Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[carkrazed]

Looking for one. Anyone have any suggestions?

I have an aem tru-boost but it fucking sucks.

Was looking at the turbosmart shit.

[highroller54]

I was looking for one also but they dont come cheap, I gave up and went with a manual in my beater.

[ratcityrex]

For a manual boost controller I rock the good old Kolbalt air compressor regulator. The one that comes with the gause. And then I use the gauge for my fuel pressure on top of my filter. All that for under $30


I was going to run a mac valve for a manual boost controller. I know there is a lot of people running them with great success. But you need to have something to run it. I was going to use ectune to control it. Im sure you could use ms to control it.

[92CXyD]

For a manual boost controller I rock the good old Kolbalt air compressor regulator. The one that comes with the gause. And then I use the gauge for my fuel pressure on top of my filter. All that for under $30

Do you need an adapter for the gauge to fuel filter?

Is there a comparable compressor regulator out there? Or where do you get the Kobalt version?



 

[ratcityrex]

Yes I needed the adapter I got online for $10, and the regulator I got at Lowe's

[highroller54]

I used a old stock smc (made in china now) 1/4 npt regulator off your advice, works ok, changes a bit with weather though. I always hit boost cut when its cold, the cobalt might be a touch more acurate.

[carkrazed]

I don't think I could ever go back to a mbc after using an ebc. Being able to set the gate crack pressure is the shit. Very noticeable difference in spool.

My AEM works good except that the gate opens and closes under full boost with a fluctuation of like 4 lbs. Wonder if it just needs a new solenoid? Its done it ever since I bought it but it was used so idk.

[highroller54]

I have never heard anyone say the aem works well. Not sure what they do wrong but everyone hates them.

[DSMR]

I posted this on another forum and got no responses. Maybe you guys can make some use of this!

 

I have been messing around with building my own EBC now that MAC valves are becoming very cheap. I am in brainstorming and design phase of this project but I thought it would be cool to share what I have so far. I do not have too much time invested in this so do not be shocked if this is rough so far.

Found a hardware diagram that will work PERFECT for this project. I have already simulated this circuit and tested it for feasibility with standard MAC solenoids used (AEM, Hondata, Neptune, eCtune, ect) and it looks good.



Hardware needed would be the driver circuit above, arduino, solenoid, potentiometer and to do something about voltages. No biggie.

Here is the code I busted out really quick. This is not tested so be surprised if it does not work!

Code: [Select]

// DSMR Boost controller concept code

void setPwmFrequency(int pin, int divisor) {
  byte mode;
  if(pin == 5 || pin == 6 || pin == 9 || pin == 10) {
    switch(divisor) {
    case 1:
      mode = 0x01;
      break;
    case 8:
      mode = 0x02;
      break;
    case 64:
      mode = 0x03;
      break;
    case 256:
      mode = 0x04;
      break;
    case 1024:
      mode = 0x05;
      break;
    default:
      return;
    }
    if(pin == 5 || pin == 6) {
      TCCR0B = TCCR0B & 0b11111000 | mode;
    }
    else {
      TCCR1B = TCCR1B & 0b11111000 | mode;
    }
  }
  else if(pin == 3 || pin == 11) {
    switch(divisor) {
    case 1:
      mode = 0x01;
      break;
    case 8:
      mode = 0x02;
      break;
    case 32:
      mode = 0x03;
      break;
    case 64:
      mode = 0x04;
      break;
    case 128:
      mode = 0x05;
      break;
    case 256:
      mode = 0x06;
      break;
    case 1024:
      mode = 0x7;
      break;
    default:
      return;
    }
    TCCR2B = TCCR2B & 0b11111000 | mode;
  }
}
long previousMillis = 0;
long interval = 1;
void setup() {
  setPwmFrequency(10, 1024);
}


void loop() {
  unsigned long currentMillis = millis();
  if(currentMillis - previousMillis > interval) {
    previousMillis = currentMillis;
    int DC = analogRead(0);
    DC = map(DC, 0, 1023, 0, 100);
    analogWrite(10, DC);

  }
}

I will update this later on if I get a chance to work on it. Code is kind of tricky and requires some massaging of a few timers. I calculate it to be at ~31hz but will check with an O-scope to verify.

What is next: All I need to do is program a board with this code and check it with an o scope. I will use the potentiometer to change duty cycle. 

Once that is checked and works like it should, I will build the high side of the circuit and check to see if it makes some clicky clack on the solenoid while I check all voltages with o-scope and multimeters. 

If all of that works it will be tested with shop air for a proof of concept. If that works all that is left is packaging. Its that easy.

For the over achievers, this would be a perfect foundation for someone to make a fancy boost controller that uses some PID trickery to home in on a selected boost pressure. MAP sensors are cheap cheap cheap! I won't be doing any of that. The only reason I am making this is because I am lazy and am tired of getting out of the car to tweak manual boost controllers on cars that I tune (and its cold outside).

