Jump to content
SAU Community

Borg Warner EFR Series Turbo's V 2.0


Piggaz

Recommended Posts

Went with link as that is what tuner preferred and there is a bit of stuffing around to setip closed loop flat shift on the PPG (which is mental!)

After I sell the platinum pro and it's accessories I shouldn't be out of pocket too much. 

Link was a straight plug in and the speed sensor is hooked up, although I'm not sure how difficult that was. 

Balancer is stock.

I might do the PRP trigger kit in the future. Looks like there is plenty of gains to be had. kind of just want to get it back on the road and enjoy it for now. 

Link to comment
Share on other sites

11 hours ago, usmair said:

Went with link as that is what tuner preferred and there is a bit of stuffing around to setip closed loop flat shift on the PPG (which is mental!)

After I sell the platinum pro and it's accessories I shouldn't be out of pocket too much. 

Link was a straight plug in and the speed sensor is hooked up, although I'm not sure how difficult that was. 

Balancer is stock.

I might do the PRP trigger kit in the future. Looks like there is plenty of gains to be had. kind of just want to get it back on the road and enjoy it for now. 

Sweet as, yeah if I had no ECU and it was between Link and Haltech I would definitely go Link - just an unusual option to swap between them so was curious :)  Oh PPG, very nice... that should be a hell of a fun setup, I look forward to hearing how it goes at the drags.    Be also interested to hear how reading the signal from the speed sensor goes, being able to wire and config the sensor isn't what I anticipate as a potential issue - it's the fact that the DI frequency is too low to accurately count the blades, even though the Borg speed sensor divides by 8.   Or has it been wired to an analog input?  That could be interesting...

If you aren't going to do the PRP kit I would suggest the NZWiring thing at the very least is a good way to go, takes like half an hour to setup and install from memory - and will be a significant improvement.

Link to comment
Share on other sites

1 hour ago, Lithium said:

Sweet as, yeah if I had no ECU and it was between Link and Haltech I would definitely go Link - just an unusual option to swap between them so was curious :)  Oh PPG, very nice... that should be a hell of a fun setup, I look forward to hearing how it goes at the drags.    Be also interested to hear how reading the signal from the speed sensor goes, being able to wire and config the sensor isn't what I anticipate as a potential issue - it's the fact that the DI frequency is too low to accurately count the blades, even though the Borg speed sensor divides by 8.   Or has it been wired to an analog input?  That could be interesting...

If you aren't going to do the PRP kit I would suggest the NZWiring thing at the very least is a good way to go, takes like half an hour to setup and install from memory - and will be a significant improvement.

Yeah the link DI's cannot support raw TSS data, my guess would be using the road rage converter box.

And i can't agree with Lith any more, i wouldn't think twice about buying the NZ kit and putting it on now, it would probably be the best vaule for money mod you could possibly do.

4 minutes ago, Dose Pipe Sutututu said:

Platinum Pro to Link is a big step

Elite to Link is sideways

Agreed-ish

  • Like 1
Link to comment
Share on other sites

9 minutes ago, Dose Pipe Sutututu said:

Platinum Pro to Link is a big step

Elite to Link is sideways

Ohhhhh... yes.... right, makes sense - cheers (Y) 

4 minutes ago, iruvyouskyrine said:

Yeah the link DI's cannot support raw TSS data, my guess would be using the road rage converter box.

And i can't agree with Lith any more, i wouldn't think twice about buying the NZ kit and putting it on now, it would probably be the best vaule for money mod you could possibly do.

Obviously agree, but I feel this really needs to be underlined.  I'd sooner not use the car with the cylinder pressures that high than continue using the stock trigger setup - ESPECIALLY with an identified trigger error issue.

  • Like 1
Link to comment
Share on other sites

7 minutes ago, Lithium said:

Obviously agree, but I feel this really needs to be underlined.  I'd sooner not use the car with the cylinder pressures that high than continue using the stock trigger setup - ESPECIALLY with an identified trigger error issue.

