Is it at all possible to remove the engine tin over the oil cooler and temp sensor without dropping the engine? I need to get at both the oil cooler top and I need to replace the temperare sensor.

If so what should I look out for?

Yes you can .
You will have to remove all fuel injections components, intake manifold and a lot of patience. The end by the distributor abd by the transmission will kick your butt.

Two questions.

1. Why do you need access to the top
of the oil cooler?

2. Which temp sensor? Oil or head?

I put a custom fuel injection setup and am using a megasquirt engine management system on it. It works fine but the cylindar head temp sensor is very nosy electrically, I will be replacing it with a type K themocouple together with a themocouple amp with offset correction and some op-amps to make it act like an thermister with a negative temp curve that the megasquirt expects. This will isolate the temp sensor from the engine ground that electrically noisy.

I had put an extra duct in the heater blower to go to the other side. It slipped off its mounting and got sucked into the impleler. I pulled a mess of wire with sheaded plastic out of the impeller and since that event the oil temp has been a bit warmer that before. Since I need to access the temp sensor on that side I can inspect the top of the oil cooler for plastic bits that may be blocking some of the area,


That front tin pieces you sold to me a few years ago has been installed and is working fine.

In that case, get both sides. There's likely debris in the cooling fins on both sides.

--DD

Why would you :

why wouldnt you just replace it with a new thermistor "that meagasquirt expects" that is not grounded to the engine case like the stock version?

Sounds like a lot less work.

Take a stock CHT, and drill out the core and replace it with a two wire (or 4-wire if you are really anal) and hook it up.

Converting from thermocouple back to thermistor-like thing sounds like an experiment that isnt required.

Stu

Do you have an other than stock temp sender?

I will be replacing it with a type K themocouple together with a themocouple amp with offset correction and some op-amps to make it act like an thermister with a negative temp curve that the megasquirt expects. This will isolate the temp sensor from the engine ground that electrically noisy.

Thermisters are qustionable above 300F i think . A thermocouple is a nice linear sensor but the only issue is making something with a postive thermnal coefficent to be negative.

yes I willl be using a thermocouple not the standard thermister.

True.

Can the megasquirt read from this digitally?:

http://www.adafruit.com/products/269#Description

Direct TC to digital output.

Stu

I thnk drilling out an old thermister and tring to glue it back together with some sort of epoxy that can transfer heat in order to get an isolated ground sensor is questionable in its final performance. I'd go the electronics route and take a sensor that is primarly linear from 0 to 500/deg C where a silicon based sensor is questionable after 300 def F.

The megasqiurit has a 0-5 volt AD in it Just take two op-amps with the first being non-inverting and add a gain of two after the termocouple controler giving it a postive temp coeffecient of 20mv/deg C followed by a unity gain inverting amp. Thus getting the same negative themo co-efficient as the thermister ( but linear). I don't care about lower temps so the layout is good from 40 def F to 420 deg F. The themocouple is the isolated ground type so it does not couple the engine ground to the sensor ground.

The megasuirt uses an AD that reads from an anolog signal. The normal configuration uses a pull up resisitor of 2490 ohms to five voltes coupled the the thermister to ground. I am just replacing the the thermister with some electronics that drives out a signal from 0-5 depending on temp. ( 5 volts = 0deg C and 0 volts 250 deg C)

Only possible way is to use CANBUS but that is too complicated and expensive. The accuracy of the tempature need is not that great.

I have been experimenting with the innovate TC-4. It can read 4 channels of CHT or EGT, but the readout is still digital.
All it requires is a RS-232 (3-wire) interface and new devices just daisy chain.

I know there is a digital input to megasquirt for this purpose because I found others doing this exact reading even though my application is not megasquirt oriented.

Stu

Four channels would be nice.... I hope one is representative. One is all that is necessary for the megasquirt. I'm doing this on the cheap. $20 thermocouple, $35 worth of electronics.


Do you see much variation in head temp?

I havent hooked up the CHT yet or run the engine much since break in.
I have been using it to watch EGTs and it works fine.
I am currently trying to figure out how to read the digital data stream with a Droid tablet so I can have a generic instrument display.

Also - You can get a thermistor that works to 200C: http://www.omega.com/Temperature/pdf/55000.pdf

And


It would work just fine.
That is how we install them on spacecraft.

The 4-wire measurement is more accurate than the 2-wire measurement, but the source/meter must support it.

Stu

IME, it's much easier to R&R the engine and be able to easily access everything.

The Cap'n

" since that event the oil temp has been a bit warmer that before "

Please correct me if I'm missing something. This statment sounds like you want to access the oil temp sensor. Yes or no.

How much noise are you talking about? I assume you are using the CHT for the coolant temp signal in Megasquirt. If so then you most likely are only using this for warmup enrichment in which case you would need a pretty major swing in the sensor reading to actually have a problem, so aside from getting nice pretty graphs in your data logger I don't know what else you would be accomplishing here by changing your sensor..

How bad are the fluctuations in the reading? I run the stock CHT sensor grounded its normal way through the block without issue.

I've noticed general 70-80 deg F swings and occasional enough of a change to start the warm up enrichment when the CHT is over 300F.

I've added extra grounding from block to frame but have not tracked down noise any more.

Nope the cylinder head temp (Temp II)

This could also be partly (or completely) due to the thermistor table you are using, depending on how it is setup if the resolution is in the wrong place you could see large jumps in readings from a minor change in the sensor resistance.

which temp sensor (part number) do you have and what values did you use when you created the sensor table?

Did you measure the values of your own sensor or did you copy someone else?

Thank you for clearing up my confusion.