NA6 Miata troubleshooting(yet again)

TL;DR: Is that a throttle position sensor attached to my throttle body, and is there any special procedure for replacing it?

On the way to the NOLA autox this morning the car was running fine, but seemed a bit down on power - yes, even for a 1.6 - when climbing handful of tall bridges over the waterways between here & there. By my 3rd run I noticed the car was running out of breath before 4krpm, though it still seemed to be running fine. By my 6th/final run I had the car floored the whole way around course except for the two tightest corners, so I knew there was a problem. 

On the home on the twin-span over lake Pontchartrain I was in the right lane struggling to keep 50mph. I stopped at a parts store in Slidell & bought tools to pull the plugs. They all looked fine & it seemed to run fine on the surface streets in town, so I decided to take the slow/scenic route the rest of the way home. 

This gave me plenty of stop & go traffic, and I discovered as long as I didn’t push the accelerator over about 1/4 of the way, it ran fine. As soon as I went past that it would completely fall on its face - regardless of gear, speed, load or even with the clutch in. So I think the TPS sensor(or whatever it is) may be the culprit?

The stock tps on an na6 is more of a switch than a sensor, doesn't send any more information than wot or not.  Maybe your afm?

In reply to cmcgregor :

I considered that, but have no idea what it’s typical failure mode is?

Also check the obvious: fuel filter and air filter.

Do you have any codes? I've had similar symptoms when I have a big leak in the intake tract, so that's where I'd start. Also sounds like it could be not getting enough fuel to keep up. 

Is the catalytic converter still on it? You might check and see if it's clogged.

In reply to cmcgregor :

No codes. Fuel filter is new, but I don’t think it’s a fuel problem - I ran it up to 5k driving, I just can’t tip into the throttle more that 1/4-way. 

I’ll check for a cracked/torn intake boot, as that seems plausible.

In reply to NermalSnert:

It has a new high-flow Magnaflow cat. 

I had a few minutes this morning, so I pulled intake apart from the filter to the TB inlet. 

• The air filter, while dirty, is definitely not plugged. 
• The AFM door moves freely, though I need to see how to test/measure it’s readings. 
• The rubber intake boot is intact & I flexed it all directions to confirm there were no cracks between the bellows. 
• The plastic intake tube & resonator are intact with no cracks, as are the 3 hoses that connect to it. 
• The TB moves smoothly, and the device attached to it does function as a switch, with the center & top pins reading open until WOT, then going to almost zero ohms. The bottom & center pins read in reverse, which makes sense.

Sadly I apparently don’t have a spare AFM, so hopefully that isn’t the problem. I’ll research test procedures for the AFM and TB switch today, then see what those results provide. 

I just tested the AFM, and I found a bit of a dead-spot about 1/3 way through the flapper movement. The resistance starts around 300-ohms and slowly climbs up to around 600, then drops back to ~340 before climbing up to around 800-max. 

Would that cause my symptoms though? It seems like if I opened the throttle quickly it should climb past the dead spot quick enough to not be an issue?

If it's like a potentiometer, it could be a dirty spot in the 'run' or whatever you call it.

In reply to RossD :

I pulled the cover off, and while it’s spotless inside, there’s visible wear on the wiper trace. This also allowed me to actuate it more smoothly, and I was able to see the resistance is jumping all over the place throughout its actuation. So I’ll be ordering an AFM shortly...

I've always had trouble measuring the AFM output myself.

Did you check for codes using the paperclip in the diagnostic connector, or just rely on there not being a light? Does the car run smoothly at idle and just struggles when it's asked for higher power? Have you put a fuel pressure gauge on the car to see what it looks like both at idle and under load?

Pete, I have a spare AFM that came with my car. I can't call it a known good one - more of a "known unknown" one - but I'd be happy to send it over to you if you suspect the AFM and want to test with another one.

In reply to Keith Tanner :

I’ve not done the paperclip yet.

It’s really odd that it runs perfectly smooth all through the rpm range - as long as I don’t open the throttle more than about 1/4-way. Load, gear, speed, or clutch in/out have no effect. 

I’ve not checked fuel pressure, but I’m not sure that could be an issue since it’ll pull fine up to(at least) 5k in gear, and hit redline in neutral?

In reply to BoxheadTim :

Thanks for the offer Tim, but a reman was under $70 after core, so it’ll be here in a couple days. 

The paperclip test is basically free and quick, so I'd do it just in case there's something interesting in there.

Fuel pressure check is also low effort, high return. It gives you the ability to sign off a fundamental system without having to change parts. Having a car that can't handle load (no more than 25% throttle means no real load) and the way it failed makes me want to ensure that you don't have a fuel delivery problem.

In reply to Keith Tanner :

Awesome, thanks Keith!

Well apparently reman AFMs are out of stock from the vendor, so now I’m searching for a used one. I may take up Tim on his offer, but he won’t be able to ship it for a while, so I’m searching locally too. 

Is your car an early short nose crank car? I had similar symptoms (down on power) and it turned out to be the short nose crank issue.  This was accompanied by an increase in fuel consumption. If so, a quick way to check is to pull the spark plug from cylinder 1 and put the dipstick down the spark plug hole, resting on top of the piston. Turn the crank pulley while watching the dipstick rise/fall until the dipstick is at it's highest height, then look at the timing marks on the crank pulley. The timing mark on the pulley should be right at the TDC mark. If not, the crank keyway and/or key may have worn and allowed the timing sprocket to rotate out of alignment.

A failed damper will also show a misaligned timing mark on that test, but won't affect the way the car runs. So it's possible you could end up with a false positive on that test. I'd follow up by pulling the valve cover and checking the cam position at that measured TDC if you see a problem with the marks.

In reply to mainlandboy :

It was originally a short nose car, but the current engine is a later long nose one. I replaced the timing belt, WP, gaskets, etc. before I dropped it in 1000-miles ago, and the snout was undamaged. 

It’s not the AFM. It still seems very throttle-position dependent  

Hopefully I’ll have time tomorrow or Sunday to try the paperclip test & check the fuel pressure. 

I don't know if you've done this or not, but you can pop the cover off that throttle position "sensor" and verify behavior by eye quite easily. If it's showing a closed throttle when it's not, you could probably get some weird behaviour. IIRC there's something about 2200 rpm in that case.

In reply to Keith Tanner :

I’ve not, but I’ll ad it to the list - thanks!

I just tested fuel pressure, it holds a steady 32psi at idle. Under acceleration is weird though. If I slowly open the throttle, pressure drops to around 22-24, but if I crack the throttle open quickly - to/beyond the point where the engine bogs - it climbs up just over 40 initially, then slowly drops when the engine bogs.

What does that indicate?

Edit: Also, after not having driven/started the car for 10-days, there was zero fuel pressure when I opened the line to install the gauge. Is that normal?