto be obtained. If each individual bear-ing clearance is correct, yet the crankis still stiff, then this points towards adistorted crank. There is also a verysmall chance that the block could bedistorted which would also cause thecrank to be stiff. If the crank you areusing has been reground or is a new

any signs of ovality or taper, then thecrank should be reground. The crankpin diameter standard is 1.8750"/1.8755". Here again we should treat

any discrepancy from the true size byregrindingThe cranks from the 2500 and the

later 2000 c.c. engines differ inas-much as that they are endowed withmuch larger main bearing journals.These are some 3/16" bigger in di-ameter than the lesser or earlier en-gines. The mains journals should fallbetween 2.3110' and 2.3115" diam-

Big end standard Mains standard1600 & 2000 (up to 1966) 1.8753' 2.0008"2500 & 2000 (after 1966) 1.8753" 2.3113'

Big ends 0.010' O/S Mains 0.010 O/S1600 & 2000 1.8653' 1.9908'2500 1.8653' 2.3013'

eter, and should the limiting tenth of athou. below bottom limit exist, then aregrind must be done. If you are se-lecting a new crank or having one

reground, then you should go forcrank pin and journal sizes about0.0002" below top limit. When usinga reground crank, we should avoidgoing below 0.010' undersize if anyreasonable degree of tuning is envis-aged. The following list gives the idealsizes of crankpin and journals to usewith a standard crank and a 0.010"regrind.


17/159

17


18/159

18

one, the chances of the crank beingout of true are very small but never-theless should not be ignored. Themost likely case of a distorted crank

will arise with those which have seenuse but not enough to cause them towear outside of the tolerances re-quired. This means a regrind will berequired to rectify the inaccuracy.

Block distortion

Lastly we can on very rare occa-sions, get block distortion. When

blocks are cast they have stressesset up in them due to differing ratesof cooling when casting. Over a pe-riod this causes the casting to movei.e. small changes of shape occur.This means that if the block was ma-chined a short period after it had beencast, then over a period of a fewmonths it can distort itself by a num-ber of thou. whilst the casting

stresses settle down. There are twoways to overcome the effects of in-ternal stresses in castings. The firstmethod is to leave the castings toage, then rough machine them andthen give them a further ageing pe-riod. After this period, which can beas much as a number of months, thefinal machining is carried out. Thesecond method is to heat the blocks

in a furnace after they have been castto allow the stresses to quickly settleout. Just once in a while it is possibleto find you are the proud owner of a

block which has not been properlystress relieved. Unfortunately thisdoes not show up immediately butusually makes itself felt some twelvemonths after, or the first time the en-gine is completely stripped. In theunlikely event that you have a dis-torted block, the only remedy is to re-place it.

We have now carefully prepared all

the major components connectedwith the bottom end. Things like oilseals, timing chains, etc. should bereplaced as a matter of course. Thecrank, rods, pistons, flywheel, clutchand crank pulley should now be bal-anced. This will not only make an al-ready smooth engine smoother, butgive that little extra reliability neededfor high performance motoring.

Valve guides

The only other points connectedwith the mechanical well being of theengine are associated with the cylin-der head. The wear on valve stemsand guides must be checked. Thevalve stem sizes for the three enginesare as follows:

Inlet Exhaust2000 & 2500 0.3107'-0.3112' dia. 0.3100 -0.3105' dia.1600 0.3100'-0. 3110 dia. 0.3090"-0.3080" dia.

The bore of the inlet guides arecommon to all engines, this being0.312"-0.313" dia. As you can see theclearance on the larger two enginescan, on the inlet, vary between 0.0023'and 0.0008'. The clearance to achieveis between 0.001 and 0.0012'. Onthe exhaust side the clearance canvary between 0.003 and 0.0015'. Forour purposes 0.002" clearance isneeded. On the 1600 engine the stan-dard clearance on the inlet valve can

be between 0.001' and 0.003". Herewe need 1 to 1.2 thou. clearance(.001" 0.0012"). The standard clear-ance on the exhaust valve on 1600engines is by normal standards large,this being between 3 and 5 thou.(0.003"-0.005"). One should endeav-our, where possible, to select com-ponents which will set this clearanceat the lower limit of 0.003".

You may wonder why the valve to


19/159

19

valve guide clearance has been goneinto in such detail. The main reasonis that the motion of the valve mustbe controlled in such a manner as toallow the valve to move in a straightline only. The more clearance one hasthe less likelihood there is of achiev-

ing the required motion. Excess clear-ance will cause an increase in oil con-sumption, a small drop in power andan increased valve seat wear rate.The last point is especially relevantwhen we consider the use of radiusvalve seats which will be thoroughlydiscussed later in the book.

Before winding up our dimen-sional check of various engine com-

ponents, we should check out therockers and rocker shafts for wear.Normal clearance is between 1.8thou. and 3.3 thou. (0.0018" - 0.0033')If the clearance is more than 0.005'above the upper limit, steps shouldbe taken to remedy it. Rockers which

are too loose can set up a mode ofvibration which does nothing to en-hance valve timing at high revs., es-pecially if the cam is of a hotter vari-ety.

Assembly

We should now come to thetime when we are able to put the en-gine together. Assemble the crank,rods and pistons into the block usingengine oil on bearing and rubbingsurfaces. Having assembled the en-gine, we now come to the test whichwill verify your dimensional checking.Rotate the crank assembly about fifty

times. Then using a torque wrench,establish the torque required to turnthe engine over. The absolute maxi-mum torque required to turn the en-gines over are 20 lb. ft. for the 2500,16 lb. ft. for the 2000, and 12 1/2 lb.ft. for the


20/159

20

1600. If the build up has been donecarefully, these torque figures can beas much as 30% lower. If the figure issignificantly higher, then something is

amiss, so you should go through whatyou have done to locate the cause ofstiffness. Excess internal friction in anengine is a sure way to lose power. 5lb. ft. of frictional torque over andabove our minimum will cost us 6.7b.h.p. at 7000 revs. As you can see,the moral is to build up an enginewhich is as free as possible whilst stillfailing in with our component dimen-

sional limits.With the exception of a couple of

points, the engine assembly fromhere is straightforward. The couple ofpoints which we must bear in mindare connected with the accuracy andconsistency of our ignition timing. Thedistributor is driven in a conventionalmanner by a skew gear coming offthe camshaft. As with any gear train,

backlash is necessary but on theseparticular engines we have a measureof control over the amount of back-lash present. The total backlash which

can be present at the distributor driveshaft is not only the backlash of theskew gears but also an amountcaused by the end float of the cam-

shaft. Our camshaft end float is con-trolled by a thrust plate at the timingchain end of the engine. The normalcam end float is between 0.004' and0.008". By selecting a suitable platewe must get the endfloat down to0.004'. We can, if we are going to beultra fussy about it, reduce this to0.003' but we must definitely not gounder this figure.

