Coolant Pump Behaviour?

[unL33T]

I am considering doing some modifications so the car can take advantage of the waste heat from the electrical drivetrain and REx by putting a heat exchanger between the coolant circuit and heater circuit.

I am wondering if I need to limit when coolant flows through this heat exchanger to avoid having the heater circuit waste energy heating the drivetrain coolant or dumping its heat through the radiator.

Does the coolant pump run all the time or only when a certain temperature is reached? If so, does anyone know that temperature the coolant pump starts running? Once it starts running is the radiator also fully open or is airflow somehow controlled across it to control cooling?

Basically I am trying to determine how the cooling system operates. I know from the literature the coolant pump has variable speed, the REx cooling has a thermostat which isolates it from the rest of the cooling until operating temperature is reached, and apparently there may or may not be actuated fins that control airflow through the radiator.

Any other reasons this is a dumb idea? It seems odd to me that in a car designed for efficiency BMW chose not to utilize any waste heat to heat the cabin. Maybe some electrical risk of conduction through the coolant and heat exchanger I am not considering?

Thanks!

 

[kd7iwp]

I believe there are 3 different cooling tracks and 2 different types of coolant in the car, might not want to mix those. How often are you running the rex? I turn mine on ever 2 weeks or so to keep it more active (I find engines that are used more regularly work better) but not that often, that it would provide me heat. Maybe if I was doing a longer drive and running it for an hour or something. This seems like a lot of work and you could also have pumps competing against one another.

 

[unL33T]

I am aware that there are 3 coolant loops and 2 different types of coolant. Hence the heat exchanger I mentioned. I would not mix them. Strangely, even the coolant for the EV drivetrain is isolated from the REx via a heat exchanger even though they use the same coolant. The pumps would not compete since all the loops are/would be isolated via heat exchangers.

I live in Canada so it gets cold a lot. Even with the heat off the REx cannot keep up with highway power consumption. Dumping all the excess heat from the EV drivetrain and REx for hours is wasting range. The computer tells me ~15% of my power is going to heat with it set to 18C and it hasn't even gotten below -10C yet. For 90% of my driving (just commuting to work) it's no issue but for long drives where I need/want more efficiency, I think it would help. At highway speed even with the REx running and heat off I am losing ~3% every 10 km. If I have a several hour long drive with no charging at the destination, this is an issue.

This would capture the EV drivetrain heat, as well, not just the REx.

If the coolant pump only runs when the coolant is hot and the radiator doesn't get airflow all the time, I would only need to install a single heat exchanger to accomplish this. If not, then I may need a few more components. Hence why I am asking for details of how it works, what it does at what temperature, etc.

Part of it is also just for the fun of doing it.

There is a guy on here that supposedly installed a diesel heater for similar reasons. I've thought of doing the same. That'll be way more work and cost. But I figured first step is to capture the waste heat and see how that goes. I assumed the car already did so (like almost every other car) until recently.

 

[mojo]

It seems odd to me that in a car designed for efficiency BMW chose not to utilize any waste heat to heat the cabin.

There seems to have been a fundamental design requirement that the core of the BEV and REX should be the same, and use the same (effectively shared) core heating and cooling systems. When you think about it, it wouldn't have been acceptable to have a design where the heater only worked properly in colder climes if you bought the REX. So they built a heating system for both that would be sufficient in both BEV and REX without needing heat input from the engine. You can argue that it would have improved overall efficiency when the REX is operating, but it's also true that running the REX is its least efficient mode of operation.

Above all, BMW seems to have wanted to reinforce the concept that "this is a range extender and not a hybrid". The engine appears to have been deliberately classed as a "bolt-on" and not a fully integrated part of the vehicle's design.

The really big problem for that sort of project is that BMW does not publish in detail how their cars are designed to work. You can get diagnostic data, and you can pay for access to their repair procedures, but the way that the car is intended to operate remains hidden. There are other examples: you want to know exactly what complex rules the car uses during charging? You won't find it documented anywhere. You want to understand exactly how the BMS works? Not a chance. You need to know what rules the car uses to operate cooling fans and pumps? That won't be in any externally-published BMW documents.

It's frustrating, because it means that the only way to get the information you need is through observation and measurement (of your own car, unless someone else has already done it).

And... having said all that... my gut feeling is that the extra efficiency you might gain is *probably* not worth the tens of hours you will end up devoting to the project. Electricity isn't (yet) so expensive that it's too much of a problem just to recharge the battery a little more often!

 

[alohart]

Above all, BMW seems to have wanted to reinforce the concept that "this is a range extender and not a hybrid".

Thatʻs my feeling as well. BMW designed a "MegaCity" EV with a REx option for those who might occasionally need to extend the range enough to reach a nearby charging station. BMW did not intend the REx engine to run for hours while driving an i3 like a gasoline-electric hybrid. Occasional REx usage would not warrant the cost and weight of implementing the transfer of engine heat to the cabin heating coolant.

And... having said all that... my gut feeling is that the extra efficiency you might gain is *probably* not worth the tens of hours you will end up devoting to the project.

