I've been looking at old/new topics about dyno runs recently, because I've been considering using the Air/Fuel graph to adjust the jetting on my carbs.  However, when thinking about it more objectively I came to the decision that using a dyno may not be the best way to tune the carbs for street riding.

Consider this dyno run:

http://gstwins.com/gsboard/index.php?topic=21789.0

What does it really tell you?  On most dyno runs I've seen, they run the bike full throttle in the highest gear until redline/rev limit and then shut it down.  Therefore, the dyno really only gives information about A/F ratios at WOT.  It doesn't give you any direct information about A/F mixture while cruising or at parital throttle (which is arguably where most riders on the street spend most of their time).

As far as I can tell, it will give you main jet performance under full accleration only.  If the optimum A/F ratio is somewhere around 12.5 to 13, then the graph shows that a stock GS is lean until about 5200 rpm, and then hovers between really rich and right on for the rest of the rev range....at WOT.  Most people tune off a dyno so that the A/F mixture stays as close to optimum as possible throughout the entire rev range (which never seems to happen).  I would argue that the mixture at 5 to 7000 rpm would have to be rich at WOT because the engine isn't turning enough RPM's yet to pull the air necessary to balance the mixture.  However, at partial throttle (when less fuel is flowing through the carbs) the mixture may be closer to the optimum ratio.

Granted, I don't know much about the interworkings of carbs so I may be way off.  Any thoughts?

Bang on. You're looking for torque, an early starting curve, and a long lasting one. Trying to dyno a bike to do so will likely lead to a huge dyno bill, and little success.

Just tweaking out the 'flat' spots in the carburator would be a good start. Check the rejetting matrix, use it as a starting point, and get yours working nicely.

WOT power is quite pointless, of course except for WOT circumstances. The lean and overly rich spots are what makes ridability poor. Once they're gone, the need for 'more power' should go away. If it doesn't, consider a higher displacement motorbike.

Good job pointing out the faults with dyno runs !

I figured I'd use a dyno to zero in on a main jet, then figure the rest out from there.  When I tried reading the plugs to determine where I was with the mains, I was wrong; I think a dyno would be more accurate. 

We'll see what happens...

Drt,

One thing to conceder is that the slides are vac operated meaning that you would also be testing the full range of the needle as well as the main. at idle the slide would be all the way down. When they apply WOT the slide will still be near its lowest setting and will rise with RPM until. So you would know how your needle is affecting output as well as main. When the bike is at idle and they get ready to start the WOT test the dyno is already applying a load so you can get a idle up reading that gives you an idea of how your pilot is affecting output. Now do I think it is worth it? Depends on you purpose. If you are racing it is an advantage to know where your bike starts to making power (which BTW you can figure out using your butt dyno) and what changes do what to the motor. Just my 2c.

Quote from: ohgood on October 19, 2007, 04:42:28 AM
Bang on. You're looking for torque, an early starting curve, and a long lasting one. Trying to dyno a bike to do so will likely lead to a huge dyno bill, and little success.

Just tweaking out the 'flat' spots in the carburator would be a good start. Check the rejetting matrix, use it as a starting point, and get yours working nicely.

WOT power is quite pointless, of course except for WOT circumstances. The lean and overly rich spots are what makes ridability poor. Once they're gone, the need for 'more power' should go away. If it doesn't, consider a higher displacement motorbike.

Good job pointing out the faults with dyno runs !

I've already rejetted my carbs via a dynojet kit, and then the jetting matrix.  I couldn't get it running right with the dynojet kit so I bought new slides and tried it with stock jets and shims.  I've got it running well now, with just a little flat spot at about 1/4 throttle when it's cold that disappears very quickly.  I'm still considering adding a washer to try to correct for this.

I'm also already considering a bigger bike.  Just waiting for my disposable income to improve.

Quote from: Oklahoma_Mike on October 19, 2007, 06:31:58 AM
Drt,

One thing to conceder is that the slides are vac operated meaning that you would also be testing the full range of the needle as well as the main. at idle the slide would be all the way down. When they apply WOT the slide will still be near its lowest setting and will rise with RPM until. So you would know how your needle is affecting output as well as main. When the bike is at idle and they get ready to start the WOT test the dyno is already applying a load so you can get a idle up reading that gives you an idea of how your pilot is affecting output. Now do I think it is worth it? Depends on you purpose. If you are racing it is an advantage to know where your bike starts to making power (which BTW you can figure out using your butt dyno) and what changes do what to the motor. Just my 2c.

The slides should be opening to WOT very quickly, if you whack it at 3k  RPM and start your run, the slides should already be all the way up before you hit 4k RPM.  One of the tests you can do on the dyno is a slide speed test, which I believe is done by starting a run, going up through the gears into say 4th, closing the throttle completely, then opening it back up as quickly as possible.  If the resulting run shows you didn't jump back to full output quickly your slides aren't working fast enough.

Is anyone here really that familiar with a carb

You adjust your main jet to be optimized at WOT for all RPM ranges. Why, because if you optimize it for part throttle, then when you goto full throttle it may be way to lean or way to rich. You have to size the main jet for the worst case scenario, WOT.

