Ah. I suppose I'll have to take some readings on that next time I'm screwing around (as in tomorrow after work).

Well, it's bedtime for tonight. I definitely want to keep looking into this and get some readings, since it's hot outside and I like having power on tap. My fall semester at school has a lot of free time, so I may end up fabricating a water injection system & intake air temperature sensors at goofy places. Putting on a K&N cone filter & eliminating the stock intake system probably wouldn't hurt either. I'd like to do that after the sensors are in place so I can get numbers though.

-=Russ=-

 

Ok, got some numbers on the way to work today.

Here's the conversion table. I don't have my TI-83 with me, and I really don't feel like doing the regression manually, so if some bored math geek wanted to help, I'd love it. Quadratic or exponential, whatever fits best.

(FSM Table)
20C/68F : 41.5 +/- 4.15 kOhm
50C/122F: 11.85 +/- 1.19 kOhm
85C/185F: 3.5 +/- 0.35 kOhm

(My readings)
28 kOhm (cold, 80F or so)
21.5 kOhm (warmed up, 5 minutes of gentle driving)
8.8 kOhm (hot, after 15 mins of highway driving)

So I'm probably sucking 150F air or so (plus heat gained in the intake runners) into the engine. I was able to notice the lack of power, not as dramatic as last night, but I wasn't driving as hard either. If I go for a spin tonight (sustained high RPM running), I'll take readings after that as well.

My intake setup is bone stock, 88 SE NA.

Would these kind of temperatures be enough to cause the power loss I'm noticing?

-=Russ=-

 

Using a quadratic interpolation, your temperatures come out as:

28 kOhm = 31 degC (87.8 degF)
21.5 kOhm = 37.7 degC (99.86 degF)
8.8 kOhm = 55.2 degC (131.36 degF)

 

Wow... that's one of the most useful first posts I've ever seen! Thanks!

Any chance you could post the equation you came up with for that? My calculators are unfortunately in a box for the summer.

Now, the key question, would that sort of intake temperature increase be noticeable? And, more to the point, since I suspect it gets a lot hotter with hard driving, would it be worth playing with a water injection system for race use?

-=Russ=-

 

ok, time for me to rejoin this discussion...

yesterday i went out and put a fuel pressure guage on my car inbetween the high pressure filter and the fuel rail, and rigged up the guage so i could see it while driving.

I didnt seem to be having any fuel issues, the guage would sit at about 31 psi on cruise and light throttle, and then rise steadily to 42 or so psi at half throttle and stay there up till WOT. This was rather independant of engine rpm... And it would stay at 42 for as long as I was at WOT (or basicaly as long as i could stay there without crashing into a car in front of me, i think the longest i could pull off was about a minute)

This continued for the whole ride... 31 at idle, 42 at WOT...

So that seems to rule out any of the filters or lines being clogged...

Also, as far as air temps go.... i remember reading somewhere that a rough estimate for HP loss due to heat was 1 HP per 10 degrees F.

So in your case you would be loosing about 4 HP... and i'd doubt you would be able to feel that

But then again i dont know if that 1 to 10 ratio is right , i just remember reading it

 

I'd agree.

What resistance are you seeing on your dynamic chamber temperature probe after a hot run?

-=Russ=-

 

using a R = a + bT + cT^2 fit using the temps/resistances given to solve for a, b, and c (and completely ignoring the errors given) the equation ends up being:

R = 72.8014652015-1.79576923077*T+0.011534798535*T^2

or

a = 72.8014652015
b = -1.79576923077
c = 0.011534798535

Oh, and the units are the same as given (kOhms and degC). Thank the TI-86 for the ridiculous precision

Will

 

Or you can use this equation:

T = 103.141256539-5.47609361024*R+0.08367927321*R^2

or

a=103.141256539
b=-5.47609361024
c=0.08367927321

for solving for temperature more easily.

 

anyone have any ideas what i should try next?

 

solder a resistor into the maf wires, a couple of guys on the forum sell these, with the resistor in it will tell the car that the air is 50-70 degrees cooler, depending on the resistance used. just search the forum and you should find some. this supposedly fixes the problem of the horrible inaccuracy the stock setup.

 

Right, but wouldn't that lead to rich running? Or is the sensor just horribly inaccurate? It feels like it's running lean to me, from descriptions I've heard of self-tuning NAs (light pedal, lack of bite).

I realized today that part of my problem may be fouled plugs. I haven't checked them lately, and I noticed this morning that I have a fairly pronounced misfire at moderate to high RPMs. I have new plug wires on the way, too. Hopefully that'll take care of the misfires.

