Re-jetting with a wideband

loudhvx wrote:

I notice when I go WOT quickly the EGA goes off scale rich (richer than 11:1), but if I go more gradually, the meter stays on scale between 11 and 12.5. The bike seems to be jetted properly by all other indicators. (I'm building the analyzer to use on other bikes to speed up the jetting procedure, mine took days of experimenting and tuning.)<br><br>Post edited by: loudhvx, at: 2007/07/26 02:36

First, if you're using a narrow-band sensor, you're not getting a very precise reading.

Second, staying between 11 and 12.5:1 at WOT is still a bit fat. This is where the dyno comes in. Every engine is a little different, but, that "magic number" of 12.8 - 13.2:1 is almost always money. You just change jets to lean it out until you either stop making more power, or hit 13.2:1, then back it up one notch.

Third, when you get on the throttle, the increase in airflow is immediate; it takes the fuel a bit to flow through the jets, atomize, and flow through the intake tract. That's why many carbs have accelerator pumps. When you mash the gas, they throw extra fuel into the mixture to compensate for the momentary over-lean condition. If you have carbs with no accelerator pumps, you will not be able to tune as clost to that 12,8-13.2:1 number as you will if you do have accelerator pumps.

Fourth. Under normal, part-throttle driving conditions, you are actually on the idle circuit. Using a wideband tuner, you are able to look at part-throttle and WOT as completely different animals; tuning the idle circuit much leaner for optimal economy, and richening WOT up, but not TOO MUCH, so you can make optimal power, safely.

If a carb is a sledgehammer, and Sequential-Fire EFI is a scalpel, jetting your carb(s) with a wideband tuner is somewhere in-between. It's much more precise than the old way of tuning carbs, but, it's nowhere near as precise as EFI.

The only real problem I have with the stationary EGA is that I can't sustain WOT long enough for the mixture to stabilize.

I don't have an accelerator pump, and it's not a CV carb, so it's a realtively simple tune procedure... jets and needles only.

I believe the quick-throttle results in a rich mixture because the accumulated fuel above the jet (in the emulsifier tube) empties much quicker than when the throttle is opened gradually. But either way, I can't sustain WOT long enough for the mixture to reach its final reading.

I'm not sure if it's a wide band or narrow band, but it is definitely precise (precision and accuracy are different things). I can clearly see the difference in 1/8th turn on the air screws. I could probably even detect 1/16th turn.

It also seems to be surprisingly accurate. When I use the arduous lean-drop method, and repeat it several times, while the engine is hot and stable, then check it on the EGA, the reading comes in right at 13.25 on the meter (which is indicated as the target on this EGA). A quarter turn gets the mixture to 13.0, if I recall.

I just have no similar way to verify sustained WOT, which is why I'm interested in making one I could mount temporarily on the bike. I know my bike is vey close, if not right on, but it would be nice to verify it.

Also, like I said, it would also save a lot of time for tuning other bikes. (I have 2 more of my own to do.)

Post edited by: loudhvx, at: 2007/07/26 03:49

That old chart that's lurking on my site dates from when I had a narrow band sensor on my bike. Narrow bands really are not accurate at all except for at stoich. Either side of stoich they just go way off, although they are still better than nothing. On normally aspirated bikes all the tuners I know tune to 13.0:1. On turbos they go for 11-12:1. I actually run my bike at 11.0:1 (the WBC includes a gauge) for safety because of the sustained high speed high boost runs it does. As Sean says, it would make more power weaker, probably 250-260hp, but probably would not survive the experience.

Lou, for a cheaper wideband control check out www.14point7.com/JAW/JAW.htm

Personally, I won't buy anything from Inovate (LC-1, LM-1, etc) since they started to sue people for selling their stuff for less than MSRP.

The company views sales by unauthorized sellers (i.e., those who sell too cheaply) to be not only a violation of its minimum-price policy, but, surprisingly, as an infringement of its intellectual property rights. ITI’s eBay “About Me” page explains that the sale of its products by anyone but an authorized dealer constitutes patent and trademark infringement. Moreover, the company claims the right to prohibit all use of its copyrighted “technical data, photos, graphics, software, product literature, catalogues, product specifications, installation guides, user guides, promotional material and other types of information” without its permission. In other words, the company claims it is copyright infringement to read its user guides and manuals, browse its catalogs, or look at its pictures without its “express written permission.”

LINK

Post edited by: steell, at: 2007/07/26 10:08

Cool, thanks all! 13:1 it is then.

How much should it vary typically? If it was jetted correctly, what would I typically see as a range of mixtures? I know it's not going to just lock in at 13:1 and stay there under every condition. It seems during acceleration it is typically richer, then when cruising at steady throttle opening and RPM it gets leaner.

That JAW seems reasonable, but even that seems too elaborate for what I want. I just want to have an array of 5 or 10 LEDs that light up to represent the mixture. Basically a voltmeter, I guess.

