Hi,
Decrease R2 the series LED resistor to 180R.
If the anode of D1goes down to 8.5V.
Then 12 - 8.5 = 3.5V drop across relay coil.
3.5 / 0.02= 175R so 180R close enough.
Tom... 
If the anode of D1goes down to 8.5V.
What makes you say that Tom?
Given the instructors in emitter will not have much resistance then it is unlikely that the collector is going to be about 5V.
Anyway you should still see the LED with a 470R if 180R gives you 20mA.
ullisees:
The coil is a solenoid that opens or restricts oil flow.When the coil is energised the voltage at the switched 0v to the supply 0v is 8.5v
The permanent supply to the coil is 12v
I understand that that point which is the anode of D1 is the switcvh to "0V" point that doesn't switch to "0v" but 8.5V.
So drop across relay when it is activated is 12 - 8.5 = 3.3V. (We assume)
Now if relay coil is really 12V, then something is wrong.
Tom...
Can you please post a picture of your project? Relay and ECU..
There's a dirty big 30V zener limiting voltage on that coil, so your LED is getting 18V of reverse bias
which is probably frying it (LEDs are normally quoted as 5V reverse voltage max).
Put the 180R in series and the led lights.
Next problem now is I i am using a rgb led and i want it to light green when the coil is energised and go red when not energised.
My thought was to use a uln2004 to turn it ground onto the red led by pulling the input to 12v via a 1k resistor. Then put a signal diode to the switched ground to the coil so that when the coil was turned on the led would switch from red to green.
Testing the circuit on the bench works but attached to the coil and controller in doesn't. Led just stays red.
Before anyone asks I am using a uln2004 because i will beed to attach 8 leds to 8 coils eventually but for now am just working with one.
Thank you