Reeeeeeally weird boost/waste gate control issue
 

Ok, so car specs

1993 1.8, 99 head with vics (operating)
Logish manifold with directed flow/ full 3" exhaust
Subaru Impreza STI TD05 turbo (oem internally gated)
Evo 610cc injectors
Walbro pump
Evo 6 intercooler (2.5" piping)
Ms3 managment


Ok, had car at tuners today for mapping, had massive boost creep, ended up removing wastegate arm and jamming open the wastegate valve, even with it like this, the car was still producing waaaaaaaaay too much boost (200kpa 15psi) by 4500rpm and this was obviously causing the ms to boost cut, and was impossible to finish the mapping.

Things we've tried:

Wiring open the wastegate fully.

A divorced setup downpipe, 3" main with 18" long 40mm wastegate pipe and a 3" bellmouth type downpipe. (Both downpipes have same results)

Inspecting the turbo, everything seems to be in order, wastegate works correctly and wastegate hole and port are not blocked in any way.

The tuners are very experienced having tuned lots of vehicles over a ten year period, and they've never seen this issue and have no clue as to what the cause is.

We've scoured the interwebs, the only reasonable answer we can possibly find is that the exhaust is "too free flowing" (which I find hard to accept) and is just bypassing the wastegate port, thus causing no boost control as the exhaust gases are all just rushing round the exhaust wheel. I think this is unlikely, but I'm happy to be told otherwise!

Any ideas from the MT geniuses?

sti turbo is a vf made by ihi, not a "td0" anything. so which turbo do you have?
many of them creep

What they told you is right. it happens often on low elevation sti's in dense/cold climates
I wouldnt' expect it to creep on a miata, but I guess yours does
the wg needs porting, or ideally just run an EWG

Pretty much. It's called boost creep, and you have it in spades. When the velocity of the air through the turbine housing is high enough that the air isn't able to make the turn into the wastegate path, you'll get excess air through the turbine wheel and boost will be virtually uncontrollable.

The fix is a restricted exhaust (ew), an external wastegate on your current manifold, or upgrading to a manifold/turbo combo that has a properly-designed internal wastegate that doesn't exhibit this issue. Our new turbo kits use the EFR6258 and they're proven to exhibit zero boost creep, even with a full 3" exhaust at boost levels down to ~6psi.

Many thanks for your reply and apologies for my slow return to reply.

Ref the TD05 business being STI, I'm In the UK and we have a lot of imports, the cars were badged differently. I bought the turbo from a guy who said it was on an STI model, but I spoke with a Subaru tuner yesterday and they said it was probably from a higher power spec wrx that was essentially an STI, but not quite.

Any hoo, you're right (obvs) it's not an STI turbo, it's a TD05 from a WRX 270. Like you suggest, we'll port the wastage port area and try again. I'll document the process, and this may serve someone for the future as the td05 turbos are cheap as chips and seem to be good for 300whp. Time will tell!

Many thanks 👍

You're welcome.

And keep in mind, opening up the PATH to the port is just as important as enlarging the port itself.

Good luck.

Thursday experiment just causes more brain bafflement ��

Ok so had a spare hour or so tonight, Swapped the turbo for a Wrx td04, took it out for a spin, less midrange boost but STILL produced a bar of boost even with the wastegate wired open.

The confusing thing for me is that on the last setup, with the 2.5" exhaust, that same exact turbo would hardly produce any positive boost AT ALL (wastegate wired open).

Can the new 3" exhaust really flow so much more that it's causing these boost control problems!?

Going to fit a stock oem exhaust tomorrow and see what happens. I'm more interested to understand the problem rather than just engineer around it with an external gate etc.

Cheers, Dan

2.5 to 3 inch exhaust is a 44% increase in cross sectional area. That's a lot.

Imagine that the engine is putting out exhaust gas one molecule at a time. Each molecule arrives in the turbo and needs to pick a path to take, either through the turbine or through the wastegate. It's going to pick the one that's least congested, where the two things that determine congestion are the geometry of the path and the number of other molecules currently in that path. If it goes through the turbine it will increase the boost, if it goes through the WG, it won't.

If you make the exhaust freer-flowing, you decrease the pressure at the turbine exit, and that makes the turbine path less congested. Thus, a freer-flowing exhaust tends to lead to boost creep for the same size wastegate. To fix it, you need to bring the flow characteristics of the two paths back into balance, which you can either do by restricting the turbine path, or opening up the WG path. The boost control system does the latter, by opening up the WG flapper further, but once it's all the way open there's nothing else it can do with the current physical geometry.

If you restrict the turbine path, you hurt the spool of the turbo, whereas opening up the WG path flow really has no downsides. Thus, fixing the wastegate is a better solution, albeit probably a more expensive one. You can do this either by porting the internal gate or using an external one.

--Ian

My TD04 runs 5psi at redline with WG wired open. Segregated downpipe into 3" exhaust with a cat. Log manifold.

external gate is actually fixing the problem
all the other stuff is the engineering around it

A properly designed/engineered internal gate is arguably superior to an external gate, since the IWG separates the wastegate diaphragm from the heat of the exhaust manifold. It's easier to place an external gate in the correct place, which is why most people do it this way

Look at the BorgWarner EFRs for examples of properly-designed internal wastegates.

Aha! You see, this is my point about understanding the problem. I fully understand how to fix it, I just don't really understand why makinsons setup doesn't produce much boost, yet mine is producing a bar of boost.

Could it be down to the vics manifold?

Another swerve ball question, The actual boost gauge and pressure feed to the Ms is taken from the top rear of the vics type inlet manifold, Is there a chance that there's a freaky pressure thing going on (to do with vics) that is actually causing an erroneous false high pressure reading from the port? Or should I just shut up, and put my head up my arse? 🙈

I'm going to move the pressure feed later anyway, just for my piece of mind and test drive it, Then start porting.

Many thanks again, Dan.

No freaky pressure will be found.

I've seen no wastegate actuator failures thus far.

In my view a properly designed internal gate does not exist. Good one's that work well do, but still not proper. I'll join the argument for remote gates anytime.

Porting the exhaust housing, increasing the exit port size and decreasing the valve size only goes so far in banning the creeps. Plus, juggling the port size and the valve size can create problems getting the valve to sit properly everytime.

A further problem with trying to fix an integral is created when you do so once, then add some flow improvement to the system and you find the same problem all over again.

It is relatively clear, in my view, that fixing it once and for all will improve the level of fun you get from driving the vehicle.

If converted to a remote, be a bit careful to avoid duplicating the problems of the integral gate by just slapping it onto a tube at a right angle.

If paranoia sets in, as is often the case, use both gates.

corky

1) completely different turbo's
2) nope
3) you can keep hypothesizing, but the problem is very simple and straightforward

By the way, I've still not heard of any wrx's running a "td0..." anything. Are you able to get the actual specifications of the turbo? Or at least exactly which car/model/year it's from?

We can't help you if you can't provide at least the basic information.

Here's a link to the euro Impreza lineup of turbos

AndyForrestPerformance

ahh, 93-96 non us non-wrx's
thanks for the clarification, that's really interesting
very little info on those cars floating around