Evo 8 injector test results

[Dxc545]

I got a real good deal on evo8 injectors and got them back from the cleaners today. All 4 were pretty gummed up and i gained 16% flow back with 640cc?s I thought they were only 560s?

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[twdorris]

Tom what were the results of your test? Very interested...

If you're asking about the EVO8 injector testing on a DSM ECU, I don't think we ever got that set in for testing.

Thomas Dorris

 

[gsxeclipse97]

Well they could be the source of my problem then. I have them scaled to 565 in ecu flash but if they are flowing more like 650 then that why the car has trouble starting.. That's almost 100cc's more than they are factory rated.

 

[Diamond Star Werks]

If you're asking about the EVO8 injector testing on a DSM ECU, I don't think we ever got that set in for testing.

Thomas Dorris

Sorry had to use them. I now have an extra set though. If your still interested in testing them I can send them in.

 

[twdorris]

Sorry had to use them. I now have an extra set though. If your still interested in testing them I can send them in.

Yeah, send 'em on.

Thomas Dorris

 

[t25fury]

Thomas, is this only for this test or is injector cleaning/balancing/blueprinting a service you guys offer?

 

[Diamond Star Werks]

Yeah, send 'em on.

Thomas Dorris

Okay will do.
Actually my trans is broke so I can send in two sets for side by side comparison if you like.

 

[twdorris]

I can in teri sets for side by side comparison if you like.

I think I understand what you're saying. And, yeah, if you want, that would be even better.

is injector cleaning/balancing/blueprinting a service you guys offer?

Yeah, sorta. It's not listed on our website because I was never SURE it was making a difference. For some people, it seemed to help a lot knowing the individual deadtime values for each injector. Full V3 provides a mechanism for adjusting that per cylinder. But for other people, it didn't seem to make much difference either way.

What we do is just testing, though. We don't do any cleaning here. We test the injectors under various conditions and then produce a report like the one below.

You must be logged in to view this image or video.

Thomas Dorris

 

[Diamond Star Werks]

I think I understand what you're saying. And, yeah, if you want, that would be even better.


Thomas Dorris

edited sorry.

 

[Diamond Star Werks]

Sounds good.

ECMTuning, Inc.
1519 W Patrick St A10
Frederick, MD 21702

I'll post up results either to this thread or a new thread. If the OP wants a "retest" of his injectors with DSM drivers and defined fluid/temperature/pressure along with a full battery voltage offset table, send them in as well...

Thomas Dorris
ECMTuning, Inc.

Both sets were just set off at the part office coming to you.

 

[goldeninja20]

There's so much myth and mystery surrounding EVO8 injectors...

The EVO8 ECU code is configured to treat them like they flow 545 cc/min @ 42.6psi (base pressure on EVO8, 294 kPa). So at 43.5psi, they should flow 550 cc/min. I can't imagine a stock set really flowing 640 cc/min without some fudge factors being applied or the testing temperature being raised or the fluid being less viscous than gasoline or some combination of all these things...

I'd really love to clear this all up. Someone send me a set of stock EVO8 injectors and let me test them with actual DSM and EVO8 drivers using compatible fluid and temperature controlled conditions and let's see what numbers we really come up with.

Also, the battery offset numbers (deadtime, dwell, whatever you want to call them) in the EVO8 ECU are lower not just because of the injectors, but because of the differences in the driver circuitry on the EVO8 vs. the DSM ECU. In fact, that's actually the *main* reason you find lower offset numbers in the EVO8 ECU than you'd expect.

So moving EVO8 injectors over to a DSM doesn't mean you can just drop the offset numbers to match. You can't. Well, OK, you can because they're still pretty close and minor fluctuations in fuel trim will correct anyway. But you're still using the DSM driver circuit with a slower drive time than the EVO8 circuit (primarily due to the presence of a resistor/capacitor configuration on the output of the DSM driver circuit that's missing on the EVO8). So you'd expect to need slightly higher offset values in a DSM ECU than you'd use in an EVO8 ECU for the exact same injectors.

