Installed my 16g from a 13b. Have a few questions.

[EagleTalonTim]

I went to the junk yard yesterday and found a turbo car with a TD05H turbo in it. I just had to buy it. After purchasing it, I compared it with the 16g that I already had. For some reason the intake side turbine blades were different even though they both said TD05H on the housing. The one from the j/y had the same size blades all the way around, but the one that I bought from someone had alternating blade sizes. What is the difference and which is better?

After installing it, I took it for a test drive with my logger and set the boost at 14psi. I did notice that it takes way longer to spool up than the 13g, but that was expected. It does pull way harder than it did though. The biggest question I have is.....
Is it safe to be running hard on the 16g every once in awhile when I still have the stock fuel pump with a re-wire, 450 injectors, and a stock FPR? When I took it for a test drive, everything was ok, I still have knock just on take off and sometimes while cruising, but at full boost, 0 knock all the way to 6500 RPM.

Thanks for the help in advance!

 

[PIMking]

if your running stock boost I dont see why not.

I run 15-16 psi on stock fuel system. but my injectors and pump are on the way.

 

[rarson]

You had a 13g and you have 450cc injectors? So you're running an auto car and you swapped the injectors for 450's?

I'm not trying to insult your intelligence or anything, I'm just trying to make sure your bases are covered. I wouldn't want to run too much boost on the stock auto injectors.

With 450's, you should be good to 15 psi or so. You'll hit fuel cut before you can max them out (unless you're leaning it out with an AFC or something).

 

[EagleTalonTim]

Yes, I installed the 450 injectors and a M/T ECU to run the 450's in my A/T Talon. I feel safe running 14psi but I still have a problem with knock from a dead stop. I need to find an EPROM ecu sometime soon so I can get the DSMlink. I would love to tune out knock below 3k. When I went for the test drive, the lag was very noticeable. I did not port anything, but now I think I should have. If I got bigger injectors and a fuel pump, would that make the take off faster? Since it is an A/T, I really want to look into getting the take offs alot faster. I have no fuel management system, so I have very little to work with.

 

[kenamond]

And did you upgrade to a 1g M/T mani or 2gt mani? As I recall, the 13g mani has a much smaller collector diameter than the 14b/T25 manis.

Sounds like you have a logger. Watch for knock and low front O2v.

IIRC, 1g A/T cars have 390cc/min injectors but higher base fuel pressure (like a 2g...like 43.5psi or so) while 1g M/T cars have 450s and like 37psi base fuel pressure. So I'm not sure how that translates into maximum safe boost for the A/T fuel system...higher fuel pressure, smaller injectors vs. lower fuel pressure, larger injectors...which wins, I don't know.

As I understand it, the same housing can "house" different compressor wheels. Same goes for the turbine housing. The housing can be machined on the inside for a different wheel. My FP Big28 has the same housings as a T25 but different wheels.

 

[EagleTalonTim]

After doing some researching, I found what the difference in the turbine blades are. The turbo from the j/y is a 14b, and mine is an Evo III 16g. Yes, I did swap the manifold to a 1g M/T as well. I bought it all for $60....could not beat that since my 13b was blowing anti-freeze out the exhaust. One thing i never did was change the FPR from the auto to the m/t. would that cause the knock on take off?

 

[dsm03]

I Thought its not recommended to raise more than stock boost if you dont have fuel upgrades..isnt stock boost 10psi?? So, is it ok to swap into a 16g but as long as you dont run more than 15psi...

 

[kenamond]

After doing some researching, I found what the difference in the turbine blades are. The turbo from the j/y is a 14b, and mine is an Evo III 16g. Yes, I did swap the manifold to a 1g M/T as well. I bought it all for $60....could not beat that since my 13b was blowing anti-freeze out the exhaust. One thing i never did was change the FPR from the auto to the m/t. would that cause the knock on take off?

I'm confused. You have a 13g, a 14b (from j/y) *and* an e3b16g? Which did you replace the 13g with?

Don't change the FPR. That'll drop your fuel pressure to 37ish which will lean you out bigtime.

Not sure what's causing your knock. Might want to post some logs for folks to stare at.

 

[kenamond]

I Thought its not recommended to raise more than stock boost if you dont have fuel upgrades..isnt stock boost 10psi?? So, is it ok to swap into a 16g but as long as you dont run more than 15psi...

Many get away with 15psi on stock fuel (maybe a rewired FP), but nobody should do this without a logger. You might luck out, you might cook your motor.

 

[EagleTalonTim]

yes, I have all 3 turbos.... 13b, 14g, and an EVO III 16g. I swapped the 13b with the 16g. What should I record on the logs to ensure all areas are covered when posting it? I am using my manual shift box to shift so I can do a 1st, 2nd, 3rd, or 4th gear pull if needed.

 

[EagleTalonTim]

Hopefully you can read this log. It is the first log I took when I went for a test drive with the 16g.

 

[kenamond]

yes, I have all 3 turbos.... 13b, 14g, and an EVO III 16g. I swapped the 13b with the 16g. What should I record on the logs to ensure all areas are covered when posting it? I am using my manual shift box to shift so I can do a 1st, 2nd, 3rd, or 4th gear pull if needed.

