One of the reasons I purchased the PFC and Datalogit was to be able to change when the stwitch occurs to parallel turbo operation. Once you have several mods to increase flow (and hence VE), the stock change-over point works less well as the primary turbo runs out of steam sooner and restricts the otherwise well flowing system.

Datalogit allows you to do this so I gave it a try last weekend and here are my results and analysis. In the software, you can change it according to TPS value. Since WOT is all I care about I only changed the top setting (60%). There are two RPM settings, I had hoped the first one affected prespool of the seconday turbo and the second one affected change-over. Looks like that's not the case.

First graph is the stock settings, and the following ones are two other settings I tried (titles are the RPM values, WG duty in blue, PC duty in red, boost in green). Boost was set at 1.0 kg/cm^2. A value of 0 means the WG/PC is open, 245 means it's closed. These were WOT runs in third gear from about 3K RPM.

Analysis: During prespool, WG and PC close at 3200 RPM regardless of settings, then PC opens at 3350 RPM regardless of settings while WG is closed. PC gradually closes until change-over at which time PC opens completely and WG opens some. Change-over point is affected by the settings: it occurs about 250 RPM AFTER the second RPM value. It does not appear that the first RPM setting changes anything but I'd need to do more testing, varying only it to verify this.

With earlier change-over, there is significantly greater drop in boost at the change-over point. I assume this just means that the secondary did not have time to adequately prespool. I don't think starting the test with a lower RPM pull would have made any difference since prespool always begins at the same time. Note that boost recovers to 20000 at about 4600 RPM regardless of settings due to the greater dip. Even though boost was lower, the car felt subjectively faster during this time since it was in parallel mode.

Also worth noting it that WG is closed loop. There is always a WG spike in response and porportional to a boost dip (even though it's not effective). Also interesting is that WG gradually closes as RPM increases to maintain boost. May reflect that I'm out of stock turbo efficiency range.

For now, I'm sticking with the 2400/3600 setting. I was hoping to be able to prespool it earlier to get more parallel advantage earlier without losing sequential performance at low RPM but looks like that's not possible.

Hope these graphs show up, if not they'll follow in a reply post.




Alan

 

You can also change the pill size in the PC line beween the turbo outlet and the PC actuator. A larger size will also allow faster pressure buildup and open the PC gate sooner. This should help your prespool.

 

Yeah, that's my plan, if I don't just throw in the towel and go non-sequential.

Alan

 

Great information, thanks for putting the time in to post it!

Have you tried a higher boost setting on the first turbo, and a higher base duty %? That might produce more pre-spool. Also, unless you are lugging the engine, the theory is that you can take a little more boost on the low end. The stock settings are higher on the first turbo than the second.

just a thought.

 

One other thought - is the first number the transition point back to Primary turbo? That might make sense if it isn't the prespool trigger. I wonder of the documentation has anything to say about this....

If that's the case, setting it to 2400 or thereabouts will keep the car in twin mode a lot more - meaing you can find yourself trying to spool up both turbos from <3000 rpm.

 

I'll quit one of these days.

One last thing to consider when moving the turbo transition point. Note that in the timing maps there are some anomolies in the curves at the transition point. I'm not sure why they are there, but if you move the transition point, then the timing maps may be trying to comepnsate for the transition after it has already happened, and not compensating for it when it does happen.

 

All good thoughts, thanks. I haven't really looked at on-off throttle data to see when/where it switches back to sequential. No, you're looking at the documentation.

Alan