I have already seen that there were attempts to reverse engineer the heat pump system on the forum (seemingly successful, but this is not certain). Most likely, that guy is unlikely to share the nuances.
Therefore, I suggest that we unite and pick apart this miracle of technology. In particular, I am adressing to mr. Daveturpin, who assembles his VW E-Bug.
I will also listen to comments and recommendations from those who really know their way around climate control technology.
He only has seems to taken his hardware away from being public. Cant really blame him as its a massive project to undertake.
The heatpump itself, the compressor is CAN. The rest of the sensors and actuators on the super manifold and such are all Analogue or direct controlled.
tom91 wrote: ↑Tue Jun 03, 2025 8:15 am
The heatpump itself, the compressor is CAN. The rest of the sensors and actuators on the super manifold and such are all Analogue or direct controlled.
I am sure that it will not be difficult to run the compressor. The main difficulty is in the freon temperature/pressure sensors. That is, to adequately assess their operating range, so that the compressor does not accidentally blow up, etc.
AMP3R wrote: ↑Tue Jun 03, 2025 8:27 am
The main difficulty is in the freon temperature/pressure sensors.
Not really they are just off the shelf analogue sensors, which Wim426F has done the mapping in his software already.
The hardest part is the control logic of the system, there are so many valves both coolant and refrigerant based ones. Which if you switch them wrong you will create issues. Defining a strategy to do what when will take ALOT of effort to make it work half as good as the OEM setup.
The AC pump will have its own internal safety features, mechanical and sensor/power based ones to keep from breaking itself or the OEM system.
A little googling revealed that EXV control is fairly simple, but it need to figure out how much to open the valve. As I understand, the amount of passage is calculated based on feedback from the pressure sensor.
AMP3R wrote: ↑Tue Jun 03, 2025 6:09 pm
passage is calculated based on feedback from the pressure sensor.
EXVs are steppers, 4 pole I believe.
Correct you are looking to hit delta P's across the EXVs with various targets based on cooling demands. Which ofcourse depends on the target temperatures of the Heat exchangers, cabin evaporators or condensors.
In principle in the extreme case for summer you can just manually set EXVs LCC, Chiller and Evaporator to some average value and go for a ride. Yes, I understand that this is a redneck style
I bought this system only because no one except Tesla has yet been able to make such a compact thermal system. I like minimalism and when there are a minimum of hoses.
After another googling and studying the Wim Boone's controller, I got some valuable information.
For water pumps, he uses bldc drivers with integrated FETs MCF8316 series with the peak current up to 8A, which seems insufficient if compared with what the professor claims (15A peak for each pump at max 6000 rpm)
Tesla uses drivers with external FETs, and apparently this is done for a reason. I know it's in the old VCFRONT from M3, but most likely they didn't change anything in further versions.
It's understand that a driver with integrated FETs looks more elegant and I like this solution more, but it's unknown how long they will handle the pumps.
Most likely I'll have to buy VCFRONT from fresh M3/Y, tear down it and then reassemble on a custom circuit board taking only what is responsible for the heat pump controls
You can always cobble together something like a VSEC or other hobby bldc controllers to run the pumps. Those TI chips do look mighty fancy though and would yield alot finer control.
Key thing to make sure in the repurposing is to ensure you do not end up with alot higher flow restrictions compared to a tesla setup. Also the radiator sizing will be key to this setup to be able to reject the heat from doing cabin AC.
tom91 wrote: ↑Thu Jun 05, 2025 11:03 am
Also the radiator sizing will be key to this setup to be able to reject the heat from doing cabin AC.
The plan is to use as many components from M3/Y as possible: the radiator and hoses to it, possibly VCFRONT, to get the control as is. The only non-original parts are the condenser and cabin evaporator.
AMP3R wrote: ↑Thu Jun 05, 2025 4:31 pm
What's the danger? Will it burst
Explode as its a gas, it will rupture and shred the lot and send it flying. Strongly not suggested to risk. Note the running pressure of the condensor is atleast roughly 5x that of the condensor. I am looking for more detailed numbers at the minute.
Note how the TXV/expansion valve drop the pressure across it before it gets to the evaporator.
