Borg Warner EFR Series Turbo's V 2.0

[burn4005]

Yea split pulse manifolds and 1.05 housings on both.

[Lithium]

Nice, that'll be quite interesting - so you won't push it any harder than the 8374, just to the same boost level?   

I'm guessing you don't have exhaust manifold pressure logging, but seeing as it seems you have turbine speed - do you have any logs from runs with turbine speed data?  I'd be curious about what kind of engine rpm you hit ~110,000rpm turbine speed, and how the engine behaves with the EFR9174 at the same kind of rpm area.

[RICE RACING]

13 hours ago, Lithium said:

Out of curiousity.... with the data you have, what is the MAP/TIP like say if you look for where where the airflow is at around 79lb/min is on the GT-R - I see you are running just over 25psi, which lines up reasonably well with the compressor flow chart I listed above.   It'd be interesting to see if your TIP starts getting uglier after that 79/80lb point or if it is "already there".  

Keep in mind, its such a high flow engine that even though the 'boost' is moderate, its way off to the right of the map, as turbine is in over speed condition from 6950 engine rpm! I have blanked out the specifics of maximum speed at peak power engine speed, motor revs are restricted only cause of the limitations of the turbocharger. Thankfully we use the best ECU designed for just this application, so we can run very accurately all the time to the absolute critical speed limit of the EFR9180... I am amazed how long it has lasted to be honest as its really being pushed well out of its comfort zone.

 

 

 

Edited November 28, 2018 by RICE RACING

[Lithium]

Awesome, that's really good data - thanks. What does the colour gradient on the scatter graph represent?  It definitely shows that you are running out of compressor pretty early in the revs as I was suspecting, really after 6000rpm it is already working quite hard - by 6900 it is really going to be wondering why it let this happen.

No TIP data there but it is mega obvious that the compressor is outmatched by the engine in this setup 

 

[RICE RACING]

6 minutes ago, Lithium said:

Awesome, that's really good data - thanks. What does the colour gradient on the scatter graph represent?  It definitely shows that you are running out of compressor pretty early in the revs as I was suspecting, really after 6000rpm it is already working quite hard - by 6900 it is really going to be wondering why it let this happen.

No TIP data there but it is mega obvious that the compressor is outmatched by the engine in this setup 

 

MAP for scat,

Yeah I left some bits out as too many copy (sorry about that) but you can see the obvious I was talking about. The map/tip is ugly on this set up for sure to the point that the piston engine does not make the normal 13% to 15% extra power (on equivalent displacement metric) v's the 13B rotary when using this turbo, its just too small basically.

[Lithium]

Yeah understand not wanting to show it all - not necessary really, most of the data is there relevant to the discussion.  

EFR9280 could make a huge difference on this engine 

[burn4005]

lol at the redacted plot.. anyone have ASIO clearance that can make sense of this for us?

the Y axis can be fully extrapolated from the data in the bottom left corner (117.61) for the cross hair value.

so you're overspeeding it by ~4%, good for you.

Edited November 28, 2018 by burn4005

[burn4005]

does the life racing calculate a VE model derived mass flow?

a good plot is mass flow vs boost pressure with turbo speed as color. makes it easy to compare with a compressor map.

 

[RICE RACING]

True top end motorsports ECU like Life Racing, Pectel/Cosworth, Bosch....... don't use Ve derived 'guesses', but that is another topic

 

What I have been meaning to fit up (but its a low priority to be totally honest with you) is a calibrated flow meter onto the turbocharger inlet pipe, then and only then will you be able to show mass flow rates and back work Ve, if you are guessing/doing it estimated fuel mass and Lambda sensors then you are mental LOL. Always personally found a direct measure is far more useful than an indirect guess.

Edited November 28, 2018 by RICE RACING

[RICE RACING]

57 minutes ago, burn4005 said:

lol at the redacted plot.. anyone have ASIO clearance that can make sense of this for us?

the Y axis can be fully extrapolated from the data in the bottom left corner (117.61) for the cross hair value.

so you're overspeeding it by ~4%, good for you.

^ You may need the KGB or CIA to do some waterboarding cause your extrapolation is way incorrect

Run allot more over speed than above estimate, Geoff knows :) will leave it at that, but again only possible with LR control methods.

[RICE RACING]

1 hour ago, Lithium said:

Yeah understand not wanting to show it all - not necessary really, most of the data is there relevant to the discussion.  

EFR9280 could make a huge difference on this engine 

map/tip will still be AIDS though?

[burn4005]

as with anything, shit in = shit out.

if you have enough fit for purpose sensors (fast response, high accuracy) that are correctly calibrated, know your injector flow ACCURATELY (which is the difficult part of this) there is no reason to doubt the result.

A VE model is just an abstracted correction function using physical models. you could convert all the corrections to ms and have no difference, and would suit your "motosports". they're fully interchangable. the advantage of a VE model is you can have single variable corrections as things are coupled in the model, whereas in a ms map they would be multivariate.

sure, its not going to be within 0.01%, but in a dynamic real time system its not neccessary to go to that, its only a compressor map, and its a visual tool.

The log I have above, the compressor stonewall flow vs turbo speed matches to within 0.8% of the BW supplied compressor map. if you are telling me that isn't useful data then by all means keep "intending" to set up a test bench.

OEM manufacturers go to extreme levels to accurately model engine characterists because they REQUIRE it. emissions are getting incredibly tight. motorsport engine control is basic stuff by comparision, especially on the fuelling side. those advanced motorsport ecus aren't doing anything clever, they are just very reliable, don't change constantly, come with excellent datalog analysis and professional support, and often retardedly complicated licencing requirements on top of the hardware cost, but that's another story. the torque control and gear shifting is what you're paying for.

 the Bosch motorsport ecus (even as old as MS5's etc) use a relative fuel mass to lambda target, the air temp correction is based in a NTP relative ideal gas model and the final fuel charge is converted to an injection time later from the Q-stat flow so get off your high horse.

