I get asked this a lot so thought it worthwhile to have a rule of thumb thread for injector upgrades for various power levels . I'm basing this on personal experience rather than calculations because I've actually hit the max. output on various injector combinations over the years so have a good idea on what actually works .
As you will see , you can reach any power level the Renesis is capable of (to date) using just the stock injectors that came on the series one engines .
I haven't included any other types of injectors .Although I've tuned cars with various aftermarket injectors successfully , I've always found them to be harder to tune than the stockers (due to uncertainty of latency numbers mainly)so I really don't see the need to spend your hard earned money when you could get an excellent result using what you already have or have easy access to .
The stock injectors that come with a series 2 would be a good alternative also but to date I haven't done any experimenting with them. Edit : Have now used S2s and would highly recommend.
As far as injector outputs go , I believe the scaled output figure used in the ECU is actually a fairer indication of injector output than what you get from a flow test which is done at a lower pressure than our system pressure, so that's what I'm using here.

Injectors listed In this order: P1/secondary/P2


Stock injectors
red/yellow/yellow .................................369/475/475 x2 = 2638cc = up to 270whp
good for a stock greddy turbo kit

yellow/yellow/blue .................................475/475/626 x2 = 3152cc = up to 330whp
Upgraded greddy turbo

Yellow/blue/blue.......................................475/626/626 x2 = 3454cc = up to 360whp

For best spray pattern use S2 injectors :
green/brown/brown................................500/657/657 x2 = 3628cc = up to 380whp

yellow/blue/uncapped yellow...................475/626/850 x2 = 3902cc = up to 400whp

Yellow/uncapped yellow/uncapped yellow...475/850/850 x2=4350cc = up to 450whp

Blue/uncapped yellow/uncapped yellow ......626/850/850 x2 = 4652 = up to 480whp
Custom turbo systems


In general it's best to put identical injectors in secondary and P2 as they share the same latency map . The latency difference between yellow and blue and between blue and uncapped yellow is small enough however, that this doesn't detract too much from the result. I would split the difference between the two maps for best results.


For Uncapped yellow injectors add 10% to all the numbers in the blue injector latency map for best results .

 

OEM Red latency map:




OEM Yellow latency map:


Bosch EV14 1000cc latency map:
Estimated through extensive road testing - no official table has been published to date. Use Scaling value of 1225.



Injector Dynamics ID1000 latency table is available from their website.

 

Someone please post Blue latency table.

 

i think you got the red and bosch maps reversed !

 

Blue latency map :



Uncapped yellow map: (approximate only)

 

how do you like the ev14's?

 

Doh, fixed

 

I haven't had any issues with the EV14s after I got the latency table and scaling right. I was talked into them by an EFI company that stocks both Bosch and Denso. They told me that the Bosch have better spray pattern than ID and Denso, being a more recent design. They also thought I was crazy for considering uncapped Densos, as apparently their spray pattern is significantly worse than all others. But after seeing Brett's track record for reliability and performance I'd conclude that arguing about which has the superior spray pattern is splitting hairs - they all seem to work if you have them tuned right. Maybe better spray pattern delivers better fuel efficiency, but my fuel economy is far worse than most rx8s, so who knows.

BTW, I'm currently running stock yellows in P1, and idle is the same as it was running the stock reds. I'm running 2x Bosch and 2x Yellows in P2 and Secondaries, and since they share the same Latency table I had to work on the VE table slightly. In the future I'll sort this out by getting another pair of EV14s so I have the full set running in P2 and Secondaries (should have done this a while back but my VE table hack is working quite nicely at the moment).

 

I ran stock yellows in Primary. Uncapped yellow in secondary and P2.

I plan to go with e50 or e85 on the next setup. I have a pair of ID1000 but I'm thinking of going 1000cc all around for e85 headroom. I can't decide if I should just get the EV14s or order 4 more ID1000s. Of course my wallet is saying ev14

 

I had issues with the EV14s running in P1. Symptoms were longer crank time at startup and fluctuating rpm at idle. Car still drove fine, and it wouldn't be a deal breaker for most, I'm just a fussy bastard.

If I were in your position I'd be using stock Yellow or Blue in the P1 to guarantee a good start up and idle, then your choice of 4x matching 1000cc injectors in the P2 and Secondary. I'd just use the cheaper option ID1k vs. EV14.

