Car- Stock93MT
ECU- DIYPNP megasquirt

I finally finished building the ECU, went out to the car to load the firmware and base map and said to myself "what the hell, im here may as well try and start it"

On initial start up it started very hard and was slobbery rich. The temp on the dash did not register and the temp on the computer said I was 240* which i know is wrong cause it was a cold start. The car idled at 1500 and the TPS ( which is the stock non variable one) started climbing in percentage all the way to 400%.

I shut everything off and re calibrated and got the same results.

The car was not timed to the ECU and the MAP was not hooked up.

Should I ignore the TPS reading?

What should I look for to fix the temp sensor?

Thanks in advance

 

sdfs

 

yes, i calibrated the tps. I still reading through the forums to see if someone has had similar problems or not.

Could my TPS just be bad? My stock ECU runs the vehicle just fine with 0 problems

 

You need to put TPS sig to a ground on the DIYPNP, with a 1K resisistor in the middle. Then when calibrating, your wide open count should be one less than the closed count. I used 1 for closed throttle, zero for WOT. It will stop counting this way.

Or get a real TPS, it is the bomb.

Not hooking the MAP up is dumb, car probably thought you were trying to WOT floor it based off MAP alone.

You probably just need to calibrate the coolant temperature sensor. Have you read the directions?

 

+1 on reading the directions because:

A) 93 cars don't actually have a TPS, just a WOT switch

B) running without MAP line would cause slobbery rich condition

 

Can an early NA be made to run well without adding a TPS?

I know that TPS is used for AE and without a TPS dot it probably wont work well, so car would stutter on fast throttle input. I'm asking to see if this ecu is relevant for early Miata in ST classes (we cant add additional sensors)

 

Ask John Brown. =)

 

I did read the directions,I know they have a WOT switch although its still referred to as TPS which is why in my original question I stated it was non variable. I will be getting a variable one but this is my winter project and winter isnt quite here yet.

Also in my original post I stated that I re calibrated the sensors twice after the initial calibration. Could I have built something wrong in my megasquirt to cause the temp sensor to not read correctly? And why does my temp sensor not read to my instrument panel and just the laptop?

Like I said the map wasnt hooked up cause I was just loading software and wanted to try starting it, my nearest parts house is about 45 minutes away and I didn't have any vacuum line on hand.

 

45 minutes to the nearest parts house? You must have a big back yard.

Anyhow, if you didn't install the correct resistors for the sensors or selected the correct bias resistor value when calibrating them, it may account for the false sensor readings. Other than that, wiring error.

My car is a 95 and I believe it has two sensors, one for the gauge and one for the ecu. Don't know about the earlier cars.

 

Lol i do have a big backyard. there is a local one but they close at 3. I followed the instructions to a T and i also referenced brains thread. ill go back through and check all my connections and double check to make sure i have a resistor there. Ive gotta do the wideband tomorrow or the day after and ill run a vacuum line and do the timing then and hopefully i can get this sorted.

 

OK so I recalibrated the Temp sensor and see where I made the mistake and now that works flawless and then fans kick on as they should.

Ran a vacuum line and did another test fire this time with the wideband wired up and still pig rich. Adjusted the fuel tables down a bit and got a happy idle with a afr of 14.5.

TPS is still not registering proper though, Ive tried a couple different things and double checked my solder job on the ECU and still nothing, starts at 0% when I start the car and climbs indefinably until i reset the connection.

 

You need to manually set the closed count one higher than the WOT count in the TPS cal page.