The other area which must beclosely looked at is the end float ofthe distributor drive shaft itself. Thestandard clearance figures range be-tween 3 and 7 thou. and the figurewe want is 3 thou. on the nail. Fig. 3shows the precise method to adoptfor determining this clearance andsetting it to the required figure. As-suming we have a perfect distributor,

any erratic timing caused by excessbacklash can cost around 2-3 b.h.p.,so play close attention to these endfloat settings.


21/159

21

WITH any tuned or hard pressed en-gine, the lubrication system must bein perfect order or better still, upratedto cope with more stringent condi-tions.

The standard oil pump is capableof coping with the extra demandsplaced upon it by a modified engine ifit is in perfect order. However, to en-sure a good engine life, a little timeand trouble spent on the standardpump will not go amiss. The work in-volved on the pump includes a thor-ough check on clearance existing andrectifying these where necessary. Fig.4 shows an exploded view of thepump assembly just so that we knowwhat bits we are talking about. Themain object of the exercise is to buildup a pump assembly having the mini-mum clearances between the bodyand the outer rotor and between theouter rotor and inner rotor, A normal

value for these clearances is around0.007'. By selecting components, wecan quite easily get this figure downto 0.004' and if a great enough num-ber of components are available thiscan be reduced further still. Theseclearance values can easily be

chapter 2IMPROVING THE LUBRICATION

checked with feeler gauges as shownin Fig. 5 and 6. The only clearancethat we can readily make adjustmentsto is the end float between the innerand outer rotor and the end plate. Bymachining a small amount off thebody we can set the end float to 0.001thus reducing any leakage at this pointto a minimum. Fig. 7 shows how theend float clearance can be measuredso as to enable the correct amount ofmetal to be removed.

When machining the pump body,be very careful as there is very rarelymore than 0.003' to come off. Whilstmachining the body, you must alsomake sure that the inner and outerrotor are of identical length or you willhave a different end float for eachcomponent. Once all the pump com-ponents have been selected or ad-

justed as necessary, the pump canbe assembled. Using very light oil to

provide a little lubrication, check thatthe pump rotates smoothly and freely.Having gone to all this trouble to pre-pare the pump, we can expect it tobe about 7-10% better than the aver-age pump. This may not sound like a


22/159

22


23/159

23


24/159

24

fantastic gain for all the time andtrouble resorted to, but we must re-member it is always the last straw thatbreaks the camels back. The in-crease is sufficient to give us that ex-tra margin of safety between retain-

ing or losing the vital oil film betweenthe highly stressed components.

Pressure release valve

Still on the subject of lubrication,suitable adjustments should be madeto raise the pressure at which thepressure release valve comes intooperation. This can be achieved quite

simply by increasing the spring pres-sure which is exerted on the pressurerelease valve. A spacer made up asshown in Fig. 8 and fitted under theface of the release valve spring re-

taining bolt will do the trick. Thespacer will cause a raise of 10-12 lbs.per square inch of oil pressure beforethe pressure release valve comes intooperation. It will be found unneces-sary to so such a mod to the springpressure unless the engine spends agreat deal of its time in the upper revo-lution range as would an engine usedin competition.


25/159

25

THE 2 and 2.2 litre four cylinderTriumph engines are well noted fortheir long life and ruggedness. If theengine is carefully assembled, then

even in a relatively highly tuned state,reliability does not suffer to any no-ticeable degree providing one paysstrict attention to the rev limit. Weshould apply the same thorough nessand attention to detail as was outlinedin Chapter 1 on the six cylinder en-gines. To avoid unnecessary repeti-tion of the whys and wherefores, thetolerances that we should work to will

just be quoted with only limited expla-nation as to why.Assuming the engine has been re-

moved from the car and has beencompletely stripped, we should be ina position to carry out our dimensionalchecking. As good a point as any tostart with is the block. A close look atthe block will reveal that, unlike mostengines which have pistons running

directly in the block casting, the Tri-umph engine has removable wet lin-ers. The definition of a wet liner is onethat is directly in contact with the wa-ter, as opposed to one that is in con-tact solely with the cast iron of the

block. Anyway, to get back to thepoint, we should check the bore forwear. Since this particular engine hasbeen out of production for some time,

I should doubt if any engines whichare still on standard size pistons arenot in need of a rebore. The standardsize of bores for the 2-litre engine andthe standard piston to bore clear-ances are as follows:

Standard Standardbore clearance

Std. 3.2677" 0.0042'

+ .020 3.2877 + .030 3.2977 + .040 3.3077 + .060 3.3277

For fast road work the bore clear-ance should be 0.005' and for racing0.006'. When determining the piston/bore clearance, the piston should bemeasured level with the gudgeon pin

but at 901 to it. If the piston to boreclearance exceeds 0.008", then arebore will be required if we are toachieve optimum power. With the 2.2litre engine the

chapter 3

BOTTOM END PREPARATION


26/159

26

standard bore and rebore sizes willbe:-

Standard Standardbore clearance

Std. 3.3854 0.0043+ .020" 3.4054 + .030" 3.4154 + .040" 3.4254" + .060" 3.4454"

For road work we want a 5.3 thou.clearance, and for racing a 6.4 thou.clearance. The bore sizes alreadyquoted are those which give a stan-

dard clearance. Since it is easier toobtain extra clearance by making thebores slightly larger, than it is to re-duce the sizes of the pistons theabove bore sizes will want enlargingon. To obtain the correct clearanceusing new pistons, the following chartshould be adhered to :

2-litre engine Road Racing

Std. 3.2685 3.2695+ .020" 3.2885 3.2895+ .030 3.2985 3.2995+ .040" 3.3085 3.3095+ .060 3.3285 3.3295

2.2-litre engineStd. 3.3864 3.3875+ .020" 3.4064 3.4075+ .030" 3.4164 3.4175"

+ .040 3.4264 3.4275+ .060" 3.4464" 3.4475

The bores should be finished byhoning to 15-20 micro inch finish witha cross hatch pattern of 45".

As has been stated before, thisparticular engine has removable lin-ers which leads us to some interest-ing possibilities as far as increasingits capacity is concerned. The linersfrom the 2.2 litre engine can be fittedto the 2litre engine, thus bringing it upto the same size as its bigger brother.Also there is a 9.25/1 piston made byHepolite which will give a further slightincrease in capacity. This piston has

a standard size of 3.4252" (87 m/m)and is available at rebore sizes of+030 and +040. The 2.2 litre linerscan be bored to take this piston but

the 2-litre ones cannot. If we use thispiston (Hepolite part number 13958)in its +040 form we will bring the ca-pacity up to 2239 c.c. from the origi-nal 2138 c.c. If this piston is used inits standard 87 m/m form rather thanin oversize form, the capacity is 2187c.c. For the purposes of building up ahigh performance road or race enginethe use of this piston is to be recom-

mended. Its advantages lie in the factthat it is quite light in weight, has noring below the gudgeon pin and is ofthe solid skirt variety. The boring sizesfor this piston are:

Road RacingStd. 3.4262" 3.4273+.030" 3.4562" 3.4573+.040" 3.4662" 3.4673"

If the Hepolite piston just mentionedis not used, then it is advisable to usepistons supplied by Triumphs. Thereare some pistons on the market whichhave an extra oil ring below the gud-geon pin. The use of these is to beavoided as are pistons having splitskirts.

