Question about ECU logic for timing and fuel mixture - LotusTalk - The Lotus Cars Community
 
LotusTalk.com is the premier Lotus Forum on the internet. Registered Users do not see the above ads.
Reply
 
LinkBack Thread Tools Display Modes
Old 02-16-2013, 04:41 AM   #1 (permalink)
Registered User
 
Join Date: Feb 2012
Location: Georgia
Posts: 184
Question about ECU logic for timing and fuel mixture

So here's the back story. I'm trying to get my '89 non-SE through it's final emissions test. So, naturally, on it's final year, it fails...but just the low speed NOx, and it just barely fails. Everything else still passes. The NOx numbers result from high temps in the cylinder. I've tried a few things that have helped a little, but it's still not there.

There's no EGR valve, so nothing to tweak there. I'm already running cooler plugs. I even tried pulling the actuator arm off the wastegate so no boost would build, and I hoped, less heat would be generated. No dice. The other two easy routes to getting temps down are to run richer and to retard the timing a bit...but the ECU controls richness and timing, and attempting to reprogram the ECU for a one time test isn't in the picture.

What I'm trying to do instead, is send misleading signals to the ECU so it retards the timing for me, but doesn't cause the check engine light to come on...which would also cause the car to fail emissions. I'm leaning towards timing over richness, because an overly rich mixture will cause other emission issues and the BLM numbers look good right now.

So, my question is this. Does anyone know the logic and sequence of algoriths the ECU uses to set timing (and richness too just in case)?

I've got the engine section of the manual, but it only gives hints as to what factors are used, but nothing really about the logic. Air temp is often used as a timing driver on many cars, so I wired up a potentiometer in parellel with the MAT sensor and, with Freescan watched timing all the way from letting the ECU correctly see ambient temp of 18C all the way to well over 100C. Unfortunately, the timing didn't seem to be affected, so it seems that air temp isn't a major factor if it's used at all.

I know it retards timing if it senses knock, but a) enough knock to retard timing would trip the CEL, and b) I don't know much and for how long timing is retarded for knock anyway, so I don't want to go that route. The engine manual seems to indicate that coolant temp is used, but I suspect it's primarily to advance ignition during warmup...which is the opposite of what I need. I don't "think" the ecu retards ignition if it thinks the engine is overheating. I may be wrong, and I'll try that next, but I'd like to have a clearer direction that just fishing around.

I guess I could just get a cheap Ebay cat to try to pass the test, then resell it. I've seen relatively new Magnaflows going for under $100, and I've spent that much on retesting and other stuff trying to get by, so now it's become a matter of principle / blind obsession

I'm only looking to get 5-10 degrees of retardation. I think that should do it. If anyone's got the ECU tables and/or logical algorithms that would tell me which signals to modify for fuel and timing, it would be very helpful.

Thanks!

Last edited by dalefield; 02-16-2013 at 05:36 AM.
dalefield is offline   Reply With Quote
Sponsored Links
Advertisement
 
Old 02-16-2013, 05:56 AM   #2 (permalink)
Registered User
 
Vulcan Grey's Avatar
 
Join Date: Jul 2006
Location: Colorado USA
Posts: 3,117
Which cam pulleys do you have? What color dots are on the pulleys that are on the car now? The stock 89 are red and blue and are already retarded. Someone might have switched to the green dot advanced pulleys...

Rather than trick the car into beating the test, the car should pass if running right.

What are you getting for your BLM vs BLM cell now?

Which memcal chip does the car have? Has it been replaced with a Marcus chip or superchips?

Are you making sure the car is fully warmed up before testing?
__________________
Travis
Vulcan Grey Esprit 89 SE
http://picasaweb.google.com/lotusse

Last edited by Vulcan Grey; 02-16-2013 at 06:10 AM.
Vulcan Grey is online now   Reply With Quote
Old 02-16-2013, 06:13 AM   #3 (permalink)
Registered User
 
Join Date: Feb 2012
Location: Georgia
Posts: 184
Wrong timing. I'm talking about ignition timing, not cam timing.

In any event, I'm running the Green dots. The engine runs great. With both Freescan and Espritmon, I've checked that all the parameters are in good shape. it idles like a top, starts every time, pulls very strong, oil pressure is good. compression / leak down test results are good, etc. Of course if everything is like new, it passes. But it's not new. It's just a 25 year old car with a 25 year old catalytic converter, that I don't want to replace for a 15 minute test. If I have to, i will, but it's pretty common to retard the timing a few degrees on distributor-based iginition systems to deal with marginal NOx numbers. This is just a temporary adjustment to sensor output signals to achieve the same effect.

