Forum » General Engine Building Discussion » Building high Compression little engine

Building high Compression little engine

General Engine Building Discussion

Forum Posts



Tech Articles

Talk about engine building here. New products, tricky questions or showcase your work - If it's engine building related it's welcome here.

= Resolved threads

Page 1


I have issue to making power from a the Peugeot TU5 1.6 liter 16v N/A engine with following setup:

Increased Compression ratio from 9.2 to 13.2 by high compression pistons and cylinder head skimming

Rod to stroke ratio from 1.62 to 1.7, 82mm stroke and 139mm rod length

Piston bore from 78.5 to 80mm, Piston compression height 29.5mm

Valve lift from 8mm to 10mm and 272 degree camshaft

Converted valve buckets to solid shim

OE valve springs and valves, Valve head diameters Intake 31mm, Exhaust 24.5mm

1 kg of Crankshaft wight reduced

light weight flywheel

Exhaust headers

OE 125cc injectors, Lambda is around 0.8 at full throttle

1 step colder plugs

I have checked the cylinder compression and it is 275 Psi

The ECU is stock Bosch ME7.4.5 and by reducing 10 degree of the ignition advance with 92 octane pump fuel no knock detected

I don't have access to the dyno to measure the power at the wheel, The engine revving good but the 0-100 km/h time is about 9 sec. and when I driving it I feel It have to much powerful.

What can be wrong with the setup?

92 octane with that nominal compression ratio? Even using the US rating, it seems rather low for that setup and I would have expected some detonation. That last part 'may' be part of the problem as most (all?) modern engines have detonation sensors as part of their management protocol and there may be in-audible detonation triggering the ECU to pull timing until it is removed, sometimes it can be in substantial steps like 10 degrees at a time, so your "10 degrees" is potentially rather more than that? You may be able to read the ECU logger and check that - but an 'old school' check would be to use a healthy dose of a GOOD octane booster (check on-line to compare with what is best and available to you) and see if that makes a difference - if so, it'd suggest timing was being pulled and you need to address it.

That is a massive amount of cylinder compression - is it a typo'?

If not, might be an indication of incorrectly set camshaft timing as, even with that compression ratio, it is damned high. I'd double check that anyway, as a tooth out can make a big difference - part of the checking is verifying the true TDC position.

Hi Gord, Thanks for your attention and answer

I will use octane booster and setup ignition table again and will check detonation with Plex knock monitor, If knock detected I will use Q16 to test the performance.

I pulled 10 degrees of timing at low rpm and about 6 degree at high rpm, tested less advance and the engine revving slow.

That engine have knock sensor but I could't find a complete and useful logger for this ECU it is 1998 design.

Suggested lift at TDC is 0.9mm for both Intake and exhaust and I set lift@ TDC to 1mm intake and 0.9 mm exhaust, checked TDC by dial indicator, Will check it again, I want to change lift @ TDC to see how it will effect cylinder compression and engine performance, Do you have any suggestion about that?

0.1mm, 4 thou', seems negligible, but that may be several degrees variation from spec' and that can make a difference to how the engine runs. I'd suggest correcting that while you're checking things.

It is a common practice to check for TDC using a DTI down the 'plug hole - were you using a degree wheel and bringing the piston UP to the same point both ways and halving the difference? I've known a few people just thinking 'that's close enough, and making an error of a few degrees. Sorry if it's granny-eggs, but doesn't hurt to check.

Was there a checking valve clearance given - assuming mechanical, or are the camshafts hydraulically operated - that can also make a difference for timing by lift?

Depending on spec', from what I have been able to find out those engines should be up to ~125hp stock - with your modifications it should be a fair bit more and good for well under 9s for the 100kph sprint - even if you're really fat ;-)

OH, just noticed the lambda is 0.8 - that corresponds to ~11.76:1 which is a bit on the rich side. For a NA engine I'd suggest between 0.85 and 0.9 which are 12.5:1 to ~13.2:1, which are around 'best power', depending on fuel and engine characteristics - you shouldn't need to go below 0.85.

Lift at TDC is 40 thou equals 1mm and OE lift at TDC is 16 thou

It will be sweet to make more power by correcting the errors :D

Is it good to increase exhaust lift at TDC and overlap? How much lift at TDC and overlap will be good for my build? Can it be calculated?

Valve to piston clearance is safe 2.5mm at intake and 3mm at exhaust

According to Peugeot specs TU5JP4 making 109hp at engine.

I have checked some stock cars at same model and production lambda is 0.85 at high rpm and load

Our arithmatic concurs, which is good, I was referring to the variation from spec' for the cam as that can be important! With cam' timing, you should start with what the camshaft grinder recommends, then you can look at swinging them a little, either way, to see if it improves what YOU have for your complete setup - you have some working clearance, which is nice. You would be amazed at what difference even 4 degrees can make to an engine - I know of one instance where 7 degrees made ~35% difference in power, but that was a specific example.

I would recommend using a degree wheel for setup, though, if possible, as it's generally easier and more accurate.

There is a big variation in the "TU5" family of engines, from 89 to 123 hp, why I said "up to ~125hp" - I used it for a base figure you be a fair bit higher than ;-)

Back in the day, the 'best power ratio was said to be 13.2:1, but nowadays it seems to be, IIRC, around 12.6:1, which is why I used that spread for the lambda - something else to experiment with.

Reviewing your posts, you didn't mention the induction side and, on that topic, silly question but I've seen it overlooked - have you confirmed you're actually getting full throttle?

Oh, and you can confirm 275PSI cranking, as that is VERY high?

Good news is that I could identify the problems and solve all the issues.

Checked the shim clearance and it was't done correctly by machine shop, Clearance was 0.85mm at intake and 0.9mm at exhaust valves, This time I do the shims clearance and perfectly set it to 0.23mm for intake and 0.33mm for the exhaust.

After fixing shims clearance tried some different cam timings and get best result by 1.15mm intake lift and 1.25mm exhaust lift @ TDC.

Swapped injectors with better quality set of Bosch injectors that was used on older version of TU5 engines with 25cc more flow.

Checked intake manifold for air leak and changed the seals.

I found that MAP sensor connector had problem so changed it too.

The result was great and car's performance improved significantly and it runs much smoother, I could do better tune with nearly no knock at high rpm with premium gas and set lambda to 0.85 at wot high load and rpm

Do a drag test with a light weight Mazda 1 that has BP engine equipped with itb and standalone ECU which was about 140HP at wheel and easily passed it at strip by nice distance.

Bad news is that the customer has crashed the car, It's not a big crash and nobody get hurt but I think it will take about a month to get it back to the road.

Oh, About the cranking compression after running car for about 2000KM I do the test again and it's about 275PSI and going to more if doesn't stop cranking. The video is attached

Attached Files

Excellent result - apart from the bingle.

Cranking pressure is huge - don't recall anything close to that, even on some very high compression engines! Tried with a different gauge?