HE351VE in a 4g63

Started by EvoAuto, July 15, 2015, 12:39:07 AM

EvoAuto

Hi
I just finished installing the turbo. Curtis was nice enough to build me the board and answer all my questions. I need some help fine tuning it for my setup. The engine is 2.3L 4 cylinder. The default values are not the greatest. It's starting to spool at around 4000rpm.

Curtis has instructed me to change the default curve_rpm values to 9000 17000 26000 39000 67000
I will be uploading this tomorrow when I finish work tomorrow and update you guys.

I also need to learn how to change the "performance" settings. I was able to change the "cruise" setting to find the sweet spot at idle. It seemed to spin the shaft the fastest at position 700 at idle. I went from position 40 all the way to 960 in roughly 100 increments. At position 40 it's spinning at 4000rpm's; at 700 its spinning at 6500rpm's; at 960 it is spinning at around 5000rpms. I've also noticed the exhaust note is the quietest at position 960.

Thanks for reading

Ryan

#1
Please continue to update this thread as I have been talking with curtis for quite a while now about retrofitting one of these on my auto awd 1G.

hakcenter

I haven't really put the code in the wiki to explain sections but I'll try to make mention of how most of the stuff works now.

Normal operation includes idle.
Cruise is a mode
Performance is a mode

During idle, which is anytime the turbo is under idle rpm, or rpm has past the idle walkdown rpm, and is under that value, then it will attempt to walk the turbo rpm down back under idle rpm. This area is tricky, cause you can get it stuck on the first curve value if idle and walkdown aren't tuned properly

Cruise was made for freeway operation in mind. It just attempts to get turbo rpm around 50,000. If turbo rpm is under that, it uses a static cruise value which is a bit higher than you would expect, but will help off idle but still let the turbo eventually spool up. In cruise though the code is setup to just continue to open the turbo to prevent rpm going past 50,000 all the way up to 25cm if it has to.

Performance for the time being is quite simple. It is a static value added to the current curve. Configured by jp2 header position. If the current curve value + the performance increase is under vane position 800, it is static 800. Until turbo rpm finally reaches top_end_rpm which is set at 105,000


Outside of learning how it works, it will help to know your turbo idle rpm, when the motor is completely warmed up. Also what curve you're using: 1, 2, 3, 4 ?

Don't get discouraged, there is so much room to adjust the turbo, from the default values :D
TS2009 Deḇarim 8:2
"And you shall remember that יהוה your Elohim led you all the way these forty years in the wilderness, to humble you, prove you, to know what is in your heart, whether you guard His commands or not.

Rx7man

I am trying to convince my buddy to put one of these on his 3rd gen Rx7.. I think the logic may be fairly different on a gas engine, especially because of the throttle plates... might nearly want 2 map sensors?

curious to know how it works for you :)
'94 dually,  67/67 HE351VE, NV5600, ~600hp
'93 ECLB 47RH, new toy truck, H pump project, 1000hp goal, 300K miles
93 XCLB auto, bone stock, 350K miles
93 XCLB 5spd, bone stock, 100K miles

EvoAuto

Sorry I've been busy at work. Haven't had much time for testing. I took the car out for a drive. With my old turbo I had a pretty good tune. With this turbo and the W2A intercooler the ECU has never been in these cells before so it's running very poorly. I will be tuning when I have time this week and update you guys.

Curtis when I start logging with the Arduino it says I'm on curve 4.

Rx7man

What ECU do you have in the car for the engine?
'94 dually,  67/67 HE351VE, NV5600, ~600hp
'93 ECLB 47RH, new toy truck, H pump project, 1000hp goal, 300K miles
93 XCLB auto, bone stock, 350K miles
93 XCLB 5spd, bone stock, 100K miles

hakcenter

What is your turbo idle rpm when your motor is completely warm?
TS2009 Deḇarim 8:2
"And you shall remember that יהוה your Elohim led you all the way these forty years in the wilderness, to humble you, prove you, to know what is in your heart, whether you guard His commands or not.

EvoAuto

I am using AEM v1. I will be replacing it with an Evolution 8 ecu to run Ecuflash. In my opinion it is more user friendly. As for idle speed, it will no longer autoscroll since the last changes you sent me so I cant tell.