I 100% agree with you, i think only bad things can happen. Absolutely nuts to be trusting stock CAS at that power level and with a totally standard bottom end.

But it's not my car and he can do what he wishes ?

Link to comment
Share on other sites

On 23/09/2019 at 9:59 AM, Lithium said:

A mate has a built RB26/EFR9180 combo going on the dyno in the next day or so before being sent over to Oz for WTAC, they aren't going to go "full send" as its a drift car being built at the last minute - but I'll share some results if appropriate.

This made 570kw on a pretty conservative tune (read: a fair bit more in it if it needed) with 30psi seen before 30psi in a sweep

Link to comment
Share on other sites

On 19/04/2019 at 3:55 PM, Lithium said:

 

OK, I've just read through a bunch of comments which flowed on from here which I agree and disagree with to a varying degree but as usual there is a little mis- or half-information that I don't feel like anyone considering them are getting enough to make an reasonably informed decision and for a change I actually have time on my side today, so I'll answer with more than just the easy "I agree with Micko". 

Like always, this is just my view point - I don't profess to know f**k all and know there are people who do or will imply that I don't, so take that as you will but I think it's worth at least using things people say to base your own research before deciding whether it's bullshit or not.  ESPECIALLY when it comes to injectors, as they literally supply the fuel for your combustion - debating turbos is one thing, making a decision which could cost you a very expensive engine is a different thing altogether.

There are a lot of different things in this topic so I'll try and put some light on the things that usually are or should be on peoples minds about them.  This whole post is SPECIFICALLY about "Bosch 2200cc" / ID2000 injectors.

So I'll focus on these things for the two major areas people talk about these, firstly the relatively academic/harmless area people debate about these injectors and then will ramble about the bigger issues.

ID2000s vs Bosch 2200cc injectors

Straight up, they are the same hardware.  Injector Dynamics themselves do not hide that fact, and put a lot of work into matching them so strictly speaking they are "better" than just going with generic Bosch 2200cc brands.  Check out the link here: http://injectordynamics.com/articles/does-dynamic-matching-really-matter/

There is likely to be less pain tuning ID2000s than generic Bosch 2200cc injectors, but that is missing the point... there is likely to be problems at some point, the question isn't so much if - but how bad, exactly how bad will be under a different heading.

So yeah, that Injector dynamics page IS accurate but it's kind of glossed over things and made out like their injectors will fix drivability issues and the flow as tested has a very small range of error at higher pulse widths - which is kinda true, however they have only plotted down to 2ms total pulse width.  Below is a scatter plot showing rpm vs actual pulse width from a car using these injectors on BP98,   I've set any load points plotted which are under atmospheric to show as green - as it comes into the transition to boost area it will go through from orange to red.   Notice how much of the driving around this car does exists BELOW the 2ms range plotted by ID in their graph?  (note, the flat line around 1ms is due to the affect of a combination of short pulse adder and dead time settings).
image.thumb.png.67ff25676175be6620dfebd028353fcd.png

These injectors have quite short dead times relative to the flow that they have, and as the linked article suggests - not every injector is made equal.   A significant thing about this is that the dead times between any two injectors can vary by a small amount, if that amount is still significant compared to the average (or the configured....) dead time and you have injectors which flow a heap of fuel for a small amount of open time then the amount of variation in flow between each injector can be very significant.   Because of this fact, as the pulse width gets shorter the potential for deviation in relative flow between each injector increases significantly - the unmatched ones going exponentially different by 2ms, and the ID2000s even are starting to show the trend.   It'd be very interesting to see what would have happened if that chart was shown down to 2ms, it is a shame that they didn't show results closer to the 1ms mark - my logging essentially shows that basically all typical driving on pump gas will be done below the pulsewidths that they will be operating with.

