Author Topic: HE351VE in a 4g63  (Read 12466 times)

Offline EvoAuto

  • Member
  • **
  • Posts: 28
    • 1995 Eagle Talon
Reply #15 on: July 26, 2015, 06:59:42 PM
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? 

I'm at about 12psi at 4000RPM. I will log a bit more this week.



Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #16 on: July 26, 2015, 10:44:02 PM
If the turbo doesn't sound like a turbine, move the curves up again by 20

And if you can, save a pull from the serial output to a txt file and upload it for me to read

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline EvoAuto

  • Member
  • **
  • Posts: 28
    • 1995 Eagle Talon
Reply #17 on: July 26, 2015, 10:59:49 PM
I will give it a try. I was planing on also trying what you suggested before regarding the wastegate. Once i find out what turbine speed corresponds to what boost pressure I'd like to have the controller try and regulate the boost. Where do I enter the max shaft speed?



Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #18 on: July 27, 2015, 08:03:28 AM
It's already setup to max out at 130,000, which is about 40psi on my vehicle. You could curve it down to 80k.. etc

It is near the bottom section of the calculate curve tab

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline EvoAuto

  • Member
  • **
  • Posts: 28
    • 1995 Eagle Talon
Reply #19 on: July 28, 2015, 02:41:32 AM
Did some more tuning this evening. Got the turbo to 13 psi by 3200 RPM's in 3rd gear. I'm not sure why the turbo's position does not change at all regardless of shaft RPM. At 14 psi the shaft is spinning at around 85000 RPM's at position 800. Here is a log as well as the modified LLB code.



Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #20 on: July 28, 2015, 08:29:55 AM
Code: [Select]
* Turbo RPM Curve : 4
   * 4000 - 9000  : 40 - 900
   * 9000 - 17000  : 900 - 840
   * 17000 - 26000  : 840*
   * 26000 - 39000  : 840 - 800
   * 39000 - 1464  : 800 - 740
   * 1464 - 105000 : 740 - 700
Done.

ran out of room in the int on the last curve rpm.

change the curves to
Code: [Select]
unsigned int curve_rpm[5] = { 9000, 17000, 26000, 39000, 65500 };

That'll fix that for the most part.

Performance mode sets vane position to current curve position + the position from JP1. * When calculated vane position + performance is not at least 800, vane position is set to 800. That's why you aren't going to see it change vane position at all. Keep it out of this mode until you feel like the turbo curve is right.

Please for the sake of tuning this turbo, don't use performance at all, the curve isn't tuned well at all and having perf mode on isn't going to do anything until the curve is stable.


The turbo isn't hauling until it is @ at least 110,000rpm, you need to be able to get it there without it holding your motor back.

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #21 on: July 28, 2015, 08:47:20 AM
Here's an update for you..

I've included a lot of other changes I've made, like the accel variable is in.. etc...

I moved your curves around, let me know how it feels in the car accelerating. Is the turbo holding it back or is it just not spooling fast enough, you can find out by toggling the cruise switch and if you feel the car accelerate the moment you hit the switch, the vane positions are too constricting.

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline kapok6

  • Member
  • **
  • Posts: 21
    • 1991 GSX
Reply #22 on: July 28, 2015, 10:57:10 AM
Keep the updates coming!



Offline EvoAuto

  • Member
  • **
  • Posts: 28
    • 1995 Eagle Talon
Reply #23 on: July 28, 2015, 11:41:14 PM
Here's an update for you..

I've included a lot of other changes I've made, like the accel variable is in.. etc...

I moved your curves around, let me know how it feels in the car accelerating. Is the turbo holding it back or is it just not spooling fast enough, you can find out by toggling the cruise switch and if you feel the car accelerate the moment you hit the switch, the vane positions are too constricting.

Funny you say that actually. Tonight while driving home i decided to leave it in performance mode (position:800) and did a full throttle pull; At 4500 it was at about 15psi when I flipped the switch to cruise (p:300) and the boost just kept climbing and fast. At 6000rpm it was reaching 25psi. My wastegate couldn't control the boost at all.

I will upload your new code tomorrow and try it out. 'Idle' mode is what I should be in correct?



Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #24 on: July 29, 2015, 08:04:51 AM
Ya it should be in idle, when turbo rpm is under idle rpm, and it should walk down if turbo rpm goes over walkdown rpm once, then rpm ever goes under it

Also remember that there is priority in the modes.

Cruise > Brake > Performance.

So with Perf on, you can't get into cruise without turning perf off.

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline Rx7man

  • Crazy guy
  • Hero Member
  • *****
  • Posts: 1065
  • Jack of all trades
    • '94 Dodge dually
Reply #25 on: July 29, 2015, 11:15:52 AM
Isn't it a little redundant to have a wastegate as well as a VGT turbo?  At this point it might be a simple way to control the boost, but you shouldn't need it at all

'94 dodge, Colt Cam, head worked, NV5600, 3.55's, SB clutch, Ford manual hubs w. hawk brakes, custom flatdeck, custom aluminum bumper, cab off restoration, totally new engine, Mahle pistons @19:1..HE351VE with 67mm turbine and 67x95mm compressor
440hp/1000 ft lb @ 2400 RPM


Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #26 on: July 29, 2015, 11:16:51 AM
I tried to let him know during fab that it is not necessary. For compounding it is necessary but I don't think he'll be compounding.

No problem thou, just mount a block off plate.

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.



Offline EvoAuto

  • Member
  • **
  • Posts: 28
    • 1995 Eagle Talon
Reply #27 on: July 29, 2015, 11:45:23 AM
I had a different setup in the car before that's why I have a Watergate. I plan on using it for knock control since my ECU does not communicate with the arduino board.



Offline Rx7man

  • Crazy guy
  • Hero Member
  • *****
  • Posts: 1065
  • Jack of all trades
    • '94 Dodge dually
Reply #28 on: July 31, 2015, 10:51:28 AM
That makes sense.

Your ECU could communicate with the Arduino in a limited fashion.. you could have the wastegate control hooked up to an input on the Arduino, and some added logic could open the VGT when it gets the signal that would normally open the wastegate... just a thought

'94 dodge, Colt Cam, head worked, NV5600, 3.55's, SB clutch, Ford manual hubs w. hawk brakes, custom flatdeck, custom aluminum bumper, cab off restoration, totally new engine, Mahle pistons @19:1..HE351VE with 67mm turbine and 67x95mm compressor
440hp/1000 ft lb @ 2400 RPM


Offline hakcenter

  • developer
  • Hero Member
  • *****
  • Posts: 759
    • 1998 12v Ram
Reply #29 on: July 31, 2015, 01:13:50 PM
Not on a dsm,  there's no real output. It communicates at 1953 baud, etc

Only way would be to interface the boost gauge output and make it a knock gauge. Then measure the output on an adc on the arduino.

AMP 2 Corinthians 8:21
For we take thought beforehand and aim to be honest and absolutely above suspicion, not only in the sight of the Lord but also in the sight of men.