Welcome to Netrider ... Connecting Riders!

Interested in talking motorbikes with a terrific community of riders?
Signup (it's quick and free) to join the discussions and access the full suite of tools and information that Netrider has to offer.

Wideband tuning carbies?

Discussion in 'Modifications and Projects' started by oldcorollas, Aug 13, 2014.

  1. any tips from people that have done it?

    Have NGK/NTK wideband and a spare boss, so figured, "why not"?
    Thinking to set up camera to video tacho and AFR display, but not 100% sure on how to measure load, or if I even need to?

    1) basic plan was to run WOT in various (legal speed) gears, to set the main jets..
    2a) then either run through revs at X% throttle, or
    2b) find steep hill and vary throttle at a fairly fixed rpm.. or

    not sure on sense of fitting pressure gauge, as can't really relate intake pressure with slide movement (unless directly measure off bike = pita), and not interested in fitting EFI.
    leaning toward fixed rpm/vary throttle technique but open to suggestions.

    umm, any suggestions as to how early in RPM range the slide is fully up? or will it only be fully up at max power/WOT?
  2. #2 oldcorollas, Aug 13, 2014
    Last edited: Aug 13, 2014
    changed exhaust and wondering if now running rich/lean (fuel economy changed by ~10%) or if it can be optimised a bit.
    and also going to play around with different baffles, maybe it will make a difference? maybe not..
    I have the means to do so, and cost is minimal, so why not? :)

    if it was a bigger engine bike, I'd likely just Microsquirt it and tune myself and wouldn't bother asking here ;)
  3. If you've got access to it, why not?

    Just run it normally, if you see the AF ratio drop too low, work out where it is in the throttle opening & change jetting to suit, it'll be a lot of trial & error I expect. You'll get a good real world AFR rather than a Dyno one which is just under load, usually in a particular gear.

    I built one up from a Jaycar kit, bought the Bosch wideband sensor, made a holder, etc, then I fried the main chip first time I fired it up, haven't been back to it since... My track VFR is getting a Rapid bike module which does all that plus logging, but I'd still like to finish the other one one day.

    The slides will probably only be fully open at full throttle, & maybe only a little after that, the vacuum can take a second to open them. RPM will vary with gear & loading I expect.
    • Like Like x 1
  4. Thanks :)
    That's a good approach, I'd rather be a little rich than a lot lean! my SOTP dyno is usually a bit rich with cars (12:1 rather than 12.5-13)
    Had considered welding a boss to the H-pipe, but now thinking easier to put an adaptor between mid-pipe and the muffler/can.
    Shouldn't get too much air mixing there except at idle.

    I use an NGK Powerdex AFX, which uses the NTK sensor (of course). The NTK is a good sensor, although the Bosch are just as good these days and more available/cheaper. The AFX updates a bit slower than aftermarket/independent units, but is accurate and designed for that sensor :)

    Rapid bike looks interesting. did you consider any others?
    I've Megasquirted and tuned a few cars and am keen to Microsquirt something 2-wheeled eventually :) but yeah, lacks logging (MS3 has onboard SD card logging)
  5. I have no idea what you are talking about when it comes to tuning carbies the way you describe. But I'd be interested to hear from anyone who does.
    As for you @Uncle Greg@Uncle Greg, stop stirring please mate.
  6. McSenna, Oxygen sensor can be used with carby (now that leaded fuel no longer used) to measure residual oxygen in the exhaust, and figure out if engine needs more or less fuel. exactly the same way as for EFI, but there's no auto feedback loop. that requires a screwdriver...
    Doesn't matter if carby or EFI, the process is the same.

    Wideband gives great feedback, and you can calculate how many jet sizes you need to move reasonably well = buying less jets. i guess it could also help with selecting needle profiles, but that still involves some pointy-hat work (ie, the dark arts)

    Tuning for mixture on the road is different to tuning for power. for power of course you need dyno (are accel rates/times).
    If you just want to get mixtures closer to "ideal", ie stoich at cruise and a "bit" rich under power, then road and wideband is super easy.

    as long as you choice a location for the sensor that is free from systematic issues
  7. Thanks, if it involves more than my ear, I'm in over my head, but I did ask :)
  8. #8 oldcorollas, Aug 14, 2014
    Last edited: Aug 14, 2014
    My ears and bum are no longer as sensitive as they used to be, and tbh were never that good.

