O2 woes- what the heck?

pres589 wrote: This sort of discussion makes me want to spend time watching eBay for a cheap oscilloscope to better monitor sensor outputs than an average DVOM can offer.

The display was probably the most expensive part of the legacy oscilloscopes.

Nowdays you can get apps or plugin's for your smartphone, tablet, or laptop.

http://www.oscilloscopeapp.com/
http://www.wired.com/gadgetlab/2011/04/iscilloscope-300-kit-turns-ipad-iphone-into-multitouch-oscilloscope/
http://zelscope.com/

Having said that (maybe because of that), you can still find 20 MHz dual trace analog scopes that work perfectly well for cheap.

Carter

Okay, quick update...

I used about 142 wire ties to get the air meter suspended in place in a way I felt okay about. Actually, I probably over did it. Took it for a drive.

At idle, it sits somewhere around 13. But cruising on the highway it's right around 10.5. So, it's rich just at highway speed, which isn't ideal. The idle, however, is solid as a rock now. Ah, but, when cruising, I still get a damn CEL. I really wanted that to go away. Well, rich is better than lean, I guess. I don't have a cat on the car right now, so I don't think that will hurt anything, and I don't really drive it unless I'm trying to fix the berkeleying ECL. I checked the junk yard that lists their inventory on line. No Celicas, but a couple of trucks that might have the right air meter.

I suspect that the old air meter has a dead spot and that's messing up the idle. The miss at idle is telling the computer to send a error code. But I have no idea why the CEL doesn't come on until you take it out on the highway for a few miles.

foxtrapper, those are some good ideas and I'll check them out. Thanks.

Hum. Guess I really need to run out and make sure the CEL is still sending the O2 code. I just assumed it was the same thing.

alfadriver wrote: In reply to erohslc: Age of tech on the car. '84 is very early in using O2 sensors, and they didn't really try to run higher frequencies until they had better controllers later in that decade. The '84 is probaly a very analog style computer- as in not much real programming vs a balance of the curcit to do what they want. My guess is that at 35mph, it would not be switching much faster than 1hz. I still don't switch it much faster than. I know reading the signal on a scope has been possible back, well, probably older than you have been born (back tracking 20 odd years from '62, if that's a decent guess). It's more about how the car runs, and how far the computer changes the fuel switch vs. the O2 signal. Early fuel control means conservative gains, smaller steps, and slower switches, overall.

Limitations of older ECU's manifested as limitations in granularity of control (size of spark and fuel tables), and complexity of calculations that could be used in control algorthims, not in control loop speed or dynamics. And anyway, the limiting factor is probably the response time characteristics of those older O2 sensors, not the ECU processor speed.

Since the regulatory requirements they had were so much less demanding, hardware was capable of meeting them. Assembly language can be a wonderful and powerful tool when used correctly.

I'm guessing that at 35 MPH, the ECU was not even in closed loop mode, and so switching was no longer a factor.

As for my age, you're in the right ballpark, but I started young, not ready (or able) to retire just yet. ;)

Anyway, to get back on-topic, looks as if 'fast_eddie_72' may have found a resolution, glad to see that.

Carter

This all makes me think that I should get to work on a Megasquirt set up for the car. I'm assuming it would be a good bit superior to the OEM ECU? I'm building the car for SCCA "Street Prepared" and aftermarket ECU is legal. I just have to figure out how to get it through Denver emissions testing.

Not to get started on that, but what a drag that going to a more sophisticated and cleaner system makes the car fail emissions testing. That seems kinda...

Why are they testing a 1984 Toyota?

In reply to 92CelicaHalfTrac:

Good question. Same reason they tested my 1972 Capri.

In reply to fast_eddie_72:

Pfft, they tested the Capri? Can't you use one of the emissions drive-by stations on the 470? You've had it through a real inspection station at least once before, right?

Your car is OBD-1 and they can't connect to that with the code reader if they wanted to so at that point the MS should be fair game. I had a discussion about this very thing here about a month ago and it seemed to say I'd be just fine. There's probably some forum local to the area where people have modded cars and have dealt with this before.

