0-5v input

  • 16 Replies
  • 381 Views
*

Offline farmerz24

  • Member
  • **
  • 3
0-5v input
« on: March 09, 2017, 10:46:44 AM »
Can this be setup to take a 0-5V duty cycle input and linearly control the vein position based on the 0-5v input?

I would be controlling this from a megasquirt on a gas application.  So the control would be just based off target boost.
Ideally, a switch would also allow direct vein control from a pot when in this mode.

if so i am interested to get one.

*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #1 on: March 09, 2017, 06:22:39 PM »
It's based on an arduino so you could control the controller from mega squirt but I highly suggest against it.

Shaft speed is the ideal control, if you build off that with extra io then yeah sure. It is designed as a standalone so it really doesn't need your mega squirt input to make a logical decision on what to put the veins at.

I would suggest watching my YouTube video where I show you the difference between shaft speed and boost based on vein / braking.

Anyone here can tell you just boost control isn't going to work well.

*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #2 on: March 09, 2017, 06:54:10 PM »
Youtube video here

*

Offline farmerz24

  • Member
  • **
  • 3
Re: 0-5v input
« Reply #3 on: March 09, 2017, 07:16:56 PM »
Ok. Yes I see now how shaft speed affects it. I do like how it is acting as a standalone to control the turbo as there is a lot of logic for shaft speed you have built into there.

I would still need to feed it a target boost level from the mega squirt via a 0-5v signal or something. Perhaps even having 0v default it to full open or have it enter cruise mode at 0v. As this is a gas application I would not need the brake feature.
Then I could use my boost table in the megasquirt to control target boost as well as have it enter cruise mode during cruise and decell. Once I start commanding boost around and above 0vac I would have it enter performance mode automatically also.

It sounds like this is possible and the way to go. Let me know your thoughts

*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #4 on: March 10, 2017, 05:53:31 PM »
The switches are ground switches.

Hopefully your megasquirt has I/Os for triggering things. Just pull the cruise switch down to ground when you want the controller to go into cruise. Pull the perf to ground to engage perf mode etc.

Target boost isn't going to work without a wastegate in a gasser unless you want max. Pretty much anything under 20psi is impossible it will eventually creep. Post that... I haven't experimented trying to get 'boost' at a reasonable rate, I'm more interested in shaft speed / flow. 90k rpm 20psi or 80k 25psi, I'll take the 90k it's making more power.'

The controller has 3 5v ADCs to program anyway you want. I just wouldn't suggest using target boost at all. If all you have is 0-5v outputs on your squirt, you could certainly program that into the lbb.

*

Offline farmerz24

  • Member
  • **
  • 3
Re: 0-5v input
« Reply #5 on: March 13, 2017, 10:35:04 AM »
I am sorry hackcenter but i still don't understand.

This will be a gas application so i will be using a wastegate to control peak boost.  Also i have a 3.9 gas motor which is quite large for this turbo so a wastegate will be needed regardless.

However, there must still be a target boost input into the black box from the megasquirt, otherwise what would the control variable be? If I want to hit say 15 psi boost, wont the vanes just close to the best possible spool position (without choking out the motor) and then stay in that position until we approach 15psi.  Once 15psi is reached the veins would begin to open.  If this was controlled with a PID loop, then the control loop should be able to control this once all PID terms have been properly set.   In this case shaft speed is irrelevant is it not? The shaft speed would be whatever it needs to be to hit the target boost level...
 If the turbo was overspeeding then i could see the veins opening to lower the shaft speed of the turbo but otherwise they would be in 1 of two modes:

1.  Full open during cruise to minimize backpressure.  This would be true up until a certain throttle position in which we want to achieve boost (say anything above 50% throttle)

2.  When we are above 50% throttle and below our target boost level, the veins would be set to the best possible spool position.  Once we get close to our target boost level the veins would begin to open.

Again, this would be controlled by the MS's PID loop for boost control, giving the black box a 0-5v signal that would linearly scale the 0-5v to the vein position from fully closed to open.

*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #6 on: March 13, 2017, 11:29:30 PM »
Your megasquirt is a standalone injection system, it controls the injection based on how you program it.

The LBB / shield is a standalone turbo controller, it controls the turbo based on shaft speed also based on how you program it.

I hope that makes sense. The LBB does not need your megasquirt for anything. Target shaft speed is 115k, which is peak compressor efficiency. What boost that is, depends on so many other variables which comes down to motor VE, that honestly boost pressure is irrelevant. I can tell you that 115k-125k is about 60lbs/min of airflow so... ya it's a lot.

Now I get the issue with gassers, because of knock and all that. Which is why I would suggest a wastegate, to control total flow to the turbo which will inevitably lower dat powah.

I understand that this concept is very very new, but to control VGT you don't want to use boost pressure, unless you are not looking for peak compressor efficiency. Flow is what is important. My peak boost is 40psi by 110k rpm, which drops to 35psi by 128k rpm, I can tell you the 35psi, is a lot faster than the 40psi. All still within compressor rpm limits.