Stock memcal, fully warmed up, BLM around 128.

Last edited by dalefield; 02-16-2013 at 06:43 AM.
dalefield is offline   Reply With Quote
Old 02-16-2013, 07:00 AM   #4 (permalink)
Registered User
 
Join Date: May 2008
Posts: 242
I think you're going about it wrong. Can you explain exactly which measurements the car is failing on, what the limits are and what your results are? Running richer won't necessarily make your emissions go down. Pulling timing, same deal, your HC will probably go up a bit and it's quite likely that you'll be running more, not less boost. Timing helps to get more heat into the cat(s). I find this especially useful for idle sniffer tests, along with heat-wrapping the cat(s).

Like everything it's a balancing act. For example, retard too much your COx will go way up, advance too much your NOX will go way up. First, let's see what you're failing, then see what you might do to correct it.

-Michael
hackish is offline   Reply With Quote
Old 02-16-2013, 12:10 PM   #5 (permalink)
Registered User
 
Join Date: Feb 2012
Location: Georgia
Posts: 184
Quote:
Originally Posted by hackish View Post
Like everything it's a balancing act.
Exactly... The science behind emissions management fills text books and even entire careers, so i tried to be brief and summarize. I don't think I was either brief or successful at summarizing.

I have a good handle on the complexities of the emissions process, but this started with me just wanting to get past one last test...if I had a distributor, it would be easy to adjust timing and see if I can scrape by. With an ECU it's tougher, but still fairly easy to manage sensor readings...literally, a minute to unplug the sensor, stick the potentiomer in the circuit, replug the sensor, and start turning the dial on the POT to get the setting you want.

As you mention, it's the balancing act that makes emissions management tough. The cat can't clean CO and HC emissions without sufficient oxygen. Conversely, low amounts of oxygen are needed to clear up NOx, which as I mentioned in my original post is where my car is failing. You can't optimize for one situation and pass the other. You can't have it both ways, so you have to find the best compromise...and when it's close, a small timing adjustment can often help clear the hurdle.

I'm pretty sure my root cause is a small exhaust leak at the manifold caused by a broken stud (which is introducing the oxygen into the exhaust that makes NOx cleanup difficult). I've verified the leak by smoking the exhaust, and while I can't prove that's the cause of the fail, it typically is a major issue with NOx.

The challenge for me is that I plan to change my exhaust after the final emissions test. Anyone who's worked on the manifold, knows it's a very difficult job. To get the broken stud, and properly seat a new one, I think I'll need to pull the manifold first. That's pull it and put it back on just for the test...then pull it again right after the test. I'd rather have a fork in my eye.

My results are close enough that I'm pretty confident a 3-way cat will do the job even with the manifold leak (new cats are miles ahead of 25 year old cats with 25 year old technology in terms of efficiency and cleaning ability), but that's also a lot of engergy (and cash for a new cat) for something I don't even want to begin with. I think I could get some of the cash back, but the time to do all that work for a 15 minute test is at least equal to a poke, if not a fork, in the eye.

My easy option would be to retard timing and see if it's enough to pass...which is tough to do without a distributor. Realistically, a reprogrammed ECU just for this test isn't an option. Lying to the ECU is, however.

If I can lie to the ECU, the algorithms stay the same but results change for free and for 10 minutes of effort and a $2 POT, and after the test, it's undone just as easily ...if I can figure out what lie to tell. Tuners do this frequently to advance timing when running higher octane gas than the ECU was programmed for. I'm just trying to go the other direction for the test. The root cause exhaust leak will be addressed when i replace the entire exhaust system after the test, but I'd rather not have to do it two more times before the test if at all possible.

That's why I'm asking for ECU logic rather than emissions help. It may not work, but relative to my other options, it's worth exploring first.

Hope this makes sense.
dalefield is offline   Reply With Quote
Old 02-16-2013, 12:17 PM   #6 (permalink)
Registered User
 
Join Date: Nov 2009
Location: New Jersey
Posts: 1,458
If all you are trying to do is fool the crank sensor, maybe an adjustable bracket would do the trick.
David Teitelbaum
jtrealty is offline   Reply With Quote
Old 02-16-2013, 03:26 PM   #7 (permalink)
Registered User
 
