EWG - Boost Creep! Help Please!
 

Good morning fellow Miata friends.

Little run down of spec to give you an idea.

Fully forged engine
STD Cams
Head has been modified quite heavily.
+1mm exhaust valve.
15 thou head skim.
GTX28567R
T25 Plus 63ar turbine housing. from ATP. Flows like a T3. So I guess similar to a t25 0.86ar
3" full system
FM manifold
Turbosmart Ultragate 38mm EWG

Photo of turbo setup.

https://www.miataturbo.net/attachmen...ine=1458560833


Right.. down to the problem.

EWG has a 1bar spring. Which it was hitting fine every time.

Went down to the dyno and got it tuned. Which we did at 1bar boost.

We decided to up the boost to around 19/20psi as map sensor is only 21psi.

We had a misfire so we changed the plugs to a cooler grade 8 instead of 7. Car then over boosted way above 21psi and popped a intercooler pipe off.

So we turned off the EBC and ran it up and it was hitting 18/19psi. on spring pressure. We refitted the originally plugs but with a smaller gap. Exactly the same problem!?

I've checked everything. Vac lines. changed vac lines. checked EWG hasn't got any problems. it opens as it should.

It's only starts to creep above around 5500 up to 19psi.

I removed one of the springs and it also did the same but to around 12psi on a 7psi spring.

Since playing around last weekend and checking all pipes. it's now hitting 1.5bar which is off my map sensor so I'm running out of fuel.

The diaphragm isn't ripped either.


Does anyone have any ideas. I'm really stuck on what to try. Turbosmart have no idea either.

Could I have a faulty EWG?

How would a spark plug change effect it? As it wasn't misfiring at 1bar previous.

Chuck ideas at me please!

What's the exposed port for on the turbosmart ewg? Just curious.

The Turbosmart has two ports so that you can put boost on either side of the diaphragm. This makes EBC work better with a low pressure spring, because you can use boost pressure to prevent it from getting blown open by the exhaust manifold pressure. You don't need to use this port, you can leave it open to the atmosphere and just run it like a single-port wastegate if you want to.

If the car is making more boost on the same spring then either the turbo's efficiency has gone up (unlikely) or something is restricting flow in the external wastegate. Unless there's something really strange going on further down in the pipe where they join, it has to be a malfunctioning EWG or boost signal supply. Perhaps it's sticking? Is the EBC solenoid still installed? You might try removing that from circuit entirely, even if it's powered down it could be cracked and bleeding off pressure.

--Ian

Did you try removing the EBC all together? Going from WG straight to vacuum source? Could be a failing solenoid.

Connecting the wastegate to a vacuum source is a great way to kill the diaphram in the gate. Don't do that.

OP- hook bottom port to boost pipe, top port open, and make a pull to redline. If you stay at spring pressure no problem, the gate is fine.

Didn't jump a tooth on the exhaust cam did you?

Don't expect spring "ratings" to mean anything. If your 7psi spring gives you 12psi, it's a 12psi spring on your setup.

Thanks for the replies guys.


I will check if it has slipped a tooth. What would this cause?



It's a screamer pipe so does not go back into the system.

I have hooked it up without the EBC altogether for testing. Same problem.

Last night I did some more playing around and it's over boosting to 252kpa thats 7psi more than the spring pressure. Which is more than my map sensor is even reading.

The odd thing is it had held spring pressure before fine. But could of been due to slight weak spark and after sorting that problem out it's made another problem.

I've tested the EWG and it opens as it should when hooked up to a air line.

But it seems to be getting worst! But it is getting warmer here so maybe the weather is effecting the boost 2-3psi.


I have attached a few logs along with my map so you can have a look.

its just after the event in the log.


Sixshooter. The old thing is below 5500/5800rpm it will hold spring pressure.

Im going to ask the wrong question..

Is the creep an actual reliability or driveability issue?

Its probably not bad.

It's fine and not a problem.

But it's a problem when your 2 bar map sensor can't read any higher so you run out of fuel.


But otherwise it drives lovely.


Just I can't find the reason behind the EWG not controlling the boost at a solid pressure.

So does it ACTUALLY run out of fuel?

If its VE then it will just run at the VE on the highest row, and typically at 20+psi ve is dropping off, so if anything it should be getting fatter.

Dann

It's not as such running out of fuel. I will re word it.

It's hit the top of the map so...

2 options (we will ignore the boost creep for the moment)

Up the VE values to allow fuelling above the map sensor limits.

Or fit 3 bar map sensor and allow myself the res to input the correct fueling for set kpa.

Which I do have and will wire up.

But it doesn't explain why I'm boost creeping 7+ PSI.

What im saying is that you dont need to up the VE values.