Driver circuit differences are the main reason you can't just take injector "deadtime" numbers off flow sheets and use them in our ECUs. You have to test with our ECUs and our drivers to get those kinds of numbers.

Lots of info here if you're interested in reading.

injdata-overview [ECMTuning - wiki]

Thomas Dorris
ECMTuning, Inc.

My evo 8s have to be scaled to 545cc in order for them to reach the target A/F ratio in ecuflash. With them set at 560cc i was about 1.5 too lean in boost in comparison to my target A/F ratios. I dont know if that means they are in desperate need of cleaning.

 

[gsxeclipse97]

My evo 8s have to be scaled to 545cc in order for them to reach the target A/F ratio in ecuflash. With them set at 560cc i was about 1.5 too lean in boost in comparison to my target A/F ratios. I dont know if that means they are in desperate need of cleaning.

What did your wide band read? I have mine scaled to 565 because it wouldn't allow me to use 560. I'm also using a evo 9 maf which from what I have read runs leaner than the evo8 even with maf scaling done.

 

[Jon91TSi]

I think I understand what you're saying. And, yeah, if you want, that would be even better.


Yeah, sorta. It's not listed on our website because I was never SURE it was making a difference. For some people, it seemed to help a lot knowing the individual deadtime values for each injector. Full V3 provides a mechanism for adjusting that per cylinder. But for other people, it didn't seem to make much difference either way.

What we do is just testing, though. We don't do any cleaning here. We test the injectors under various conditions and then produce a report like the one below.

You must be logged in to view this image or video.

Thomas Dorris

Tom,

Any interest in testing a set of FIC 2150's?

 

[dskagr]

Tom,

Any interest in testing a set of FIC 2150's?

If you guys haven't tested ID 1000's I can send some in as well.

 

[xxNocturnalxx]

My FIC 1050s flow tested at FICs shop at 1140cc, just thought I'd throw that out there.

 

[twdorris]

Some results for these EVO8 injectors. First, a huge THANKS to MJcanada! You may not appreciate how difficult it is to get people to follow through on offers to send stuff in sometimes. So when someone does, it deserves a special mention.

He sent two complete sets of EVO8/9 injectors. So I had 8 injectors total to test with.

What I found kinda surprised me a little bit, but before I get into that, let me explain some stuff about injector testing. It is NOT possible to test injectors using ALL the same parameters as will be in effect when the injectors are physically installed in a car. You can get an idea of some of the different factors that come into place here:

Injector testing overview

And I'm sure that's not all of it. I'm sure there are more things to consider that we haven't even collected data on. Suffice to say, there are probably a dozen variables all interacting with each other to produce a final flow rate number. It's impossible for me to say "this injector flows 550 cc/min" and that be the end of it. There are just too many things to factor in. I might test with one particular fluid at one temperature and get a result that is off by as much at 5-10% from what you might see in your actual car with a different fluid at a different temperature.

So you can't take flow numbers as gospel truth. Mitsubishi did their testing using equipment FAR cooler than what we could dream of building and I'm sure they had some very specific test parameters that they controlled very tightly to determine the numbers they put into the DSM and EVO ECUs for injector flow rate. And even those numbers will be "wrong". Mitsubishi can control the parameters of their test, but they can't control the exact parameters of all the different fuels and temperatures these cars will see in the field.

That's why we have fuel trims. Trims take up that slop and account for any number of conditions that deviate from test conditions.

The best I can do is to normalize my testing rig so that stock DSM injectors produce numbers that match the factory specs and then measure other injectors using that setup. That's what I did here. Stock DSM injectors using a DSM ECU driver module flow 450 cc/min on our stand and have a deadtime right at 680 usecs when tested at 43.5 psi. So that matches perfectly with the DSM ECU setup.

