16 replies to this topic

[Henk Vasmel]

Don't believe too much in magic numbers. These days everybody has a phone with a calculator so "exact" capacities can be calculated. A lot of the magic numbers come from the days of the slide rule.

Could you tell me the metric capacity of the "289" engine?

I am currently working on cleaning up my database for engine capacities and searching for bore and stroke values for all kinds of engines. (This should be a separate thread, maybe I will start one soon), so I have seen a lot recently.

One of the things I found out is that some (mostly British and American) engines are designed to imperial dimensions, while others use the (much easier for me) SI units.

The "1558" engine appears to be an obvious case of imperial values. These do not use decimals, but fractions.

The bore is clear. 82.55 mm is 3.25 inch (exactly). But that is supposed to be 3 1/4", since fractions are used.

The stroke, however is not so clear. 72.746 is an odd number in millimeters, and 2.864 inch doesn't make sense either. It doesn't translate easily into fractions.

We have to go back to the original specifications to find out the real dimensions, and don't be surprised if "magic numbers" turn out to be slightly incorrect.

[Henk Vasmel]

John Aston replied to my post (when it was still in the jim-clark-museum-to-get-lotus-25-r6-on-loan thread)

 

4727cc - 289 ci  . As fitted in AC 289, Mustang and lots else .

[Charlieman]

0.864 is 7/8 minus 11 thousandths. So the Cortina stroke was a nominal two and seven eighths of an inch, less 11 thou for adjustment/bearing wear/thermal expansion.

[Henk Vasmel]

John Aston replied to my post (when it was still in the jim-clark-museum-to-get-lotus-25-r6-on-loan thread)

 

4727cc - 289 ci  . As fitted in AC 289, Mustang and lots else .

Yes, this is one example where "magic numbers" go wrong. 4727 is a very common capacity given for this engine, yet I have found what it really should have been. That number is also published many times, so I am not the only one in considering it incorrect.

The bore and stroke given for this engine is often 101.6 * 72.9. This corresponds to 4728 (not 4727). However, in inches, it is 4 * 2.87. 2.87 is very unlikely and is probably rounded from 2.875, which in fractions is 2 7/8, a very logical number for a stroke in inches. With that value, we end up at 4736, a number seen in several specifications.

All these values translate into 289 cu in, except for 4727, which is actually just 288.

[Henk Vasmel]

0.864 is 7/8 minus 11 thousandths. So the Cortina stroke was a nominal two and seven eighths of an inch, less 11 thou for adjustment/bearing wear/thermal expansion.

Thank you. This is good information. Usually I consider nominal values only, but couldn't make it work for the Cortina engine. Odd that nominal is in fractions while the adjustment is in thou. Difference between the design department and the guys in the workshop I suppose

[Ray Bell]

Oh?

A 'thou' is 1/1000 of an inch, definitely a fraction.

[Henk Vasmel]

Oh?

A 'thou' is 1/1000 of an inch, definitely a fraction.

[Lee Nicolle]

Thank you. This is good information. Usually I consider nominal values only, but couldn't make it work for the Cortina engine. Odd that nominal is in fractions while the adjustment is in thou. Difference between the design department and the guys in the workshop I suppose

11 thou for expansion? I think not, A race engine will have 1.5-2.5 thou oil clearance. The rod may make  1 or 2 thou extra though an engine should be measured cold.

The only thing that may make 11 though extra length is a nitro fuel dragtser engine.

As for the 289, at this point in history they are all overbored . Buying standard bore pistons is surprising hard and most 289s have a top and bottom ring mark where the piston rocks. And that is the reason oh so many have a pan full of piston skirts.

The only short stroke engine available is late factory 302s or aftermarket blocks. Which are better,, and not historic.And I expect in common use complete with 3" or more stroke

[dbltop]

Ford also used a 4.6 engine in the Mustangs and Crown Vics before the 5 liter was reintroduced. I'm guessing the 4.6 = 284 ci.

[Tim Murray]

Perhaps before this thread goes too much further we might start using the correct terminology - this is TNF, after all.

The cubic capacity of any internal combustion piston engine is the total volume available for the fuel-air mixture to occupy, which comprises the volume displaced by the pistons plus the volume of the combustion chambers. Thus the term used to refer to the volume defined by the bore and stroke should be displacement (or swept volume) not cubic capacity.

[Glengavel]

Unless you've got really odd-shaped bearings, which I think would cause all sorts of unwanted side-effects, the stroke depends purely on the throw of the crankshaft, does it not? So any intolerance is purely down to the manufacture and machining of the crankshaft?

[Michael Ferner]

In my experience, older US engines are always designed on binary fractions, i.e. 1/2, 1/4 etc. up to 1/64 - the famous 300 CID Mason-Duesenberg, for example, intriguingly had a 3 63/64 bore - everything else you find are figures rounded to decimals, for example 2.87 is actually 2 7/8 i.e. 2.875, I have even seen 3.8 for 3 3/4 i.e. 3.75! Roughly from the sixties onwards, some US designers use thous, making it more difficult to detect rounded figures. I am not so sure about British engines, I believe Sunbeam e.g. designed engines on the metric system even before WW1!?

[Allan Lupton]

Those cubic inch numbers are a bit coarse cf. cubic centimetres - there are 16.387 cc to the ci - so that 289 number can mean anything from 4727 cc to 4744.
Another potential muddying of the waters is the value used for "pi" - in the days before electronic calculators it was common to use 22/7 as a good approximation. That would cause nearly 2 cc difference in 4727.

 

ETA anent Michael's point, some UK engine sizes were metric - e.g. almost everyone who made a 1½ litre four cylinder used 69 mm × 100 mm

Edited by Allan Lupton, 23 October 2019 - 08:15.

[Tim Murray]

Another potential muddying of the waters is the value used for "pi" - in the days before electronic calculators it was common to use 22/7 as a good approximation. That would cause nearly 2 cc difference in 4727.

Indeed, and even the more accurate approximation of 3.142 that I was taught to use from (UK school) O level onward gives a difference of over 0.6 cc in 4727.

[Collombin]

it was common to use 22/7 as a good approximation

Or 355/113 for a very good approximation.

[Vitesse2]

I think Allan's probably more likely correct, given that in the days before pocket calculators there are many examples of British - and probably American - writers using 'divide by 8, multiply by 5' to convert km and km/h to miles and mph. A lot easier than dividing by 1.609344 when you're doing it on paper ...

[Henk Vasmel]

What's actually surprising is that so many old specifications turn out to be correct.