1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Evening out turbulence in a 3 pipe

  1. Jun 12, 2010 #1
    Evening out turbulence in a 3" pipe

    Hi folks,

    I have a small problem. In my car I'm building new pipes for the pressurized air. On a straight pipe there is a MAF sensor (Mass Air Flow). It consists of a hot wire and a temperature sensor. After the pipes were redone and the MAF sensor was moved to the pressure side and after the intercooler I experience problems with an uneven signal from the MAF sensor. This is certainly caused by turbulence in the air passing through it as these sensors are very sensitive to that.

    This is causing the engine to have problems injecting the correct amount of fuel in low load scenarios causing the engine to run uneven and almost stall in some cases.

    So basically I need a way to straighten out the air as easily as possible. Honeycombs usually perform this job well, but I can't imagine gluing a honeycomb to the intake of the MAF sensor can become any good. There is already a steel grid in the intake for correcting minor turbulences.

    My idea is welding two plates in the tube before the MAF sensor somewhere. I'm also open for other suggestions. I've made a drawing to illustrate the problem:
    [PLAIN]http://img718.imageshack.us/img718/4475/turbulens.jpg [Broken]

    Are you physics folks able to guess what the best solution is? I have really no experience or know any theory about turbulence, so I need some help on this :).

    Thanks in advance,
    Christian
     
    Last edited by a moderator: May 4, 2017
  2. jcsd
  3. Jun 12, 2010 #2

    jack action

    User Avatar
    Science Advisor
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    Are you sure you're MAF is not dirty?

    According to http://en.wikipedia.org/wiki/Maf_sensor#Conditions_that_affect_operation":

    It sounds a lot like what you are describing.
     
    Last edited by a moderator: Apr 25, 2017
  4. Jun 12, 2010 #3

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    the trick is to use gentle bends with as large a radius as you can stuff in the engine room..45 degrees is ok but max radius
     
  5. Jun 12, 2010 #4
    Re: Evening out turbulence in a 3" pipe

    Based upon recirculation wind tunnel designs, rather than any particular knowledge, I would say the best place to locate the turning vanes would be within the bend.
     
  6. Jun 13, 2010 #5
    Re: Evening out turbulence in a 3" pipe

    Yeah, I'm sure it's not that. Because it worked fine when it was at the previous location. But due to new turbo and some stuff, there is no way to have it there anymore. The MAF sensor is also pretty new and the inside looks to be clean.


    It isn't possible to use gentler bends than there is now. Before the MAF sensor the intercooler is also causing lots of turbulence, plus the hard bend welded to the intercooler.


    Unfortunately that isn't possible due to the bend being a silicone hose. The only way to fix it practically is mounting something in the straight pipe.
     
    Last edited by a moderator: Apr 25, 2017
  7. Jun 13, 2010 #6
    Re: Evening out turbulence in a 3" pipe

    After a little though, I think you want to hit the turbulence at close to the source as possible. But why not construct an X-shaped cross-section for the entire length and tack weld it at both ends? It doesn't have to fit snuggly for effect. Vibration considerations may say otherwise.
     
  8. Jun 13, 2010 #7
    Re: Evening out turbulence in a 3" pipe

    Yeah, I also think it should be hit as early as possible. But my concern with having a cross section all the way is if there is more air in one of the sections than in the other because of the bend. Then them merging back together before the MAF might also cause turbulence that a straight pipe afterwards would've fixed.

    But then again that is my logic :p, could be wrong :).
     
  9. Jun 13, 2010 #8
    Re: Evening out turbulence in a 3" pipe

    No, you're right. I didn't want to say too much at once. You would want a full length divider to be a bit leaky. Build-in some leaks with a few half moon cutouts toward the bend end.
     
  10. Jun 13, 2010 #9

    S_Happens

    User Avatar
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    So, you don't believe a typical honeycomb laminizer will do the trick or are you concerned with how you'd mount it to/near the MAF?

    Not to be too nitpicky, but you can't be "sure" that it's not a MAF issue just becuase it worked in the past, especially since there has been plenty of modification and relocation. Did you check the voltage (I'm assuming Bosch hot wire style from what you said and playing the numbers) or do you have access to read with a scan tool/software? You stated that this is actually your problem, so I hope you have some data instead of just a hunch.

    I certainly don't agree with your idea of the plates (even in theory), especially just fabbing up your own. It's probalby more likely to cause problems than alleviate them.

    I doubt there is any need for controlling turbulance anywhere other than at the MAF. Most likely you only need to fix the consistant measurment problem and won't be facing any sort of restriction (when you run out of boost you can start worrying about that).
     
  11. Jun 13, 2010 #10
    Re: Evening out turbulence in a 3" pipe

    Mostly concerned how I'd mount it to/near the MAF. As long as the air is evenly distributed in the pipe, then the honeycomb laminizer should do the trick. The MAF relies on evently distributed air by design as it only measures the air flowing in a fraction of the pipe. And ofcourse relies on no turbulence as a good repeatability is absolutely neccessary to not cause problems.

    Not sure how I would go about securing it. Must be cutting out a honeycomb circle with very tight fit and apply epoxy glue to the wall around it. But the consequences of destroying the honeycomb could be catastrophic. Originally the MAF has a wounded steel net in and out. I could glue it before that, but not sure if that's ideal with the wounded steel net right behind the honeycomb. I guess it would be best with just the honeycomb.