Those variations in flow DO matter if you give a shit about drivability, it is the kind of thing that causes shitty starts, hesitations at transient throttle, less than ideal idle quality and such forth.   They can be glossed over by tuning them to run richer mixtures, but still bear in mind that this is a band aid - and one that's not lovely.  The variation in flow between cylinders in different situations is still going to be there, it's just been masked.   This may not bother everyone, and will obviously be less of an issue with E85 where you immediately add >30% to your pulse widths for a given amount of airflow - but then this is the most trivial concern with using these injectors, so I'll move on to the bigger concerns.


These injectors are CNG injectors

The implications of this seem to be mostly poorly understood or communicated.  I'm not going to profess to know heaps about it to be honest, I am not an injector engineer and don't know a lot about how they work *but* this whole tuning lark is like that.   Everyone knows some of the picture,  doubt the f**k out of anyone who is 100% confident on everything they are talking about.  I do feel there are some basic points which aren't questioned or discussed and they are pretty meaningful here.  Hear me out...

When the "CNG injectors" thing comes up, it is usually when discussing E85 use.  That is kind of true, they are not designed to use E85.   People talk about them being "OK" if they use stainless internals.... welllll..... that is only part of the picture and that certainly doesn't make them safer.   

It is true, they aren't designed to use on E85 - but that's more to do with the fact that E85 is a fluid and they are NOT DESIGNED TO WORK WITH FLUIDS.   Any fluid.   Here is something from the horses mouth:


image.thumb.png.04f655344c1f8ed758b5b755cb7a2806.png

So what does this mean?   I can't say exactly what the behaviour inside is, how long it takes, how likely it is to happen etc but I have had the misfortune of having to deal with cars running these injectors and I am finicky enough to analyse the shit out of everything and identify things that are not working the way they should be and try and work out what or how the problem is occuring.   A lot of people don't run into issues, from what I've seen they are either lucky or just are not paying enough attention to realise what is going on.  Likely a bit of both.

What I do know is that these injectors can change how much they flow over time when used with fluids.  That can be a mixture between immediate, and longer term.  There are PLENTY of cases where people have run into tuning issues or even engine failures and returned them to a supplier who has tested them and found that where they previously were ~2200cc/min at 3bar they are now <2000cc min with possibly a greater margain of error between the injectors.   That has been previously deemed not enough to cause the issues seen, they've been given some love and back to their previous flow level and everyone has been happy.  At least temporarily.

This is where things get scarier though, their flow behaviour can change on a much more minor time scale.  Like in the period of half an hour, with no tune changes, the same IAT/ECT and basically all typical variables within a range where you would expect a minimum variance in flow.   I'm talking well over 10% in flow, with a significant error between each injector.   It seems like it can be associated with fuel temperature, and likely is - however this is not a consistent error...  corrected volumetric to mass flow calculations which take into account fuel temperature will not fix this, as it isn't about the density of the fuel changing.  It's about how the actual injectors are delivering the fuel, essentially delivering an inconsistent fuel volume for the same pulse width at different times.   

It is not a huge leap to make to assume that this can be attributed to the whole "swelling behaviour of elastomer seal in liquid media" thing - which is not a one way street with this kind of material.  Basically they can swell and contract depending on temperature and exposure to fluids, the elastomer seal is designed to stop a gas from seeping past the injector when the injector is not meant to be allowing fuel through - and the upshot of this is there is a thing which is not intended to be used in a way which can cause it to swell and contract means that the rate which it does it is not consistent in any way.   They are meant to be dry.    

The upshot of this is that when you hear about these cases where a car has had a lean out on these injectors and hurt an engine, the injectors have blamed for being "blocked" due to E85.... that absolutely can be the case, but every now and then you hear about the injectors being tested and they've turned out fine and things get messier - this is where the horror starts.   You don't need E85 to cause this issue.  You don't need foreign contaminants.   The injectors can return back (more or less) to the flow behaviour they had before and not give any hint that one or many of the injectors at one point during operation may have dropped to  <2000cc/min of flow on a car tuned assuming 2200cc/min flow potential.  A cylinder which was happy running 30psi on E85 at 11.5:1 (petrol scale because most people seem to talk in that even on E85) may not be so stoked with running 13.0:1.   The same car can richen up or lean out further depending on the mood of the injectors, and there is no way for you to know how when, if or how bad the problem will happen but based on what I've seen it would be foolish to use these injectors and not expect it.