Rx7man

#8
A lot of settings may be a lot different on a 2.3 L gas engine than on a diesel... there's no doubt you can make it work, it may just take a bit more work... The throttle plate is going to throw a wrench in your spokes possibly as well.

*Maybe* you will need another map sensor between the turbo and throttle plates to deal with the turbo, and the one after the throttle plates for the engine mapping?... there may of course be workarounds
'94 dually,  67/67 HE351VE, NV5600, ~600hp
'93 ECLB 47RH, new toy truck, H pump project, 1000hp goal, 300K miles
93 XCLB auto, bone stock, 350K miles
93 XCLB 5spd, bone stock, 100K miles

EvoAuto

I don't think a second map sensor is needed for my setup. I have a water to air intercooler with almost no pressure drop from Turbo to manifold.

hakcenter

Make sure when you're making changes in the arduino IDE, that you save them, before uploading, so things are constant.


// Turbo
const int minimum_turbo_rpm = 4000;

to

// Turbo
const int minimum_turbo_rpm = 1;

^ Will make sure the serial output is continuous as long as the turbo is actually spinning.


const unsigned int idle_position = 480;

to

const unsigned int idle_position = 720;

^ smaller idle position, may need more


const unsigned int turbo_curve_1[5] = { 800,740,700,640,600 };
const unsigned int turbo_curve_2[5] = { 820,760,720,660,620 };
const unsigned int turbo_curve_3[5] = { 840,780,740,680,640 };
const unsigned int turbo_curve_4[5] = { 860,800,760,700,660 };

to

const unsigned int turbo_curve_1[5] = { 840,780,740,680,640 };
const unsigned int turbo_curve_2[5] = { 860,800,760,700,660 };
const unsigned int turbo_curve_3[5] = { 880,820,780,720,680 };
const unsigned int turbo_curve_4[5] = { 900,840,800,740,700 };

^ is moving the curves up by a 1cm^2


const unsigned int max_position = 940;

to

const unsigned int max_position = 960;

^ is going to let the turbo use 960, which in my testing was the absolute smallest the turbo would go physically. It might goto 980 /shrug, but you're going to need more room to close the turbo up.

I imagine your curve will probably include that position, especially if you've used it and it sounds nothing like my videos ie: turbo whining at idle.
TS2009 Deḇarim 8:2
"And you shall remember that יהוה your Elohim led you all the way these forty years in the wilderness, to humble you, prove you, to know what is in your heart, whether you guard His commands or not.

Rx7man

Quote from: EvoAuto on July 20, 2015, 12:57:12 AM
I don't think a second map sensor is needed for my setup. I have a water to air intercooler with almost no pressure drop from Turbo to manifold.
I'm not concerned about the pressure drop through the IC, I'm thinking at part throttle you're going to have a BIG pressure drop across the throttle plates, and if you try and spin the turbo too hard you'll surge it.. or the BOV is just going to be venting a whole lot depending on the setup
'94 dually,  67/67 HE351VE, NV5600, ~600hp
'93 ECLB 47RH, new toy truck, H pump project, 1000hp goal, 300K miles
93 XCLB auto, bone stock, 350K miles
93 XCLB 5spd, bone stock, 100K miles

hakcenter

I'm sure he has a blow off, most gasser turbo cars have them. And since its rpm the rotor should have inertia and not spin down incredibly fast and bark.
TS2009 Deḇarim 8:2
"And you shall remember that יהוה your Elohim led you all the way these forty years in the wilderness, to humble you, prove you, to know what is in your heart, whether you guard His commands or not.

EvoAuto

Went for a tune last night. The car pulls very hard at 10psi. After about 30 minutes of driving though the car started to blow blue smoke. I limped it back to the shop thinking it was the engine but when I started this morning there was no blue smoke at all. Did a compression and leak down test. Engine checked out ok. I think the oil pressure might be too high coming from the oil filter housing so I decided to feed the Turbo from the head instead. I also have a poorly positioned drain due to the controller. I'll report back soon. I hope i didn't blow the turbo's seal. Very excited though the Turbo is spooling around 4000rpm and pulls very hard.

Ryan

When you say "spooling", what do you mean?  It's just starting to spool and build boost at 4000, or it's at "full" boost (10psi) by 4000?