Tuning with O2 sensor

[b4black]

OK, I have a voltmeter hooked up to my oxygen sensor. What values am I looking for and when. At WOT, they will be ~860mV and slowly drop as I accelerate.


To lean or fatten the mixture, which is better to change, the rods or the hangers? Are the rods corse and the hangers fine tuning?

 

[Freddie's Buick]

Rich,

I'm hoping that you have an OTC Monitor 2000?

If it's anything like the one I have, lean is anything below 450 mV. Rich would be anything above 450 mV.

Mode 4 gives an exact mV reading on the O2 sensor. All ECM Buick engines will read 450 mV at 'Open Loop'.

I've often wondered why is it that when the engine is in Open Loop and the carb is idling rich, the O2 sensor is still emiting a lean signal? Though it's contradictory, I assume that one has to wait till the O2 sensor is hot enough for it to emit a proper reading, (600 F).

From there on, (in Closed Loop), one can see that the mixture is constantly changing on an OTC, digital voltmeter, or a mixture control gauge.

As for the rods and hanger, WOT is the only point you're going to be tuning at. On mine it is set for a very rich mixture. When I floor it I can see the gauge light up to the max yellow bars. So I know it's working. Good signs for a decent mix are no black smoke, no charred plugs and the O2 sensor is in great shape and not fouled.

I guess tuning is a touch and go type situation. Besides, my feelings are that most of the more elaborate tuning methods are really for full out racing. This is were constant high speeds depend on an exact and specific mix.

 

[Freddie's Buick]

One thing I did forget to mention. It is unusaul for your reading to drop off during acceleration. It should only increase. Your engine is leaning out at WOT. Is that normal for turbos?

 

[b4black]

I have an OTC, but the update rate is too slow. I can see my O2 sensor go lean-rich-lean so fast (on the digital multimeter) that the scan tool misses it. When they scan tool says 0.820, is that peak, average or a random reading during the 2.5 seconds? The DMM is much more precise and accurate.

When too rich at WOT, the Buick V6 stumbles like it's out of fuel. I don't ever see black smoke.

When cold, the O2 sensor can only generate low volatge (lean) signals.

I'm not worried about closed loop. I'm worried about WOT. I want to know what mV to target and when for power without knock. I have read that 0.805 is 12:1 air/fuel. My readings vary throughout the run. I would guess that I should target 0.805 at the top of each gear, but I'm not sure.


More than likely, the O2 sensor isn't accurate enough, but it's all I have right now (and I'm racing Saturday )

 

[gofstbuick]

IMHO, the O2 is ignored during WOT, but if the readings are dropping it is going lean. Prior to WOT, the readings should be high (as you're accelerating). The refresh rate of a DMM is too slow to tell if your O2 is functioning properly. A better method is to use a graphing multimeter or digital storage oscilloscope. Using one of thest tools will give accurate O2 response times in milliseconds. And the test proceduretakes around 10 minutes. Also, you can see if your engine is misfiring. If you have access to a tool like that, then I can explain a O2 sensor test procedure.

 

[b4black]

The oxygen sensor is only a couple of weeks old. How can this DMM be too slow if it's easily 100x faster and has 10x the resolution compared to a scan tool. The precision/accuaracy of the DMM far exceeds the temperature dependent O2 sensor.

The DMM is connected directly to the sensor, so it's not going thru the ECM. The Carb's seconardies are controlling the fuel mixture.

During a run at WOT, the O2 mV are going to vary some as rpm's change and gears shift and boost levels change. I just want to know what A/F ratio is good and when.

 

[gofstbuick]

The DMM refresh rate is too slow. Here is the test using a fluke 98 (or simular tool) and propane enrichement tool: Insert a paperclip between the O2 sensor weather pack so you can obtain a reading. Connect the positive lead to the paperclip and the black lead to a good ground source. Leave the sensor connector (between the sensor and ECM) connected, because you want to see what the computer sees. Set time line to 200 milliseconds/division(ms/div) and set voltage range 0 to 1000 millivolts (and 200 mv per division), trigger down slope @ 560mv. This will allow you to see misfires on the graph once the O2 sensor is warmed up and to measure response time. A misfire will resemble high frequency hash marks though out the O2 signal (graph) line. Now to test response time, flow propane into the air cleaner snorkel and the graph should flatline above 800mv, after it flatlines, shut the propane off and let the readings drop, just as it is starting to level off (below 200 mv), snap the throttle hard (this will initiate fuel enrichement) and drive the graph up. Look at the graph and take your measurement between just where the line starts to turn up and just before it starts to level off at the top. The time to go from full lean to full rich should be less than 100 milliseconds. A new sensor should respond in around 50ms (about 1/4 of the space between one time division).

I did check with a carb buddy of mine and he indicated that changing the rod are a coarse tune and changing the hangers is a fine tune (will change the rate or duration of richness). You'll probably get the response you want by just changing the rods.

 

[gofstbuick]

I also try to set my paddle travel to the 5/32 mark (on the higher side of the specification). Try fine tuning the lean and rich stops, if you haven't done so already.

 

[b4black]

OK, now that's pretty cool. I guess I can see where my DMM (it's a Fluke 87 III, btw) would be too slow to pick up misfires. A graphing MM would be very cool for what you're suggesting (and recording ¼ mile runs).