Slash has confirmed the ID1k works well and rates them highly...2 engines and counting, but I think we can blame the mad scientist more than the injectors here.

I've confirmed the EV14 works well. I've been running 13 psi for a while now. Not sure about ID1ks, but the EV14s do requiring different injector plugs, so you'll just need to cut them off at the original plug and crimp a new plug in there. Don't use solder cos the heat can melt it. In short, you'll be fine with any of these two 1000cc options.

 

here is a file with dead times for ev14 that i found online. not sure how helpful this is

 

Regarding Latency table and Scaling values - the tune relies heavily on the Latency table at startup and low-mid load, so Latency is a key factor for drive-ability around town. As you get into higher loads, Latency matters less and the tune relies much more on the Scaling value.

When you change injectors you need to check that you have the appropriate Latency table and Scaling value, as listed above.

Your AFRs will probably be off, and you'll need to correct them by adjusting either your VE table or your Open Loop fueling tables. It's up to you which method you use - 2 different ways to skin the same cat - both methods are effective if done correctly.

It's very important not to jump right in and go WOT at 8psi after an injector change. You need some foreplay, so start at 0psi and make sure the AFRs are good and safe before going to second base. This may require a few tunes to get there. Step it up to 5psi. From here, make the boost increases smaller and smaller because the risk of a lean AFR blowing your engine increases significantly above about 5psi, and only gets worse the higher you go. Personally I corrected AFRs at 0psi, 5psi, 8psi, 10psi, 11.5psi, 12.5psi and 13psi, then backed the boost back to 11.5psi. I found that by the time I was hitting about 10psi, the tune didn't need much adjustment at all, as the VE doesn't change much above 100% load.

If you have a lousy turbo setup it's quite likely you'll blow your motor by attempting the above, so don't do it, and if you do, don't go past 8psi.

 

I hadn't seen that particular table, but Deadtimes (latencies) vary quite a lot between injector sizes, generally increasing with flow (pintle takes longer to open), so I'm pretty sure this 300cc table wouldn't work for the 1000cc units. FWIW I've also seen (and tested) another white paper for the EV14 that shows you how to use formulas to build your own latency table. While the theory and formulas in the white paper were probably correct, it just didn't work. I suspect a lot of the rx8 tune is just arbitrary values. E.g I'd be surprised if the millisecond units in the latency table are actually real world milliseconds, but they still relate to other areas of the rx8 pcm tune, resulting in accurate AFRs, which is all that matters.
I'm with Brett in that I trust what's been tested our cars much more than what the theoretical formula says it should be.
If I were using a stand alone ECU with a MAP sensor, I'd be all over the manufacturer and flowbench published tables, but I don't think we can relate them to the rx8 PCM tune.

 

Dropping some good advice and info there Jimmy.


I guess you opened that PDF file and didn't scroll down to the 1000cc injector dead times...

 

I know I sent a set of yellows and reds off to be cleaned and flowed and they sent them back to me quite larger. Only cost me $150.

Yellow: 880cc x 4
Red: 800cc x 2

If anyone is interested...
RX8 & WRX Injector Modification

 

^ i have the same yellows
that's what we've referred to as "uncapped"

 

Interesting to see what an uncapped red flows .

I noticed that the injectors pictured on that site show the plastic cap is also removed when they take off the steel cap . This is a bad idea as the injector then has clearance on the rubber ring that is supposed to seal it off .




place injector at angle shown and rotate slowly while feeding it into the grinding wheel.




Finished product with side of plastic cap still in place.

 

I had to go back to stock injectors. Was waaayy too much fuel for the half bridge. It was never my intention for them to flow them so heavily. I only wanted them cleaned and all the flow rates matching for consistency...

 

did you end up selling them on ebay?

 

That was a bit of a ***** up !

 

I still have them in case I ever had an application that I needed larger injectors for.

 

Added a couple more popular combinations

 

ah, because of the info on this thead, i can run reds primary and 2 uncapped yellows on my latest ms3-pro -FC build...mainly because i have these injectors just laying around...thanks again for your contributions to the rotary kind of all types.

 

Added S2 injectors to the list. If you are going to buy new , I strongly advise buying S2 injectors as they have superior spray patterns.
Been using Brown in P1 lately and found I needed to increase idle to 1000 rpm to avoid surge.

 

Or for superior fuel injection characteristics; Injector Dynamics fuel injectors ...