Pro-drive EBCS - w or w/o pill?

[Deer Killer]

What did you finally get up to?

 

Seems I can run a good deal more boost but who knows how much is too much for the bottom end. According to Rao my engine is dead anyway, might as well find out how far it goes.

[LittleBlueGT]

What did you finally get up to?

 

 

Tuning 101, blow up your own engine, you might learn a lot!

 

I got up to 26 psi tapering to 21 psi, but I have a feeling it was only 24.5 tapering to 19.5 if the OEM MAP is right.

 

I am currently running 22.5 tapering to 19 psi (w/ alky injection:wub:), as measured by the OEM MAP.

[SeeeeeYa]

Tuning 101, blow up your own engine, you might learn a lot!

 

I got up to 26 psi tapering to 21 psi, but I have a feeling it was only 24.5 tapering to 19.5 if the OEM MAP is right.

 

I am currently running 22.5 tapering to 19 psi (w/ alky injection:wub:), as measured by the OEM MAP.

 

Hmmmmm..... verrrry interesting. That is exactly what my AVO380 alky tune runs. Seemed a sweet spot somehow.

[Deer Killer]

I was thinking 22 to 19 psi would be a good idea without meth. But then again I blew a ******* PCV hose out from under the intercooler and freaked out. Those things don't have barbs or clamps! WTF?? Good thing it wasn't the FPR hose. Instant death, I would imagine.

[mickeyd2005]

The compressor map is scaled correctly. I double checked the scale and plopped in 300 g/s. It's the same.

 

The spreadsheet ASSUMES a VE to generate the rpm plots. You have to enter in your own data for VE or just plot your datalogged CFM data.

 

The collection of compressor maps is only used to compare one versus the other.

 

FYI - the original VE estimate was based on a bone stock LGT.

[Deer Killer]

eh I don't know why you go through all these conversion since we meter in mass anyway. But for my real data to match that map, I'd need a VE of 93% at 6700rpm. :-/ But the CFM to lb/min conversion doesn't match the AVO map using standard temp and pressure anyway.

[mickeyd2005]

Did you read my statement?

 

It is for COMPARISON.

 

The compressor maps that I found all used different X-scales so I picked the one that came up the most frequently.

 

I don't know what you mean by it doesn't match. I checked it and it matches.

 

Also, it doesn't account for pressure drops across intake and intercooler.

[Deer Killer]

the last point on the chart I posted is 6700rpm, if that makes sense to you, *shrug*

[LittleBlueGT]

Here is just a real quick plot using DK's formula. It makes sense to me.

 

http://i248.photobucket.com/albums/gg194/littlebluegt/pr380.jpg

[Deer Killer]

confirms what I thought before. The more boost I add at the top end, the more efficient it gets.. I'm in the 68% island with 17psi.

[LittleBlueGT]

confirms what I thought before. The more boost I add at the top end, the more efficient it gets.. I'm in the 68% island with 17psi.

 

Who knows how the smaller hotside affects back pressure?

 

Either way I think you are fine running higher then boost as long as your fuel can take it.

[Deer Killer]

But what does that really mean? If it's exhaust gas dilution then you need more timing and less WGDC until it's not worth it anymore. IMHO it seems that even the avo380 has a bigger turbine scroll that most Subaru turbo's. So this should not be that much of an issue.

 

You can make that compressor map into a total system efficiency map by changing the boost curve in 1psi increments both ways and then overlaying the plots. The optimum point (s) should become apparent. Should be easy for you with street tuner.

 

93 octane FTW. But you can run a crapload of boost as long as you retard enough timing or blow something. Finding the optimal points without a load-based dyno is extremely hard if you don't have a good starting point.

[LittleBlueGT]

But what does that really mean? If it's exhaust gas dilution then you need more timing and less WGDC until it's not worth it anymore. IMHO it seems that even the avo380 has a bigger turbine scroll that most Subaru turbo's. So this should not be that much of an issue.

