There is some interesting information about 96-97 obd and the drive cycle. It even includes a vin number change IIRC. Definitely different from the 99-03. I will try to find the relevant information.

 

Ok I know this picture sucks. It was the only way I could get it, phone pic of phone


At the bottom is specific information for 96-97.
Also this, in case you hadn't seen it before.


Ignore the stuff in red for 96-97

 

It does suggest that the earliest obd2 cars need to run 13 minutes between 13 & 36 mph to set some of the monitors.

 

Thanks @sonofthehill , very useful information, some of which, of course, conflicts with what I've read/seen.

What strikes me the most is that the drive-cycle graph clearly puts the (Mode 4,5) drive cycle component before the (Mode 2) component. The flow diagrams and text explanations I've read all say that the modes must be completed in order: Mode 1, then Mode 2, etc. If I can just get one more readiness monitor to complete, I can take it for a smog.




I drove the car a bunch the other day and managed to get it to set off a CEL. Unfortunately this was for a cylinder 2 misfire, which I hit while stomping on it and bouncing off the rev-limiter out of a combination of joy and frustration at not having set any monitors. At the same time, I did get a pending code for the idle air controller. The code essentially resolves to - the actual idle is too far from the commanded idle. So I went back and reset the idle speed and base timing. The idle was off, not hugely, but it did take a couple of turns on the idle screw to bring it down to the right value. I then set the base timing again (parallax error sucks when reading the crank pulley), and set the idle a second time.

I borrowed a compression test tool from O'Reilly and went ahead and measured compression, even though I don't think there's actually a misfire problem. The numbers look good, if anything cylinder 3 is a little lower than the others, not cylinder 2. I can pull out the pictures of the actual results, but basically the compression test didn't reveal any problems, IMHO.

It looks like this afternoon may be clear here, so I'll try the EGR mode test again. I found the flag in the computer that I believe is what Mazda refers to as the "RFC FLAG" above and I can tell it's been set, even after I cleared the code for the misfire.

Oh, wait, ****.

I just realized that I read that flow diagram wrong! I have to do the Mode 3 test before Mode 2! Holy ****. ****. OK. Now we're onto something...

Thanks!

 

Also also:

My car is a very, very early 96, like within the first 100 units. However it came with a replacement ECU already. And I've swapped to a different ECU that is still part-number compatible with the 96. So that chart is super useful.

 

I also remember 18psi (The Prophet) once said, "Never touch the screw."
On the misfire, I see more miatas with NGK wires, that are not fully seated on the spark plugs. Often they say have been driving fine like that for years. I always take a big *** screwdriver, turn it upside down and use the handle to push down on the center of the plug boot until it clicks, shouldn't be giving away this one, top secret.
Also have seen plugs not fully tightened give a misfire, folks always scared to torque plugs in aluminum heads, just use a ratchet no longer than your palm.
Good luck with the monitors.

 

Or use a torque wrench? Spec is 14-18 ft-lbs, IIRC.

--Ian

 

Mission accomplished!

Somehow, despite not following the drive cycle at all, I got the monitors to set today! It wasn't even a weird or odd or long drive at all.

I'm telling myself that the act of re-setting the base idle and timing caused things to go better. So sure, I'll agree with 18psi - don't mess with the screw! Unless you're un-messing it. What I think was important about this change was that instead of hitting some value from the factory service manual, I used my OBD scanner to read the "commanded" or "target" RPM from the ECU. I did not know that the ECU transmitted this value! With the TEN terminal grounded, this becomes almost, but not quite fixed. I put the "target" and "actual" values on the screen of my scanner and got the actual value to match as closely as possible to the commanded value. Ten to twenty RPM or so, IIRC. Then I went and reset the timing to as close to 10 as I could get it. Then back to the idle speed.

My results are above-average in a couple places. I'm not thrilled with that, but I can live with it.



Now I can park this Miata on the street and (once I reorganize the garage) put the 928 in the garage for some much-needed fixing (and eventual registration).

I also went ahead and bought the plane tickets, hotel stay, and rental car for my DirtFish school. So with the nearly $1k in back-fees, penalties, and smog check, today was not a good day to be my bank account.

 

Congratulations This is what he meant for sure.

 

I'm installing some prototype engine mounts this weekend and happened to glance at my brake pads.



I should really be keeping better notes, but I believe I used these R12 pads for a total of two days at Sonoma and two days at Thunderhill. I really need to get those ducts hooked up. I had the hood open and thought to myself "oh ****, that brake fluid looks low, I hope I don't have a leak". I know where it went now! Rear pads look perfectly fine.

I also found a new bolt that wiggled loose and disappeared. I have a catch can't.

 

Ducts did better than tripling my pad life.

