Using a Gas Analyzer with AFR to set TPS

Lorenfb · 09:04 AM

I decided to set the TPS on my '04 BCR for better starting, idling,
and throttle response. In the past, I just adjusted it using a voltmeter
and finding the ideal setting by trial & error which was an improvement
over the .40 volts setting.

Recently, I obtained a OTC Stinger (TIF5GA) Gas Analyzer which has
CO, CO2, HC, NOx, O2, & AFR displayed. I disconnected and removed
the O2 sensor on the BCR and used the hole for the analyzer probe.

The following are the results of the test at 3000 RPMs:

TPS set to .40 volts

AFR = 18, CO2 = 9.0%, CO = .5%

TPS set to .70 volts

AFR = 10, CO2 = 7.5%, CO = 3.2%

TPS set to .50 volts

AFR = 13, CO2 = 12%, CO = 1.6%

Setting the TPS to .50 volts results in a better cold start & idle.
Also, the engine has better throttle response over the full range.
Using this approach can further optimize the TPS setting, which
should be done, when different exhausts are added.

Obviously, a better test would have been performed using a dyno.
This test, though, indicates that performance improvements can be
made by setting the TPS to a value other than the standard setting
of .40 volts always used. Also, it's always better to be on the rich
mixture side, which results from higher TPS settings, than the lean
side to minimize pinging effects.

roger albert · 03-01-2004, 08:49 AM

It's important to actually understand what the engine computer does hwoever. Just realize you're forcing the motor to default into open loop when going >>400mV. Not that I disagree at all about the better idle and response possibilites with mild bumps, such as up to 450-500mV, but there are of course fuel economy and emission penalities to be paid.

Loren, I'm curious about the measurements. Were they repeatable? I'm surpise that a TPS of 400 and 700mV both yield an AFR of 18, but that the intermediate 500mV yields a much richer 13. The CO readings look about right though.

Interesting. I also agree with the comments that power and safety are both served by erring slightly on the side of richness than leanness.

Note also that folks shouldn't expectd the same results (or steady state mileage) if the Lambda sensor is left out.

I'm still puzzled by the 18AFR at .700v. I'd say it clearly indicated the engine computer was faulted to a fixed value, but I'm not sure how to reconcile that with the CO and CO2 figures.

Loren or anyone else have any ideas?

Rapid Dog · 03-01-2004, 09:02 AM

I think we should have a SoCal TPS Setting Meet so that one of you in the know can set all our bikes to peak performance.
What say Loren?

Lorenfb · 09:27 AM

The AFR at .70 volts was a typo, S/B 10. You're correct about when it goes
closed loop. The AFR is right on at 14.7 and the CO comes way down below
.5%. The key, though, is that the throttle response is not that affected by
the O2 and closed loop. Also, better starting results, since the system is
also open loop. It may well be that the .40 volts setting at idle & other
RPMs places the AFR on the lean side (as can be seen AFR = 18) and the
O2 "pulls" to center (14.7).

More testing needs to be done on a dyno, though. At least this is a starting
point.

Lorenfb · 03-01-2004, 09:10 AM

A simple test for all is to use a voltmeter and:

1. check the present setting
three wires - ground, +5 volts, TPS output (varies with throttle movement)
2. adjust it to .50 volts
3. do some cold starts and riding
4. if unhappy or negative results, reset TPS to initial setting

If you like, I'll ride north and bring a voltmeter & my BCR some weekend.

repoe3 · 03-01-2004, 10:01 AM

could you layman=ize the acronyms for us non-tech geeks, please...

repoe3

Rapid Dog · 03-01-2004, 10:17 AM

Yeah Loren, that would be swell.

EZ-RIDER · 03-01-2004, 02:07 PM

Hey Loren;

I have found with my bike 2000, If you do the zero= Zero
And instead of adding the .006 volts leave it set at .004 volts
and lock it down, And adjust it bewteen .360 and .380 with the set screw
Then start the bike without the throttle cables hooked up, adjust the right side to match the left with the set screw. Reconnect the
throttle cable, Set the air by pass screw and call it good

Dr. Curve · 03-02-2004, 04:10 AM

Loren, take care with the mods you have suggested. Your bike will surely carbon up over time with those settings. Your readings are most likely misleading because they are taken with the engine off load. Rich settings can be deceptive. Take the bike out on a 150 mil e zip, bring it home, and check the plugs and exhaust. You don't want them too black.........just a nice light medium brown. EZ is describing a procedure that sure has worked for lost of people. I wonder what his numbers would be on your new sniffer? Best,

Lorenfb · 07:29 AM

Setting the TPS to .50 volts is not too overly rich. I've been using this setting
for about six months without any problems. When running closed-loop with
the O2 sensor, the system runs just as if it were a stock setting, i.e. the
O2 sensor "pulls" the AFR to 14.7 and the CO back down. The stock settings
are a little too lean open-loop from my experience.