LinersIf by some chance, your engine has

already been bored to its maximumsize and is yet again in need of arebore, then it will be found neces-sary to fit new liners. This in itself is asimple enough job, but one whichmust be done properly if we are toavoid any possible leakage from

joints.First off, one should remove the old

liners. It is definitely not advisable totry bashing out the liners solely withthe aid of a hammer. To do the jobproperly without incurring any nastyside effects, one


27/159

27

should make up a special tool to dis-tribute the hammer blow evenly overthe end of the liner. The dimensionsfor such a tool are shown in Fig. 9.Place the block, head face down, onsome blocks of wood. Place the liner

removing tool into the bore of thelinerthe crank end of the liner. Placea suitable stout punch i.e. 1 dia. steelbar on top of the removing tool andhit the punch with a heavy hammer.In most cases the liner will come outwithout undue trouble. Occasionallyan obstinate liner is encountered, butthe application of a little penetratingoil will ease the situation. Having re-

moved all the liners, thoroughly cleanthe area on which the bottom jointinggasket seats see Fig. 10. This areamust be perfectly free from foreignmatter which may prevent a perfectseal being formed. Smear the newliner sealing gaskets with a non-set-

ting jointing compound. Fit the newliners in position and as soon as pos-sible afterwards, fit the head togetherwith an old head gasket and torquethe assembly down to about 6070 lb.ft. Fitting the head effectively pulls

down the liners, seating them firmlyonto the gaskets. The head can nowbe removed so that we can check theliner protrusion above the block face.Place a straight edge across the topof each liner, and using a feelergauge, check the gap between thestraight edge and the block face seeFig. 11. This clearance should be be-tween 0.002"0.005'. In all probability

the protrusion of each liner will varybetween these limits. If the protrud-ing height of each liner is the same,then you have just been lucky. In thegreater likelihood that each linerheight varies one to another, a littlemachining is called


28/159

28

for. The top faces of the liners shouldbe skimmed to bring them all to thesame height above the block face.This height, however, must still remainwith the 0.002" - 0.005" tolerance. Ifyou should find that the protrusionheight is less than 0.002", then the

liners will have to be removed and afew thou. taken off the top of the block.Having done this, refit the liners asdescribed previously, and machinethe tops of the liners to bring them allto a common height, within the speci-fied tolerance.

Rings

Having, we hope, successfully pre-pared the bores, we can turn our at-tention to the piston rings. As usual,for high performance engines, weshould check the rings for flatnessand for ring to groove clearance. Thefactory limit for ring to groove clear-

ance is 0.001 -0.003' We definitelyshould not allow the clearance to ex-ceed these figures and ideally if a se-lection of pistons is available, thisclearance should be at 0.0015'. Thering gaps for all rings should be set to0.012'-0.015' and the corners of the

gap stoned off to a 0.005"0.010" ra-dius.The tappets or cam followers must

be a free fit in their bores but on theother hand should not be sloppy. Themaximum tappet to bore clearancewe can tolerate is 0.0016'. Since mostof the wear seems to occur on thecam followers rather than the bore,any wear problems can usually be

overcome by fitting now cam follow-ers. If a new standard cam or a highperformance cam is to be fitted, thenew cam followers are essential nomatter what condition the originalones are in. If this Is not done youhave a 90% chance


29/159

29

of having a cam life of less than a fewthousand miles. Whilst we are on thesubject of cams, the camshaft must,when fitted into the block, rotate freelyand smoothly. A bearing clearance of0.002' being ideal and 0.0037" clear-ance is the maximum allowable clear-ance for all the bearings bar the front

one. The clearance on the front bear-ing is best around 0.003" and themaximum that is permissible is0.005".

To avoid undue power loss, the rodsmust be true within fine limits. Themaximum bend when checked overthe length of the gudgeon pin shouldnot exceed half a thou. (0.0005") andtwist should not be greater than 0.001

. Chapter 10 will clarify the conditionswe are ideally seeking to minimise.

Big ends MainsStd. 2.0861 -2.0866 2.4790-2.4795"0.010, 2.0761 -2.0766 2.4690-2.4695"0.020' 2.0661 -2.0666 2.4590"-2.4595"

The little end of the rod must be asnug fit on the gudgeon pin. This usu-ally means a clearance of 0.0002"/0.0004". Although these engines arenot prone to rod break ages, it is agood idea to play safe and have themcrack tested and as another precau-tionary measure, fit new big end bolts

each time the big ends dismantled.As has been said before, we aredealing with an engine which hasbeen out of production for some time.It is therefore reasonable to assumethat in most cases the majority of theuseful crankshaft life has been used.This will mean that nine times our often a crank regrind is called for. Thelimits of crank journal sizes are shown

below at standard size and 0.010' and0.020" undersize.


30/159

30

For a road going engine, a crankregrind to -0.040" is passable. On theother hand, if the engine is to be usedfor serious competition, the 0.020"regrind should be regarded as maxi-mum. It is always difficult to knowwhere to draw the line as far as crankregrinding goes. Each step of 0.010"does not reduce the crank strengthby a vast margin, but going from stan-dard to 0.060" does. Since the crankin standard form is only safe to 6500continuous or 7000 instantaneous, wemust conserve all the crank strengthpossible. These engines will seem-ingly last for ever at 6000 r.p.m. butat 7000 the life can be measured inminutes.

Assembly

We should now be in a position toassemble the bottom half of the en-gine. The first move is to fit the crankand check for stiffness, The methods

of pinpointing the cause of stiffnessand its remedies were outlined inChapter 1 concerning the six cylinderengines. The fitting of pistons and conrods follows normal procedure. Wecan, at this stage, assess the turningtorque of the crank/rods and pistonassembly. If all is well then the assem-

bly should rotate when a torque notexceeding 24 lb. ft. for a road engineand 20lb. ft. for a race engine is ap-plied. If the torque figure is signifi-cantly higher than this, then some-thing is amiss, in which case, if youare wise, you will strip the engine tofind out why. The difference betweenthe turning torque for a road engineand a race engine, arises from the

different bore clearances. Once theengine has run in, this torque figurewill drop and the difference betweena road and race engine will drop from4 lb. ft. to about 2.5-3 lb. ft. This dif-ference may sound like a trivialamount to worry


31/159


32/159

32

about, but it is worth about 4 b.h.p.and 6500 revs. and with a race en-gine every little bit counts.

As with the six cylinder engines, we

must pay close attention to the cam-shaft end float, and the end float ofthe distributor drive gear. The frontcam bearing is removable and hason it a flange which controls the depthto which it can go into the block seeFig. 12. The normal cam end float is3 to 7.5 thou. (0.003" to 0.0075").Since we want the most accurate ig-nition timing possible and since cam

end float effects the ignition timing,we must reduce the cam end float toa minimum. The minimum is at 2 1/2. to 3 thou. (0.0025"-0.003"). In vir-tually every case the existing end floatwill be more than we want. To reducethe end float down to the figure weneed, some machining will be calledfor on the abutment face of the flange

on the front cam bearing. This facewill need skimming on a lathe to givethe end float required.

