Heat shield manufacture

I was planning to work today, honest. But when I came to work, the computer system was down, with no time of availability. So I went home to do some long-postponed work on Bozo.

First was to put in an elbow pipe to put my air filter in a better position. But the elbow the local metal shop had made me was way different from what I had sketched for them and hard to cut with a hacksaw, so I ended up improvising something with rubber plumbing sleeves. I'll try to replace one part by plastic pipe later.

Next was to manufacture a heat shield around my header to try to cut down on convection of hot air from the header through the engine bay and into my air filter. I first made a prototype from cardboard and masking tape, and then I copied that using the aluminum plate I had bought earlier.

Unfortunately, I found the aluminum the metal shop had given my to be very tough to machine. The non ferrous blade on my radial arm saw would cut it OK, but drilling holes in it was an agony. I found that I could not go up more that two drill bit sizes each stage, a major pain since the holes I needed were the maximum drill bit size. Even then it took about half a minute to get through the plate. *Each* drill bit! Also, I broke my three smallest drill bits drilling the initial holes.

When drilling the last hole, to add insult to injury, the on button on my drill also was stuck. I recognized the reason must be I had touched the select button. This made me wonder, however, whether the drill was rotating the right way. It was not! While giving no clear warning sign at all, this drill had been stupidly rotating the wrong way all the time. You would think that Sears would be able to figure out that if you do not provide a clear warning, people will end up with their drill rotating the wrong way and they not even knowing it! Very stupid of Sears not to show a red light or so to indicate the drill rotating the wrong way.

In any case, I hope that the heat shield will cut down on the hot air in the intake. I have my doubts though; if the air filter cannot draw the air from the header, it will probably draw it from the radiator. The engine bay is pretty much sealed elsewhere, with no obvious route for cold air to come in.

Leon

Reply to
Leon van Dommelen
Loading thread data ...

I dont think Sears is at fault here.

Reply to
BJsMX5

On my '02, there's a bulkhead with a couple of relays mounted to it that's left over after removing the airbox. This bulkhead looks like it was intended to shield the stock airbox intake horn from hot air coming from the radiator, and there's a fresh air inlet behind the left headlight. So, now I suspect the bulkhead is not helping.

Do you really not have some kind of fresh air inlet? Mine is a hole in the sheet metal right behind the headlight.

So what I'm thinking of is fabbing a bulkhead, basically a vertical shroud to shield the air filter from radiator air and channel more of the 'fresh' air to the filter. I need *something* more stable than the current plastic shield since it's flimsy without the airbox to support it.

I'm actually not that concerned about header-heated air rising to the air filter as long as the vehicle is in motion, but that could be naive of me.

Dana

Reply to
Dana Myers

A fresh air inlet? I will look once more. Thanks.

The nastiest pre-ignition occurs when accelerating from rest.

Even if driving, the airflow through the engine bay is not clear to me. The header seems to be very well shielded by the supercharger. Also, since there does not seem to be a place for the air to leave except at the very bottom, I would think that the air in the engine bay may be pretty much stagnant most places.

Leon

Reply to
Leon van Dommelen

I've been thinking of the same thing. Saw this:

formatting link
Sort of a pain to install, what with removing the supercharger et al. I'm also wondering how much heat a blanket will divert from the engine bay. Seems like the inevitable slit between the blanket and the engine block header flange will allow all the heat the blanket has trapped to simply flow up and through the slit (there's not a seal between the blanket and the block) and right back into the engine bay. Not a thermodynamacist, but simple reason tells me the blanket-trapped heat will move UP; not DOWN.

Still, I'd like to divert header heat away from my new, kustom-maid, PAPER air filter assembly.

Oh, on drills and stuff, Clockwise means you're going IN, Counterclock means you're going OUT.

Steve McMahon Green JRSC '00LS

Reply to
McMahon

Don't bother, Leon. That's an NB-only feature.

You might direct your creative curiosity toward designing a cowl induction setup, enclosing the filter in an airbox and ducting outside air through a hole in the firewall. I don't have a supercharged Miata to study, so I can't tell you where you'd find the room. But instead of trying to shield the intake from the header, it would let you ignore the header entirely.

Reply to
Lanny Chambers

Ah, I suspected this might be the case.

Dana

Reply to
Dana Myers

My worst pre-ignition takes place near-WOT from 3000 to 3300 RPM, at least this is where I *hear* it the most. Accelerating from rest doesn't seem to present a problem for me since I usually get above 3300RPM quickly.