I set up my Ducati using the gas analyzer with a mixture control pot within
the ECU. The Ducati has no CAT/O2 sensor and runs open-loop. Here are the
readings from it, which is basically where Ducati indicates an
idea race setup:

3000 RPMs

AFR = 12.75 CO2 = 11% CO = 4.25% O2 = 1.5% HC = 172

The CO is a little rich @ 4.25%, but the bike runs strong, no pinging,
runs cool, no backfire @ de-accel, & no burnt valves.

The benefits, as I mentioned, of setting to the TPS to .50 volts are:
1. better cold starts - not effected by the O2 sensor
2. reduced pinging effects during accel
3. better throttle response - O2 response/control to increased fuel lags
the steady RPM O2 response/control

roger albert · 03-02-2004, 07:33 AM

Dr. Curve is however right here. That rich will carbon up a good deal over extended intervals. I'd strongly recommend increased oilchange frequency with that rich of a setting. (as well as tanking

Rapid Dog · 03-02-2004, 07:35 AM

me want!

brad black · 03-03-2004, 02:57 AM

loren - have you tried relating the setting to a throttle fully closed setting? like 700 at idle being 350 closed then wound open another 350.

just makes it repeatable and transferable to other bikes.

i did a zero = zero today on a gs and made it worse! the owner came back to say it surged and wasn't very nice. it's the nicest gs with pipe i've ridden by far! of course, telling him that didn't help. bugger.

Bob Hancock · 03-03-2004, 04:08 AM

Loren,
This sure is an interesting discussion, but I wonder how relevant it is if the bike is running fine. My new (to me) BCP starts and runs well as stock. I bought a cat elim pipe mostly for a little more sound but haven't installed it yet. Am reluctant to spoil a good thing.
OK...my question....is all this ECU setup business to fix problems you guys have or to make the bike run even better? Just wondering if I should consider doing something or just enjoy the tech talk.
Thanks,

repoe3 · 03-03-2004, 04:33 AM

on the heel of bob's question...one thing that should be considered by all is location of machines and available fuels resulting in differing running conditions.

basically the further west you go, the fuel gets worse (in PC talk, more refined/regimented formulas) and can result in pinging, surging (more or less), etc.

if your bike runs fine, then great. just understand that as soon as you start changing things, then you may need to adjust things accordingly. switching out the cat to a Y pipe should not do anything but get rid of the feet oven currently mounted to the bike.

my last S, i played around with different settings with the TPS as i had mods, but not yet a chip. it was a good temporary fix, but ultimately the proper chip and the TPS at the factory setting+ was the trick.

repoe3

Lorenfb · 03-03-2004, 06:09 AM

Brad;

The throttle body butterflies are at the stock setting which would be a .40
TPS. I just adjusted the TPS for an additional .10 volts off the stock setting.
This is a very very small increase in TPS movement.

This adjustment may solve a number of problems w/o having to use a Power
Commander or a new chip after an exhaust replacement or because of stock
problems such as; hard starting, rough idle, flat spots, or pinging. As I said,
just try it and if there's no improvements, it takes less than 2 minutes to go
back to the stock setting.

roger albert · 03-03-2004, 06:22 AM

Agreed, little to lose outside of economy and a bit of extra carbon. My concern, in general, with these kludges, is that they do nothing at all for wide open, where one is more likely to incurr engine damage due to an overlean condition (though at WOT the factory isn't bad, and therefore there is some margin for safety) Stated another way, many folks hear the popping go away, and figure their system is better matched/tuned in general, when in reality, it is only better at idle/closed-throttle, and relatively small throttle openings.

One just needs to keep that in mind.
If you like what it does to throttle-response, idle, and starting, go for it.

Bob, If your beast is running well, this won't do much. It's just curious, and interesting, experimentation to see if a bit more can be eked out. In that spirit; cool beans.

Lorenfb · 03-03-2004, 07:03 AM

Speaking of kludges, my feeling is that if I put some of the aftermarket mods
on my BCR (a PC, a chip, or an exhaust), I'd first put the bike on the dyno to
monitor the AFR over the full RPM range. You never really know what effect
the major mods will have.

As an example, we all know that a new exhaust will more than likely require
an increase in fuel mixture. The key, though, is whether the engine now runs
too lean, over the full RPM range, causing pinging or burnt valves even with
the mods to richen it up.

My approach is to only do engine mods which you know the true/actual
final effect of, or you've fully tested for, e.g. a dyno monitoring AFR.

roger albert · 03-03-2004, 08:14 AM

A fair and safe approach, but, neglects that one normally needs to tune a bit rich on the dyno, to compensate for their being less, and often warmer air, than at speed on the street, especially if a low restriction intake is installed. In general, I agree however. That's why quite a few folks have bought AFR logging devices. They are, by definintion when used on the street, more accurate than a static dyno run.