 

I'm still unemployed, but at least I'm enjoying it a bit. I don't have quite as much runway as I did the last time I lost my job, but I feel like I've got a much stronger sales pitch than I did before. I've been enjoying some time in the garage working on a prototype that I can't show you yet - it's been awesome working with people who actually know their **** and give a damn about supporting the community,

While awaiting some prototype revisions, I sat down in my front wheel well with my Aircraft Spruce ducting and tried to figure out how to route it and account for the compression/elongation necessary to allow the wheel to turn. After some head scratching, I decided that I needed to come up with a place to anchor the intake end of the duct. I remembered Emilio's post here where he mentions the brake ducting being in the radiator ducting. So I decided I needed to build radiator ducting, finally. I've been running without it for at least as long as I've had the K-swap, maybe longer. Note that this is how simple projects go sideways. I eventually realized that I was over-complicating **** and went back to just designing a radiator duct. I can fit the brake inlets just outside the duct, but still within the "mouth" of the stock bumper.

I used the tried and true poster-board mockup technique, though it left something to be desired. Just a little too bendy. So I ended up gluing in some reinforcements. And as usual, because I was using superglue, I ended up bonding my fingers together a couple times.



I even stopped to make sure that there was enough room to get the riveting tool into the duct to set the rivets!


The idea is that I will weld a couple of tabs to the cross piece that sits in front of the radiator. I made mine removable a while back, so this is convenient. The duct will hang from these, and I'll weld some tabs onto the area I had to trim just inboard of the "baby teeth".

The duct opening will be 5" high by 20" wide. It starts just about flush with the flat face of the frame rail ends. It'll fit nicely inside the stock bumper, for now, and when I eventually go to an air dam, I can craft an extension that sits up against the dam.

I designed the radiator end to sit nicely on the end tanks of my cross-flow, as well as hitting the structural lip at the bottom of the finned area. There's an inch or so of finned area up at the top of the radiator that won't get any airflow, but it's hidden behind a lip on that cross-piece. So the exit of the duct is 13" high by 22.5" wide.

I've got solid templates in poster board now, now I all have to do is add tabs for rivets and get to cutting!

 

Noice. Always love seeing some well-thought out ducting plans. I was also thinking of integrating brake duct inlets into my radiator ducting, but figured I was better off locating them outside of the bumper inlet rather than taking a potential hit in cooling performance. You weren't running into any cooling issues on track previously though, were you?

Semi-finalizing my track schedule for the year now. Might see you at NASA Sonoma in May if you're gonna be there

 

One of the joys of the K-swap is that it runs nice and cool. I've had no cooling issues without ducting, so it's only going to get better.

Unless something serious comes up, I'll be there working Tech at the very least. If I have a job, I'll be driving too. Swing by Tech and ask for Peter.

 

I have a bunch of updates to share - pics and video from Dirtfish, pics of the new AWR K24Z3 engine mounts I've been prototyping, thoughts on those mounts, but what you get first is the radiator ducting that I don't need, which will allow me to then install the brake ducts I actually need.

You don't want to know how long it took me to get to this point. The shapes are simple, the bends are fairly simple, but somehow my penchant for exactitude made it difficult to get here. The duct sits flush against the SuperMiata Crossflow, pressed up against the end-tanks and the U-channel at the bottom of the radiator.



Here it is, stuffed into the stock opening. I had to trim the little "horns" off the body - those little curves of sheet metal directly inboard of the "baby teeth". Straight shot into the radiator, flanges pressed against the end-tanks and bottom cross-bar. Flanges will get edge trim applied to them at the last moment to prevent metal-on-metal contact. As you can see from the bottom sheet, I clearly need a set of stretch/shrink dies. The inlet size is 5" x 20".

The upper edge stops just shy of the lip on the back of the radiator cross-piece. I could have trimmed this part to allow flow to the top inch or so of the radiator, but since my hood is attached here, I really didn't want to weaken the cross-piece by removing the stiffening effect of this flange.



This duct was designed to fit the body, not the bumper, since I will eventually be going to an air dam. The stock nose is good enough at directing air into the duct for now. When I go to a dam, I'll extend the duct. As you can see, there's plenty of room for 3" brake inlets on each side.

 

ducting looks super clean, nice work!

You'll ideally seal around the brake ducts too so there's no open air - you want ALL the air coming in the bumper opening to be forced into some sort of cooling duties, no room for it to bypass into open air.

 

I've heard that a "mid plane" in ducts like this can help a lot, as the air essentially passes through the lower half the rad at higher speed (least resistance), instead of spreading out and passing through the entire rad at lower speed. Maybe a thought

 

My conundrum is this:

if I want to attach my tow strap to the fasteners for the "baby teeth", I have to move the brake inlet down a bunch. This is OK as I think this is more at the same level as the duct will run, but then the inlet is shadowed by the curve of the stock mouth.



Where else would be a good place to attach the front tow hook? I have a soft strap on hand, takes a M10 bolt.

 

I decided last night that as the car needs to be in the trailer on Friday morning and something is better than nothing, I'm going to go with progress over perfection.

This was pretty easy to throw together, blocks off most of the air flow with a simple part, and feeds air to the brakes. The other side is bent and ready to mount, it just needs the hole cut and the flange inserted.



Later today I'll find a place to tie the tow strap up so that it's out of the way of the brake duct.

 

super clean solution!