I see they mention the Bosch lambda sensor. Is there any other one out there that is really small and works well?

It's just an idea I'm kicking around.

You want to be at your leanest under cruise conditions, at steady throttle. The closer you can get to stoich (14.7:1), the better your economy and the lower your emissions.

13:1 would be the number you'd shoot for at WOT, and maybe 12-12.5:1 under transitional conditions, between cruise and WOT.

I had a long conversation with wiredgeorge the other day about carbs for my KZ650. I have a 79 model, with accelerator pumps. He recommended I go with no-accelerator pump models, as they're less troublesome.

What made me curious was how you would handle those transitional conditions when you go from cruise and add throttle without accelerator pumps. I'm a newb to streetbikes, and every car I've ever dealt with that had a carb also had accelerator pumps. When you're tuning efi, you essentially have virtual accelerator pumps, as you have multiple tables for different throttle positions, and you can dial in a "pump shot" from the injectors.

If there are no accelerator pumps, I'm guessing you'd have to jet the idle circuit fat. Wiredgeorge said the difference between carbs with accelerator pumps and those without is about 2mpg. I'm guessing that's from the necessity to tune richer to compensate for the lack of accelerator pumps.

Lorcan wrote:

That old chart that's lurking on my site dates from when I had a narrow band sensor on my bike. Narrow bands really are not accurate at all except for at stoich. Either side of stoich they just go way off, although they are still better than nothing. On normally aspirated bikes all the tuners I know tune to 13.0:1. On turbos they go for 11-12:1. I actually run my bike at 11.0:1 (the WBC includes a gauge) for safety because of the sustained high speed high boost runs it does. As Sean says, it would make more power weaker, probably 250-260hp, but probably would not survive the experience.

Any engine under boost is much more susceptable to detonation. As a safety measure, they are tuned fatter at WOT than non-boosted applications. It's worth the trade off; even tuned richer, a turbocharged or supercharged engine will still make more power than it would if it were naturally aspirated and tuned leaner.

loudhvx wrote:

The only real problem I have with the stationary EGA is that I can't sustain WOT long enough for the mixture to stabilize.

That's where the wideband comes in. The improved sensor and massively better processor gives you much more accurate information, with a much shorter sample time.

seanof30306 wrote:

If there are no accelerator pumps, I'm guessing you'd have to jet the idle circuit fat. Wiredgeorge said the difference between carbs with accelerator pumps and those without is about 2mpg. I'm guessing that's from the necessity to tune richer to compensate for the lack of accelerator pumps.

I don't know about the MPG, but I set the idle mixture to about 13~13.5, which gets the best idle, but I haven't noticed the settings affecting the acceleration. It certainly isn't any richer than the idle setting I use on accelerator-pump carbs or CV carbs. I think it's because of the following:

I believe the extra shot of fuel comes from fuel which accumulates above the jet. Remember, the jet is immersed in fuel. When it is not flowing heavily, the orifice is not really a restriction and a static fuel level resides above the jet. When you open the throttle, the fuel flows faster and the jet becomes a restriction, but there is already a small amount above the jet which sees no restriction. That small amount goes up the emulsifier tube into the carb throat, sort of like an accelerator-pump shot. When it gets used up, the mixture leans out because the jet now acts as a restriction, and not as much fuel accumulates above the jet when the throttle is open. At WOT, probably no fuel accumulates above the jet.

However, the shot of fuel still takes time because it has to travel a long way. It's not like an accelerator pump which has an immediate exit at the venturi. Because of this, you'll notice, on slide carbs you can't whack the throttle to WOT instantaneously without a stumble or stall. You have to roll it on. (Or limit the whacks to a small amount of opening.)

The accelerator pump helps fill in, especially on bigger engines and carbs. CV just control the throttle opening so whacking it WOT won't cause a problem.

Mechanical-slide carbs require you to "feel" the engine vacuum. When you pull the throttle, you can feel the engine vacuum pulling back. If you open too fast, there is no pull and you've opened the throttle too fast. It's noticeable in performance. If you chug along and roll it to WOT without feeling the vacuum, the bike will accelerate smoothly, but not quickly. If you roll the throttle while feeling the vacuum, it definitely accelerates faster.

It's fun when you hit just the right throttle roll. I like mechanical carbs because they are so simple and easier to tune with way fewer variables.

Post edited by: loudhvx, at: 2007/07/26 15:18

14.7:1 is generally not the best for economy at cruise, it's usually around 16.5:1 (or higher).

To tune for economy with a wide band oxygen sensor is relatively easy. Look up the sensor output at ~16.5:1 for your sensor (~2.65 volts for the DIY-WBO2, for example). Put that value as your EGO trigger point in MegaTune, and if your EGO ± Limit (%) is sufficient (given the state of tune of your VE table), you should be getting good economy.