I installed the Aquamist/JR/ERL water injection kit and, now that I seem to have cured an issue with the injection nozzle leaking and very carefully tweaked the pressure switch to come on at near-WOT just below 3000 RPM, my pinging appears to be totally cured, even when I don't add Outlaw MMT octane booster to name-brand 91-octane fuel.

Of course, as Lanny pointed out, you're in an NA and I have an NB. Nonetheless, have you considered water-injection for you pinging control? It won't do anything for intake air density but it probably would cure pinging in a heartbeat.

Dana

Reply to
Dana Myers

I very much hope you mean knock, not preignition. Otherwise, park the car immediately and fix it.

Water injection is another thing to worry about. I would not want a setup where I had to depend on having water to avoid knock.

I used the Link ECU to kill off all knock. (Well, accelerating from 1500 at full throttle in 4th, there is still some at 2500 or so.) Otherwise, zero.

I could probably get some more power with a more aggressive setup, but it is no big deal to me.

I need to figure out why the thing is suddenly running rich. It used to show about 14.7 on the wideband cruising, but now it is more like 13.0 on cruise. And 12 on idle.

Leon

Reply to
Leon van Dommelen

Right. It's pinging, presumably the same as knock.

I can appreciate this. In my situation, I really want to modify the stock ECU arrangement as little possible given the stringent emissions testing program we have in CA, and water injection seemed to be the hot ticket. I mean...

... replacing the ECU would count as another thing to worry about

*for me*. You're eliminating knock through the use of a modified timing curve, correct?

I really just wanted to enable the stock ECU's knock sensor and timing program to be effective.

I'm most pleased by the seat-of-the-pants sensation of increased mid-range torque with water.

Ugh. 13 on cruise! Have you verified that the wideband sensor is indicating correctly? My first hunch would be to replace the stock O2 sensor. When you're using the Link ECU, do you get engine check codes?

Dana

Reply to
Dana Myers

Removing the SC is not such a big deal.

I do not think a blanket would do much good either. My heat shield is just a contraption of aluminum plate. Lots of holes, and it will not stop buoyancy-induced flow, but I hope it will cut it down by a lot.

You think that is going to help? It takes me quite some amount of serious thinking to figure out which direction is clockwise, and then try to remember whether clockwise is closed or open.

I have had trouble with a Sears timing light before. Sears simply does not recognize that the normal state should be locked in, and you should have to set an abnormal state explicitly.

Leon

Reply to
Leon van Dommelen

So, the distinction drawn here is between pre-ignition, where the air-fuel mixture is ignited (typically by a very hot deposit) before the spark plug is fired, and knocking/pinging where the air-fuel is ignited by the spark but, before the air-fuel is completely burned, it spontaneously burns all at once but otherwise would not do so ?

Dana

Reply to
Dana Myers

Correct. I also added a bit of fuel here and there.

Knock sensors are not very effective; they retard timing too much.

The stock O2 sensor (just replaced) is not the problem. It properly indicates a rich mixture too. No check codes.

Leon

Reply to
Leon van Dommelen

In knock, it does not burn at all once. Nor in pre-ignition, for that matter. In knock, some locations start burning before the flame front hits them.

Knock is a corrupted normal burning process. In pre-ignition both time and location of the ignition is completely wrong, producing major out-of-spec forces.

Leon

Reply to
Leon van Dommelen

My experience with my '02 SE is that the knock sensor seems to try to advance timing as much possible, it doesn't seem to "remember" very long after a knock event and it doesn't seem to retard a great deal.

Ah! My next question. So both O2 sensors agree and one of them is new. Does the Link ECU run closed-loop under cruise?

Dana

Reply to
Dana Myers

The problem is that knock sensors wait for knock and then react. At that time more retard is necessary (Flying Miata give a number of 30%) than if the timing was set to prevent knock in the first place.

I guess the problem is that the glowing deposits caused by the initial knock require more timing retard until they cool down.

All three agree.

It is supposed to do lambda l3 tuning. I think the idea is to rapidly vary the mixture about 14.7 to allow the cat to do its work.

I think the problem started when I replaced the link chip with the latest, but I am not sure. I did so much work on the car recently. Maybe I should not just have dumped the settings of the old chip onto the new.

Leon

Reply to
Leon van Dommelen

Obviously, I should not have. Two of the memory locations were given a different function, making their contents nonsense.

Currently the Link ECU does not do L3 tuning below 128 kPa MAP pressure. However, it does do some serious air temperature correction unless it reads 128 in two's complement. :)

I guess I will be retuning the car from scratch tomorrow.

Leon

Reply to
Leon van Dommelen

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.