Join Date: Apr 2011
Location: Nevada City, CA
Posts: 202
I understand what you are trying to do, but short of creating a variable digital delay line from the crank sensor, your looking for some other analog sensor who's output you can vary to cause the ECU, itself, to retard the timing. I have no experience with the 4 pot ECU, but in the V8 ECU, ECT advances the timing when the engine is cold and only the knock sensor retards the timing. There is no external sensor which can be used to fool the knock sensor output in the V8 ECU, it uses a digital time count algorithm to create a retard command prorportional to the magnitude of the knock signal pulse level. There is logic in the V8 ECU to implement Boyles Gas Law, the output of which, is also used to make timing corrections based on MAP, Baro and IAT, but changing any of these analog inputs would also change the fuel mixture. You could contact John Welch at WC Engineering, he's the most knowledgable person I know on the internals of the 4 pot ECU.
Wayne Ellison is offline   Reply With Quote
Old 02-16-2013, 04:09 PM   #8 (permalink)
Registered User
 
holmz's Avatar
 
Join Date: Sep 2006
Location: Antipodes
Posts: 1,986
Images: 5
NOX is the result of combustion tempurature.
The temp is the result of the mixture and the advance.
Adding fuel and retarding the ignition will both reduce the temp which will reduce the NOX.

If you are at a stochiometric mixture, or really a mixture where you are maximum exhaust temp, then whether you go rich or lean, you will reduce the combustion temp.
The problem is that if you go too rich you will increase the CO, and generally lean has higher temp than rich.

Usually you will want to increase the throttle at idle, and reduce the advance.
Reducing the advance gives you less NOX, and the increase in the throttle allows it to keep running by providing more air (and fuel).
holmz is online now   Reply With Quote
Old 02-17-2013, 06:37 AM   #9 (permalink)
Registered User
 
Join Date: May 2008
Posts: 242
I suggest purchasing an AEM FIC and using that. It will allow you to retard the timing because it has a programmable delay in the crank/cam signals. Of course the cost of fixing the exhaust to cure the problem probably doesn't make this a cost effective solution.

Don't forget the cat saturates and with a working O2 sensor. Unless you're throwing a system too rich or system too lean code, normal cycling of the sensor should be good enough to keep the last stage of the cat fed with O2 and HC. A small exhaust leak typically does not introduce enough into the system to affect this.

I have a Ferrett 16 5 gas analyzer I typically use to diagnose these sorts of issues and frequently replacing the cat with a quality one fixes problems on an otherwise healthy engine. I prefer Random Technologies cats as the OEM whitebox specials are as good as the dollars you paid. In my experience the stainless core units also light off quicker and give slightly better performance.

Hope that helps

-Michael
hackish is offline   Reply With Quote
Old 02-17-2013, 02:23 PM   #10 (permalink)
Wingless Wonder
 
carbuff's Avatar
 
Join Date: Jun 2004
Location: On the Apex
Posts: 2,265
What happen to NOx if you dump a high percentage of alcohol (isopropanol) in to the gas tank? We used to do that "back in the day" to pass emissions? (I know the engine will run like cr*p but it's only for the test, right?)
__________________
Atwell Haines
'88 Esprit
Succasunna, NJ USA


"Not all angels have wings." - Turbo R
carbuff is offline   Reply With Quote
Old 02-18-2013, 05:19 AM   #11 (permalink)
Registered User
 
Join Date: May 2008
Posts: 242
I've been present when customers have tried this. I'm not an e-test technician but I frequently get hired for tuned and tunable cars to adjust them for a pass. Off the top of my head I can't remember what happened to NOX per se but I've never seen a car do better with these guaranteed to pass things or methyl hydrate or E85. Why, I'm not 100% sure but I suspect it may have to do with cat temp.

-Michael
hackish is offline   Reply With Quote
Old 02-18-2013, 05:57 AM   #12 (permalink)
Registered User
 
Join Date: Feb 2012
Location: Georgia
Posts: 184
Thanks for all the replies. This was helpful, but I don't like the answer. I think I'll just have to suck it up and pull the exhaust manifold. The idea about using a bracket to move the the crank sensor a few degrees is intriguing and low cost, but it's a fair amount of work and I don't know how drivability would be affected or if it would work. The FIC is a really cool idea, but it's expensive and the new cat would "probably" work but is also a little costly, and with either of those options it still might not pass after doing a lot of work. Tricking the ECU is almost no work, so I thought I'd explore it and if it didn't work, I wouldn't be out anything. The level of effort and cost for the others, though, tells me I'd be working hard to find a way to avoid working hard, so it looks like I'll be pulling the manifold soon.

