Tweecer/emissions Tuning Teaser

[shonut]

I had the 89 (non EGR) e-tested Friday. The test is 2 parts - one is a steady loaded 24 mph, and the latter is an idle test. This discussion is about the 24 mph test.

For the 24 mph test, initially I wasn't doing anything special, except holding the temp at 190F by setting the fans to basically stay on during the test. (Normally one might allow the temp to rise to help complete combustion, but that may cause excessive NO, and with no EGR on this SHO, we don't need any extra NO). No special timing, no special fueling, but adaptive was disabled (though acheiving closed loop was no problem, the car had no trouble achieving stoich, and, in fact, the Lambse's were in the 14-15 range, so adaptive would not have been much help anyway). Timing advance at steady load and speed was 35 degrees, based on stock settings. Readings were taken with the TwEECer RT Calcon logging screen - load, HEGO, timing, coolant temp, air temp, etc. etc. were all visible in real time. Here's a screen shot.



Here are the results of the test (my result/limit):
78/67 ppm HC
0.05/0.37 CO%
544/750 ppm NO

As you can see in the results, HC failed, so I figured I had a little room to add some timing to burn a little hotter to burn the HC a little more, at the expense of NO, where I had some room. There was no point in adjusting the mixture - CO was damn near 0. So +5 on the PT spark advance adder ... and I got 40 degrees advance during the test load/speed:



The results were:
75/67 HC
0.06/0.37 CO%
321/750 NO

Hmm, this doesn't make sense, NO went down, but, well, let's try more spark, we are heading in the right direction with the HC, right? Well, this time, I added 5 more degrees of PT spark advance DURING the pre-conditioning stage of the test instead of before the run like I usually do (long story). With the TwEECer, you can write changes with the car running, on the fly, and it takes about 1 second to write a change, then the EEC restarts, and away you go with the new settings. Thus I had the tech hold the throttle constant while I wrote the change during the pre-conditioning stage of the test, where you hold 25 mph constant for 90 seconds prior to the test. However, after the car got back into closed loop, the timing was fixed at 19.5, when it should have been about 45 degrees.



Well, nothing we could do about it, so we finished the test, and lo and behold, the car passed, now at:

57/67 HC
0.03/0.37 CO%
238/750 NO

I figured we were onto something, so I had the tech set the emissions testing dyno up to do a steady state test, where we could see the HC/CO/NO readings on the screen, instead of the normal test, where the screen only shows the test speed and the timer (all that the tech is concerned about during the test, i.e. hold a steady speed, and watch the timer count down). I tried various settings, all with the tech holding the throttle steady while I wrote changes, and no matter what I did, the timing was fixed at 19.5. I even tried turning the car off and back on (still with the throttle steady and the car at 25 mph), and STILL 19.5. Finally I gave up and went home. About 1/2 an hour later it hit me what was happening. Let's see if you can figure it out.

I'll reiterate the important points ...
- throttle held steady during writing and test
- EEC restarts after write, or key cycled, both during test
- timing is fixed at 19.5 after write/key cycle

What's happening? Why couldn't I control the timing?

If nobody gets it, I'll give one more hint.

Joshua Teixeira
SHO NUT Performance
www.shonutperformance.com
sales@shonutperformance.com

[luigisho]

So much of this is over my head but I'd really like to learn. So keep this tutorial stuff going. Does KAM get erased when you reset?

[shonut]

So much of this is over my head but I'd really like to learn. So keep this tutorial stuff going. Does KAM get erased when you reset?

'reset' was a poor choice of words - perhaps 'cycled' or 're-started' would be a better choice of words. I'll go and edit the post. In this case, the EEC is restarted after a write.

BTW, since I have Adaptive disabled via the ACT/Adaptive scalars, my KAMRF is always 1.000. KAMRF is the long term fuel trim that the EEC 'learns' over time, and is one of the components stored in KAM (Keep Alive Memory). The value you see on the log comes from a table of KAMRF that the EEC develops over time, based on the results of closed-loop operations, and the required Lambse's to get the HEGO's to switch, i.e. the ratio of Lambse to 14.64 during closed loop. The axes of the KAMRF table are believed to be RPM and load. This is the 'learning lean' or 'learning rich' that people talk about.