I think 1g guys log rpm, timing advance, knock, front O2v. But nothing's jumping out at me about your knock problem.

You might check your fuel trims during cruise and see if they're out of spec. A 1g guy might have to correct me, but on 1gs I think 100% is perfect. On 2gs, you should be within +/-4% or so. Not sure about the +/- on a 1g. I'm also not sure if you can log the different long-term trims separately with your logger or if it just shows you whatever trim is currently in effect. If you have one LTFT trim on your logger, it'll automatically show you the trim the ECU is currently using to correct the fuel map value.

Could also be a bad knock sensor or phantom knock. I don't know a whole lot about it, but I've heard folks say the Mitsu knock sensors could be better. Also, if brown goo is coming out of the knock sensor, it's shot. I'm not clear on all the possible sources of phantom knock (knock signals that aren't actually pre-ignition or detonation). Might want to search around.

So given all I've typed (I'm in a typing mood), I don't know if I have helped.

EDIT: Forgot to ask...how many counts of knock? Knock of 2 or below on a 2g does nothing to timing.

 

[rarson]

Am I just retarded, or can I not enlarge the picture you posted? I'm using Firefox with Adblock and Noscript and Peer Guardian and blah blah blah so it's sometimes a headache when I run across stuff off the web. Speaking of headaches, browsing the web on a 36" SDTV kind of hurts after a while with the fuzzy text.

Anyway, the EVO III 16g is probably pushing a bit more air than the standard 14b or 16g. You might want to try keeping the boost down a bit. From what I can see of the tiny numbers on the log you posted, it doesn't look like you're getting much knock. I was never too educated about 1g knock values, but I'm pretty sure they have to get into the double digits before knock becomes significant (if you're just logging raw knock counts). I wouldn't worry about it too much. But it's also possible that you're running a bit too rich using the higher base fuel pressure of the auto FPR (assuming kenamond is correct and the auto is higher... I personally don't know anything about 1g auto fuel pressure). So the lag could be a bit of a rich bog, but also keep in mind that you're going from a 13b (not sure if you ever felt it running right, but I'm assuming it would spool quite a bit faster) to the largest 16g there is. It might just be the difference in lag that you're noticing. Adding fuel won't help the situation. If anything, you'd probably want to lean it out some, and that might help acceleration, since you're running a higher base pressure, and DSMs tend to be kind of rich anyway.

As far as porting goes, here are my thoughts:

Theoretically, porting the turbine housing might make the turbine spool slower, as you're basically giving it a little more room to breathe. Practically, I can tell you that porting a 6cm2 housing really wakes the bastard up. It's possible that the max power increase just makes the car feel faster, but it's also possible that the porting helps make more power across the board. I don't know, but I will definitely say that I think anyone running a 6cm2 exhaust housing should port the sucker and brace for some good times. There's really no good reason not to port it if you have the opportunity. Go ahead and do the exhaust manifold outlet and O2 housing while you're at it. Personally, I think the 6cm2 housing is a little choked for a good 16g setup.

 

[kenamond]

Heh...you posted a log while I was responding.

Go do a search and find out what knock count threshold causes pulled timing in a 1g. The 2g doesn't retard timing until it goes past 2 counts, and there's a formula to compute the retard from knock.

If the 1g is the same, you only see 2 counts in that log, so it wouldn't affect *anything*. Find out for sure what the 1g situation is. If the 1g is similar and you're only seeing 1-2 counts of knock when cruising, you can ignore it.

I agree that you should port out the mani and turbine housing to match a 7cm^2 gasket. One problem is that you have to shitcan the sealing ring. This puts more of a demand on the gasket, and I've seen some of them do very strange things (I think romeen posted something recently where the gasket warped *bigtime* and was actually puckered in on the sides and obstructing exhaust flow...after 3k miles). If the gasket doesn't seal or stops sealing, you get pre-turbo exhaust leaks which can hurt performance.

Some folks think porting is overrated, but I think that most of this is related to porting the WG entrance in the housing to battle boost creep (which you might encounter if you go to a catless 3" turboback).

From my modest understanding of 1g LTFT (long term fuel trim) values, yours look good for the Lo and Mid. The Hi might be okay, too (+5%), but maybe it's a bit lean for 1g standards. I'm "pretty sure" a value of 105 on a 1g means that the ECU will add 5% fuel when airflow per rev (load) is in the "Hi" range which means it's running a bit lean.

Finally, I'm not as familiar with "typical" 1g O2v values, but as I recall, they're lower than 2g values (2g O2v is normally in the 0.96+ range for stock tune, but 1g values are - as I vaguely recall - in the 0.8 range). Check into it and find out what a good O2v value is typically (but keep in mind that narrowband O2v are extremely unreliable for interpreting AFR anywhere but stoich). As an option, you can run WGA boost (vaccuum line straight from j-pipe to WGA) and log it to see what O2v you get there. Hopefully, that'll be the O2v you want when all's well-and-good, because you'll be running 8-10psi boost where the injectors should be able to easily keep up with the airflow. Then you can up the boost until your O2v drop from that value and/or knock becomes a problem...then drop the boost a bit.