    I've had quite a few hours troubleshooting this. At first the pipe design was very different. With the MAF being about 5 cm from the end of a long radius 90 degree bend. With this setup the car wouldn't run any good at all. It was lugging and shooting when driving with any kind of normal throttle and the idle was varying very much up and down.

    After a while of troubleshooting I figured out that the problem was due to turbulence. Because it ran nicely if I moved it back to original position (the turbo isn't changed yet, this is a part of the preparation for turbo change) with the longer cable I've made for it. It also ran nicely at the same position but with the pipes before the mass meter removed and replaced with an air filter directly on the MAF.

    So I rebuilt the pressure pipes so that I got less bends total (total of 135 degrees less bends before MAF). Also with the new design i got 25cm instead of 5cm with straight pipes before MAF. Now the car is drivable, but it still suffers slightly noticable during normal driving from the same problem. With very little gas during cruising, the Air/Fuel ratio varies a lot and the car feels a little bit lugging/uneven. On idle it varies with about 100 rpm, which it didn't do before. Basically the same symptoms as with the first design, but 10 times less noticable, but still there.

    So there is basically no other explaination for this behaviour than turbulence. Slight turbulence causing fluctuating readings from the MAF.


    Ok, so basically the plates could cause more turbulence or simply not do anything?
     
  12. Jun 13, 2010 #11

    S_Happens

    User Avatar
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    I am familiar with hot wire style MAF sensors.

    My inclination would be to try and figure out a way to install a honeycomb screen. I would not want to simply epoxy it to the inner wall upsteam of the MAF as you suggested. If you're willing to try and install horizontal plates into your air intake, why not consider something like making two plates to hold a screen in between? Without seeing what you have I can't say whether you might have to modify another MAF housing, or be able to fabricate something on your own. It depends on whether your current screen is held in place by something like a snap ring, if it's just sandwiched in, and parts/tools you have available.

    That would be the first route I would pursue as far as trying to alleviate the turbulance.

    Is the vehicle equipped with MAP sensor as well (quite a few vehicles had both)? If you can't get the MAF working right, you may be able to use an air density type control.
     
  13. Jun 13, 2010 #12
    Re: Evening out turbulence in a 3" pipe

    Okay. I will try a honeycomb screen then :). No, unfortunately my vehicle has just a MAF sensor. It's a Bosch 0280 213 012 Air Mass Meter.

    The original screen is held in place by a lock ring, but there is no space for locking the honeycomb filter using that I think. But the tube is a bit wider in diameter first, then it gets slightly smaller, to keep the original screen in place. So if I cut my honeycomb filter so that it precisely fits the first big part, then it shouldn't be able to go further in. If I then use lots of Epoxy glue around the walls and make a line of epoxy outside the screen, then it should hold. Assuming degreased and sanded surface and that an epoxy glue can work permanently in those conditions (temperature variations).

    You think that would work or should I try to figure out a more mechanical solution?
     
  14. Jun 13, 2010 #13

    jack action

    User Avatar
    Science Advisor
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    I will agree with Phrak. Why can't you just put 4 or 5 tubes of a smaller diameter in your silicone bend?

    Also, here is another design for your system that, I think, would help you. It consists of a constant diameter bend followed by a tapered pipe (note that the taper of the pipe is tangent with the "inside" of the bend):

    intake 01.JPG

    It might be too complicate (or costly) for you to modify your system, but it might give new ideas.
     
  15. Jun 14, 2010 #14
  16. Jun 14, 2010 #15

    S_Happens

    User Avatar
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    Those screens look great. I'm glad you found something. This is certainly the most practical way to go about it (not that the other ideas were bad, but making one-off parts can be complicated and expensive).

    I'm concerned about using epoxy though, especially inside a nicely machined, highly sensitive MAF. I'm not sure I would want to just wedge it inside the MAF like that guy did, but at 1/2" or 3/4" thickness it will probably be pretty secure.

    So, now that you've at least got a prospective fix, how about some details? I haven't used HPTuners much (are you using it or did you just stupid across that thread you linked), but I've done some tuning on NA OBD1 camaros.
     
  17. Jun 15, 2010 #16
    Re: Evening out turbulence in a 3" pipe

    I think it could be wedget into a slot in the MAF. So that combined with epoxy after sanding the surface a bit (mine is plastic), it has no way of getting anywhere :).

    I haven't used HPTuners either, I just stumbled upon that forumthread after searching for a while. I think I have ODB1 aswell. This is for my Volvo 940 Turbo street project. Due to the new and big turbo I'm about to install and that I need to use an external dump valve with it, I needed to reinforce the MAF sensor and move it to the pressure side of the system. That caused some problems with turbulence causing some false readings so I had to try out a few designs to get it work OK. Now it works pretty good, but still not perfect, so that's why I'm going to try the MAF screen as there is still a little bit of turbulence left :).
     
  18. Jun 16, 2010 #17

    S_Happens

    User Avatar
    Gold Member

    Re: Evening out turbulence in a 3" pipe

    Wow, I just noticed this. I meant "...did you just stumble across..." It wasn't a Freudian slip, trust me :uhh:

    Let us know if the screen works, and if not we'll go from there.
     
  19. Jun 16, 2010 #18
    Re: Evening out turbulence in a 3" pipe

    Yeah, I'll let you know when I get it. Will probably take about 2 weeks.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook