240SX exhaust temperature

[mgp240]

Hi guys,
I was wondering if anyone out there can tell me what exhaust temp I should be seeing??
I have melted the #3 exhaust valve on 2 engines... the first was a stock motor with over 190k miles on it so I wasn't real concerned with it. But then in June I blew the same valve in a stock rebuild with only 5 race weekends and about 3-4 test days on it. I finally got another motor together and went to a test day 2 weeks ago and got some numbers...
I drilled a small hole in the #3 primary tube about 10 inches away from the head and hose-clamped a thermocouple into it. Recorded things with a Mychron 3.
I decided to see if I could get any change in temp by trying to trick the ecu (using a stock ecu from an automatic car) into putting more fuel into the cylinders by lowering the voltage reading it sees from the O2 sensor.
with the stock O2 hookup I got up to about 2000F in full throttle on the straights, with my modification installed it only dropped to 1950F and I cut the voltage going to the ecu down to 75% of normal, so it pretty much seems like my modification did nothing.
Any ideas? Are these temps normal? If so, what brand of valves are you guys using to withstand this?
Thanks,
Mike

[Eagle7]

I don't know much about boinger temps (rotary guy here), but I know they're lower than a rotary, and your 2000 degrees is way too high for a rotory.

Most (all?) ECUs ignore the O2 sensor at WOT, so messing with that signal isn't likely to gain you much. Sounds like you're running way too lean or your timing is way off. One of the best investments I ever made was a wideband O2 sensor so I can really see what AFR I'm getting.

Marty Doane

[lateapex911]

Right. Gotta get some real data there. I think 10" might be too close, and is elevating your numbers a bit, but even so, 2000 is too freakin high. Too bad you didn't get two sensors, so you could compare and contrast.

I'm a rotary /carb guy so I am NO expert here, but I would suspect a bad injector as well. Might want to either replace, or have them cleaned /tested. It's not something you want to screw up, as you've found out. Also, as mentioned above, timing will play a role as well.

[kthomas]

EGT should be measured about 6 inches from the port and be in the 1250 F to 1450 F range.

[mgp240]

EGT should be measured about 6 inches from the port and be in the 1250 F to 1450 F range.
[/b]

Only 6 inches into the exhaust piping???? Or are you saying 6 inches past the merge

I put it 10 inches down the pipe because I was interested in #3 so I installed it before the merge.
As for timing, I run about 12-15 BTDC(15 is factory)

[JamesB]

6" off the head is usually the recomended point for the thermocouple.

[Greg Amy]

WAY WICKED lean. No exhaust valve can stand up to those temps; hell, nickel alloys in a turbine would have a fun time with 2000F...

Sumpins' big-time wrong.

[MMiskoe]

As Marty said, the O2 sensor will have no effect on things at WOT. You might as well just unplug it.

If you can try cranking up on the fuel pressure instead. Sometimes fooling the ECU into thinking the engine is cold will help richen it up.

I'd also be looking for a manifold leak and trying the the thermocouple on the different cylinders to compare/contrast. At least a pyrometer on them to see if there are any wild swings.

6" from the head is similar to what I have seen on other motors.

Matt

[Ron Earp]

I'm in agreement - you've got a big issue. Therocouples are always installed around 6" out, planes, cars, etc. I've seen the same all over. 1400 would be nice to see, 2000F means you need to stop straight away - it won't last long at all.

The O2 sensors on cars are the transistion type crossing 1V output going rich to lean and visa versa. The ECU ignores the input anyhow at WOT as a couple have pointed out.

Get one of these:

http://www.aempower.com/ViewCategory.aspx?CategoryID=67

$250 on Ebay all day long. Have one for my Lightning, works well and just as good as higher priced ones.

You don't really need it though, you've got EGTs and that will tell you what you need to know - and it's telling you that you are extemely and dangerously lean.

Fuel pressure might help, but you might have some other issues at work with the injector being bad or something similar - no pulse signal to it, etc.

How are EGTs of other cylinders? If they are fine then you know you have a injector problem with this cylinder. If all are 2000F then you've got pressure and other issues.

R

[mgp240]

Well, the fuel pressure I know is good. I checked it when I installed the new motor. Will check it again of course and look into the injector and wiring first, as this seems like it would be easiest.
One thing I'm curious about... the 1400F ballpark figure, what type of driving conditions is that under? On track, dyno pull, tame street driving? I don't have the info in front of me now, but when I was off the throttle setting up for corners the temp dropped significantly and quickly (I want to say to somewhere around 1300, but I'll double check)
Thanks everybody for the help so far

[JamesB]

~1400 at WOT is ideal at any time, street and dyno are about the same quick bursts, and you know racing your WOT all the time. Without another cyl probed for comparison there is no way to tell if the FP is just too low overall or if its just a clogged/bad injector or a wiring issue.

