compressor surge under full throttle..
Turbo Mini Forums

I am experiencing something odd. if i load the engien up in 4th or 5th i will get full boost(say .8 bar) by approx 5000 rpm this will hold flat til 6000 rpm, then it will spike to anything between1.3 and 1.5 bar and very quicly settle again to the regulated boost level all the way to the soft cut limiter at 7900.

even with the boost controller disconected and the wastegate fed straight from the outlet of the compressor housing this occurs.

this has jsut happened as i have changed cylidner heads after suffering a burnt out exhaust valve., the lower half of the engien and the cams inlet exhaust etc remain the same.

would i be right in thinking i am experiencing compressor surge? it is hard to hear any audiable signs of surge, as the car is quite noisy and frantic when the spike occurs.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

Surge woul;d do the oposite. Pressure would be seen to drop then rise.

You have to understad what surge is. Surge is where the difference beetween compressor inlet pressure and compressor discharge pressure, excedes the compresssors ability, tthe discharge gasses then backflow through the compressor untill the compressor recovers, then it starts all over again. This is pressure/ impellor speed related. The faster the impellor spins the higher the presssure differential will be before surge, then you are restricted by the mechanical limits of the unit itself.

Baisicaly surge happens at high boost, low compressor speed.

I would have thought high pressures were due to tthe wastegate more than anything

Thatts my thoughts anyway

Sounds more like a boost control issue.

Does the actuator work properly ? not sticking at all ?
How do you bleed off, to achieve higher boost ?

Why did it burn an exhaust valve ?

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

sticking wastegate spindle when it gets hot ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I agree about it being a boost controll issue but i was starting to doubt myself:

first off i had a good old bleed valve between the throttle body (side that doesnt see vacuum) and the actuator. this required quite a legnth of hose but worked fine for me with the old head.

then i installed an AVCR at the same time as i swapped the heads. however i have tried bypassing the avcr and have also now moved teh pressure feed from the throttle body to the turbo outlet pipe, reducing the hose legnth to about a foot all in at the most. i know an excessively long hose to the actuator can cause spikes as can any restriction in teh hose (avcr solenoid) so removign these was the first thing i did.

next up i started on the actuator, pressure check and the diafram isnt ruptured. the arm on the turbo moves througha full 90 degrees without stickign or any oddness when feeding the actuator with a foot pump. i tried ajsuting the preload lighter and heavier (crack at .5 and .9 bar), ddint seem to make any difference to the spike. its always there.

exhaut valve burnt coz im a twat stevie :) when i was opening out the chambers on that head i was a little too enthusiastic near that exhaust seat. but decided to just run it anyways as i didnt expect the engine to hold together for anything more than a few weeks anyway. as it happens, it lasted quite some time.

this spike is doing my head in though been at it 2 days solid now. cheers for prompting tho gents, will take a wee step back and think if i have missed anythign obvious on teh boost controll side of things :(

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

here she is...



know its a bit unclear, yellow = RPM
light blue = Manifold pressure in kpa (absolute)
Magenta = injetor duty cycle.

spike was actually quite tame in that run. thats with only the actuator connected nowt else :)

Edited by evolotion on 19th May, 2008.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

How about ignition advance ? Backing-off ignition too much makes the opposite of the expected: it creates after-burning condition (ALS like) and makes the pressure rise.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

did you try the waste gate spindle movement with it hot, evo ? is there an internal faliure of the pipe to the wastegate letting it collapse in a sneaky way ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

The fact that the boost soon goes back to where it should be would indicate that the wastegate is sticking.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

alpa, the boost rise results in a very pronounced increase in power. i do map my car on teh road rather than the rollers, so i know my timign is quite conservative, i will try throwing in some advance anyway. this cylidner head has given me slightly lower compression ratio than the last one so a re-tune is on the cards, but i wanted to sort this boost issue first. for obvious reasons!

Robert, tried it when stinking hot, replaced the entire actuator feed pipe and tried feeding the actuator from just before the throttle and just at the outlet of the turbo!

Paul, i agree, but it doesnt seem to stick:(

iv hung back in work to do some playign with it. my main thing is i dont consider the angle the wastegate moves through to be ideal for quick opening, if that makes sence. it does open fully, but not in teh most ideal way. going to see if i can bias it towards opening more quickly at first. going to compare it to a few turbo's kicking about too..

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

well its sorted... i had the actuator attached to one of those wee handheld pumps instead of my foot pump, so had finer controll over it. noticed it was a bit notchy, didnt notice this before. it would appear that the rod is hitting teh wee hole it passes through. to remove the head i have to remove the turbo, to do that i have to remove the acuator bracket (its very tight down there!) and there is enough slop in the mounting holes so it can be mounted touching or not! cheers for keeping me in that area, always the simple things.

Edited by evolotion on 19th May, 2008.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.