[Minor Threat]

If you don't have a PID loop, all you've really got is a fancy MBC that will hold the gate closed longer. PID control/boost targeting is what makes a good EBC shine over an MBC.

[HiProfile]

It would be very simple to have it read a MAP and adjust boost. Program it to activate at 1psi (~2.9 volts on a stock honda), hold shut til 6psi (~4 volts), then do a constant sample input voltage & adjust PWM. One dial would be faster spool ("gain" iirc), the other for max boost. Mine would be labeled min & NOG setting. Spool dial would just select a percentage of voltage between 1psi's voltage and max boost voltage.

This could be done with $5 in Picaxe components and a couple dozen lines of code. With a larger or 2nd chip you could probably interpret the VSS and RPM to make it boost-by-gear too. Or with 3 magnetic reed switches and a magnet on the shift lever.


BTW with a small mosfet, you wouldn't need a heat sink. I have one that deals with 4+ amps driving injectors, and it doesn't even get warm.

[DSMR]

If you don't have a PID loop, all you've really got is a fancy MBC that will hold the gate closed longer. PID control/boost targeting is what makes a good EBC shine over an MBC.

Adding closed loop control is trivial. I am laying down the foundation for someone to home brew something for cheap if they desire. I could easily add this later but I really want to use it for tuning the shit boxes I deal with on a daily basis. They all come in with ebay manual boost controllers that do not work. I am building a dyno rig that will hook up to their car fast and make fast adjustments and let the customer deal with setting the desired boost.

[DSMR]

It would be very simple to have it read a MAP and adjust boost. Program it to activate at 1psi (~2.9 volts on a stock honda), hold shut til 6psi (~4 volts), then do a constant sample input voltage & adjust PWM. One dial would be faster spool ("gain" iirc), the other for max boost. Mine would be labeled min & NOG setting. Spool dial would just select a percentage of voltage between 1psi's voltage and max boost voltage.

This could be done with $5 in Picaxe components and a couple dozen lines of code. With a larger or 2nd chip you could probably interpret the VSS and RPM to make it boost-by-gear too. Or with 3 magnetic reed switches and a magnet on the shift lever.


BTW with a small mosfet, you wouldn't need a heat sink. I have one that deals with 4+ amps driving injectors, and it doesn't even get warm.

Thank you for the input! Good idea on integrating a "gain" setting. I was thinking of that also. If I were to make this "polished" and fancy I was thinking of doing a LCD for target boost with some other nobs for gain or other PID type parameters.

I chose Arduino because I am comfortable with it and I have several boards laying around. This can be done on any controller with coding. Arduino is also noob friendly with coding so it is good for the DIY crowd. You can easily burn the chips with the dev board and do a stand alone.

Thanks for the tip about the mosfet. I found that diagram searching online and simulated to check it against my project. It is overkill (I calculate a bit over 0.5 amps sunk from the solenoid) but I will test the hardware soon.

If anyone else wants to expand on what I have feel free. I have calculated the cost of what I am wanting to build at under $45. Add a few more dollars and you could add the map sensor and other goodies easily.

[DSMR]

Made some progress!



That is really the hard part of getting this to work. The foundation is laid. Next step is to add the potentiometer and get it to control the PWM.

Here is the code:

Code: [Select]

//DSMR boost controller test code 11/10/2012

void setup() {

  int ClearREG = 7;
  TCCR4B &= ~ClearREG;

  int myPrescaler = 5;         
  TCCR4B |= myPrescaler;
}

void loop() {
  analogWrite(7, 128);

}
This is for a Arduino Mega 2560. It can be adapted to an uno if needed. I am integrating the potentiometer next!

[DSMR]

I got a bit further with this. I have integrated my potentiometer code with the PWM code. Now I can adjust duty cycle with a knob. Here is the vid:

Arduino Boost Controller

Here is the code:

Code: [Select]

long previousMillis = 0;
long interval = 1;

void setup() {
  // initialize serial communication at 9600 bits per second:
  Serial.begin(9600);
  int ClearREG = 7;
  TCCR4B &= ~ClearREG;
 
  int myPrescaler = 5;
  TCCR4B |= myPrescaler;

long previousMillis = 0;
long interval = 1;
}
void loop() {
 
   unsigned long currentMillis = millis();
  if(currentMillis - previousMillis > interval) {
    previousMillis = currentMillis;
    int DC = analogRead(A0);
    DC = map(DC, 0, 1023, 0, 255);
    analogWrite(7, DC);

 
 
  Serial.println(DC);
 
  }
}
The only thing left is to incorporate a control circuit and make some noises with the solenoid. After that it is packaging and finished project. Easy peasy.

[highroller54]

who was it on here years ago that went into designing and building a ebc? Did it ever work?

[DSMR]

No love for a RHMT boost controller? My post count is not high enough or something? Anyone want a programmable EBC at  prices?

[rawr]

whats the rise and fall time on that squarewave? Shit looks better than most function generators, but it could be the time setting on your scope.