 

[rarson]

Yeah, I always thought the 1g guys were crazy running .8x volts because I'd lean the car out like crazy and never get below .9V.

 

[EagleTalonTim]

so I take it, I am safe for now till I get some sort of fuel management system, injectors, and fuel pump. I will run at 14psi for a few days and see how things look with the logger to see if I am running lean at any point. I would really like to learn about the fuel trim numbers and what they mean.

 

[kenamond]

so I take it, I am safe for now till I get some sort of fuel management system, injectors, and fuel pump. I will run at 14psi for a few days and see how things look with the logger to see if I am running lean at any point. I would really like to learn about the fuel trim numbers and what they mean.

As I mentioned in my previous post, you need to verify the 1g behavior w.r.t. knock. 2g ECUs will not pull timing unless you get more than 2 counts of knock. A 1g might be different. You get to do your own homework on that one

Trims:

The ECU has a fuel map. It stores the target amount of fuel to deliver for every combination of airflow and rpm. The MAS and CPS provide the airflow and rpm values, and the ECU looks at the map at that airflow and rpm and gets a fuel value. It injects that amount of fuel. But this is only the "open loop" mode of the ECU. It does this when you're WOT.

When you're not at WOT, the Mitsu guys made a safe bet that you could safely run 14.7 (stoich) AFR without damaging the motor. This is where closed-loop and learn modes come into play. The way the ECU does this is as follows. The front O2 sensor will put out 0.5v if you're running right at stoich (AFR=14.7). The ECU adjusts fuel until the O2v signal oscillates back and forth around 0.5v. If it stays above 0.5v, the ECU knows it's running a bit richer than 14.7. If it stays below 0.5v, it knows it's running a bit leaner than 14.7. So it gets a fuel value off the map just like in open-loop mode, then looks at the O2v signal and decides if it's rich or lean or perfect. If it's rich, it adjusts a short-term fuel trim (STFT) to be a bit smaller. If it's lean, it adjusts STFT to be a bit larger. This STFT value is added to the fuel from the map and injected. If the O2v start oscillating at 0.5v, the ECU is happy. If conditions are right, the ECU will be in learn mode and will then "memorize" the STFT value in a long-term fuel trim (LTFT). If the airflow is small, it gets stored in the LTFT-Lo, if it's medium airflow, it goes into LTFT-Mid, etc. After it copies the value over from STFT to LTFT, it sets STFT back to a "don't add/remove anything" value (which is different between 1g and 2g). I lied before a bit...when the ECU calculates fuel, it looks at the map, gets a fuel value, then adds STFT *and LTFT* corrections, then injects that fuel. It will grab either the LTFT-Lo/Mid/Hi depending on airflow and add it to the current STFT and add that to the fuel map value, then inject that fuel. Then the ECU goes about the same process of looking at O2v, correcting for rich/lean via the STFT, then memorizing or learning the STFT by copying it into the LTFT and resetting STFT. So while you're in learn mode, the STFT and LTFT values can be constantly changing. But usually, the ECU figures out a good LTFT set, and it doesn't change.

So if something's whacked out on your car during cruise (closed-loop and learn modes) and this causes you to run lean or rich, you'll see it in your LTFT values. If you have a clogged injector or fuel filter, you'll run consistently lean, and your LTFT will be above 100 (has to add more fuel). If you run larger injectors without a fuel controller, your LTFT will be below 100 (has to remove fuel).

In your log, your Lo and Mid LTFT look almost perfect, but your Hi looks a bit on the lean side (105).

On a 2g, the trims are perfect if they're 0. So negative is removing fuel from the map value while positive is adding fuel to the map value.

Hope that helps.

 

[EagleTalonTim]

wow, that was really informative! Thanks! Is there anyway to adjust the high side fuel trim without a dsm link or an SAFC? What is the most effective, less expensive way of being able to modify fuel trims? My take-off really suck and I want to get them way better. I did find that on a 1g, each count of knock is 3* to 5* timing pull. My average on take off is 2 to 7 counts. It happens just as I press the gas pedal. If I let off, the knock goes away and I can mash on it again without any knock. Strange......

 

[kenamond]

wow, that was really informative! Thanks! Is there anyway to adjust the high side fuel trim without a dsm link or an SAFC? What is the most effective, less expensive way of being able to modify fuel trims? My take-off really suck and I want to get them way better. I did find that on a 1g, each count of knock is 3* to 5* timing pull. My average on take off is 2 to 7 counts. It happens just as I press the gas pedal. If I let off, the knock goes away and I can mash on it again without any knock. Strange......

The ECU is constantly adjusting the trims. You can't change them, because they're not constant. I'm not positive, but I'm pretty sure the trims are not used in open loop mode. I'm out of ideas for the knock you get at the start of a pull. I'd search for phantom knock, and you'll probably find lots of threads discussing mysterious knock. That ought to give you some things to check.