This leaves us with only the distribu-

tor drive gear end float to contendwith. On a standard engine this canvary between 0.003" and 0.007. Forour purposes, the 0.003' setting isrequired. The technique for settingthis is virtually the same as for the sixcylinder engines and is shown in Fig.of Chapter 11. Building up the remain-der of the bottom half of the engine isa straightforward procedure. All gas-

kets and oil seals should be replacedduring the rebuild. The timing chainshould also be renewed. If all thestated tolerances have been adheredto, and a great deal of intelligent careapplied, the end result will be wellworthwhile as far as power and reli-ability are concerned.


33/159

33

THE 1600 Vitesse Mk. 1, 2-litreVitesse Mk_ 1, G.T.6 and early 2000saloons use a head which is basically

common. The Mk. 11 Vitesse andG.T.6, TR5, TR6, and the 2.5 P.I. sa-loon use a different head. The heads,as far as fitting is concerned, are in-terchangeable between the early andlate models. The major differencebetween the two basic types of headlies in the chamber design and themanifold joint face. It any headchange from early to late models is

contemplated, then the manifoldsmust be changed to suit. The laterheads also sport a larger inlet valvewhich assists breathing. The largervalve, together with the redesignedchambers gives an increase in powerof about 7%. Since cylinder heads areso expensive, one should only con-sider changing the head if the ultimatein power is required because the early

heads can be modified to give resultsvery nearly as good. In this chapterwe will deal specifically with the earlytype of head. It is quite simple to de-termine which type of head you haveby comparing your own head with thehead drawing in this book. The early

heads have the manifold joint faceterminating about 1 from the edgeof the casting when viewed end on.

The later heads, however, have themanifold joint face terminating flushwith the edge of the head casting.

We will start off with a Stage 1 typeof head for the 1600 and Mk. 1, 2-litre Vitesses, Triumph 2000, earlytype, and the Mk. 1 G.T.6.

Removing guides

Strip the head of all parts so thatyou are. left with the bare head cast-ing. I personally do not like the use ofa hammer to remove valve guides. Ifyou are using a heavy hammer to re-move the guides you only have tomiss the guide removing drift or catchit at a slight angle to do the head aninjury. By far the best way to removeguides is by using a press. Of course,

if you find that none of your local mo-tor machinists or garages have one,then the only alternative left to you isthe large hammer. Either way, pressor hammer, a guide removing drift willbe required, the relevant dimensionsof which are given in Fig. 13.

chapter 4

MODIFYING THE EARLY 6 CYLINDER HEAD


34/159

34

Thoroughly clean the head faceusing emery cloth as required, so asto achieve a surface suitable for mark-ing on. Apply marking blue to the headface. Marking blue, by the way, isavailable at any hardware store deal-ing in engineers tools. While you are

at it, you will also need a tin of engi-neers blue. For those of you not fa-miliar with either of these compounds,marking blue is a marking agentwhich dries very quickly, in fact, in onlya number of seconds. Engineers blueon the other hand is a marking agentwhich virtually never dries out and isused to detect and show up points ofcontact, i.e. high spots between mat-

ing surfaces. We find the engineersblue virtually indispensable later whenlapping in valves.

At this point we will need tomake up a template to accuratelymark out our new chamber shape. Inthis particular instance, the best bet

is to make one from perspex of 1/16thickness. The drawing of the cham-ber Fig. 15 is to scale so the cham-ber shape can be traced on to theperspex from the drawing. When trac-ing this shape, do so as accuratelyas possible, as there is very little room

for error. A badly traced template anda little erroneous grinding later cancause a scrapped head. Trace outboth the original shape and the newshape on the perspex and if possible,trace out the shape on both sides ofthe perspex to reduce to a minimumany effects of parallax due to the thick-ness of the material. Inspection of Fig.15 will show that the chamber is only

modified over part of its perimeter. Ifwe now cut the perspex only alongthe reprofiled part of the chambershape Fig. 14, then we can use therest of the chamber outline, whichremains unchanged, to line up ourtemplate. The chambers in


35/159

35


36/159

36


37/159

37

the head are handed. To mark outan opposite handed chamber wemerely turn the template over onto itsother face. Once you have conscien-

tiously made your template, the headmarking out procedure can begin.

To do so you will need a very sharpscriber, and also one which has a longtapered point with, say, an includedangle of about 300, not a short stubbypoint. The only reason for this is thata short stubby point will inevitably pro-duce a line which does not coincide

with the edge of the template but is ashort distance out from it. As I havementioned previously, we have virtu-ally no room for error on the ,cham-ber shape, or we shall infringe, if thatis the word, into the domain of thegasket.

We can at last start grinding, butfirstly some more words of caution. Ifyou are the proud possessor of asteady hand you will find that on thishead you have enough room to beable to avoid serious contact betweenthe grinding wheel and valve seat. Ifthis is the first time you have at-tempted any serious head work or youare not at all sure that you have asteady enough hand, then it is best totake a few precautionary measures.

First off, take a couple of scrap valves,one inlet, one exhaust, and machinethem across the head face sufficientlysuch that when they are fitted in posi-tion the valve head face is flush withthe chamber roof. Next, tap the valveguides a short distance into the head,

just sufficiently to locate them. By in-serting the valves into the chamber

we are working on, we have effectivelycovered the seats and protected themfrom accidental damage.

If you look closely at Fig. 15 show-ing the chamber modifications, youwill see that although the chamber

outline over its greater part remainsunaltered, the walls of the chambersrequire fairly extensive work to bedone on them. Basically what we are

trying to achieve is a chamber wallhaving no sudden changes of form orany corners formed between angledand vertical sections of the chamber.The area of the chamber wall aroundthe spark plug, should not be touchedexcept for polishing. Although a lineis used on the drawing to depict theextremities of this area, a sharp edge

should not be formed at this line.Machining the walls of these cham-bers requires a good eye to blend thechanging curved section into thestraight walls. The crossed sectionsshould therefore be regarded as areasonable guide as to where the wallsection remains standard.

When all the grinding has beencompleted, the chambers should bepolished to give a smooth finish. It isquite unnecessary to polish until youhave a mirror finish as all the work indoing so will be completely wasted.The type of finish required is one thatis smooth to the toucha good com-parison for this would be a piece ofchina ware.

This now leaves us with the ports

to tackle, and we will deal with the in-let ports first. At this stage one mustdecide whether or not radius valveseats are to be used. Let me warnyou, it is a long and tedious task do-ing radius valve seats unless you arethe proud possessor of some sophis-ticated valve seat machining gear.Doing radius valve seats by more

menial methods can take 12-15 hourswork, and sometimes longer. The endresult of such labour is 2-4 b.h.p.more, depending on the state of tuneof the engine, than conventionalseats.


