29 June

The MAP mystery continues. Did some more troubleshooting and while it’s not complete, it helps clear up a few things, albeit also posing more questions.

Checked out the blowoff valve, which dumps air out of the intake tract when there’s vacuum downstream of the throttle plate. It works fine, at least when tested statically with a handheld vacuum gauge, opening at about -20 kpa*. I checked every fitting and hose for looseness and holes – nothing. What keeps bugging me is how MAP will stay at 200 kpa for a few seconds, then leaks down to 150 kpa. Why it’s not a steady leak at 150 kpa I don’t know.

I then tested the wastegates, feeding 20 psi into the control lines to check their operation and to make sure they aren’t leaking. Pressurizing the backside of the diaphragms (which pushes the valve open, helping the spring) showed there were no leaks in that circuit. Pressurizing the front side of the diaphragms and… hmm, they leak, though the valves still opened as they should. The leaking isn’t good, but I’m unsure if that’s the problem. That is, if some of the MAP pressure was leaking out, it would make the wastegate valve LESS likely to open, not more likely. That’s what’s so strange about the problem, that running only on 200 kpa springs works for several seconds, then MAP magically drops to 150 kpa. Where is that pressure going? Blowoff valve? Intake track hose? A wastegate that suddenly gets weak? Something internal to the engine? A cylinder that stops firing? (I don’t detect any roughness in the engine when it happens)

Though it’s a bit of effort, I’d like to pressurize the entire engine. That is, feed about 30 psi into the turbo inlet and plug up the exhaust pipe. That would certainly locate any obvious leaks. However, it won’t find internal leaks, such as a wastegate opening early and bleeding air around the turbo. Still, it would help compartmentalize where the problem is.

* For the units-challenged, 101 kpa is 14.7 psi. I use kpa because that’s what the ECU uses.

21 June

Longest day of the year – it’s all downhill from here!

Went to a local Saturday morning car get together, sort of a mini Cars and Coffee. Same thing though, anything and everything shows up, from new Ferraris to an authentic Ford Model T. I talked to the owner and learned that not only are the brakes cable-actuated (stretching as they’re applied), they’re also only on the rear axle. It reminded me of a story my mom told us how when she was a kid, her family moved from the midwest to the west coast, passing through Colorado. She said that to this day she still remembers them nearly going off a steep cliff and her seeing a river down in a canyon way too close. I think they were in a Model T and it wouldn’t take much to overheat the brakes – the drums looked like they were about 6″ diameter.

Also at the car meet was a BMW M6 with the total “hella flush stance” (Google that to see what it takes to be one of the cool kids). He had about 1.2 millimeters of clearance between the tires and fenders, never mind 8 degrees of camber – no, that’s not a typo or exaggeration. He got a lot of admiring glances and comments from those his age, one guy saying, “That’s the most awesome car here.” I suppose.

Took the car out 5:30 am Sunday morning, expecting to find a deserted freeway for doing an engine pull to set the knock sensor noise floor. There were more cars than expected, so I had to slow down until those behind almost caught up, giving enough room ahead so that I could complete “the mission.” For this sort of thing, the rules are: no passenger, no passing, and no swerving across lanes. The idea is that from way behind it makes it tough to say how fast a car up ahead is traveling, or that’s the theory. Anyhow, I got a surprise, just not what you’re thinking. I got up to 7000 rpm and felt the engine go a bit weak. That’s always a bad sensation, so I called it a day and headed home. The logs were downloaded and the answer was – unclear. The problem is in two parts:

First, not expecting to have this problem, only engine speed, MAP, lambda, and the four knock sensor variables were logged, and sure enough, at around 7000 rpm, MAP dropped from ~200 kpa to ~150 kpa. That’s odd because I had removed all boost control, so the turbo was running solely off its ~200 kpa wastegate springs.

Second, for some reason, the ECU datalogger app didn’t display lambda, even though it was on the list. So not only don’t I know what the fuel mixture was at the time of the problem (identifying whether it was cutting fuel), but I also didn’t log spark timing. It has to be one or both of these, since engine rpm didn’t drop (as would be expected if an rpm limit was hit). Another variable that should be in the logs is air temperature coming out of the turbo. If that got too high, the engine protects itself. So first I have to figure out why lambda didn’t get logged, then log the other variables to figure this out.

Now that I think about it, it’s possible this rpm problem has been around for a while. I remember at the last track day event feeling something similar, but a lot has changed since then. Different ECU, no water/methanol, possible knocking, it’s too many variables. It’s mostly my own fault, I keep changing things and introducing new variables. That said, things will settle down now that the engine’s is fully “done”, it’s now just a matter of squeezing out the last few bugs in the system.

7 June

Joined up with the local Mini Club for a drive out to the back country for pizza. It was a really nice day but is getting downright warm out there; I was surprised that the grass was still green. Midlana did fine, no nonsense or fussiness, but there wasn’t any horsing around either, what with the older Mini’s having around 50 hp or so :). I need to start thinking about a trackday now that the car’s running well on E85. Unfortunately, since the tracks are in the desert, it’s already getting unpleasant out there.

Oh yeah – the trend continues. Got 6 thumbs-ups from sport bike riders, the guys bombing up and down Palomar Mountain dragging their knees and all. Got one thumbs-up from a touring bike, and exactly none from any Harley riders. Hard not to draw conclusions about why…