My car was falling flat in WOT 2nd and 3rd gear. I found a GM service bulletin that suggested this could be from running too rich. Scan tool was showing over 0.900 mV. I changed the stock rods and now it's usually about 0.860 WOT, but it tends to fall off a little during a run, not much, maybe down the 0.810 mV. It doesn't fall flat anymore. (I found it ironic that I was too rich after hearing for years that the Q-jet could never be richened enough to run more boost.) I bring my collection of rods and hangers and maybe do some fine tuning tomorrow at the track.

 

[gofstbuick]

If you're close, then maybe the hanger is the answer, because then you can change the duration of rich or lean.

 

[bossman]

b4black, could you "theororetically" increased your boost in you rich condition and gain more performance?

 

[b4black]

Yeah, raising the boost would make sense. With the stock AC box and full exhaust, I doubt these could get past 9 psi (in other words, the wastegate isn't even needed, the retrictions limit boost). Now it's about 10-11 now with a K&N cone and gutted cat. Changeing the WG actuator rod length won't help. I might try unhooking the converter shell from the DP and run an open exhaust.


I'm also more concerned with getting it running great and in tune, before upping the boost too much.

 

[b4black]

Made 7 runs on Saturday My best didn't improve by much (15.88 @ 82 mph), but I was much more consistent. Last time I only had one run in the 15's. This time 3 of the 7 were 15's and the others were 16.0's or bad traction (cold day!)


The stumble was still there, but I have more information. Changing the metering rods didn't do much. I even went as lean as DD's which are very thick. Upping the boost to 13 didn't do much. I see some light knock on the gauge, but it doesn't seem to correlate with the stumbling. Ambient air temps didn't have a noticable effect.

I had the DMM connected to the O2 sensor and saw someting I don't see on the street. The O2 reading were all over the place going thru the traps. Rich-lean-rich-lean. I think what's going on is misfiring under high rpm/high boost causing erratic oxygen levels. The ignition is all stock except the plugs (R45TS @ 0.050? gap). I have tried there diffent plugs:

1. R44TS @ 0.040
2. R43TS @ 0.035 stumble improved, idle rough
3. R44TS @ 0.035 stumble same
   [/list=1]
   
   I think the tighter gap helped. I what to buy a MSD 6AL BTM, but they are expensive. Maybe I will try a hotter coil first, since I think you need a coil upgrade with the MSD.

 

[Freddie's Buick]

Rich,

I swear by MSD. You could run a new coil with a hotter spark I suppose, but it is'nt a 'must have'. MSD kicks butt either way. With the turbo you have, the MSD BTM would be an awesome asset.

I know it's expensive, but it is truly worth it. On my N/A application the differance was radically noticable. (And that was just the MSD-6A).

It was one of the very few upgrades I 've installed with no regrets in investment. And that's coming from the cheapest bast@#d you'll ever know!

 

[bossman]

I have to ask this guys. Why buy an MSD with the retard when there is already a retard system in our cars?

 

[b4black]

The ESC reacts (overreacts) to knock after it happens. The BTM allows you prevent it in the first place.

It also allows you to alter the timing curve to some extent. This is even more important on the ECM cars since they can't be altered any other way.

 

[Ttype83]

Rich,
I'm thinking(stop laughing) that the lean-rich-lean-rich maybe due to the M/C not keeping up with the ECM or vise-versa...

If you are running rich(as we all are) at WOT hotter plugs should help...could that be the R43TS???? I was running these before the rebuild. I'm now using with Autolite 666 gap @ .40 and I'm running pig rich all the time I'm going to check Cap, rotor, wires aswell as timing this weekend just to rule them out.

Jim

 

[b4black]

At WOT the M/C isn't doing the bulk of the metering. The secondary rods are. At part throttle, the M/C goes into power enrichment OK, so I believe it's working. (I could try another good carb I have to rule it out.)

The heat range made little difference. I ran 44's this weekend anfter changing them from 43's earlier this month. The tighter gap did help.

Playing with the secondary rods did little. CP to DD gave little change.

Seems to be pointing to a weak spark?

 

[Ttype83]

OH that's right...M/C does it's main job before WOT.....
I changed my cap/rotor and checked the wires and timing.
The wires tested good, but the timing was retarded 6 degs.
This helped get me passed 7 psi with out falling on it's face untill 10-11psi. I bought 43's and will install them in before saturday(BOP) I can to remove the wastegate that I got with the other turbo due to leaking from the pressure side to the vac side. It still will build boost but I want to save it for saturday. I reinstalled the one I made, but it's to short....turned all the way out I still can boost 18psi or more at WOT...

 

[b4black]

If the ignition is the problem, any improvement should help. I went ahead and ordered a coil from Summit. It's there own brand, but it should be an improvement over the stock coil:
http://store.summitracing.com/partdetail.asp?part=9611

It's $35 with free shipping, but a $8 "handling fee".
The MSD stock style coil is $70. The MSD 6 BTM is $335. It's actually cheaper to combine a MSD 6A for $132 and an add-on BTM for $180.



Jim, are you racing Staurday? I'm glad you have your problems sorted out. Racing is the best way to tune, IMO. I'd go every weekend if I could. You get can get 18psi? I definately have a problem. 12 is tops for me. I hope this coil does the trick. :fingers crossed:

I'm not sure I follow on the actuators. A leaking actuator shouldn't keep you from building boost - it will stay closed. A blown actuator won't open and will overboost.

If it's opening too early, maybe the spring is week. Try adding an external spring to help it. Most hardware stores have a pretty good selection.

GOOD LUCK!!