38/159

38


39/159

39


40/159


41/159


42/159

42

bar. Distribute the wear on the bar bymoving the tool bit up and down thelength of the bar. Whilst accuratelyforming this radius, be sure to hold totool bit at a slight angle, approximately100, to the bar to give the tool a cut-ting clearance. Once the tool hasbeen made up, set it in the tool holder

to the dimension shown in Fig. 21.Using a pillar drill set on a slow speedto avoid tool chatter, cut the seats.Now back to the -grinder and blendthe radius seat into the port as shownin Fig. 22. At this stage in the proceed-ings you will have to leave the headand turn your attention to the inletvalves. The idea here is to reshapethe valves to assist our gas flow. It

may seem strange at first sight, buttulip valves do not always give thebest gas flow characteristics. To getbetter cylinder filling we must reducethe tulip shape to something morealong the lines of the head of a nail.Fig. 23 shows the shape we should

aim for. Except for the actual seat it-self this shape applies to both the con-ventional valve seat and the radiusvalve seat.

Lapping in the valves

Once the valves have been re-

shaped, we can start the job of lap-ping them onto our radius seats. Themethod of lapping and correcting anyerrors of concentricity are as follows.Using fine grinding paste, lap thevalve into its seat for a short periodi.e. a couple of minutes. Wipe thegrinding paste from both the valve andseat and apply a thin smear of engi-neers blue to the valve seat. Drop the

valve back onto its seat and rotate it.If we now remove the valve, the engi-neers blue will show where the valveis making contact with the seat. If theseat cutting operation was really ac-curate then the seat will exhibit a com


43/159

43


44/159

44


45/159

45


46/159

46

plete circular line. In all probability theseat will be slightly out of true and thiswill be shown up by a blue line over

only part of its circumference. In sucha case the seat will have to be dressedwith the cutting tool. Using the radiuscutting tool by hand, lightly cut the partof the seat showing the blue line. Hav-ing done so, lightly relap the valve andcheck it again with the engineersblue. Continue the lapping, cuttingand blueing until you achieve a com-plete circle on the seat, and a seatwidth of about 0.030".

The conventional 450 angle valveseats are lapped in the normal man-ner. It is a good idea to check the va-lidity of the lapping with engineersblue to make sure the job is up toscratch.

Skirnming the head

The final machining operation is toskim the head to achieve the desiredcompression ratio. With the Triumphheads we usually have a fair marginof metal that can be taken off, thereby

giving us any reasonable C.R. that wewant. For road use, a 10/1 C.R. isplenty and for racing, about 11.5 willdo the trick. It is not easy to quote theexact amount one should removefrom the head to achieve these C.R.,but the amounts will be about 0.070"and 0.085' respectively. The best way

to determine what should come off thehead is to fill the chambers with par-affin or petrol from a burette. If we fillthe chambers with fluid to the volumerequired for a particular C.R., then theamount of metal left proud of the fluid

is the amount to remove. Here are twoexamples showing the volumes re-quired for the chambers on the 1600

and 2-litre engines.

C.R.= (V+C) / C where V = sweptvolume of cylinder in c.c. and C =total chamber volume in c.c.

For the 1600 engine 1600V = 1600 / 6 = 266 c.c.

Therefore C.R. = (266 + C) / C

Therefore C.R. - 1 =-266 / C

For a C. R. of 10 to 1 we have:10 - 1 = 266 / C

Therefore:

C = 266 / 9 which = 29.5c.c.

There is, however, approximately4.5 c.c. in the gasket and block whenthe piston is at T.D.C., so this mustbe subtracted from our 29.5 c.c. togive us the required volume in thehead. Our head volume therefore is

29.5-4.5 25 c.c.This volume will give us a 10 to 1

C.R. on a 1600 engine. An 11.5/1 C.R.will require a head volume of 20.8. Onthe 2000 c.c. engines to get 10 to 1or 11.5/1 we will require 32.5 and 27.2respectively.

This leaves us only the assembly

of the head to do apart from any portmatching operations that may be re-quired. The methods to adopt for portmatching are examined in a laterchapter as is the choice of valvesprings for a given application.


47/159

47

FOR an even greater increase inpower, the use of oversize inlet valvesshould be contemplated. The use ofa big valve head will come into its own

when other components directly re-lated to engine power have seen at-tention. For this reason we will findthat the use of larger valves is a wasteof time unless the carburation andvalve timing have been improvedupon i.e. a hotter cam and better thanstandard carburation set-up. If theengine is to receive considerable at-tention in such quarters, then the useof a big valve head is to be recom-mended. It does, however, entail alittle more than just dropping in thelarger valves. As with the Stage 1head, we have the alternative of ei-ther conventional seats or radiusseats. The choice is yours, but for thehighest power the radius seats mustbe considered the first choice.

Inlet valves

When fitting a larger inlet valve boththe inlet port and the combustionchamber are modified to a slightly dif-ferent shape to those used for the

chapter 5

BIGGER VALVES FOR EARLY HEADS

Stage 1 head. This, of course, makesthe assumption that you are using avalve about 1 /10" larger in diameter.Let us assume at this point, that the

valve to be used is 1.395" diameter,which is, incidentally, the one whichis supplied by S.A.H. Accessories ofLeighton Buzzard. The procedure toadopt to modify the head will only bediscussed regarding those pointswhich differ from the Stage 1 head.

First off, we will need a perspextemplate for the big valve chambershape which can be traced directlyfrom Fig. 24. The procedure from hereon to complete the chambers is asfor the Stage 1 head in Chapter 4.

We can now turn our attention tothe inlet port. With the guides out, pro-file the inlet port to within I/. of thevalve seat as shown in Fig. 24. Whenthe port has been finishpolished to thisstage, fit the guides which, inciden-

tally, should have been previouslymodded as in Fig. 18. A tool will berequired to bore the valve throats outto take the larger valves. If we are touse 451 conventional seats, then atool bit


48/159

48


49/159

49

fitted into the tool holder as shown inFig. 21. should be used. The settingdimension (A) will be 1.030. The toolsetting to bore the correct size for a1.395" diameter valve is shown in Fig.25. This will give us a throat diameterimmediately under the valve of 1.310"diameter. Boring out the throat alsomeans we stand a very good chanceof completely removing the small stepwhich occurs just under the valve seaton the majority of heads. When theport has been bored to a depth of 1/8"to 3/16" blend out any irregularity withthe finished part of the port. Cut thevalve seat and lap in the valves. Nowcomes a tricky bit; put a small radiuson the inner edge of the valve seatthus breaking the sharp cornerformed by the seat at the port (Fig.26). In doing so, this should reduceour seat width to about 0.030'.

Radiusing seats

For those of you who are braveenough to tackle radius seats, you willneed the tool holder and radius cut-ting tool shown in Fig. 21. For thisparticular application, the radius of thecutting tool should be 0.140" insteadof the 0.125" shown in the drawing.Our setting dimension will also be dif-ferent.

