So I do have both ERWIN versions for the vehicle,, the first 2019 edition has the mention of the heater bleed hole when doing the actual routine. The second 2020 version leaves out any and all mention of using that heater hose bleed at all.
Strange, but ive tried both methods and still cant seem to get every bit of the air out.

Before running the bleed routing- which for my vehicle I have just one VCDS option under basic settings for "bleed cooling circuit".. So before i do that basic setting routine,, with engine off and heater high fan low, I've been running the output for "heater support pump" and "coolant pump 2" ,10 cycles each,, as soon as a cycle is done, I have been by hand squeezing every coolant line for air bubbles. Rinse. Repeat. I have no option to run main pump, just the "switching pump" which only makes the click sound.

Hell, I even lifted the front end and did the entire hokey pokey dance as just described and still have some air pockets after driving.

I did find this random hose hanging out pointed below the vehicle that looks like it has some intentional filter at the end, but cannot find in any manual wth it is.
https://cimg6.ibsrv.net/gimg/www.aud...862a0df278.jpg






https://cimg6.ibsrv.net/gimg/www.aud...0cf6872ed6.png
https://cimg2.ibsrv.net/gimg/www.aud...b7f17b855c.png

So its confusing, but ive tried both methods and still cant seem to get every bit of the air out. Before running the bleed routing- which for my vehicle I have just one VCDS option under basic settings for "bleed cooling circuit"
https://cimg0.ibsrv.net/gimg/www.aud...20c6cdd3ed.png

Yeah, cardaign.com only has typically the initial releases and never updates from there. I don't have a self copy for CWGD. Sometimes the updates are good, sometimes they are bad (recent structure and leaving out a lot of diag test procedures), and sometimes they are critical (updated torque specs for the changed crank bolt on the EA888 Gen2, e.g.). Wonder why they elected to remove the reference (assuming the employees doing the documentation edits have ever even been under the hood of an Audi vehicle).

It is interesting that the VCDS label just says "test". I wonder if this is the same as 0119 - coolant circuit vent routine in ODIS. Would be interesting to run a CAN sniffer during the VCDS action and the ODIS action and compare the actual commands forwarded to the ECM. Seems to be no quick search hits on bleeding a CWGD or DECA out there.

No search hits brought me here hoping for some luck. You’d think there would be something out on the web by now, given how long this engine has now been on the market. Not to mention the vast amount of pumps, valves and thermostats that have gone bad on it.

I have been videoing everything I’ve done so far (including the half-wit idea to drive only the front passenger wheel up onto a ramp 🫨), so if I accidentally figured out the magic method to bleed this thing correctly, I can post up the instructional.

Since there's no sight glass to see air in the coolant system, how are concluding you have air that still needs to be removed?

Oh, and that tube. I assume it's the drain/vent line for the headlamp, the way it looks. Should slip into a notch in the air guide along the radiator.

I think that there is air in the system because after bleeding, I’ll go on a test drive and as soon as I engine gets to operating temp 207c, I get the dash pop up for “coolant temperature too high” when I’m a bit aggressive with a WOT. I’ll then get some nice coolant boil over.. At that same time I have vcds open and watch the coolant temp go from normal 98C to 108-116c for a minute.


Then, as after I let off gas, it’ll make its way back down to the normal 98-ish. But I’ll still now obviously have to do a top up of coolant right afterwards, aggravatingly.

So I am guessing in its finest form, that the issue alert would be the un bled air that made its way back through the cooling circuit, but I can’t explain why it happens only when engine is hot .

Bump

So here is something strange that may help diagnose my issue I hope. In VCDS I have this output test option to run "IDE05036- Switchable Water Pump". Typically, when I run this output, I only get the "click, click" sound and nothing else ever happens.

Today however, I ran this output on a dead cold engine that sat for a day, with the electrical connector to my upper switching valve removed,, but left the vacuum lines on it alone. Now i hear an actual pump pushing coolant through the system. What is up with that?

(connector in reference)
https://cimg5.ibsrv.net/gimg/www.aud...3e370d217a.png
https://cimg6.ibsrv.net/gimg/www.aud...71f8fe8b9f.png

Well,, I vac n filled and bled the circuit overnight by running a Microsoft power automate session that ran the heater support pump and coolant pump 2 every hour while leaving the reservoir lid off. Went to test drive this morning and got the wonderful “coolant temp too high” alert… So I pull over and turn off the engine and let cool down for 20 minutes.

Started back up— with a puttering whole car shaking vibration and a new code for P0203… sprayed all the injector contacts with electronic cleaner and reconnected with no luck. Now I have a whole new damn problem and can’t find any solutions on the web past what I did cleaning connections.

Frustrated to say the least.

It appears that the factory procedure is calling for a hot/running engine to do this. This probably activates valves that normally not working for a cold engine. If this is true, then the vacuum fill probably won't work... unless you can find all the valves and pumps involved and activate them by force (applying signal or power) on a cold engine for the vac and fill to work.