To answer your question, no, KAM was not reset, nor is it reset during a write or re-start. At this point in time, clearing the KAM, including setting the KAMRF table values back to 1.000 requires you to disconnect the battery.

[shonut]

I figured we were onto something, so I had the tech set the emissions testing dyno up to do a steady state test, where we could see the HC/CO/NO readings on the screen, instead of the normal test, where the screen only shows the test speed and the timer (all that the tech is concerned about during the test, i.e. hold a steady speed, and watch the timer count down).  I tried various settings, all with the tech holding the throttle steady while I wrote changes, and no matter what I did, the timing was fixed at 19.5.  I even tried turning the car off and back on (still with the throttle steady and the car at 25 mph), and STILL 19.5.  Finally I gave up and went home.  About 1/2 an hour later it hit me what was happening.  Let's see if you can figure it out.

I'll reiterate the important points ...
- throttle held steady during writing and test
- EEC restarts after write, or key cycled, both during test
- timing is fixed at 19.5 after write/key cycle

What's happening?  Why couldn't I control the timing?

If nobody gets it, I'll give one more hint.

Joshua Teixeira
SHO NUT Performance
www.shonutperformance.com
sales@shonutperformance.com

Well, no takers, so here's what I think the answer is.

If you compare the (relative) TPS for each of the runs, you will see that the final run is a near-zero value. From what I understand, when you first start the car, the EEC looks at the (actual) TPS voltage, and sets this as the base voltage, i.e. closed throttle, and subtracts this voltage from all subsequent TPS measurements to get a relative TPS measurement. At this voltage, relative TPS is now zero. Any voltage ABOVE this value is considered to be a measure of throttle angle.

So when the EEC was re-started with the throttle held open, it set this value as 'closed throttle' - hence the near-zero (relative) TPS value.

Of course, closed throttle spark strategy is far different than part throttle spark strategy. Thus, I feel that the EEC was running in closed throttle spark strategy, which gave a spark advance of 19.5 degrees, instead of the part throttle spark strategy that would have given 45 degrees ...

So, if I were to do it again, I would have to write the change/restart the EEC/cycle the key with the throttle released, then go back to the previous throttle position for testing. Then the EEC would show a significant TPS, and be in part throttle strategy and allow the 45 degrees that it should have (barring detonation-provoked spark pull).

[shonut]

I took the wife's 170K mile 95 MTX winter beater in yesterday for e-testing.

The baseline test (stock timing etc., no other TwEECer mods besides keeping the fans on during the test for consistency in ECT and having a well-tuned MAF curve) gave:

HC: 10 ppm
CO: 0.05%
NOx: 360 ppm

The timing commanded by the EEC was 40 degrees for these test conditions.

This would actually have passed, but why not try for more?

So I worked through 10, 15, 20, 25, 30, 35, 40, 45 and 5 degrees of advance during a manual test, allowing the EEC to re-start with a closed throttle each time.

As expected, the HC and NOx increased with increased timing. CO stayed basically the same (remember that the HEGO's are switching, so the mixture is at stoich ...). By the end of the test, we had ended up fouling the sample train from constantly restarting the car (a restart would produce a spike of >10,000 ppm HC), so the HC had drifted up (i.e. the last test at 5 degrees showed a higher HC than expected). So the filters were changed, the e-tester was re-calibrated, and we ran the actual test at 10 degrees of timing, which gave ...

HC: 3 ppm
CO: 0.02%
NOx: 210 ppm

If you do the math, that's quite an improvement ....

The more observant ones here will note that pulling the SPOUT plug (the one near the cam sensor - NOT the octane plug near the tranny) will also produce 10 degrees of timing, so if you are having trouble passing an e-test, you can get the same improvement as I did - all you need to do is pull the SPOUT connector. Remember that MTX SHO's do not have EGR, so they sometimes have trouble with the NOx portion of the test, which EGR does reduce.