Before you remove the inlet guidesfrom the head to grind the ports, theradius seats should be roughed out.The setting dimension for roughingout the seats, dimension A in Fig. 21,will be 1.003. Once the seats havebeen roughed out, the guides can beremoved and the ports ground to therequired form as in Fig. 27 and pol-ished. New modified guides are thenfitted and the seats finished off. To doso, the radius seat cutter, dimensionA, must be reset to 1.01 3" and care-fully recut. The finishing procedure em


50/159

50


51/159

51


52/159

52

ployed on the seats is then as detailedin Chapter 4.

The grinding of the exhaust portsand the raising of the C/R remains the

same as explained previously. Withthe big valve head it will be found thata little more is required to be skimmedfrom the head than for a Stage 1 headbecause of the greater amount ofmetal removed from the chamber.The extra metal to be removed isroughly going to amount to 7-10 thou.but this figure is intended as a guideonly. The only sure way to get the C.R.

to the required level is to use the bu-rette and paraffin to measure the vol-umes and determine the correct

amount to take off.The use of this large inlet valve type

of head gives a power increase ofaround 7% over the standard valve

size modified head. It also raises thepoint at which peak power occurs byabout 200-250 r.p.m. on the 2000 c.c.engines and about 300-350 on 1600engines. If used on a 1600 engine,the rest of the engine should bebrought to a fairly high state of tuneto bring it in line with the characteris-tics of the head. Even on a 2000 c.c.engine other modifications will need

to be done to make full use of such ahead.


53/159

53

THIS chapter applies to the headsused on the T.R.5 and 6, the 2000

saloon from 1966 onwards and theMk. 11 versions of the 2-litre Vitesseand G.T.6. Comparisons of Figs. 15and 16 with Figs. 28, 29 and 31 willshow the basic differences in theseheads. The later head is generally ofa superior design to the early one. Theports have a more gently curved con-tour to the valve than the earlier head,thus aiding gas flow. The inlet port

terminates at a valve some 0.144' big-ger than its predecessor which,coupled with the better port, shows asignificant advantage in power. Thecombustion chamber in the later headis also changed for the better and toget the best from it without completelyredesigning, there is little we can doto improve matters. A glance at Fig.28 will show just how little work is re-

quired on the chambers. In plan viewthe chamber shape remains un-changed. The sum total of the grind-ing required is all performed on thechamber walls. The end result of themodifications to the chambers shouldbe a chamber wall which changes

chapter 6

MODIFYING THE LATER2000cc AND 2500cc HEADS

smoothly from a vertical to an angledwall, leaving no projecting edges in

the chamber. This tidying up of thechamber shape gives us improvedcombustion efficiency but the gainsachieved are only marginal. No doubtif we compared the difference inpower gain by the slight change inprofile, the net result may only be inthe order of 1 A-2%. Since every en-gine tuner is after every little drop ofpower within a given budget, it can

be argued that although the gain issmall, the modification is justifiable.It is surely this type of attitude towardsthe preparation of a motor that makesthe true enthusiast stand out from therest of the crowd.

Porting

Anyway, back to the subject in hand

which is now the ports. These requirelittle more than polishing. Do not, un-der any circumstances, open themup, as if anything, they are slightly toolarge now. As with the earlier type ofhead, we have the presence of thesmall step just


54/159

54


55/159

55


56/159

56

under the valve seat, the completeblending out of which is a little diffi-cult. On some heads it will blend out,but on others it will not. It just dependson the casting Another point which isa little undesirable but one we can donext to nothing about, is the pocketaround the valve guide. It could betermed a negative boss in as muchas it is a boss recessed in rather thanone projecting out. The presence ofthis inset boss means that we cannotget a smoothly contoured port in thearea either side of the guide and im-mediately behind it. From our point ofview, the presence of this depressionis a little annoying, but the TriumphEngineers in their wisdom, did not putit there for nothing. Consider the factthat the guide is a press fit in the head.If one side of the hole into which the

guide is fitted is substantially longerthan the other side, the forces pre-vailing from the press fit impart un-even loads from one side of the guideto the other. This means that the guideitself becomes bent. Also the machin-ing of an accurate guide location holeis nearly impossible by productionmethods if it breaks through on an

uneven surface. That explains whythe depression is there but providesabsolutely no remedy for it. On occa-sions attempts have been made to fillsuch depressions with plastic metal,but the practice seems to be a littleunreliable. Should the plastic metalbreak away, it will cause serious dam-age to the valve and possibly the pis-

ton in that particular cylinder. Alterna-tively one could have the depressionfilled with weld to reduce any adverseeffects it may have on gas flow. Hereagain, this is quite a tricky job as thewhole head will have to be heated to

a relatively high temperature to avoiddistortion. The answer to all this in-decision is to leave the depression asit is unless you are bent on gettingevery little bit of power possible. Oncethe inlet port has been fully blendedand polished as in Fig. 29 we can fitthe inlet guide modified as in Fig. 30and cut the valve seat. Any sharpedges left by the valve seat cuttingoperation should be blended out byradiusing the edges. The sharp edgesof a seat can have a significant effecton the gas flow into the cylinder, es-pecially during the period when thevalve is only just off the seat. Thecareful blending of a valve seat intothe walls of the port and the roof ofthe chamber can have the same ef-fect as a cam with a faster lifting rate.The use of a cam with a higher flank

acceleration rate involves higherstresses in the valve gear, whereasthe seat blending has no mechanicaldrawbacksassuming the seat widthhas not been reduced to a point whereit can no longer sustain the loads im-posed by the valve.

Rather more work is required on theexhaust ports than that needed on the

inlet. The amount of grinding requiredthough, is still a lot less than neededon most heads. Grind the exhaust portto the contours shown in Fig. 31 andavoid taking metal from the tight cor-ner opposite the valve guide.

One should avoid making the ex-haust port any larger than is neces-sary to achieve a good finish. Polish-

ing is an utter waste of timea goodsmooth ground finish will do the trick.With the guides machined and fitted,as in Fig. 32, cut the seat and removeany sharp corners formed. When re-moving the corners,


57/159


58/159

58


59/159

59

job to be done this has one other dis-tinct advantage, this being that we canaccurately form the area just belowthe valve seat. The most critical pointson any head on which one is attempt-ing to substantially improve breathing,is the area 1/2" before and 1/2" afterthe valve seat. since, for our pur-poses, there is a distinct lack of metal

just before the seat, at a point we re-quire some, the obvious answer is toput some there, hence the seat insert.To bore the head to take the insertsyou will have to call on the services ofa motor machine shop, and if you lackfacilities, he can also make up and fitthe inserts.