If a 351 is too big for your application maybe a 341 is a better choice (smaller but same controller stuff), cause you really want to get to maximum flow.

If you still have questions keep asking :)

*

Offline Rx7man

  • Crazy guy
  • Hero Member
  • *****
  • 908
  • Jack of all trades
    • '94 Dodge dually
Re: 0-5v input
« Reply #7 on: March 18, 2017, 11:47:31 AM »
What are you basing your decision on regarding what the desired boost is? TPS? RPM? Those can all be inputted to the Lil'bb and it can make it's own decisions, leaving the megasquirt out of it... Alternately, if you REALLY want the MS to do the 'thinking', you could have it output a 0-5V output into the analog input of the lil'bb and program it from there... 0V = 0PSI, 5V = whatever your desired peak is.. it's very adaptable if you put the time into programming it (and the MS I guess too)
'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 65fpvmustang

  • Sr. Member
  • ****
  • 211
    • 65 mustang 83 ranger
Re: 0-5v input
« Reply #8 on: March 21, 2017, 09:45:15 AM »
If your running a waste gate i would use ms3 to run that and little black box to control the turbo. I read your post on ms3 forum.

*

Offline punisher454

  • Member
  • **
  • 13
    • 74 TA 454TwinTurbo
Re: 0-5v input
« Reply #9 on: April 03, 2017, 03:24:04 AM »
To have the shaft speed running at 100k+ ALL THE TIME isnt going to work out on a gas engine very well. Diesel engines get away with it because they run wide open max airflow all the time, and vary engine power with fuel flow (over simplified obviously). Gas engines regulate power by controlling airflow and the fuel gets added based on current airflow. basically the opposite way a diesel works.

 For sure one of the last things you want is 30psi on top of your throttleboby butterflies at closed throttle.

The DSM guys figured out years ago that even on a 2.2L you'll need a wastegate on a VGT, they had boost climbing even with the turbine all the way open.
  At idle, part throttle driving and cruise I dont want to waste any horsepower (and fuel) by trying to keep a compressor wound up all the time building a bunch pressure thats going nowhere.

 I have 2 other car projects to finish up before I can start on my twin turbo big block (next winter I hope) but I plan to start writing some modified code more appropriate for gas engine use. The current version should work mostly, but it needs a few different modes of operation I think. The control scheme farmers24 described is similar to what I have in mind, with the addition of a wastegate and BOV.

*

Offline 65fpvmustang

  • Sr. Member
  • ****
  • 211
    • 65 mustang 83 ranger
Re: 0-5v input
« Reply #10 on: April 03, 2017, 09:11:59 AM »
well i'm about to hit 40k on my vgt GASSER build. around 15k with pneumatic / spring.
max boost is 324 kpa reduces to around 250 kpa at redline lpg injectors max out.
turbo is run off 2 generic ms3 outputs that pull a 5v signal low.
arduino with canbus shield run the turbo.

rx7man and hakcenter have been very helpful and knowledgeable about control.
If I was you I would try to listen more and say less.
You can't run 30 psi boost on closed throttle plate there is no exhaust flow.

there is tps control in lilblack for cruise.
best vgt spool position mostly follows airflow and pressure.
That's why shaft speed works so well. 
The best spool position changes with shaft speed, or map and engine speed.
Shaft is the easy way to go.
This was discovered through testing.
speaking of testing make sure you don't run to much exhaust pressure with vgt closed down trying to spool turbo.
one the compressor goes in to surge 2 the vgt veins warp.



*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #11 on: April 03, 2017, 07:07:24 PM »
To have the shaft speed running at 100k+ ALL THE TIME isnt going to work out on a gas engine very well. Diesel engines get away with it because they run wide open max airflow all the time, and vary engine power with fuel flow (over simplified obviously). Gas engines regulate power by controlling airflow and the fuel gets added based on current airflow. basically the opposite way a diesel works.

 For sure one of the last things you want is 30psi on top of your throttleboby butterflies at closed throttle.

The DSM guys figured out years ago that even on a 2.2L you'll need a wastegate on a VGT, they had boost climbing even with the turbine all the way open.
  At idle, part throttle driving and cruise I dont want to waste any horsepower (and fuel) by trying to keep a compressor wound up all the time building a bunch pressure thats going nowhere.

 I have 2 other car projects to finish up before I can start on my twin turbo big block (next winter I hope) but I plan to start writing some modified code more appropriate for gas engine use. The current version should work mostly, but it needs a few different modes of operation I think. The control scheme farmers24 described is similar to what I have in mind, with the addition of a wastegate and BOV.

You can't get 30psi hitting your throttle plate unless you don't run a BOV. Secondly, you can't have 100k all the time, that's pretty much impossible. You need energy in the exhaust to spool these turbos, they don't magic spool. Watch my video. You can clearly hear turbo whine like it is 'spooled' up. It is only doing 30k rpm.