VE works in such a way that you dont, because total turbo/cylinder head efficiency is dropping as boost rises at 20+psi, so despite VE not going up with boost, the car WILL get richer up there anyway.

You need a better ported manifold, better flowing screamer pipe or a smaller spring, or free-er flowing vacuum tubes to the gate.

Dann

Trying to run the system above the limits of the MAP sensor is a stupid idea, IMHO, and likely to break the motor.

The spring values are nominal, the precise boost that you get with them depends on the efficiency of the turbo and the relative flow levels of the turbine and wastegate/pipe. For my FM2R system, a "7 psi" spring gave me 6, two "7 psi" springs gave me 12, and a "7" and a "3" gives me 9.

If the boost level is consistent (i.e., with the same tune you always get the same boost level to within a psi or so) then I'd say what happened is that having tuned the car you're now flowing more exhaust energy and the minimum boost level with the spring you have has gone up. OTOH, if the boost level *isn't* consistent, then you have some kind of mechanical/pneumatic problem with the boost control system. You need to figure out which of those two it is, but if it's boosting past your MAP sensor then that's not safe to do with it as-is. You could fix that by going with a bigger MAP sensor, or you could change the EWG spring.

The 38mm ultra gate looks like it works similarly to the the 45mm one FM sold me, in that it can have up to three different springs inside it. Did you install the spring in it yourself, or did you buy it as a "1 bar" unit? If the latter, then you might want to open it up and check what springs are actually in there (they have paint marks on them and the color tells you what the spring rates are).

IMHO, the right approach is to put in a smaller spring (aim for something around 7 or 8 psi) and see if the system behaves consistently. If so, then just find a spring combo that gives you the base boost you want, if it's still inconsistent at that point then you know you need to hunt down a boost signal leak or mechanical failure in the EWG.

--Ian

Take out the spring and do a pull. Does it creep to the same boost pressure? Then you need a large wastegate. Does it maintain a low (or no) boost? Then your boost reference is bad.

It's possible that you were operating right on the verge of the wastegates limits, and new plugs plus a more efficient tune put you over the edge.

Thanks for the input guys.

I just got a 10 - 3 - 5 psi springs turn up so I have something to play with.

I also have a 3 BAR map sensor to fit so I can at least be able to correctly control the fuel and ignition in the higher PSI areas.



the EWG has 2x 7psi springs inside of it. But as you have mentioned having a correct spark and a good tune looks like it has made flow a lot more for the EWG to control at stated spring PSI.

So really I need to forget the PSI rating of the spring and set the combo of springs that gets me to the set PSI I'm after rather than fitting the spring that is rated to my PSI goal.


Plan is to fit 10psi spring and see how it acts and use the EBC to move it up to my goal.


I will test without any springs to see if it makes any boost at all. But I suspect it won't make much. As with the 7PSI spring to topped out at around 10 psi at redline.

I've also been reading that the boost source needs to be the correctly sized vacuum pipe.

so I've ordered up 5mm barb and some 5mm lines to match the 5mm barb on the EWG.

As I'm running around 3mm at the moment. I wouldn't of thought it make any difference if I'm honest. But it's the recommended size from turbosmart so can't do any harm and it's cost peanuts to change.


Another thing I may do is port the manifold EWG hole. As the metal is over a inch thick. So could be worth smoothing and rounding the port in the inside to help promote flow directly to the EWG to control boost.

Will update with my findings.

Why did you go with a 38mm EWG anyway? That seems smaller than desirable.

--Ian

I brought the manifold second hand.

Could fit a bigger one but not sure about the stud holes.

Otherwise option is fit V-Band but have a bolt on bracket and mill out the hole bigger.

https://www.miataturbo.net/attachmen...ine=1458827255

The 38 mm guy should be big enough, in my view.

If the spring is removed and it does not overboost, then clearly, the gate is fundamentally big enough.

If it does OB, the stroke of the valve could be the problem. It then needs to be determined why the stroke is restricted.

Lack of stroke could be limited by the diaphragm, the spring or hitting the cap. If the diaphragm is the problem, change the static position of the dfram relative to the valve with a spacer between the base and the dfram. This will incsease the up stroke by the thickness of the spacer.

If the valve stroke is limited by interference at the top, again, fit a spacer in the same place.

Check coil bind. If that limits the valve stoke, fit a spring with fewer coils, or thinner wire. (reduce the coil diameter to get the rate back)

Wastegates always work better as the spring gets longer with a lower spring rate accompanied by a higher preload. If a 1.0 spacer can fit, try stretching the spring to match.

A higher preload is always desired, so don't let that be reduced.

I would be surprised if any manufacturer tests a gate to 20 psi. A really proper 20+ psi design will have such a long spring as to make the cap look ridiculous. It would resemble an old TV cone head.

corky