As you have probably guessed bynow, the inserts are not going to be

just a plain straightforward ring set intothe head. Fig. 33 shows the profileset on the insert and one should at-tempt to meet the specified dimen-

sions as close as possible. The reallycritical one is the diameter of the ra-dius upon which the valve seats. If thisdimension is out by more than 0.005'(5 thou.) the valve will not seat in itscorrect position. Before fitting the in-serts or in fact, machining the headto take them, the inlet port should befinished within 3/8" of the valve seatand the guides fitted. It is also of para-mount importance that you haveguides whose bores are on or verynear bottom limit (0.312 dia.). Thesize of the guides needs to be accu-rate for the pilot of the counterboremachine to locate when machiningthe head for the inserts. To get thevalve seat of the inserts accurate, inrelation to the guide, we must ensurethat both the location diameter and thevalve seat of the insert are truly con-centric. Also the counterboring of thehead


60/159

60


61/159

61


62/159


63/159

63

THE T.R.4 head, from the home-tuners point of view, is one that isfairly easy to modify. The head is big

enough to make just about every partof the port easily accessible. The in-let port is big enough to practicallyclimb into, and the same goes for theexhaust ports to a lesser extent. It isunfortunate from the power outputpoint of view, that the inlet ports are alittle too big, but there is very little thatwe can do about that. A partial solu-tion to the problem is to avoid taking

any more metal out of the ports thanis required to give a good smooth fin-ish. As with the other Triumph heads,we are plagued by the small stepwhich inevitably exists just under thevalve seat. 50% of the time this stepwill blend out when the ports are modi-fied as in Fig. 35. If the step is greaterthan 60 or 70 thou., then you just haveto live with it, or a small amount of it,

even after grinding. Basically the workon the inlet port can be summed upas a little reshaping and blending anda lot of polishing to get a smooth uni-form finish. We do not want a finishakin to chrome plating, as this will onlyserve to exaggerate the fact that the

port is too big. It will also serve to en-courage fuel separation from the airbecause of the lower gas speed pre-

vailing in an overly large port. The typeof finish we should go for would bemore along the lines of a vapourblasted or fine shot blasted finish. Infact if you have access to or knowsomeone who can fine shot blast theinlet ports after grinding and light pol-ishing, so much the better. If you havethe port shot blasted, then the last partof the port just under the valve seat

will need to be polished. The lengthto be polished will be about 1/4" fromthe valve seat down.

After fitting the guide, cut the seatand remove any sharp edges from theseat by radiusing off. In the interestsof retaining a valve seat of reason-able width i.e. about 30 thou., any ra-dius applied to the sharp edges of theseat should be small, in the order of

40 thou. or so.

Chambers

As can be seen from Fig. 36, thereis a fair old chunk of cast iron to betaken out of the chambers. To

chapter 7

MODIFYING THE TR4 HEAD


64/159

64


65/159

65


66/159

66


67/159

67

enable reasonable uniformity ofshape from chamber to chamber oneshould make up a template fromwhich the marking out of the new

shape should be done. As describedin previous Chapters, the templateshould be made from perspex for anumber of reasons, primarily though,because it is easier to make and usethan most other materials. Whenmarking out the chamber, we docome across a slight snag. Part of thesquish area on these heads is not flat.

Since our template is flat and will beon the head face, we will find that mostof the surface on which we are mark-ing is falling away from the template.This means that you will have to useyour skill to judge by eye the line youare marking directly under the edgeof the template. After marking out thechamber, it is just a case of hogging

out the cast iron as shown in the draw-ing. One thing that is a little difficult toshow in the drawing is the fact thatthe undercut in the chamber wall,shown dotted, blends in smoothly withthe newly ground shape.

Ports

Inspection of Fig. 37 will show thatthere is not too much metal to comeout of the exhaust ports. Unlike mostof the other ports on Triumph engines,this one is best ground with the valveguide in place. This then brings thevalve guide flush to the roof of theport. This practice of bringing guidesflush to the port especially on the ex-

haust side is one which should not bepractised without due consideration ofthe consequences. The shortening ofthe exhaust guide is usually accom-panied by a rise in exhaust valve tem-perature. In this particular case the

shortening of the guide is smallenough to have a negligible effect onthe exhaust valve temperature. Ashas been said many times before,

there is no point in polishing the ex-haust port as the first few minutesrunning will completely coke up themost perfect polish. A good lump andbump free ground finish will suffice.

Apart from removing any suddenchanges in section like the bump atthe point the stem grinding termi-nates, there is little valve reshaping

to do. So long as the contour changeon the back face of the valve issmooth then the valve will perform itssealing function whilst presenting theleast resistance to flow. As with theseat in the head, we should blend theunder head face of the valve smoothlyinto the seat to increase the valve ef-fectiveness at low lifts.

It is now head skimming time. Solong as we bear a few points in mind,we can get the C/R to a reasonablyhigh level. For a road car 10 to 1should be considered about as highas we want to go. For racing, we canup this to 11 to 1 since the intervalsbetween decokes are usually shorter.When skimming the head, we must

avoid making the thickness betweenthe head face and the undercut in thechamber adjacent to the inlet valvetoo thin. If we do make it too thin, thiscan lead to several undesirable con-sequences.

1. If it overlaps the bore slightly, itmay cause a hot spot, leading to deto-nation.

2. If it does not overlap the bore butinstead rests on the gasket, the pres-sure of the gasket caused by tighten-ing down the head may be just suffi-cient to cause the edge to crack andbreak away. This means


68/159

68

we have a small but very danger-ous piece of cast iron floating aroundin the relevant cylinder.

3. If the area of the head face di-

rectly over the pocket has been ma-chined excessively, then when thehead is fitted we will find that part ofthe gasket is exposed. Numerousblown gaskets will be the result of this.

To get a 10 to 1 ratio we will need achamber volume of 55.5 c.c. Since we

have approximately 6 c.c. in the blockand gasket, the volume in the headwill need to be 49.5 c.c. For 11 to 1this figure will need to be 46 c.c. To

get these ratios the head when modi-fied as in Fig. 36 will need about 70-80 thou. off. When the head has beenskimmed, remove the sharp edgesfrom the chamber and it is ready toassemble.


69/159

69

CAN we say that any of the heads sofar described for the six cylinder en-gines is the ultimate within our limita-

tions, imposed by using, as a startingpoint, a standard casting? Would thehottest head as described in Chapter12 be good enough to produce thepower required by our hypotheticalVitesse racer mentioned in the intro-duction? In all probability it would falla little short of the mark, but this ismerely incidental.Do we need bigger valves? The an-

swer to this one is no. The valve areawe have, should suffice up to about8000 r.p.m. This means that any de-ficiency in breathing is not due to thevalve sizes employed. The ports cer-tainly are not too small for the size ofvalves used. If this is the case thenwe are left with only three areas inwhich drastic improvements can bemade. These being the chamber,

valve seat and the entry direction ofthe port. If we can make substantialimprovements at these points, thenwe could be well on the way to achiev-ing that 195-200 b.h.p. that is neededto make a sure race-winner

Let us look at tile chamber design

chapter 8

THE ULTIMATE HEAD

first, and consider what happenswhen we open the valve into a cham-ber with a flat roof and walls well clear

of the valve.To do this we will look at tile effectof fluids in the port instead of air, sincefluids exhibit the same qualities asgases except for compressibility. Fig.38 (a) shows part of a port, a valveand a chamber. The fluid in the portis coloured differently from that whichis already in the chamber thus en-abling us to see the flow pattern when

the valve is opened. Drawing (b) inFig. 38 shows the flow pattern whenthe valve is opened, and this is char-acteristic of the flow pattern through-out the opening and closing cycle. Itcan be seen that turbulence startsright on the edge of the valve seat.This has the effect of impeding theflow of gas just before the seat andthe turbulence itself is absorbing valu-

able kinetic energy which otherwisecould be assisting flow into the cylin-ders.