As for alcohol in the fuel, if it's 99% (lower % may contain too much water), it burns very cleanly (but also a little hotter), so it can help with HC and CO levels if that's where the problem is. It can help NOx levels in that if you do plan to retard timing to improve NOx levels, it can help offset the rise in HC and CO levels that would come from the timing change. I've known of folks with hollowed out cats using high ratios of alcohol or acetone to pass the test with flying colors, but I don't think it's as helpful for NOx except to offset other engine adjustments.
dalefield is offline   Reply With Quote
Old 02-18-2013, 11:26 AM   #13 (permalink)
Registered User
 
eriks4's Avatar
 
Join Date: Jan 2011
Posts: 24
Just s quick reply to a part of the first question regarding ECU control over the richness. It is possible to change the 14.6 reference AFR. It's documented in the GM P4 document and is part of the mode 4 input message. You can only do this with the engine running (as far as I know), and with Espritmon you need to go to the menu in which you control the quad drivers. There are two buttons in this menu (inc AFR and dec AFR) with a readout in between. With these buttons you can change the reference AFR and hence make the car run a bit richer or leaner. If your version of Espritmon does not yet (correctly) support this, let me know and I will email the latest version.
Have fun
Erik
eriks4 is offline   Reply With Quote
Old 02-18-2013, 12:03 PM   #14 (permalink)
Registered User
 
holmz's Avatar
 
Join Date: Sep 2006
Location: Antipodes
Posts: 1,986
Images: 5
Quote:
Originally Posted by dalefield View Post
Thanks for all the replies. This was helpful, but I don't like the answer. I think I'll just have to suck it up and pull the exhaust manifold. The idea about using a bracket to move the the crank sensor a few degrees is intriguing and low cost, but it's a fair amount of work and I don't know how drivability would be affected or if it would work. The FIC is a really cool idea, but it's expensive and the new cat would "probably" work but is also a little costly, and with either of those options it still might not pass after doing a lot of work. Tricking the ECU is almost no work, so I thought I'd explore it and if it didn't work, I wouldn't be out anything. The level of effort and cost for the others, though, tells me I'd be working hard to find a way to avoid working hard, so it looks like I'll be pulling the manifold soon.

As for alcohol in the fuel, if it's 99% (lower % may contain too much water), it burns very cleanly (but also a little hotter), so it can help with HC and CO levels if that's where the problem is. It can help NOx levels in that if you do plan to retard timing to improve NOx levels, it can help offset the rise in HC and CO levels that would come from the timing change. I've known of folks with hollowed out cats using high ratios of alcohol or acetone to pass the test with flying colors, but I don't think it's as helpful for NOx except to offset other engine adjustments.
I believe that the combustion temp is lower, so NOX would be lower.
Whether it is enough or not is unclear.
I would think more than a bottle to 20 gallons would be needed - so that is where the snake oil is at. It is was 20% ... then maybe.
holmz is online now   Reply With Quote
Old 02-18-2013, 01:08 PM   #15 (permalink)
Registered User
 
Vulcan Grey's Avatar
 
Join Date: Jul 2006
Location: Colorado USA
Posts: 3,117
I've remove the exhaust manifold in situ several times on my 89SE. It's really not that difficult.
__________________
Travis
Vulcan Grey Esprit 89 SE
http://picasaweb.google.com/lotusse
Vulcan Grey is online now   Reply With Quote
Old 02-19-2013, 04:40 PM   #16 (permalink)
Registered User
 
Join Date: Feb 2012
Location: Georgia
Posts: 184
Quote:
Originally Posted by eriks4 View Post
Just s quick reply to a part of the first question regarding ECU control over the richness. It is possible to change the 14.6 reference AFR. It's documented in the GM P4 document and is part of the mode 4 input message. You can only do this with the engine running (as far as I know), and with Espritmon you need to go to the menu in which you control the quad drivers. There are two buttons in this menu (inc AFR and dec AFR) with a readout in between. With these buttons you can change the reference AFR and hence make the car run a bit richer or leaner. If your version of Espritmon does not yet (correctly) support this, let me know and I will email the latest version.
Have fun
Erik
Hi Erik,

I think I have your latest, but I'd appreciate the email anyway just to make sure. Also, if you could point me in the right direction for the relevant documentation, I would really appreciate it! This seems like something good to know regardless of the emissions test.

Thanks,

Dale
dalefield is offline   Reply With Quote
Sponsored Links
Advertisement
 
Reply

  LotusTalk - The Lotus Cars Community > Lotus Discussions > Lotus Esprit



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On



All times are GMT -8. The time now is 05:52 PM.


Powered by vBulletin® Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Search Engine Friendly URLs by vBSEO 3.3.2
Garage Plus vBulletin Plugins by Drive Thru Online, Inc.