I havenʻt paid attention to the motor and motor controller coolant temperature. However, based on the small size of its radiator and the efficiency of the motor and motor controller, my guess is that the coolant temperature on a cold day would be considerably lower than the temperature of the cabin heating coolant, so a heat exchanger would transfer heat the wrong direction.

A heat pump would make much better usage of this waste heat. I believe that Tesla extracts waste heat with its heat pump system.

 

[unL33T]

There seems to have been a fundamental design requirement that the core of the BEV and REX should be the same, and use the same (effectively shared) core heating and cooling systems. When you think about it, it wouldn't have been acceptable to have a design where the heater only worked properly in colder climes if you bought the REX. So they built a heating system for both that would be sufficient in both BEV and REX without needing heat input from the engine. You can argue that it would have improved overall efficiency when the REX is operating, but it's also true that running the REX is its least efficient mode of operation.

Above all, BMW seems to have wanted to reinforce the concept that "this is a range extender and not a hybrid". The engine appears to have been deliberately classed as a "bolt-on" and not a fully integrated part of the vehicle's design.

The REx already has both a different heating system and cooling system at least in North America. Non-REx has a heat pump and in the back there is a heat exchanger between the REx and the electrical components.

The really big problem for that sort of project is that BMW does not publish in detail how their cars are designed to work. You can get diagnostic data, and you can pay for access to their repair procedures, but the way that the car is intended to operate remains hidden. There are other examples: you want to know exactly what complex rules the car uses during charging? You won't find it documented anywhere. You want to understand exactly how the BMS works? Not a chance. You need to know what rules the car uses to operate cooling fans and pumps? That won't be in any externally-published BMW documents.

It's frustrating, because it means that the only way to get the information you need is through observation and measurement (of your own car, unless someone else has already done it).

Yes, I understand. I wouldn't be asking here if it was published.

And... having said all that... my gut feeling is that the extra efficiency you might gain is *probably* not worth the tens of hours you will end up devoting to the project.

From a recent long drive I am leaning the same direction as even with the REx running, which is connected via heat exchanger to the electrical drive train's cooling system, it still took about 45 minutes before whatever needs to warm up would give me full power. It's not even that cold out yet. Temperature was -7.5C that day. Today was -11C but I was able to park in a heated garage before departure. I believe the battery is not connected to either cooling loop so REx and drivetrain heat is not going to help it warm up any faster.

Electricity isn't (yet) so expensive that it's too much of a problem just to recharge the battery a little more often!

My issue is long drives with no charger at the destination. Most of our small towns only have a few of them and they are always 10+ minutes away from where I am going so far.

I've noticed that using public charging stations where I live are so expensive that it's actually barely cheaper than running a gas car. I guess my own fault for not researching that enough beforehand especially since I can't charge at home. At least I get free charging for 85%+ of my use.

 

[unL33T]

Thatʻs my feeling as well. BMW designed a "MegaCity" EV with a REx option for those who might occasionally need to extend the range enough to reach a nearby charging station. BMW did not intend the REx engine to run for hours while driving an i3 like a gasoline-electric hybrid. Occasional REx usage would not warrant the cost and weight of implementing the transfer of engine heat to the cabin heating coolant.

I havenʻt paid attention to the motor and motor controller coolant temperature. However, based on the small size of its radiator and the efficiency of the motor and motor controller, my guess is that the coolant temperature on a cold day would be considerably lower than the temperature of the cabin heating coolant, so a heat exchanger would transfer heat the wrong direction.

A heat pump would make much better usage of this waste heat. I believe that Tesla extracts waste heat with its heat pump system.

I don't think it would add that much weight or complexity. Another thermostat to avoid the issue you mentioned in your 2nd paragraph and another heat exchanger.

I guess I'll have to investigate it myself. Can you read these temperatures somehow with an OBD scanner? I did buy one that is compatible with BimmerLink but haven't tried it yet. If it's viable using the heat to help heat the battery would probably be even more beneficial but messing with the AC loops seems way more risky for someone not familiar with them than the regular cooling loops.

 

[mojo]

The REx already has both a different heating system and cooling system at least in North America. Non-REx has a heat pump and in the back there is a heat exchanger between the REx and the electrical components.

Agreed, there are differences between REX and BEV, but the differences are not huge. The heat pump in the BEV is simply an option - not all BEVs have one. So the essential difference (ignoring the heat pump) is the addition of one coolant-to-coolant heat exchanger so that the engine coolant circuit can dump its heat into the main coolant circuit - which is otherwise exactly the same as on a BEV.

For the battery assembly, the heating is done electrically via heating elements bonded to the base of the battery casing - as far as I can tell there is no simple way that you could employ engine heat for that purpose directly.

 

[unL33T]

.... The heat pump in the BEV is simply an option - not all BEVs have one. ...

It is standard in North America. Just saying.

 

[alohart]

It is standard in North America. Just saying.

This was true in the early years but no longer. Our 2021 U.S. BEV doesn't have the heat pump option which is fine in Honolulu. It might have become optional with the 2018 LCI.