Edited November 28, 2018 by burn4005

[burn4005]

32 minutes ago, RICE RACING said:

^ You may need the KGB or CIA to do some waterboarding cause your extrapolation is way incorrect

Run allot more over speed than above estimate, Geoff knows will leave it at that, but again only possible with LR control methods.

sorry I assumed the max speed for a 9180 was ~1119/120k rpm from where you've started blanking out. discovered now its 116k so you can see where I was coming from.

Edited November 28, 2018 by burn4005

[RICE RACING]

Ve is a guess, but lets not argue it hey? you like yours I like mine, what was the last Bosch motorsport ECU you personally used? Cosworth? Life Racing?

Edited November 28, 2018 by RICE RACING

[RICE RACING]

6 minutes ago, burn4005 said:

sorry I assumed the max speed for a 9180 was ~1119/120k rpm from where you've started blanking out. discovered now its 116k so you can see where I was coming from.

No stress, eitherway turbo is not right on that engine, not even a 13B to be honest, its more for small engines when run to full power.

[RICE RACING]

47 minutes ago, burn4005 said:

OEM manufacturers go to extreme levels to accurately model engine characterists because they REQUIRE it. emissions are getting incredibly tight. motorsport engine control is basic stuff by comparision, especially on the fuelling side. those advanced motorsport ecus aren't doing anything clever

You do realize LR do OEM development with that exact same 'motorsport engine control' that exceed Euro5&6 standards? https://www.liferacing.com/products/ecus/gdi-series/ so not sure where you are going with that one to justify Ve based ECU's? I think its just marketing or justification of principle of operation getting confused with reality but hey.

I know a LR is extremely clever, but I use it daily so I am biased, yes. Will dismount hobby horse now.

Edited November 28, 2018 by RICE RACING

[burn4005]

Euro 7? Life racing's charge temp correction and transient enrichment/decay is far too simplistic for that. there is zero chance they are using their ECU in an OEM situation and meeting that standard.

a one-off non-performance tune of a car to scrape through euro6 maybe. but oem e7, nah.

enough internet debate, and I know you love them as a dealer/partner but they're just another option in the middle of a sea of many.

Edited November 28, 2018 by burn4005

[Lithium]

9 hours ago, RICE RACING said:

True top end motorsports ECU like Life Racing, Pectel/Cosworth, Bosch....... don't use Ve derived 'guesses', but that is another topic

What I have been meaning to fit up (but its a low priority to be totally honest with you) is a calibrated flow meter onto the turbocharger inlet pipe, then and only then will you be able to show mass flow rates and back work Ve, if you are guessing/doing it estimated fuel mass and Lambda sensors then you are mental LOL. Always personally found a direct measure is far more useful than an indirect guess.

Weird flex, but ok.  I'm going to restrain myself from going off topic on that one as well haha.  The estimated airflow thing is a fair call, however - while it's nice to get a thumb suck it definitely is an estimate which relies on all kinds of things being right which probably aren't always bang on.  It's one of the reasons I was interested in TIP, MAP and turbine rpm on the setup you are using, as really it was going to be a more reliable way of getting a gauge of how hard the compressor is working and how much the engine is getting choked by the turbine.

9 hours ago, RICE RACING said:

map/tip will still be AIDS though?

I can't answer that at all, have not seen any TIP data  

The data you have given would make me expect TIP to get AIDS above 6000rpm purely because of the compressor flow, regardless of how well the hotside flows... however I don't know how the MAP/TIP is *before* the compressor starts choking.  Maybe the hotside would *also* be a limitation, but if the MAP/TIP isn't bad before 6000rpm then I'd hazard a guess and say that it might actually improve a heap with the bigger compressor and then you'd start finding out if/when the hotside legitimately becomes a restriction of it's own. 

Edited November 28, 2018 by Lithium

[RICE RACING]

8 hours ago, burn4005 said:

sorry I assumed the max speed for a 9180 was ~1119/120k rpm from where you've started blanking out. discovered now its 116k so you can see where I was coming from.

NP and I agree with you that enough debate, I appreciate your input and some of what I put up is of use so lets keep at that cause I know I can't test everything myself, so its good if we can work together since there is not many people who spend 30k in electronics and development time/resources per car, and in multiples its even more rare, not just one off personal hobby projects (nothing wrong with that though) but the more you do the better potentially the data collection is.

[RICE RACING]

1 minute ago, Lithium said:

Weird flex, but ok.  I'm going to restrain myself from going off topic on that one haha.  The estimated airflow thing is a fair call, however - while it's nice to get a thumb suck it definitely is an estimate which relies on all kinds of things being right which probably aren't always bang on.  It's one of the reasons I was interested in TIP, MAP and turbine rpm on the setup you are using, as really it was going to be a more reliable way of getting a gauge of how hard the compressor is working and how much the engine is getting choked by the turbine.

I can't answer that at all, have not seen any TIP data  

The data you have given would make me expect TIP to get AIDS above 6000rpm purely because of the compressor flow, regardless of how well the hotside flows... however I don't know how the MAP/TIP is *before* the compressor starts choking.  Maybe the hotside would *also* be a limitation, but if the MAP/TIP isn't bad before 6000rpm then I'd hazard a guess and say that it might actually improve a heap with the bigger compressor and then you'd start finding out if/when the hotside legitimately becomes a restriction of it's own. 

Lith it sounds like your do a bit of this?

What is your limit for map/tip or what do you find in experience on the types of engines you do starts to yield lower results and increase negatives in relation to such parameters like say knock or another metric.