If you want to adjust various other operating points then you need to adjust the taper of the needle(good luck with that one). You can modify the "effective" operation of the needle by shimming the needle with a washer. And thats why people shim the needle, for the part throttle corrections. This however, only shifts the entire curve up or down, and does not allow for variation between various part throttle corrections.

Also so no one is confused, the jetting matrix is a rough starting point. And will vary from bike to bike even with the same setup, its not 'fact'. If you want proper jetting then go to a dyno and rejet until its right.

Quote from: coll0412 on October 19, 2007, 10:42:58 AM
Is anyone here really that familiar with a carb

You adjust your main jet to be optimized at WOT for all RPM ranges. Why, because if you optimize it for part throttle, then when you goto full throttle it may be way to lean or way to rich. You have to size the main jet for the worst case scenario, WOT.

If you want to adjust various other operating points then you need to adjust the taper of the needle(good luck with that one). You can modify the "effective" operation of the needle by shimming the needle with a washer. And thats why people shim the needle, for the part throttle corrections. This however, only shifts the entire curve up or down, and does not allow for variation between various part throttle corrections.

Also so no one is confused, the jetting matrix is a rough starting point. And will vary from bike to bike even with the same setup, its not 'fact'. If you want proper jetting then go to a dyno and rejet until its right.

I'd agree that shimming the needles only moves the entire curve up or down, rather than actually affecting the curve at specific ranges of RPM.  I would argue that if you know that your mixture is right at WOT, then what do you know about the mixture at parital throttle?  If you are right-on with the mixture at 6000rpm at WOT, but super lean at 6000 rpm at partial throttle then you've got problems.  Espcially since you'll spend maybe 1/8 to 1/4 of your ride time at WOT, and the rest at partial throttle around 6000 rpm.

I would also argue that you can't dial in the main jet for all rpm ranges because the main jet delivers a specific amount of fuel when not plugged by the needle, but the amount of air flowing through the carbs would be dependent on RPMs.  So you'd have to be rich at mid range to be right on in the top end, or lean at the top end to be right on at the mid range.  That is, unless the amount of fuel delivered by the main jet is also dependent on the amount of airflow through the carbs.  Again, (like I said earlier) I'm no carb expert.  I'm just trying to critical think from the dyno runs that I've seen.

Quote from: DrtRydr23 on October 19, 2007, 12:22:55 PMI would also argue that you can't dial in the main jet for all rpm ranges because the main jet delivers a specific amount of fuel when not plugged by the needle, but the amount of air flowing through the carbs would be dependent on RPMs.  So you'd have to be rich at mid range to be right on in the top end, or lean at the top end to be right on at the mid range.  That is, unless the amount of fuel delivered by the main jet is also dependent on the amount of airflow through the carbs.  Again, (like I said earlier) I'm no carb expert.  I'm just trying to critical think from the dyno runs that I've seen.

The main jet responds to low pressure generated by air flowing through the carb body.  More flow (more RPMs) means more signal, which means more fuel gets sucked through the main.  In an ideal world, motors would be 100% volumetrically efficent across their rev range, at which point there wouldn't be changeable main jets, just a fixed oriface.  You could swap carbs amongst different motors and never touch the jetting.  Real motors are far from 100% VE, (Sub-optimal port design, restrictive airboxes, etc) so you have alter the main to account for strong or weak flow by moving the main down or up to match.  In most cases, once you've got the main right for WOT on top, it's close for WOT across the board.  If you're finding big lean or rich spots along the rev range, you've got some interesting flow dynamics showing themselves.  (Harmonics, air box interations, etc.)

DrtRydr --

Why not just buy a A/F meter?  You can ride around with it at all throttle settings under different circumstances and really get a feel for what's going on.

Afterwards, loan it to me.

Quote from: Chuck on October 19, 2007, 01:52:47 PM
DrtRydr --

Why not just buy a A/F meter?  You can ride around with it at all throttle settings under different circumstances and really get a feel for what's going on.

Afterwards, loan it to me.

I didn't know an A/F meter existed.  Even so, I'm more curious than I am worried about my mixture.  I think I'm pretty close as is, but was just curious about dynos and some of the theory behind carb function.

I've been thinking of doing some carb tuning with A/F meters.  Really good ones run like $400, with data logging and whatnot.  I'm to cheap for that.  You could get a wideband O2 sensor for like 50-60 bucks and connect it to a multimeter for a cheapo alternative.  You'd have to weld some bungs into the header to get left/right readings, or you could make a tailpipe probe (which is what I'd do, being lazy and all) and hope the cylinders are in sync.

If you want to make your own, which is actually pretty easy, you need a heated wide-band 02 sensor($50.00) and a simple comparative circuit hooked to some LEDS and really thats it.
If you use a heated 02 sensor than you could probably just hang it off the pipe.

Just remember, an A/F meter will only get you close on partial throttle.  Some motors give better throttle response on the lean side of ideal, others prefer to be slightly rich.  Once you get it close, you should be able to go by feel more than a direct A/F read.