There are other things which E85 and methanol, or even MTBE (which is used in plenty of other fuels and have similar properties to E85 which can negatively affect fuel systems ) which can cause issues - that is a whole different topic which does cross over into this one, but is actually less scary that what I've actually seen.   I've been unfortunate enough to see the shenanigans these injectors can offer, owners of cars I've been behind the laptop where these injectors have been used have been fortunate enough to have someone that picky that we've noticed things before they became a problem, and as such I'm fortunate enough to not have melted an engine when using these as I've pulled the plug and insisted they get changed before something goes wrong.   I know others who haven't been so lucky.

So yeah, this is my opinion and based arguably on anecdotal evidence - but if anyone wanted a more substantial explanation of why not to use these injectors than "throw them in the bin" or "they're shit", I at least feel like I've done my part to ensure there is something to chew on... take it as you will.

PS.  Just don't.

PPS. Sorry for the overly long rant, I look forward to the TL;DRs and hiding under my rock again until next time I get bored and put my neck out with an opinion. 

I had been trying to remember where I put this rant, thought I might dredge it up seeing as it's been talked about in here and I should probably quantify things a bit.  I basically worked a lot of this out after tuning my first car using the Bosch 2200cc injectors and struggling with some drivability and getting the charge temperature correction right.  Normally I find this kind of thing goes quite smoothly and decided to sanity check the setup and my process, so sat down and carefully checked over the logs, did some maths and worked out that the variation I was seeing was impossible based off ideal gas law, then dug deeper and discovered a bunch of the info I shared in the above rant.

The car I was tuning when I found this now has a set of XSpurt 1550cc injectors and immediately tuning the car became a different experience, and the car drives and behaves quite differently.  It's alarming how differently in behaves- in some ways.  After talking to other tuners, and based off my own experience I would NEVER use a CNG injector on a car using a liquid fuel.  

The car using these injectors uses a Borg Warner Airwerks turbo, btw - so here's a result from the poor cousin of the EFR:

RB30E block with S2 R33 RB25DET head 
CP 8.5:1 pistons
MRP rods, arp2000 bolts
ARP main studs
Oil pump drive adapter
RB25DET oil pump
Tomei 256x 8.5 poncams (NVCS)
Lightly ported r33 s2 head
Genuine greddy intake manifold
VH45 throttle body
6x XSpurt 1550cc injectors with flex fuel sensor
150mm intercooler
Sinco twin scroll exhaust manifold
66mm precision waste gate
Borg Warner S300SX 8375 w/ twin scroll .91 housing
4inch exhaust from turbo to back
Vipec i88 ECU 
NZ Wiring cam trigger kit

69697916_3484973428195168_1424323077906366464_o.thumb.jpg.36cd22e0bb0c5e7a208a38658c50e1fa.jpg

69507975_3484958871529957_4317882935094018048_o.thumb.jpg.548ac1a31c7885cc660313afcad5ba92.jpg

Edited by Lithium
  • Like 1
Link to comment
Share on other sites

17 minutes ago, Dose Pipe Sutututu said:

@Lithium do you also have a graph with the manifold pressure? the torque sure drops quite heavily resulting is very flat power up top, nearly looks like valve float (if pressure remains constant).

It was more or less by design, I had limited dyno time and was just collecting as much data as possible in each run to make sure I could configure the boost control thoroughly etc - I didn't bother trying to make the boost curve completely flat so this was the highest power single run we got.  The BCS duty needed to start going for the sky to hold much more than 17psi at high rpm and you could audibly hear the turbo working hard so I didn't bother trying to push it harder at high rpm as heartily delivered 600whp was coincidentally a bit of an unofficial target (and we expected it to choke around 18psi at full rpm) and it hit that so definitively, there were no surprises so we just decided to leave it there and tidy up the other bits and pieces.