[DSMR]

whats the rise and fall time on that squarewave? Shit looks better than most function generators, but it could be the time setting on your scope.

Rise time is really small. At 200ns/division it is 48ns.



Looks like a nasty ring up there! Its not really that bad for this application though. The base resistor eliminates this ring.

[Tim]

Good project. Kind of want to build one just to build one.

[jdmhatchracer94]

if I knew how i would build this. integrate bbg and hook a nog up.

[carkrazed]

Badass!!

Ive thought about using an arduino to run a mac valve on several occassions. Just never had the time.

Nice work man.

[rawr]

Rise time is still really good and even the rings not that bad. Better than I expected from an arduino.

[Joseph Davis]

What the fuck?  Make your own thread for this.  Also, conside having multi-dimension adjustment.... the ability to drop boost if EGT goes too high or O2 lean.

[PhilStubbs]

I like this thread. I would like it even more if I didn't use a valve and Neptune for my boost control.

[DSMR]

Thanks for the kind words.

Since I have gotten a good response with this, I am going to implement stage 2 of development which is a closed loop control system. I just got my power supply from the house which will replicate a map sensor voltage that I can shift and test responses.

I will also incorporate a quick spool function into this (Very easy to incorporate) 

JD, Way ahead of you on dropping boost based on input. Already got the code for that figured out in my head.

BBG is a no go because there are too many different vehicles out there. I could integrate boost vs speed with a gps module if enough people were interested.

I guess I will start up a new thread with my new findings when I get a chance. Please keep the ideas flowing so we can have the best DIY RHMT  EBC!   

If anyone wants to build this, I say go ahead and do it! I tested it and it works!

[Joseph Davis]

Easier to offer an 05+ CANbus only OBD2 plug in, IMO.  Earlier scan speeds on critical PIDs suck on some vehicles without bribing the OEM for internal highspeed loggin protocols. Anyway, CANbus; set up an N/V gear judge and also be able to select safety trip points based on internal calculated parameters like inferred exhaust flange/O2 tip/catalyst temps, exceed programmed MAF limits, etc.

[DSMR]

Easier to offer an 05+ CANbus only OBD2 plug in, IMO.  Earlier scan speeds on critical PIDs suck on some vehicles without bribing the OEM for internal highspeed loggin protocols. Anyway, CANbus; set up an N/V gear judge and also be able to select safety trip points based on internal calculated parameters like inferred exhaust flange/O2 tip/catalyst temps, exceed programmed MAF limits, etc.

Thank you for your response. I have all the hardware to fiddle with CAN bus but have not had the time to dedicate to protocols and such. You have a really good idea that I will have to look into. Are the protocols all the same for the years you mention (Ford, Chrysler, Hyundai, Nissan)? Other issues would be complexity of my rig being set up for any car. This will have to evolve to some sort of LCD that the user can program. Only issue is how complex it would be. I would rather use a droid app to program this maybe? 

I would develop your idea as maybe stage 4 of my trailer park EBC. I am finished with stage 1 so now I am at stage 2 which is getting closed loop to work correctly. Stage 3 I will need help with. It is packaging everything on a custom PCB and doing some sort of UI other than telling people to change it in the code. At that point I would need someone other than myself testing this and breaking it before it can be nog approved. 

[Joseph Davis]

You might be onto something with the droid (and iPhaggot!) app - why load up the device with expensive extra hardware when everyone has a smart phone?  Bluetooth and go, let the app+phone be the interface.  Also script a simple interface for laptops with bluetooth so the tuner doesn't have to kludge around with multiple bits of gear while tuning the car.  BT modem and CANbus chip is cheaper than a screen.

Before we get in too deep - avoid providing too much support as a standalone boost controller.  If you can do it so it works great as a simple standalone EBC then it will sell as one, but don't get too deep.  There is already a huge market for this, and the competition is innovative as far as that goes.  The multi-dimensional (EGT, AFR) angle is already covered by the Plex unit, IIRC.  But tailoring yourself to the CANbus market lets you do practically all vehicles in the last eight model years correctly off of a generic API.  Tailoring your product to newer cars is the market where the money is at, and the last $50 you think you are worth isn't quibbled over if the product works as advertised.

Anyway, read up on J2534 (Drewtech's site is best) and go from there.  It's not a big deal to do most everything you need via generic PIDs.  I can possibly sniff a couple juicy enhanced PID addresses from Ford/GM/Honda/Scoob/Mitsu/other with the gear I have laying around, and if I get the EASE software I can also do Chevy, Chrysler, and probably a whole lot more.  I'm happy to help as much as possible with stage 3, but that is slightly exceeding the limits of my ability.  Happy to put you in touch with folks that would probably steer you right, or pitch a hand in, for the sake of geekery.

[Joseph Davis]

Jorsher on here is your guy for phone apps.  I don't think he works free, but he might work for a percentage if you want to make a business go of it.