logalinipoo

The only latency info i could find from id was in milliseconds. So no direct copy and paste, that i could find.

Oltmann found that the milliseconds column in coil dwell was actually milliseconds multiplied by 256. Im guessing that calculation carries over for anything else that uses milliseconds.

Im pretty sure that the injector size is cc/min at pressure. So id 725's are probably based on 3 bar. At our 4 bar system they will flow quite a bit more.

Be careful how big you go in p1 it could give you idle issues I'm going with stock yellows and 4, 750's in the secondary and p2.

slash128

Thanks for the feedback. It could be divided by 256 but the resultant values seem off to me. For instance, on the Sec/Aux table if we pick on the 1.07 baro column at 14v the value is 553.00. Divided by 256 would result in 2.16 msec. That seems like a long dead time, where dividing by 1000 results in .553 msec, which seems more in line with Denso latency figures. I just don't know enough about injector technology.

I agree with not going too big in the P1's. I am leaning towards stock yellows in P1 and either ID725 or ID850 for P2/Sec. My previous setup was stock yellows in P1 and uncapped in P2/Sec. They were flow tested at 820cc/min @ 2.25 bar. I could just go back to this configuration with new injectors. I believe I had one or more failing uncapped injectors based on all the oddball fuel issues I was logging.

logalinipoo

Try dividing stock latency by 1024. Its still a multiplier of 256. And it did seam right. Ive been playing with my dw700's getting them tweaked in and that seams about right now that i think about it.

Where did you find denso latency data? It would be interesting to look at.

Also lookup what oltmann found about latency. There is another sub table based on expected intake manifold vacuum in our latency tables the far left is wot/boost area and going to the right is when the engine is at low load.

slash128

I was looking at this table as a reference. Our specific injectors don't seem to be in the mix, but the values seem more in line with dividing by 1000 vs 256. But I am definitely not basing a decision on this, just trying to work backwards when I don't know specifics.

http://injector-rehab.com/shop/lag.html

Is this the thread by Oltmann you are referring to?

https://www.rx8club.com/showthread.php?t=229295&styleid=22

logalinipoo

I have the latency data in ms for my dw injectors so ill try multiplying by 1024 to see how that works. Ill look at the those denso and the id tables to see the same thing. That is the thread I'm referring to.

Don't the suabru guys use the same yellows. I wonder what their tables look like.

slash128

Thx man. Worst case I can go with an all new set of yellows and uncap them. I just don't trust them at this point

slash128

I haven't seen the ATR tables for our yellows in Subaru, but I found this doc. Look at the latency numbers in the image from the PDF.

https://static.cobbtuning.com/cobbtu...ide_Subaru.pdf

9krpmrx8

iTrader: (46)

It looks like the WRX uses the blues, interesting info.



It looks like five-o (where I got all my new OEM injectors) sells an upgrade too.

http://www.fiveomotorsport.com/impor...jectors/subaru

slash128

Nice, thanks 9K!

logalinipoo

Based on that table thats about equal to our scale/512 but there is going to be some variation due to fuel pressure and i would imaging they're running a 1:1 regulator so there wont be a pressure correction.

Brettus

iTrader: (2)

ID 1000 latency table from Injector dynamics ...... all care no responsibility

slash128

Thanks man That is similar to the info I downloaded from their website. There are two questions I have tho:

1) The x-axis values in ATR are completely different. Do I assume to just edit to match, or is ATR using a different unit than kpa? ATR won't let me paste in that section of that table and won't let me directly edit. If I edit one of the cells, no matter what number I type in ATR sets it to zero. However, it seems I can multiply those cells.

2) The table values in the file from ID are in milliseconds. The values in ATR are not. Am I safe to assume ATR is in microsecond and just multiply the ID table by 1000 or is there something more to it?

http://help.injectordynamics.com/helpdesk/attachments/15319865

slash128

Injector Dynamics support has been helpful. After some discussion with them they said that the values in ATR look like microseconds. Also, they suggested that the delta across the table for a given voltage is small enough that it should have minimal impact if the X-axis values are off. This is tempting to just bite the bullet and try.

However, given that it is still an unknown I am also considering another alternative. The link that 9K posted had some interesting looking replacements for our yellows:

2006 MAZDA RX-8 R2, 1.3L, GAS Fuel Injectors

Until I have a motor built up for 17psi I may look at running 5psi for a while with a set of these all around. With our 4 bar fuel pressure system this should give me 1600cc/min per rotor...

EDIT: looking at the photos of these injectors I wonder if they will fit in the P1 location...

No boost is no bueno

Brettus

iTrader: (2)

The numbers in the table are the same for the stock injectors for both M/E and ATR, so don't worry about changing the x axis. I can edit each cell in the table no problem ?


I used the table above for a tune and it worked ok ...........If i were tuning in ATR i would enter the same numbers .....

slash128

Thanks Brettus. As far as editing the cells maybe it's a quirk of my version of ATR. If I click a cell in the top row for Baro, hit the 'e' key to edit and put in a value ATR automatically just changes it to a zero. Weird. I can edit the values in the table itself just fine. This is the only table where I have run into this behavior so far.

Brettus

iTrader: (2)

You are taliking about editing the x axis ...right ? Which you don't need to do ....

Anyway , I just tried that on mine and i can edit the x axis no problem ..... you do have to choose a number lower than the next number along and higher than the preceeding number for it to work though .... that is the same with any editing of the x axis in ATR..........

slash128

Yes, I'm referring to the Baro x-axis that I now know I don't need to edit. I just found it odd. It seems to be only the values that are less than zero. I can't put in negative values, even if it is a value between the adjacent cells. It's a non-issue at this point, just found it interesting.

Brettus

iTrader: (2)

ok ... just tried putting a negative in and got the same thing , so it's just a quirk by the look

slash128

Thanks for verifying. I haven't seen it in any other tables yet.

logalinipoo

I just did some looking at the table I had from there. Here it is. They just multiplied their figures by 1000. This excel file had the milliseconds for all of their injectors set up for the rx8.

slash128

Cool, that's the conclusion I came to as well, as did the ID support guy I discussed with.

slash128

Regarding injector sizing for staging, how much does it matter when looking at the capacity delta between P1 and the others? I was pretty happy with idle and drive ability of my previous setup with stock yellow in P1 and uncapped yellows in the rest. I'm just skeptical of the consistency of the uncapped yellows.

What I am considering now is stock yellow in P1 with ID850s in the rest. This would be pretty close to what I had before. However, stock yellows are pricey from the dealer ($260) and aftermarket doesn't seem to get a whole lot cheaper. The Five-O's looked interesting but I don't think they will fit in the P1 hole due to how low the connector is on the body.

So my next thought was stock reds in the P1 and 4x ID1000s. Five-O has new reds for $65 ea. Is this big of a size difference going to cause significant issues?

logalinipoo

I don't know. I just know I've seen a few posts about keeping the same split.

I had the same concern over uncapped yellows thats why i went with dw's but the latencies have taken a little playing with. I have to get home and get my car running then try the multiplier we were talking about. I think it is going to work well.

slash128

I'm interested in your results!

logalinipoo

i checked my tune and 1024*ms is exactly what I'm running with the scaling set to flow rate at 60 psi.

That put them at a nearly identical tune as i had on stock injectors.