In reply to pres589:

Thanks- you may be right. I moved here from CA and you couldn't do anything there. Megasquirt would fail on the visual inspection before they even started the car. Guys on the local autocross forum should know.

Yeah, they tested the Capri. I registered it as an "Historic Vehicle" or whatever, and would never have to have it done again if I kept the registration current. But I didn't, so if I ever get that one running again it will have to be tested again. But the bar is very, very low. It had it's tired old bazillion mile engine in it last time and passed by a huge margin.

In reply to fast_eddie_72:

I don't know how you actually use one of those drive by stations on E/C470 but that's how I'd do it. I had relocated to Denver so had to get the Olds checked, if I had stayed longer I would have done the drive-by.

fast_eddie_72 wrote: This all makes me think that I should get to work on a Megasquirt set up for the car. I'm assuming it would be a good bit superior to the OEM ECU? I'm building the car for SCCA "Street Prepared" and aftermarket ECU is legal. I just have to figure out how to get it through Denver emissions testing. Not to get started on that, but what a drag that going to a more sophisticated and cleaner system makes the car fail emissions testing. That seems kinda...

Far superior.

And easy to put into the original housing. Well, pretty easy....

Plus, since you have to pass a sniffer test first, super easy to be clean.

Between the better driveability and cleaner emissions, you'll like driving it better than you do now.

It's all about the tuning ...
Eric Clapton playing the oldest, most broken down $5 pawnshop 6 string would beat me on a $100K Fender Strat every single time.

erohslc wrote: It's all about the tuning ... Eric Clapton playing the oldest, most broken down $5 pawnshop 6 string would beat me on a $100K Fender Strat every single time.

Getting a '84 Celica to pass a sniffer test is not hard to walk someone though. But it's a whole lot easier to make the car drive better with the hardware he wants on an MS...

I'll be happy to walk you through it. The fact that you have a WB sensor is a good start. A good logger that goes with recording the computer data- and you are good to go. For the MS- two days of auto tuning will dial it in, and then a couple of days tuning the closed loop fuel- done.

F-E-72- do you know the details of the test you need to pass? No hurry- but I bet the rules are not too strict.

BTW, humor me for just one more off-topic post: I use one of these data modules with my laptop:

http://www.dataq.com/products/startkit/di145.html

Does the job of a 4 channel recording scope, plus the software is amazing. Lets you capture and replay as much data as your hard disk can hold.
Easily forms the basis of a DIY Data Acquisition System.

NFI, etc, just thought some might find it useful.

Carter

In reply to erohslc:

Makes my kit I'm putting together a little weak. Although, we are planning GPS, too....

I don't know the details, but it is NOT hard. The car had over 200,000 miles on it when I brought it into the state. Passed with ease. The cat that's (not currently) on it is almost new- shoot, they put it on the car so it would pass in CA. I don't recall the details but I do remember thinking it wasn't a close call by any stretch on any aspect when I did the test here.

Driving better and all is great, but mainly I want to wring all the power I can out of it. There isn't a load to be had, but the car is an absolute pig for F Street Prepared. Only chance of making it work is exploiting the power advantage. The engine itself is out of an '85 Celica. I do have the whole '85 set up- different EFI. But it's running on the '84 electronics. And yes, I had an '85 engine running on '84 EFI before. That's not the problem. (I have a very embarrassing story of why I had to build two engines for the car. Short version, everything worked awesome first time... until it didn't) Anyhow, stock, the 85 engine made like 116 HP and 140 lb ft. of torque.

It's got a .040 overbore, balanced, cranks scraper, under drive pulley set, header and big exhaust, so it should be doing a little better than that now. The big air meter and throttle body are supposed to help, so that's a few more. 9lb. flywheel, which doesn't really make power, but sure makes it feel faster. Not sure how much HP it can make, but the torque should be significantly better than anything else in the class.

A friend of mine did an AE86 for FSP. Stock that thing didn't make 100 lb. ft. and barely made 100 HP. Yeah, it probably responded more to the limited mods we're allowed, but 50% more torque is a good head start. Everyone says torque is king in autocross. Guess we'll see. He did well in his 86, so I'm hoping the Celica can keep up. It weighs about 100 lbs. more than his car right now. I still have some weight to get out of it, but it will be probably end up 50 lbs. more than the 86.