 

You can make that compressor map into a total system efficiency map by changing the boost curve in 1psi increments both ways and then overlaying the plots. The optimum point (s) should become apparent. Should be easy for you with street tuner.

 

93 octane FTW. But you can run a crapload of boost as long as you retard enough timing or blow something. Finding the optimal points without a load-based dyno is extremely hard if you don't have a good starting point.

 

What I am saying is I don't know how far you can push the comp chart. If this turbo was in fact a GT turbo, we would be able to (on race gas or alky) run very close to edge of the chart and see gains all the way.

 

I don't think that is the case with the AVO380 because of the hotside.

 

I am interested to see how well your 93 octane gas does as you increase the boost. I am cheering for you.

[Deer Killer]

What I am saying is I don't know how far you can push the comp chart. If this turbo was in fact a GT turbo, we would be able to (on race gas or alky) run very close to edge of the chart and see gains all the way.

 

I don't think that is the case with the AVO380 because of the hotside.

 

I am interested to see how well your 93 octane gas does as you increase the boost. I am cheering for you.

 

I'm saying you should be able to plot this and see that you're approaching the edge. Very time consuming and difficult without realtime tuning.

 

All I keep thinking of is the sound of a piston rod leaving the engine bay at a high rate of speed

[SeeeeeYa]

I'm saying you should be able to plot this and see that you're approaching the edge. Very time consuming and difficult without realtime tuning.

 

All I keep thinking of is the sound of a piston rod leaving the engine bay at a high rate of speed

 

RPM, not boost, related, oil problems absent.

[Deer Killer]

That was not a serious comment.

[Deer Killer]

back to the original topic - looks like you might not be alone. Your mods might make the issue show up more. Basically it seems the mechanical response of the wastegate maybe slower than the response of the turbo. Or it could be that the path through the prodrive is very narrow. Doesn't completely explain this behavior. I'm going to remove the prodrive and put the direct connection between the compressor and wastegate back, and see what it does.

 

This makes it very hard to tune boost for dynamic conditions. Could be perfect on a 2nd, 3rd, 4th, 5th gear run. Go WOT above 4k and you're in trouble.

 

Graph below: I'm expecting ~14psi across the board, no more at least!

[Deer Killer]

My conclusion: mechanical problem.

Pics, after I tried to fix it:

1. The angle in the linkage is >90*, this means that as the door opens and it nears the right angle, it's going to slow down and possibly bind a bit. I believe this is the hump in the avo380 dyno graphs, not the boost leak.

2. You can see that the actuator does indeed fit between the oil line a turbo, and could (and needs to) drop down even further. Take a look at how far away the original mounting hole was.

3. You can also see my previous hack to slot out the plate, because otherwise the actuator wouldn't clear the oil line. Also the top hole is enlarged for more play again to fit the oil line.

 

The previous combined angle was enough to cause some binding at the linkage, not to mention creating a hump, which only got worse with more preload. Boost curve is much smoother now! Still could be better though.

[LittleBlueGT]

My conclusion: mechanical problem.

Pics, after I tried to fix it:

1. The angle in the linkage is >90*, this means that as the door opens and it nears the right angle, it's going to slow down and possibly bind a bit. I believe this is the hump in the avo380 dyno graphs, not the boost leak. What boost leak are you referring to? My Bump was not due to a boost leak.

2. You can see that the actuator does indeed fit between the oil line a turbo, and could (and needs to) drop down even further. Take a look at how far away the original mounting hole was.

3. You can also see my previous hack to slot out the plate, because otherwise the actuator wouldn't clear the oil line. Also the top hole is enlarged for more play again to fit the oil line.

 

The previous combined angle was enough to cause some binding at the linkage, not to mention creating a hump, which only got worse with more preload. Boost curve is much smoother now! Still could be better though.

 

Thanks for the pics.

[Deer Killer]

You said your original compressor cover had an extra hole in it.

[LittleBlueGT]

You said your original compressor cover had an extra hole in it.

 

Yea it did.

 

My boost bump was after I had the leak fixed.