Radius valve seats, can these help?The answer to that must be a con-servative yes. By using radius


70/159

70

valve seats we can delay the onset ofdrastically turbulent flow while thevalve is at low lifts. The radius seatscan actually increase gas flow into the

cylinders by as much as 20% at lowvalve lifts. As the valve lift gets higher,so the effectiveness of the radius seatdrops off. By the time full lift isreached, the improvement gained bysuch seats is down to about 3-5%.This is by no means a small gainwhen you consider it has beenbrought about solely by a little reshap-

ing. The basic problem remains as tohow to reduce turbulence as much aspossible, at least until the incominggases are well and truly in the cylin-der. Fig. 39 shows a chamber wallsection which goes a long way to solv-ing the problem. Virtually half the tur-bulence is eradicated until the basesare well within the cylinder. Tests on

such a wall contour indicate gains asmuch as 6-8% on a normal engine

when compared with a flat chamberroof. Add this to the gains achievedby venturi ports and radius seats andwe could well be 11 % up on a con-

ventional headed engine.

Down-draughting

Down-draughting the inlet portshould also produce a worthwhile in-crease in power. This could be in theorder of 3-4% when used with suit-able carburation or fuel injection. If we

combine all these ideas, the headshould look like Fig. 40.

So far everything looks quite goodon paper, but what about the problemsof making such a head. Down-draughting the head is reasonablystraightforward for any well equippedtuning establishment. It is just a caseof setting up the head in the correct

position and boring a large hole in it.Having done this, a suitable size of


71/159

71

pipe together with some sealing com-pound is press fitted into the hole. Thevalve guide locating bore isremachined to remove the obstruc-

tion at one end, caused by fitting thepipe, a bit of port blending is done andwe have a down-draught head. Do-ing the venturi type radius valve seatsis straightforward enough even if it isa delicate operation. This just leavesus with the chambers.

To get the chambers to the shapewe require, it is obvious that metal will

have to be put into the chamberswhich is, to say the least, a bit beyondthe scope of most home tuners. For-tunately there are firms who special-ize in welding cast iron so the prob-lem is simply alleviated. To fill thechambers, the whole head has to beheated red hot so that stress cracksdue to temperature differentials are

avoided. A quick look at the relevantdrawing will reveal that it is not nec-

essary to completely fill the chambersbut only build up the areas around andbetween the valves. We will obviouslyhave to build up more than we want

to so that machining and grinding canbe done to bring it back to the correctsize. Another problem is going to con-front us in as much as filling the cham-bers is going to raise the compres-sion ratio a little higher than is reallydesirable. The answer to this is to usethe head from the 2.5 litre enginewhich is identical in every respect bar

chamber depth. Having filled andmodified the chamber, we will find thatthe C/R is at a reasonable level andcan be brought up to spec. by a littlehead skimming. With this type ofchamber design C/R in the order of12.5 to 1 would be beneficial.

With such a head, probably the bestsequence in which to modify the head

is: (a) weld up the


72/159

72


73/159

73


74/159

74

chambers, (b) fit valve seat inserts,(c) down-draught the head and, (d)grind and polish ports and chambers.

Using this type of chamber design

together with a down-draught port, aslightly tulip shaped inlet valve is likelyto give us better results. The shapingof the inlet valve can have a greaterinfluence on gas flow than is com-monly suspected. Tests have indi-cated that there is a family relation-ship between chamber design andvalve head profile. This relationshipcan be used only to give us a guide to

the valve profile as it is a little toovague to give us any concrete figureson which to base the valve profile.

This head design should go a long

way to giving us the power output re-quired of our hypothetical 2-litre Tri-umph racing machine. By itself, it isprobably not enough to make a cer-tain race winner. There are fortunatelyother areas in which substantial im-provements can be made and weshall discuss these in the relevantChapters.


75/159

75

TO many people it seems that portmatching is something of a mystery.It is in fact quite a simple job that

needs just a little care and patience.The simplest method, ensuring areasonable match between the portand manifold is to match both itemsto the manifold gasket. To do this, boththe manifold head joint faces arecleaned and marking blue applied toeach face. Two manifold studs arethen fitted into the head, one at eachend, and the manifold gasket located

on them. We then scribe round theport apertures onto the head thus giv-ing us a clear indication of the posi-tion the gasket will take up. Some-times the gaskets have large clear-ance holes for the studs, which meansthat the gasket is rather a sloppy fiton the studs. This can be remediedby winding a few turns of maskingtape around each of the studs, thus

giving us a more stable location.Once the head has been markedout, we set about doing the samething to the manifold. By using somenuts and bolts the same size as thesecuring studs or bolts in the head,the gasket can be secured to the

chapter 9

PORT MATCHING AND DOWELLING

manifold and the port aperturesmarked out. From here it is a straight-forward job to cut away the metal on

both the manifold and gasket up tothe previously scribed line.We have, of course, made the as-

sumption that the ports in the gasketsare the same size as the ports re-quired in the head. Most manifoldgaskets are about 3/32" larger thanthe port sizes which in all the caseswe have covered remain standardsize as far as the inlet are concerned.

This means that ideally the line thatis scribed for working to must be 3/64in from the edge of the gasket. Theport size at the joint face between themanifold and head is not critical within1/64" so when we say the port shouldnot be enlarged, it should not be en-larged any more than is necessary tomatch it up. This method of portmatching is the simplest but the ac-

curacy obtained is limited by clear-ance on the manifold location studs.This clearance allows a certainamount of movement which directlyaffects the lining up of the ports. Al-though


76/159

76

this method is not super-accurate, itdoes leave things far better than theyare as standard.

Dowelling

For a super-accurate port match-ing job, we have to be a little moresophisticated about things. Firstly weshall require some 1/8 dowel pinsThese can be made up from a lengthof 1/8 dia. silver steel as this is veryaccurately sized. Silver steel inci-dently, is available through most con-cerns dealing in engineers tools andis commonly sold in 13" lengths. Fromthe silver steel we make up somedowel pins about 3/8 - 7/16 long.These should then have their endschamfered or radiused. The number

of dowels required will depend on thetype of manifold being fitted. A one-piece manifold will require two dow-els; a two-piece manifold will requirefour dowels; each part of the mani-fold requiring two dowels to locate it.

The next step is to fit the manifoldStuds in the head and bolt the mani-

fold in place with two or three gas-kets between the head and manifold.Be sure to line all the gaskets up withone another. To take up any play onthe studs we can, as before, usemasking tape. With the manifoldbolted securely in place, drill two holesin a convenient position in ea

Documents

Tuning Standard Triumphs Vizard