N13 Turbo - Round 2... Fight!

[Charlescrown]

That diagram is a std power/torque for a Camira and when I saw your VE table it looks either very rich down low and high or very lean in the mid range. I use the torque curve to plot my VE tables.

[vlad01]

The graphs on that site are just estimated based on assumptions and the engine specs. None of them are real. I've compared a few real to the ones on the site to confirm, and they were all way off.

[Charlescrown]

Looks pretty realistic to me from figures I found on other sites.
Certainly wouldn't get max torque @ 6000rpm which is where the max VE is showing.

[BennVenn]

Ahh I see what you mean.

So that table is Boost VE, from 100kpa (atmospheric) to 1Bar. It definitely doesn't reflect torque.

I just set the entire table to 95% VE (pig rich in most places) then run it up on the dyno or road tune, log what CalcVE says it should be, then go back and adjust to suit. I figure that is the safest way to approach a boost VE table, always on the rich side of safety.

As for that torque curve, I agree with vlad that it's an estimation and not at all based on a real dyno graph. I'll attach my stock NA dyno graph for comparison, stock memcal, stock tune, with 280,000kms on the motor.

stock.JPG (75.06 KiB) Viewed 2233 times

Ignore that small hump in the torque, i was still dialing in the PID for the dyno. Torque is almost dead flat from 1k to 3500 where it begins to taper down. This is the same cam used in the unleaded camira's so I'd expect the same torque curve.

[Charlescrown]

The pic I posted is for a NA Camira but the VE should peak around the same as a boosted one. I use that as a starter to determine wher I should set the max VE and drop it off either side similar to a std torque curve. Obviously a booted will have a much wider torque curve and usually starts off at a lower RPM but as you said the wideband will tell all. I think you said it leaned ot so maybe set the injector scaling higher then that will give you room to play with the VE numbers. I'll be interested to see if you have broken any ring lands.

[BennVenn]

That pic is of a NA 18LE. It should be almost identical to the 1.8L camira engine. There's certainly no torque peak on a stock, NA, 1.8L pulsar motor like your graph. At least not on my pulsars.

It leaned out because of the 0 in the VE table... Interesting to see how fast the wideband responds though. At 95% VE on boost I hit 10:1 AFR

Rings lands are good, bearings are good, pistons are out now. About to run a hone down the bores and my new set of rings will be here at 3pm. I'll gap them with the ring file to .47mm as per ACL's recommendation for up to 15psi boost.

Camira pistons have thinner ring lands than the pulsar which I thought was interesting. Thinner rings too.

[vlad01]

Ahh I see what you mean.

So that table is Boost VE, from 100kpa (atmospheric) to 1Bar. It definitely doesn't reflect torque.

I just set the entire table to 95% VE (pig rich in most places) then run it up on the dyno or road tune, log what CalcVE says it should be, then go back and adjust to suit. I figure that is the safest way to approach a boost VE table, always on the rich side of safety.

As for that torque curve, I agree with vlad that it's an estimation and not at all based on a real dyno graph. I'll attach my stock NA dyno graph for comparison, stock memcal, stock tune, with 280,000kms on the motor.

The attachment stock.JPG is no longer available

Ignore that small hump in the torque, i was still dialing in the PID for the dyno. Torque is almost dead flat from 1k to 3500 where it begins to taper down. This is the same cam used in the unleaded camira's so I'd expect the same torque curve.

Yep, even in the graph it says estimation at the top.

Theirs vs my graph. HP curve yeah closeish, but the torque curve is not even close.

wykres_power.jpg (62.77 KiB) Viewed 2186 times

[Charlescrown]

In no way am I saying the graph I posted is correct but the point I'm making is that maximum VE should correspond with maximum torque and the VE should follow the torque curve which on that curve a drop off from 3750. Any increase in VE after that is running too rich.
Oh and I just noticed that graph is for a Commodore which does have a very wide flat torque curve.

[vlad01]

That's the thing too, my VE tables don't match the torque curve either. My VE for the above dyno pull peaks from 3600 to 4800 with the absolute highest values at 4000. There is another medium but sharp peak at 1000.

There is a trough from 1400 to 2800 and it starts to roll off after 4800 too, but smoothly. I have seen similar things on other engines. My hotted up VP has a really wavy torque curve but the VE is way more linear than stock to redline but torque fall off similar to the above well before peak VE.

A 355 I did years ago was a big hump on the dyno for the torque but the VE iirc was pretty flat with no particular peak. The VE is calculated by air pressure and fuel and the AFR read out the exhaust against a bunch of mathematical models. How the engine extracts the power from air/fuel going in isn't always ideal once the AFRs are read on the way out, even if they are read ideal. A lot of this has to do with resonance interplay between overlap, intake and exhaust and a whole bunch of other factors. Not all of the air/fuel can be converted to work at an ideal way across the entire engine's working range. Doubly true for engines with heaps of overlap.

Best way to think of it is literally as the name says, the ability for the engine to ingest air (VE), but this says nothing about if the engine at said operational point is going to make the most of it or not.

Maybe you are thinking of BMEP ?

[BennVenn]

I find the same when tuning these 4 cylinders. If I tuned for the torque curve I'd end up very lean way up high where torque is dropping off hard. On the Gt17 turbo build even when torque and power fell off a cliff I had to keep VE in the 90's to keep AFR at 12:1. I'm not really sure where all the fuel is going or what it's doing - probably ending up in the crankcase judging by how worn this old engine is haha

I just start with a fairly high VE table (80's across the board for NA) and let CalcVE do the rest.