-=Russ=-

 

Heh, I just rejoined this, amazing what you can accomplish on this car with a simple meter, huh? Wait until you have some fun with the ECU inputs/outputs while driving, you'll have a much better understanding of everything that's going on after this...

I would think that installing that resistor you guys are talking would make her run richer, which by extension would kill your gas mileage. I guess someone claims she runs better like that, hmm?

Kudos on the math, captmrwill, I get to deal with a lot of A&P mechanics fresh out of high school that can't spell worth a $hit, and the math basics were obviously lost somewhere back in 3rd grade (no wonder this country's going to hell, lol)

Vectorminds- it's time for you to get to that ECU and see what kind of inputs your various sensors are giving you as she warms up, then compare to the FSM "normal" readings. Start with your water thermo sensor, that sucker seems to give a lot of people (including yours truly) a lot of grief...

 

Hm. I haven't checked my water sensor. I should get numbers on that today.

What does the ECU do, specifically, when it thinks the car is getting too hot?

-=Russ=-

 

yeah, the meter was very useful... i have a feeling when i have some money im going to invest in a bunch of gauges...

as far as buzzing out the ecu goes... currently i have a cheap archer kit analog mulitmeter that i built myself, with questionable accuracy.

Should I go and buy a nice True RMS Digital Multimeter for this?

 

Another silly thought I had, as I'm trying to track down my car's running so darn hot...

Is your radiator fan clutch working properly? If it's totally seized up at hot temperatures, then your engine is going to be dragging the fan as well. That could take a significant amount of power.

Also, I cleaned my plugs today with a wire brush, and it seems like it's running a lot better. It surprised me merging onto the highway - I put the pedal down like I normally do, and got a LOT more of a kick than I normally do. Might be worth trying.

Finally, how is the ground on your car? I got around to poking with it today, and realized my main ground on the shock tower is fairly dirty. I'm going to redo it this weekend when I'm doing all the other stuff. Also, coil grounds might be worth looking into cleaning - if they're dirty, things may have problems as they heat up.

-=Russ=-

 

yeah, im going to look into regrounding everything on the car... as far as the clutch fan locking up, how would i test that?

 

Spin it when hot. It should spin a bit. If it spins freely, it's probably not doing a good job cooling. If it doesn't spin at all, it's not going to slip at higher RPMs as it's designed to do.

I don't think the fan should make a huge difference, but it's something to check. I'm going to put my 3rd fan clutch of the week in tomorrow morning (first one failed, second one I put in sucks, 3rd time's a charm, right?).

-=Russ=-

 

hopefully... as far as i remember when the engine is hot the fan spinns a little then stops, but i will recheck the next time i go out driving

 

How about clogged exhaust?

 

Why would a clogged exhaust only cause problems when hot?

-=Russ=-

 

at least in my case i doubt a clogged exaust is too likely... i dont think theres much in the mazdatrix true duals to get clogged

 

Vector- you don't need a fancy digital meter to troubleshoot this car (and you don't need an "RMS" one- that only applies to AC voltage, and you don't want to be chopping your DC square waves down 70%, which is what happens to the AC voltage wave with an RMS meter). The digital Fluke meters we use on the planes at work confuse a lot of "rookies", because it provides them with too much information- it jumps around a lot, and it's accuracy is far better than you'll ever need on a system such as our cars (i.e. instead of displaying something right around "infinity" such as you would see on a standard analog meter, the Flukes display Megaohms, making guys think they have continuity in a circuit when in fact it's open- they're reading through their fingers, back through the circuit, etc...). That, and the significant digits on the Fluke extend to the hundreth- who the hell cares if you have 12.34 volts? It's 12 volts, that's all you need to know

Go buy a cheap $10 unit from radio shack- it's all you'll ever need, I promise

 

More importantly, buy a cheap $10 unit you have no problems with leaving in your car. I'd think twice before leaving a nice digital multimeter in my car (heat, theft, etc). If a $10 multimeter gets cooked? *shrug* Give it to a kid to play with.

-=Russ=-

 

ok, thats good to know... i just didnt want to be spending money on a lesser meter to decide i wanted a better one in the future....

and yeah, ive used flukes before... they can get a bit confusing, i liked my old huge analog meter, but sadly its now broken

maybe ill get one of the ones that connect to a laptop, so i can just go for a drive and then look at a graph later

 

Well, When the car has sat for a while and the exhaust gasses have eventually escaped. Basiclly when there is restricvtion that builds up the exhaust pressure slowly. IE: banana in the tailpipes. Car will run fine, then slowly die out. Possibly a baffle moving around an one of the mufflers. I've had it happen before on my truck and the muffler ended up being the culprit.