Just for fun, let's look at the effect of vehicle mileage/burning of oil ...

This car has 170K miles and burns practically no oil. The previous owner tells me that the stock cats are fairly new (?!) The 89 has 235K miles, and burns 2 quarts every 3000 miles, and has 2 year old (or so ..) SHO Shop cats .. so they are probably the same age.

At 40 degrees, they got:
75 (89) / 10 (95) HC
0.06 (89) / 0.05 (95) CO%
321 (89) / 360 (95) NOx

at 20 degrees, they got:
57 (89) / 25* (95) HC *high due to sampling error, should be more in the range of 10 or less
0.03 (89) / 0.01 (95) CO%
238 (89) / 220 (95) NOx

As you can see, the CO% and NOx are comparable, but the HC is wildly different ... In general, since NOx is the tougher one for the cats, I can presume both cats are working, so the remaining difference must be the oil burning by the 89 .... nearly 8 times the HC when running at normal conditions. Unfortunately, there's not a whole lot you can do about that with a car that burns a lot of oil, other than pulling the SPOUT and hoping for the best ...

[luigisho]

That first comparison run was impressive. I'd like to see the results of your SC car.

Lucky for me there is no emission testing in my area.

[shonut]

That first comparison run was impressive. I'd like to see the results of your SC car.

That's classified

[shonut]

That first comparison run was impressive. I'd like to see the results of your SC car.

That's classified

Actually ... it passed.

I took the blower car in for e-testing yesterday. The car has +40 cams, straight up, fairly fresh HEGO's, and all of the rest of the goodies.

I changed the part throttle spark to 10 degrees (same as pulling SPOUT) (note that this leaves the closed throttle timing stock), changed the fan-on temp to ~210, and (this is the biggie), put cats on it from my now-dead 89. (Previous testing 2 years ago without cats was a total disaster, we won't go there) This is an equal-length SHO Shop Y with cats, with a few years on it of an oil-burning 89. Other than limiting the airflow into the engine so I didn't over-run the 25 mph running test, that's it. No cam timing changes, nothing. Gas used was 6-month old Sunoco 94. We ran the test manually to see if it would pass, and once it looked like it would (just one try, no fiddling was done), we went to the official test.

Other than extreme difficulty in keeping the car below 25 mph (86 mm is WAY to big for a TB, took two tries to get the car to run a steady 25 mph ...), and me shutting the car off at the wrong time and having to re-start during the idle test, the test went well, and it passed.

Here are the results (measured/limit)
Running (25 MPH)
HC 37/58
CO% 0.05/0.32
NO 77/435
Dilution 14.4 (VALID)

Idle
HC 186/200 (byproduct of flooding the car on startup, our 'pre-test' was 60)
CO% 0.02/1.00
(NO not tested)
Dilution 13.4 (VALID)

Next time, I'll bolt on the 55 mm 5.0L TB though, that 86 mm TB is just a
little too big to be able to run 25 mph with any degree of control.


To compare, this is the running test for the 89 with the same set of cats, and stock timing. The 89 burns oil (so does this car, but not so bad ...), and the limits are higher, because this was a while ago.

78/67 ppm HC
0.05/0.37 CO%
544/750 ppm NO

Reducing the timing to 19.5 on the 89 gave:

57/67 HC
0.03/0.37 CO%
238/750 NO

I never did test 10 degrees on the 89, it got dismantled before the next round of e-testing came up.


To see what the 10 degrees of timing is worth, here's a stock 95 that can go 5000 miles before burning much more than a quart of oil, even when beaten, with stock timing:

HC: 10 ppm
CO: 0.05%
NOx: 360 ppm

Here it is again with 10 degrees of timing, as you can see, big % improvement.

HC: 3 ppm
CO: 0.02%
NOx: 210 ppm


So, would the blower car have passed with stock timing? Maybe not, but it's an easy mod - just pull the spout plug (or run the car at 10 degrees with your TwEECer, like I did).


Now who said +40 cams could not get through an emissions test?

[somedude]

its amazing what a good tune will do to a car and how clean a highly modded car can run. thanks for this thread another one to save for the books.