The point being, the whole idea behind the project is to have a big gob of torque to pull off corners faster than the 86. So anything I can do to maximize the power is important.

But right now, I just want the damn CEL to stay off.

pres589 wrote: In reply to fast_eddie_72: I don't know how you actually use one of those drive by stations on E/C470 but that's how I'd do it. I had relocated to Denver so had to get the Olds checked, if I had stayed longer I would have done the drive-by.

I don't know how accurate it is, but there's a weird check thing on the ramp from I25 to 6th East. It just says "Good!" or "Fair" or "Bad". With the cat on it says "Good! Saving you money!" Without the cat it says "Fair". Ah, but if I speed up a bit before, clutch and coast through, it says "Good!" Not sure if that would work for the 470 deal or not. I don't really know anything about that. But might be worth a shot.

With the cat on I'm sure it should pass fine. Now, if someone can show me how to do a MS set up that will give me 300 lb. ft of torque, I'll win Nationals!

Okay.

Fiddled with the air meter a bit and decided I'd done what I could do there and moved on to try to get this annoying CEL to go away.

I figured the TPS was the next logical spot. Tried to adjust it and it wouldn't adjust properly. So I replaced it with another one I have. It did adjust properly and the car drives much nicer. I still get a CEL, but not nearly as often. Went for a long drive with my son and it only came on twice, and only for a couple of seconds on a drive of probably 90 minutes. So it's improved.

Looks like there were/are several issues. Air meter fixed the idle problem but had no effect on the CEL. TPS seems to have made a big difference, so I'm hoping I may just not have it adjusted 100% and a little more fuss there will take care of it.

Anyhow, if anyone has any other ideas, love to hear them. But thanks for explaining that it likely wasn't the 02 sensor itself. I would have chased my tail on that forever.

Ed

it doesn't say on your profile.... but where the heck do you live that requires an air quality on a 28 yo car ? thank goodness here in NC it's only '96 and newer....

'course I didn't even know cars that old had CEL's ... my '86 Civic didn't have one.. first time I ran it them was on one of my Accords .. either the '88/'89/or the '93

I'm in Colorado. I live in Denver. I think if you live outside the city you don't have to mess with it. But in Denver they check it.

(edit) I posted over at CelicaGTS.com and those guys seem to agree that it's the TPS still wanting attention. So I'll have another look.

I fiddled with the TPS some more. Even the second one won't really adjust correctly, so I ordered a new one. Hey, they're 27 years old. Actually, the "good" one came off an '82 Supra, so it's probably 30 years old. So that seems reasonable. If I knew how cheap they were I'd have just ordered one yesterday. Seems a lot of cars used the same sensor. It crosses to some Fords and Suzukis. Several more that I don't recall. Hopefully that will fix the CEL.

In reply to erohslc:

No PC laptop here, no iPad, no interest in those things vs. an actual scope. $200 and Craigslist or eBay seems to be a magical price point, or for about $320 there's this;

http://www.mcmelectronics.com/product/VELLEMAN-SA-HPS50-/72-8770

Dug up this old thread - so looks like waaaay back in early Dec. I bought this new TPS based on what I learned in this thread. Well today, nearly two months later, I finally got it on the car.

I'm juggling a lot with a sick kid and work and yadda yadda, so haven't had a great shake down, but it made a clearly noticeable improvement. Not sure if it's all fixed or not, but I think that may have been the problem down under a few others. I don't think the timing can be set right with the TPS out of whack, so I probably will now need to go over all the basic tuning again. But there is no doubt that it runs better and the idle is much more smooth.

I put the Supra AFM on it when I was messing with it before, but I think I need to put the stock part back on to see what I have. I'm starting to think the Supra AFM just makes the engine run rich and people think it's faster. Not sure. I'll get the battery unhooked for a few minutes after my wife gets home and get the codes cleared. Tomorrow I should be able to set the base timing and get the stock AFM on. Then we'll see if the CEL is cleared. Fingers crossed.