Squish and Turbocharging

[Alan Whitaker]

Hi Guy,
How important is squish on a turbo charged engine, is there a maximum squish distance from the head to the piston.
I have s squish of 0.100 inch, I will be running a boost of 16 psi, compression is 9.5:1.

Many thanks for all your help and excellent advice

Alan

[Guy Croft]

A brief explanation of this 'squish' effect for those who don't know about it.

Controlled turbulence in the cylinder during the upper part of the compression phase encourages the development of a homogenised mixture - in other words air and gasoline droplets are mixed thoroughly to give fast and complete burn when ignition takes place. I say ‹Å“controlled¢ž¢ because too much, or so I understand, can make the engine 'rough', though I am not sure how intrusive or damaging this phenomenon is. Noisy in a road car is probably close enough.

Squish is radially-inward motion of the charge (to concentrate it) in the spark plug region - and it aids good combustion on spark ignition engines. Apart from good burn a well-known benefit of squish-induced turbulence is its beneficial effect on reducing the tendency to 'knock' - detonate. This is because the radial-inflow helps purge pockets of end gas. There are other ways to get good mixing - using ramps, offsets and deflectors in the inlet port to get the air to flow in a controlled way round the cylinder wall as it enters, and a thing called 'tumble' - I think invented by Ricardo some 10 years ago but which I know nothing about. (The trouble with some of the port-induced effects, as far as I¢ž¢m concerned, is that whilst I know they make the engine more economical, I don¢ž¢t know what the net effect of removing some of the features will be. Will the less-effective burn impact more significantly on power than the reduced volumetric efficiency that we get from leaving them in? Difficult one that..)

An engine with closer proximity of crown to head in the squish-band area will have more squish than another - and assuming the motors are identical in other respects, it will be less prone to detonation, so can tolerate, say, higher boost on the same fuel. That said, a well set-up competition engine won¢ž¢t detonate anyway - unless you let its operating regime drift outside the acceptable envelope, ie: running it over-hot. Squish bands ¢‚¬Å“ built in for that purpose ¢‚¬Å“ are there to protect road cars where the manufacturer hopes to offer the widest possible operating safety margin.

How squish works (in basic terms) is this. The scraping of the piston rings against the bore creates high speed vortices during compression, and near tdc the close proximity of piston crown to opposing flat 'squish' bands on the head face forms powerful jets of mixture that travel in towards the centre of the cylinder. Squish-induced turbulence is called just 'squish' for brevity. The amount of squish can be increase by reducing that proximity or increasing the squish bands - areas of opposing flat surface areas of piston and head. Because this is all to do with good combustion it's worth pointing out that compared with a carburetted engine a fuel injected one already has far better mixture quality because of the finer atomisation of the gasoline particles. I know that say, on units with individual throttle bodies, injecting onto the throttle plate also yields an additional power benefit by adding some turbulent mixing to the fuel stream.

Turbocharged engines generate far higher inlet valve exit flow velocities than normally-aspirated ones; this in itself aids mixing. Turbocharged engines only have squish bands to stop them detonating. The 16v Integrale has no bigger band than my full-spec Vauxhall XE heads, produces any bhp you care to name with same and doesn't detonate if calibrate it right and look after it. The 8V Lancia Integrale has bigger squish bands than any engine I¢ž¢ve seen, and I have never known one detonate, great. But, on the flowbench that head doesn¢ž¢t flow well at all. Does it matter? Not as much as it would on a n/a engine. As an n/a head it¢ž¢s hopeless and those huge squish bands have got to go.

High speed n/a competition engines don't need as much squish as low speed ones. That's a fact but very hard to quantify. What it really means is that if you're building a race engine with a powerband 5000 + you might as well ignore squish. The head will have an amount determined by the manufacturer and you certainly will not want more, and maybe, with 2 valves per cylinder you more-often-than-not want to get rid of some of it to deshroud the valves. Give me flow against squish any day. Remember I said that.

Regarding how much squish you can build into an engine, the first thing to note is that you've got to have a reliable working clearance between the piston and the head, to allow for stretch, expansion and natural 'take-up' of running clearances. I reckon the closest you ever want to be is 1mm, thinner than most gaskets, so in effect you'd have to run the piston crown out of the block to achieve it. I have run as little as 0.5mm. You'll find out if your drive for maximum squish has sacrificed reliability when the piston starts hitting the head at high rpm.

Squish is extensively modelled at industry level and is indeed one of the hardest things to determine, how wide the clearance can be before it starts to impact on power I wouldn¢ž¢t care to say, but bear in mind that older engines like the Jaguar 6 cylinder had no squish at all and they ran pretty well even if they were notoriously ‹Å“thirsty¢ž¢.

Do you want an engine that will never knock (even if you run it on the lowest grade gasoline you can find and on the hottest day of the year) or one that flies? That¢ž¢s the call you have to make.


GC

[Alan Whitaker]

Hi Guy
What a powerful read, I would like to add to my first post.

I have a set of R&R rods which along with the TRW forged pistons that should be 9.5:1 compression with a squish of 0.040; I should have asked for a compression height of 1.660, I have sent the order for a compression height of 1.560.

This is my fault so I must replace the pistons with the correct compression height or replace the rods; the cost of pistons is less than the rods.

I do not want to build the engine up just to take it apart to replace pistons along with the build up costs if there are goint to be problems.

Thanks for a informed read

Alan

[Guy Croft]

Alan, hi

Easily done I can tell you! But rods short by 0.1" is no big deal. I don't know what the motor is but instead of getting more pistons suppose you just mill the block down 2mm or so? I assume the valve reliefs are designed for the piston crown flush with deck? Running the piston low will have the effect of making them 'bigger' than they need to be.

Other considerations in milling the block are:

1. Gasket locating dowel counterbore will need deepening
2. Head bolts may 'bottom out'
3. Tensioner may need some er, alteration
4. Adjustable camwheels (depending on engine) will be needed.
5. The block deck needs to be thick enough (all the Fiat TC 8v and TC are).

GC

[Alan Whitaker]

Hi Guy
Some good ideas, I will look at all the options you have kindly suggested

I have wills rings in the block so I can machine down 0.090 but would have to loose the rings as I don¢ž¢t think I can go another 0.060 to put the ring groves back in
The best option is to machine the block and obtain the best gasket available to me.

Thank you for your kind suggestions and help

Alan