Standard AF meters are only good at WOT, anything other than that and the lights bounce back and forth.  Then you've got wideband O2 sensors which are more commonly used for fuel tuning and give an actual number instead of rich/lean and stoich.  Stoich is 14.7:1 AF which is safe, but not necessarily the best for power.  Generally you make more power running lean than you do with stoich or rich.

Fuel Injection makes the whole process so much easier.  Hook up a fuel controller/computer, a wideband o2 sensor and hit the dyno.  You'll see exactly what your fuel curve looks like along with the corresponding hp and tq curves.  You can basically lean it out as far as is safe without it pinging wherever you need to in order to make more power and you don't have to keep changing jets/needles, etc. 

As an example, heres the link to a dyno run on my car (prior to a few major changes), with 18lbs of boost which was about all I wanted to push on the stock headgasket, in which you can see how it leans out in the upper RPM ranges as power peaks and levels out. http://www.rs-enthalpy.com/images/dynos/sr-gt28rs-86-18psi.jpg The leaning out was done deliberatly to make power and is not a result of lack of fueling.  (I run a set of 740cc injectors with a 255lph fuel pump).  The car was daily street driven for about a year with no pings, knocks, or any other fuel related problems.

So just keep that in mind when you think about hooking up an AF meter and just making it read stoich.  Its much easier to put the bike on a dyno with a wideband and tune for max usable power than it is to do it on the street.  Taking the bike out for a run while trying to watch the AF meter and record its readings through the entire RPM range, then coming home to change jets, adjust needles, etc, then take it back out, record everything again, repeat is a hassle, not to mention not quite the safest way to do it since you really need to go WOT for the entire RPM range to tune the entire range.

Quote from: Kurlon on October 19, 2007, 07:34:37 AM

Quote from: Oklahoma_Mike on October 19, 2007, 06:31:58 AM
Drt,

One thing to conceder is that the slides are vac operated meaning that you would also be testing the full range of the needle as well as the main. at idle the slide would be all the way down. When they apply WOT the slide will still be near its lowest setting and will rise with RPM until. So you would know how your needle is affecting output as well as main. When the bike is at idle and they get ready to start the WOT test the dyno is already applying a load so you can get a idle up reading that gives you an idea of how your pilot is affecting output. Now do I think it is worth it? Depends on you purpose. If you are racing it is an advantage to know where your bike starts to making power (which BTW you can figure out using your butt dyno) and what changes do what to the motor. Just my 2c.

The slides should be opening to WOT very quickly, if you whack it at 3k  RPM and start your run, the slides should already be all the way up before you hit 4k RPM.  One of the tests you can do on the dyno is a slide speed test, which I believe is done by starting a run, going up through the gears into say 4th, closing the throttle completely, then opening it back up as quickly as possible.  If the resulting run shows you didn't jump back to full output quickly your slides aren't working fast enough.

Yea what he said. I reread my post and got the same impression, That I was trying to say the slides move up slowly as the rpm go up. No they dont, but you can get data from the slides down to up it just happens fast and to just point out(for those that don't know) that your throttle cable is not hooked to the slides in any way. The slides lower and raise based on vac, weight and friction.

Quote from: IN_DET on October 20, 2007, 10:45:47 PM
Standard AF meters are only good at WOT, anything other than that and the lights bounce back and forth.  Then you've got wideband O2 sensors which are more commonly used for fuel tuning and give an actual number instead of rich/lean and stoich.  Stoich is 14.7:1 AF which is safe, but not necessarily the best for power.  Generally you make more power running lean than you do with stoich or rich.

Fuel Injection makes the whole process so much easier.  Hook up a fuel controller/computer, a wideband o2 sensor and hit the dyno.  You'll see exactly what your fuel curve looks like along with the corresponding hp and tq curves.  You can basically lean it out as far as is safe without it pinging wherever you need to in order to make more power and you don't have to keep changing jets/needles, etc. 

As an example, heres the link to a dyno run on my car (prior to a few major changes), with 18lbs of boost which was about all I wanted to push on the stock headgasket, in which you can see how it leans out in the upper RPM ranges as power peaks and levels out. http://www.rs-enthalpy.com/images/dynos/sr-gt28rs-86-18psi.jpg The leaning out was done deliberatly to make power and is not a result of lack of fueling.  (I run a set of 740cc injectors with a 255lph fuel pump).  The car was daily street driven for about a year with no pings, knocks, or any other fuel related problems.

So just keep that in mind when you think about hooking up an AF meter and just making it read stoich.  Its much easier to put the bike on a dyno with a wideband and tune for max usable power than it is to do it on the street.  Taking the bike out for a run while trying to watch the AF meter and record its readings through the entire RPM range, then coming home to change jets, adjust needles, etc, then take it back out, record everything again, repeat is a hassle, not to mention not quite the safest way to do it since you really need to go WOT for the entire RPM range to tune the entire range.

According to the dyno graph you are slightly rich all the way across the board, not lean.  If the ratio of Air to Fuel is 12 to 1, then that is less air/fuel than 14.5 to 1.  Less air = rich.