    here's some other poor bugger with a ninja250 and an Innovate MTX-L

    unfortunately, the refresh rate of the Innovate makes it hard for his camera to record easily.. the NTK updates far less often = easier to record I guess

    '68 Honda CL450 with carbs and wideband
  9. I'd be aiming for lean at cruise. Between 1.0 and 1.1 lambda or so. No real load on the engine so shouldnt be an issue.
  10. The options were Power Commander 3 or 5, Bazzaz, earlier Rapid Bike moduels... The appeal of the RB2 was cheap & had timing which neither the Bazzaz or PC do easily, then I got in on a group buy of RB harnesses to suit the VFR, which allowed either the Race or Evo module to be used, simple choice from there! Timing, auto tune, quickshifter, pitlane limiter etc etc all as standard.

    I thought about a Microsquirt system too, but that would be way out of my depth at this point, I don't even have the bike running yet.
  11. sounds like RB was the way to go, to get those features without stuffing around with different firmware/versions/mods :)

    I reckoned the diameter of the H-pipe was too small to not obstruct flow with the sensor..
    so ended up taking the easy way out, and made an adaptor for the sensor to sit between midpipe and can.

    couldn't get exact pipe diameter, so had to slit and weld to slide into the can, then add external sleeve to go over midpipe.. no audible leaks and is solid enough. puts the can back about 3" (bracket off rear peg)

    slight difference in sound between sides, but with baffles in later, any slight flow differences due to the sensor will be cancelled out.

    NGK recommend not closer than 10 exhaust diameters from the exit, and this is close (40mm internal diameter) but should be ok.. if not, just rev more :D
    with baffles the exit will be 23mm ID, so should be no problems with back-diffusion.
    with bigger bikes, the larger exhaust flow should have no issues at all at normal riding revs

    • Like Like x 1
  12. When tuning carbies you get your idle mixture roughly right, then you start at the main jet, then move to the needle, before going back to the idle mixture.

    So your WOT approach is right. On a 250 you shouldn't have any problems doing it at legalish speeds. On big bikes it's a problem, because 2nd gear is the only gear you can use at legal speeds and then you can't hold it long enough for a decent reading.

    In the end, however, you may have do the final tune with the arse dyno, as setting your mixtures to exactly Stoichiometric may actually give you some flat spots.

    Before you start, make sure your carbies are clean, the float height is set and they are synced. Make sure you have fresh plugs and a fresh tank of petrol, of the type you are going to use regularly.
    • Like Like x 2
  13. Good enough, keep us posted on how it goes.
  14. finished the db killers, and wideband wiring but still raining so no vids yet. will play with idle and no load readings this weekend.
    agree, any bike with actual power needs to be contained on rolling road. 250 is so slow it might work on some nearby 80 and 100 roads.

    yup, carbs and plugs are in good nick and it runs well atm, so hopefully readings will be useful.

    as another tool, there is an app out there designed to use audio from exhaust to make a "power curve" for single cylinder 2 stroke scooters...
    since a twin 4 stroke has same number of exhaust pulses/rpm, will give it a crack as well, wonder if it will work? (SIP MicroDyno)

    also came across a nice clear vid of slides working. can see throttle linkage move, but no exact indication of how much. seems to not be on full throttle much.
  15. break in the clouds, had some fun :)

    choke really does richen up the mixture :p
    following pics are 3krpm, with choke or throttle.
    first pair are with 60mm long db killers, second are with no inserts.

    with no inserts, feels like it goes a little quicker and a bit better throttle response, and mixtures are also richer (and get richer faster with throttle).. need a go-pro or similar to video on the move
    [​IMG] [​IMG]


    vid is in garage testing choke and throttle before venturing out.
    the refresh rate is about 0.2sec(?) which makes it easy to read on the road, cos when you flick eyes down, the numbers are still for long enough to read them

    cruise AFR with db killers was spot on, at stoich to 15. WOT was mid to high 11's
    took db killers out mid ride..
    cruise AFR with no inserts was more rich, around 14.3ish, WOT was low 11's.
    will have to check for calibration drift and switch back to the inserts as a control

    anyway, is a start