I've exhausted a lot of control options and by far, shaft speed is the most preferred and dependable. If you want to make 3d to 5d control on top of that, sure. But you want to go, shaft speed + boost + option c = position.

What my code and controllers do, is try to get maximum airflow, without dragging the motor down. Gassers will need a wastegate if they can't handle the flow, and if that was the case on one of my gassers I would get a smaller turbo instead.

*

Offline punisher454

  • Member
  • **
  • 13
    • 74 TA 454TwinTurbo
Re: 0-5v input
« Reply #12 on: April 04, 2017, 11:02:53 PM »
OK, looks like I misunderstood one of the previous posts about keeping the shaft speed up. I had assumed (and you know what that does) that you were referring to all types of driving conditions, not just full power runs. And to keep shaft speed way up at light throttle would mean closing the vanes down pretty tight which would add up to a bunch of back pressure (obviously depending on engine size). So I had assumed you were saying there was no consideration for part throttle cruise and idle. So I should read more carefully first next time for sure. I'm not really an ass, I just sound like on on the internet :)

Anyhow, My project involves a 454 chevy with aluminum rectangle port heads (320cc intake ports), Roller cam, 16 injectors and a big intake. I have a few HE351VE's and plan to run twins with water to air intercoolers (possibly ice chilled at the track). My target power level is 1000-1100 HP. My math says that 60Lbs/min x 2 is about enough airflow for 1140-1260 HP.  Sounds about right to me, but I do have to admit that I'm more experienced with blowers and have only built up to 850-900 HP in the past. This is a whole lot more motor this time, but I'm not interested at all in exceeding 1100 hp, I would feel the need to spend a whole lot more money on better bottom end as well as more expensive trans and converter etc...
  I dont run trailer queens, and drive to any place I race, plus occasional cruise nights and such.  So to keep the HP level under control especially with only 12" wide tires, I will need to keep boost pressure down during launch and possibly more depending on traction. After that I would like more boost, I'm thinking around 15PSI should do it on this engine.

 Once I am up to the boost pressure level I need, I would like to have the turbine opened up as wide as possible, while maintaining the desired boost pressure level I need. This is because I dont want to run any more exhaust backpressure than is necessary. For the same reason most drag guys run really big exhaust housings that dont spool till way late, the less restrictive exhaust at the top end makes a whole lot better power, of course with a huge sacrifice down low.

So anyhow my thoughts are that I should want to run as much shaft speed (up to 125k or so)as I can at any operating condition, unless I'm nearing the manifold pressure limit I have set in my EFI computer at whatever point on the table I'm running at. If I'm at or near the MAP limit I will first begin to open the vane position up, If thats not enough I will then begin opening the wastegates.

  Along with that I think I will want to maintain a minimum turbo shaft speed, at part throttle cruise mainly to keep quick response time.
 So is my control concept way off base, or am I close to what will work well with this controller? and I do intend to use the 0-5v inputs, doing the boost control logic outside the turbo controller.

"But you want to go, shaft speed + boost + option c = position."
 With boost and option c being, supplied by the ADC inputs, correct?

*

Offline hakcenter

  • developer
  • Hero Member
  • *****
  • 623
    • 1998 12v Ram
Re: 0-5v input
« Reply #13 on: April 05, 2017, 06:15:49 PM »
There is a spare digital io, high or low, as an extra input, shoot.. the perf cruise and brake modes could be completely disgarded and you have 3 more high / low digital io's at your disposal.

Your setup is definitely within the parameters of the controller.

What I would do is re-code brake mode, since gassers don't need it anyways, into an optional target shaft speed within the modes code, very similar to how cruise works, almost.. if you will 'exactly' how it works but probably setup the timers to be a lot faster.

Depending on how you want to launch, if you brake launch etc.. you could relay your brake light / pedal switch (however it works), to pull the brake mode pin down to ground, and shazam, you're in your special code area when you apply the brakes.

If that's too simple and you want to drive the thing, you might want to setup a launch code in whatever you choose for EFI control and use an output there to pull the brake pin down, or if it only has high, relay to ground.

For boost control I suggest you use your EFI with an EBC on your wastegate.

Cruise could be setup in your EFI, as TPS <=, Engine RPM <=, equals cruise mode pull cruise pin to ground.



For twins, I believe you can wire the canbus's together on both turbos. But I would almost double all the vein sizes. 3cm = 960, would really be 6cm cause 2 turbos.. etc.. so you may have a few things to work through on that part.

*

Offline punisher454

  • Member
  • **
  • 13
    • 74 TA 454TwinTurbo
Re: 0-5v input
« Reply #14 on: April 05, 2017, 11:40:28 PM »
Ok thanks. I have 6-8 months to get this all worked out and put together before I'll be ready to even fire the engine up.

 I was planning to just run 2 LILBB units programmed identically and all the inputs just split parallel between them. That way each turbo can be on its own closed loop control.