You can try mixtures leaner than 16.5:1, however you may get a lean 'surge' that feels like someone is pushing the car forwards, then pulling it backwards. Generally, this is a sign that you are too lean.

This and much more is available in the tuning section of the MegaSquirt MegaManual

I think Lance had his Corvette running 18:1 at cruise, but he said the slightest touch on the throttle would send it into lean surge.
But he was getting "Excellent" fuel mileage

steell wrote:

14.7:1 is generally not the best for economy at cruise, it's usually around 16.5:1 (or higher).

To tune for economy with a wide band oxygen sensor is relatively easy. Look up the sensor output at ~16.5:1 for your sensor (~2.65 volts for the DIY-WBO2, for example). Put that value as your EGO trigger point in MegaTune, and if your EGO ± Limit (%) is sufficient (given the state of tune of your VE table), you should be getting good economy.

You can try mixtures leaner than 16.5:1, however you may get a lean 'surge' that feels like someone is pushing the car forwards, then pulling it backwards. Generally, this is a sign that you are too lean.

This and much more is available in the tuning section of the MegaSquirt MegaManual

I think Lance had his Corvette running 18:1 at cruise, but he said the slightest touch on the throttle would send it into lean surge.
But he was getting "Excellent" fuel mileage

I'm certainly no expert, but, I've been burning chips on my TBI injected Firebird for some time. Stoich is considered the optimal A/F ratio because it is a complete burn of all the fuel and oxygen in the combustion chamber, yet, does not damage the engine, as ultra lean conditions do. When Honda first introduced their CVCC engine, for example, they weren't initially required to have catalytic converters because they could beat all EPA requirements without them. The only reason Hondas have cats is because US automakers lobbied the government over it. The CVCC engine (Compound Vortex Controlled Combustion) was designed to operate at stoich most of the time.

Going leaner than stoich is very dangerous. The leaner you go, the hotter the combustion gets, and the worse detonation gets. It's much more dangerous with a carb, but, can be accomplished (to a degree) with fuel injection. The problem with running leaner than stoich is, the SLIGHTEST increase in load, even going up the gentlest rise on the highway, causes immediate detonation; especially when you're loafing along in an overdrive gear, as anyone running that lean to improve economy will surely be doing. The EFI computer deals with this by having tables which richen the mixture when they get a signal from the knock sensor. The problem is, that richening is usually pretty significant, and more than negates any gains you make by going ultra-lean. Then you have to deal with knock sensor sensitivity. Not sensitive enough, and you can start looking for holes in the tops of your pistons. Too sensitive and the sensor is giving false positives for lifter noise, etc. I and a lot of other TBI tuners I know spent years messing with those knock sensor tables, trying to find the perfect "sweet spot" that would allow us to safely run ultra lean, and ultimately decided stoich makes more sense.

Aside from detonation, running your engine ultra lean raises combustion temperatures dramatically. The engine isn't designed to dissapate that much heat. On an air cooled engine like ours, you can run into serious warpage issues. Even if they're not immediately apparent, they will definitely affect engine life.

My father was a P-38 pilot in WWII. There was a serious problem in the Pacific; most of the P-51 Mustangs were in Europe, and they were the only fighters with the range to escort the bombers on missions over Japan. Charles Lindburgh (the guy who was the first to fly solo over the Atlantic) came in and managed to literally double the range of the P-38 by using ultra-lean mixture settings (pilots could adjust the mixture from the cockpit).

My dad used to tell stories about this. The powers that be resisted Lindburgh's settings, because they were sure they would destroy the engines. After some experimentation, they found they could closely monitor cylinder head temperatures and negate that, but only under perfect conditions. If the planes were cruising at a constant altitude, under a rock-steady throttle setting, with no wind gusting, there wasn't a problem, but, if a pilot wasn't careful and let the cylinder head temps get too high, or hit a gust of wind, the engine could blow in a moment.

Cars and motorcycles don't operate under those conditions. Even the flattest road has rises and falls, and we're constantly making small, unconscious adjustments to the throttle. We also have much less airflow around the engine than a P-38 cruising at 300+mph. Unless you're driving on an ice road in Canada (PERFECTLY flat and VERY cold), the risk/reward just isn't there to run ultra lean. Beside, we're not talking about fuel injection here. Without knock sensors and incredibly precise timing tables, all you're looking at when you go over stoich on a carbed motorcycle engine is some serious repair bills.

Post edited by: seanof30306, at: 2007/07/27 10:18

Post edited by: seanof30306, at: 2007/07/27 10:39

You are stating your opinion, and unsubstantiated info as facts, but if you are correct then you should be able to post some links to verify what you are saying.

Tell me you have an eddy current dyno with a four gas exhaust analyzer in your garage, and you burn the chips for your TBI Firebird based on info provided by the above, then I'll be impressed