Compression testing

[Greenfield]

Some months ago I started a post about preignition problems with my B33. Just recently I fitted a decompression plate to see what the effect would be. I'm interested to see what the compression ratio is now, but can't get the Hawk compression tester to work as its impossible to kick over the engine without using the valve lifter as in starting. Am I attempting the impossible?
It works fine on my A50 and B40.

[Spaceman]

Although you say that you are interested to see what the 'compression ratio' is, I assume you meant you were interested in how many psi you have now got? In my experience, although it is easy to do a compression test on an engine with an electric start, I've never managed to do one successfully on an engine with a kickstarter. When I've attempted it, I've always ended up with a low pressure reading and assume it's because the engine has to be turning over at a minimum speed to generate a meaningful pressure. I'm sure you could use a roller starter in lieu of an electric starter to do a proper test?

[Greenfield]

Yes, trying to measure the compression (psi), so that knowing the current atmospheric pressure you can calculate the compression ratio. I'm just curious.

[DAVE BRADY]

Hi,

It can be done on a twin with the other plug out.  A few kicks is enough.  A bit more difficult on a single.  The first kick is usually disregarded anyway but if you can keep up the momentum of the flywheel and turn the engine over compression after letting the decompresser go you should get a reading.

Dave.

[Spaceman]

Yes, trying to measure the compression (psi), so that knowing the current atmospheric pressure you can calculate the compression ratio. I'm just curious.

You can work out mathematically what the CR now is from what it was before and from the thickness of the decompression plate - its all down to the change in geometry.  In fact, you would need to do a similar calculation trying the work out the CR from the pressure reading. Happy to help with the calculation!

[Greenfield]

Thanks for your input. I've given it another go, but you just can't get any momentum to push it over the top with any speed. I think that a roller starter would be the answer, but not available.
You never know what you're dealing with when you take on an old bike with minimal history. When I removed the head, I found an incorrect exhaust valve guide made of bronze that had been clearly sliding in the head. I've replaced this with steel which is solid. When I removed the barrel there was already a standard 65 thou decompression plate fitted with a single gasket. Did a previous owner try to solve the preignition? There was also a bottle of octane booster in the toolbox! I tracked down a 75thou decompression plate that I've fitted with a gasket both sides. I'm not a big bloke but the kickstart easily takes my weight, only slowly descending as air seeps past the rings. I found light scoring in the cylinder but there's certainly plenty of compression, and it isn't burning oil. I'll see how it goes on the road once the weather picks up.

[Spaceman]

I'm not an expert on the B33 but, if the standard CR is 6.8:1, using a 75 thou decompression plate will lower the compression ratio to 6.15:1. At that CR, I wouldn't have thought pre-ignition should be a problem but is there any evidence that this B33 is fitted with a high compression piston?

If the CR was 8.0:1, the decompression plate would reduce it to 7.1:1

If the CR was 9.0:1, the decompression plate would reduce it to 7.8:1

[Mike Farmer]

 

My thoughts: - Thereoretically the PSI in the stroke is only a "by product" of the compression.

The CR is the difference between the swept and the unswept volume of the cylinder. Not I THINK with the pressure.
If you put any plate in the cylinder, top or bottom, you increase the unswept volume by X. Thereby decreasing The CR

My very dated and somewhat suspect maths suggest, for instance let us say---You have inserted a 6mm spacer in a 70mm bore. (o.6cm x 7cm) you will have increased the unswept volume by approx 23cc (3.5 x 3.5x 3.142 x 0.6 =23.1cc).

So take the present swept Volume=stroke X piston radius squared X 3.142 that is the basis of the current CR. add the above figure.
and you now have the new unswept volume, thats the bit that piston top no longer reaches as it no longer travels as far up the bore. So the area above the piston has now increased and the relationship between the two volume swept/unswept factors has changed in direct proportion

I reckon its pretty difficult to do it any other way. I really cant see how its possible with just a pressure reading but as always I am happy to learn.

Mike

[Spaceman]

You have to rely on Boyle's law if you start with the pressure: P1 x V1 = P2 x V2. P1 would be atmospheric pressure, P2 would be the measured peak pressure, V1 is the total volume of the cylinder at BDC and V2 is the volume at TDC. If you know P1 and P2 you can then work out the CR = V1 / V2. The problem will be if you can't measure P2 because you can't spin the motor fast enough!

[Mike Farmer]

 

I see that but as you say getting an actual reading would not be easy.

Further to the actual problem my thoughts go along the lines that it is unlikely to be a CR problem cos that CR is how the engine was designed to run. Unless there are so many variables that its a guessing game, starting from a piece of hot "coke" stuck in the head and going north from there.

I would be seriously interested in the outcome.

Has your decompression plate cured it?

Mike

[Spaceman]

I would definitely check to see if the engine is fitted with a high compression piston raising the CR from 6.8:1 to probably 8.5:1. You should not get pre-ignition problems with 6.8:1 but 8.5:1 might make it susceptible to this type of problem. If it does have a 8.5:1 piston, the 75 thou (1.9 mm) compression plate should lower it to 7.5:1.

[Greenfield]

Thanks Spaceman. I was also thinking along those lines. Is there an obvious difference in the appearance of the high and low compression pistons. As far as I remember there was no stampings on the piston crown but it did have two 'cutouts' for the valves.

[Pete C]

here is the 60 oversize piston from my B33. No compression plate and no problems with preignition.
The valve cut-outs seem to be standard with a flat top to the piston. I would assume a higher comp pistom would be domed so that may be an easy identification method.

[Spaceman]

I believe the B33 in the picture and the B33 featured in this thread are both long conrod models (pre-1953?), in which case, they should look similar if a standard 6.8:1 CR piston is fitted. If the latter had a higher CR, the central part of the piston crown between the valve cut-outs should be raised.

For the long conrod models, the parts book does not list any higher compression pistons in contrast to the parts book for the later short conrod models that lists 3 (7.5:1, 8.5:1 and 11:1). I believe the pistons for the two lengths of conrod are not interchangeable and so I am not sure whether a high compression piston is available for the B33 in question? It would be interesting to know what is marked on the piston?

[Mike40M]

In my youth, riding JAP engined speedway bikes, we measured compression rate as part of the usual prerace service. Setting engine at TDC, then measure the amount of liquid needed to fill it through the spark plug hole. Then easy to calculate CR as swept volume plus amount needed divided by amount needed. Sadly all those years later forgotten what liquid we used. Something available, maybe methanol, kerosene or water. Guess petrol will work. Curiously enough I remember that 38cc gives 14:1 CR.