This run was on 66% ethanol and again it can technically has a little more flow but we got everything we needed and the car is very rapid as it sits.  Out of interest, the boost control setup works really well - has a 4-port solenoid on it with a 0.4bar spring and runs about 5-6psi in first, about 11psi in second and then gets to more serious boost levels in the higher gears.  It can hold WOT in any gear in decent conditions and just sits on the limit of traction at worst, which makes it surprisingly un-lethal to drive fast considering it's an R32 GTSt.   As the boost plot probably paints the picture of, the thing builds boost easy as hell so you don't even have to think about trying to get it on steam when driving so between that and boost per gear it's a pretty fool proof kind of package.  Definitely when that "600whp" number only tells a bit of the story when it is delivered so broadly.

Btw with the boost control fully dialled in the ramp to full boost is also quite a bit more aggressive than this plot shows if you set a 18psi target, and the power peaks at around 7000rpm.  

FB_IMG_1569309556307.jpg

Edited by Lithium
Link to comment
Share on other sites

Very good result and insight, I find with 4 ports they're interesting to get right. I currently have a 1bar spring it in and between 1~2% change on the DC it's changing as much as 0.3 bar at time.

I'm going to take it apart (maybe never) and swap it to a 0.5bar spring and see how that behaves, hopefully give the control range a bit more resolution.

  • Like 1
Link to comment
Share on other sites

Cheers - yeah, a 1bar spring with a 4port solenoid is going to make life very tricky for you as you say..... Partly because small changes in duty will result in large changes in boost. 

 

The other issue is if you are using closed loop boost control a lot of the more middle-of-the-road ECUs don't factor in dead times, which isn't a biggy with open loop but if the closed loop logic starts including the "dead time" portion of the pulse to the BCS then the attempt to correct error will always have an invalid component to it - when 1-2% duty is going to have a meaningful effect on boost you will have issues when maybe for sale of argument 10% of the duty is taken up by just getting past the dead time.  Not sure if the Haltech has the same issue but the Link does and I have to consider that and make compromises when tuning the closed loop logic, and closed loop seems pretty much a necessity with 4-ports.

 

With this setup there is an ever slight oscillation in the control but it is definitely not something you can tell when you drive, it just irks my OCD 

Link to comment
Share on other sites

17 hours ago, Lithium said:

I know of two RB32s running 1.45a/r EFR9280s being tuned in the coming weeks, fingers crossed there should be some interesting results showing up :)

I'm super interested to see the results as this is the same combo (from a high level) that I am going to run as well. I haven't bought the turbo yet, as I'm having pre-buyer's remorse, but can pull the trigger any time. Keep us posted. 

Link to comment
Share on other sites

On 07/09/2019 at 6:19 AM, Full-Race Geoff said:

Because you want more midrange than top end, 550kw / 750hp with EFR is doable in two ways here: twin EFR6258 or single EFR8474 both are options. 

The 9180 and 9174 you are asking about are good turbos.  9174 was our most popular turbo for professional drift competition for many years... the 9174 continues to work extremely well despite people who dont like the numerical rotor sizing.

the 1.45 a/r does increase top end power in exchange for a slight midrange loss, but on a RB30/26 the 8474 or 9174 on 1.45 a/r makes sense for this power target

twin gates typically provides slightly earlier spool and more midrange due to the fact that the pulses remain seperated, and more energy is imparted to the turbine

Hi Geoff,

Quick question, reading on your website in regard to 84/92 variant efrs and it states not to use a oil restrictor as the turbo has one internally. Does that mean you wouldn't run a Turbosmart oil pressure reg either which are meant to supply a max of 40psi to the turbo from my understanding?  

 

Screenshot_20190928-134443_Samsung Internet.jpg

Screenshot_20190928-133903_Samsung Internet.jpg

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...