Using that setup, I get EVO8 injectors flowing at 565 cc/min with a DSM driver module and 580 cc/min with an EVO8 driver module (lower deadtime with the EVO driver allows for slightly higher sloped flow). Deadtime at 14v and 43.5psi is 690 with a DSM driver module and 590 with an EVO8 driver module.

Here's a picture of one of the test results where I ran an EVO8 injector using an EVO8 driver module and a DSM injector using a DSM driver module and an EVO8 injector using a DSM driver module all at the same time.

You must be logged in to view this image or video.

What I found interesting is that 2 of these 8 injectors were considerably different from the other 6. I never expected factory injectors to vary so much, so I actually assumed my test setup was off. I probably spent the better part of a day working on that until I finally came to the conclusion that those two injectors were just different. Instead of flowing in the 565 range, they were more in the 585-590 range! Deadtime was different too. So even with bone stock injectors, you can see at least a 3% variation in flow rate.

Anyway, there's the data. If I were running EVO8/9 injectors in a DSM with about 43.5psi base fuel pressure (like a 2G DSM, for example), I'd use the following settings in ECMLink.

Global fuel scale: -20.3%
Global deadtime adjustment: 0 usecs
InjBatAdj (direct access):
7v: 2295
9v: 1395
12v: 885
14v: 690
16v: 555
19v: 495

I'm sure the global fuel scale can be tweaked up or down a bit to account for your slightly different fuel setup. Or you can just let the trims run a bit and once you've gotten a feel for that, move the resulting trim into the global fuel scale.

I hope that helps put to rest any claims of 600+ cc/min stock EVO8/9 injectors. They're more in the 545-565 range based on our testing and the factory EVO8/9 ECU code.

Thomas Dorris
ECMTuning, Inc.

 

[gimmie11s]

So evo8 560s are, well..... Actually 560s.

Excellent info though and as always... Thanks for taking the time on this.

 

[Shaggy_357]

Awesome info! Now, being a math reject, could anyone break down the Link settings based on 1g fuel pressure for me? I wouldn't mind doing the legwork if anyone would provide a formula in layman's terms. Thanks again TWDORRIS for doing this and to the guy that sent him the injectors!

 

[Diamond Star Werks]

Anyway, there's the data. If I were running EVO8/9 injectors in a DSM with about 43.5psi base fuel pressure (like a 2G DSM, for example), I'd use the following settings in ECMLink.

Global fuel scale: -20.3%
Global deadtime adjustment: 0 usecs
InjBatAdj (direct access):
7v: 2295
9v: 1395
12v: 885
14v: 690
16v: 555
19v: 495

I'm sure the global fuel scale can be tweaked up or down a bit to account for your slightly different fuel setup. Or you can just let the trims run a bit and once you've gotten a feel for that, move the resulting trim into the global fuel scale.

Thomas Dorris
ECMTuning, Inc.

Anytime I can help let me know!

Also this info should be added to your wicki page as it currently says you ran your evo injectors @ global 17.9%. Also I "think" the deadtime values are alil diff,as well.

 

[m_0ney_pit]

I'm just curious why the difference in latency between the Evo 8 and DSM injector drivers? More power required to open and close at lower flows the DSM driver can't provide?

Is there any danger to the DSM driver running the Evo 8 injectors?

 

[twdorris]

It's simply a difference in the specific drive transistor used in the EVO8 ECU vs. the DSM ECU. There is no risk running an EVO8 injector with a DSM ECU. You simply need to use dead time values measured with a DSM ECU driver.

 

[justacanuck]

Just to clarify, do the 1G and 2G ECUs have the same drivers for the injectors? Can we use the values above on a 1G DSM ECU, if using the Evo injectors but with 2G fuel pressure of 43.5psi?

 

[twdorris]

Just to clarify, do the 1G and 2G ECUs have the same drivers for the injectors? Can we use the values above on a 1G DSM ECU, if using the Evo injectors but with 2G fuel pressure of 43.5psi?

Yes and yes.