[DavidM]

Just curious. How do you go about isolating which cylinder has an issue if you're using something like an o2 sensor? The sensor is usually plumbed after the exhaust pipes are combined so you're getting a combined reading. How do you tell which cylinder is lean if you get a lean reading? Based on this thread it looks like you could use an infrared pyrometer to check each exhaust pipe next to the port on the head and the one running hot is the culprit. Are there other methods? Just looking to gain some knowledge for future trouble shooting.

Thanks.

David

[stevel]

Just curious. How do you go about isolating which cylinder has an issue if you're using something like an o2 sensor? The sensor is usually plumbed after the exhaust pipes are combined so you're getting a combined reading. How do you tell which cylinder is lean if you get a lean reading? Based on this thread it looks like you could use an infrared pyrometer to check each exhaust pipe next to the port on the head and the one running hot is the culprit. Are there other methods? Just looking to gain some knowledge for future trouble shooting.

Thanks.

David
[/b]

Spark plugs are good thing to check for rich/lean conditions.

s

[mgp240]

Well, I checked the wire harness and injectors last night... Harness is good, all connectors fired the noid-light, after that I removed the fuel rail and zip-tied all the injectors into it and cranked the car over a few times with the injectors spraying into peanut-butter jars (real scientific, I know). All ammounts seemed close, but I have no real way of measuring. Then energized each one individually, and they all appeared to have a similar cone shaped flow, but again, no way of measuring the pattern or the small ammount of fuel that they shot. My uncle found a company that flow tests and cleans injectors, and I already have an email in to them to see what their turn-around time is.
Looking to check fuel pressure and run the ecu self-diagnostic on the injection system either tonight or tomorrow (I've run the diagnostics before for different problems and it has never helped me yet)
Again, thanks for all your input
Mike

[JeffYoung]

I know this sounds basic, but do you have a leaky head gasket maybe? Causing the lean condition?

[lateapex911]

Or a vacuum leak in that intake runner. Could be some weird valve that opens above idle, that sort of thing.

And the tesing of the injectors is a good start, but does it tell you what happens at higher and faster duty cycles?

One test would be to get a 2nd egt probe, and make a few runs, watching the pattern. No need to get close to 2000...then swap injectors...if the problem remains on that cyl, you know it's not the injector. You could probably do this on the street, no need to hit the track.

[Scott Nutter]

If it helps here is a graph of Mike's temperatures. Temperature is of course Channel_1. And I did the math so temp is in degF not degC.
[attachmentid=623]

[Eagle7]

...My uncle found a company that flow tests and cleans injectors, and I already have an email in to them to see what their turn-around time is...[/b]

I had very good luck with http://www.cruzinperformance.com

[mgp240]

OK, I've checked fuel pressure. Rock steady at idle at 44psi, when I rev and hold at about 2500rpm it dips very slightly (about 43psi) and comes back up to 44. WOT dips to about 42.5 and starts coming back up before I let off the throttle, and throttle response is very quick(it was sluggish and hard to start when I had a tired fuel pump that I changed out last spring)

http://www.cruzinperformance.com is who I have the email in to, so Eagle7's vote of confidence in them is reassuring, and the price looks good too... it's just a matter of how quick I'd be able to get them back.

I had compression numbers within 10psi of each other right around 180 when I installed the engine, so I don't think it is a head gasket thing. The valve thing is interesting... But I checked external temps on the header tubes with a contact pyrometer and got pretty close numbers for all 4 cylinders at idle. About 360F with the engine warmed up, which leads me to believe that all 4 cylinders would be running hot. Maybe it's just the fact that #3 is in the middle of the engine, so it fails because it can't get rid of the heat.

I talked to someone at Jim Wolf Technology and he said the stock injectors are 259cc, anyone know how much time they take to pass that ammount of fluid? He also said there is something seriously wrong with my engine and\or fuel system and that their computer would not compensate for it... you gotta at least appreciate his honesty!

I checked the fuel flow and I'm getting about 146 lph out of the fuel pump with no back pressure, but because the pressure doesn't drop significantly when I rev the engine, I am reluctant to attribute it to a low-flow problem from the pump.

Still floundering... any more ideas?
Thanks,
Mike

[ScottNutter]

Here's another problem...
according to the color of the plugs, I'm running rich
[attachmentid=624]
correct??
Mike