Remote Electric Water Pump

[gongloo]

I'm thinking of replacing my water pump with a remote electric unit mounted somewhere towards the front of the car, and then running AN lines though much the cooling system (with custom flanges/fittings wherever needed). Some of the reasons I think this might be cool (pun intended):

1. Frees up space in the engine compartment, making the VoD easier to access, clean, and investigate.
2. Reduces rear weight bias a tiny bit (every pound counts!).
3. Eliminates long heater return line from front to back of car (reducing front-to-rear coolant lines from four to three).
4. Puts otterstat in return line from radiator rather than feed line to radiator, possibly reducing unnecessary fan activity.
5. Allows the water pump to be configured to run only when the engine is above a certain temperature, decreasing warm-up time to optimal temperature.
6. Allows the water pump to be configured to run for some time after the engine has shut off.
7. Eliminates failure mode of engine overheating with a shredded/malfunctioning belt.
8. Reduces unnecessary engine load at high RPM (I doubt this would be noticeable, but every horse counts!).
9. Makes water pump and/or thermostat replacement trivial later down the line.
10. AN fittings eliminate leaky hose connections and greatly reduce maintenance pain.

I created this crude diagram to help get straight all of the routing. This moves the thermostat, otterstat and water pump out of the engine compartment and close to the radiator. Though I haven't yet looked at exact options for mounting locations, I suspect I can find a place up front that's accessible with reasonable ease from under the car.


https://docs.google.com/drawings/d/e/2PACX-1vSpRDmGUbKAMXEMGuMlhy3-4MBbmZTAHvV3yJEneyyc1l5ZBx6FV_-i8uR-TicAz3Pf9JrTlAiUjyYq/pub?w=637&h=409 (https://docs.google.com/drawings/d/1AhXOGSOrOd1_4zo_EIkr6LtEvuJIz6pitU6wdthbEMk/edit?usp=sharing)


Would love to hear your thoughts, especially if anyone has made significant modifications to the cooling system like this before. Right now this is mostly a thought experiment, but if I decide to pursue this further, I'll document any progress made here.

[Mark D]

I can understand where you are coming from wanting to improve the car and make it more modern, add some control functionality, etc.

But unless you plan on keeping this car forever, it sounds to me like this is something that would need to document to the Nth degree to avoid being the 'dreaded previous owner' if you ever sell your car later on down the line. What may seem like an improvement now will most likely be something that the next owner will have to un-do to return the car back to stock. It may also devalue your car if potential buyers do not want to deal with a custom setup that nobody else knows how to troubleshoot or repair without remaking custom hoses, buying special AN fittings, etc.

Is the standard cooling system design really that bad that it needs to be completely re-engineered? There are a few obvious shortcomings like dealing with broken bolts on the Y pipe in the VOD, the plastic overflow bottle that typically gets updated to stainless, and the plastic side tanks on the radiator. But for the most part it's a well designed system that is very robust compared to a lot of modern cars. Nearly all of the shortcomings can be addressed with a one-time replacement of a part like the expansion tank / radiator / etc. I'm not reading dozens of threads on DMCTalk about ongoing cooling system issues that would benefit from completely redesigning the system.

The cooling system was 100% original on my DeLorean until around 2011 when as a preventative measure I replaced a bunch of stuff, you know, just because I was in there doing other stuff. The system lasted decades before any major parts needed replacement...Compare that to my daily driver, A 2004 E46 BMW 325i. This car has a cooling system that is nearly all plastic and has a recommended service life of 60,000 miles or 5 years. There are pages upon pages of DIY threads and guides for how to "refresh" the cooling system before a catastrophic failure occurs. The expansion tank loves to crack and eventually explode, ejecting all the coolant from the engine which leads to nearly instantaneous overheat -> cylinder head warping -> head gasket failure -> engine failure. This is obviously an extreme comparison, but it makes the DeLorean cooling system with it's aluminum tubes and rubber hoses look like a bulletproof system.

If you do go ahead with your redesign I would suggest that you at least make it easily reversible just in case you (or a future owner) decide to go back to the stock configuration. It does sound like a really cool project to watch and see how it's all going to work out. If this is something that you're really passionate about doing don't let me or anyone else kill your ambition.

[dn010]

Few things:

You will still need a belt to run the alternator so you'd need to figure that part out as well. Might be simple as getting a smaller belt for just the crank to alternator unless the angle of the tensioner doesn't allow it to work.

You'd probably want coolant to circulate when the engine is running so that you don't get hot spots in the engine and all sensors get the same temp coolant.

You'd probably not want to run the pump when the engine is off unless you also plan to have it run cooling fans at the same time [battery strain].

How often do you plan to access your VOD for maintenance that justifies removing the pump and redesigning the cooling system?

Just a preference, but I would rather have a mechanical pump that slowly indicates it's need for replacement [by weeping] over an electric pump that either works, or it doesn't, and usually the doesn't part comes at the worst times.

I'm not going to get into the whole weight reduction of removing a water pump or your HP gain. If that is a concern of yours you should probably be checking out engine swaps instead.

I can see the benefits you're trying to achieve, but personally the water pump is the LAST thing I worry about when I am driving. 17 years later and I haven't had a blown hose or coolant related breakdown, knock on wood. If you do go forward with the project, you'd certainly be the first one to do so! Good luck.

[FABombjoy]

I would take all that money you're going to spend on AN fittings, hoses, electric pump, wiring, etc., and put it toward EFI instead:


1. Frees up space in the engine compartment, making the VoD easier to access, clean, and investigate.
4. Puts otterstat in return line from radiator rather than feed line to radiator, possibly reducing unnecessary fan activity.
5. Allows the water pump to be configured to run only when the engine is above a certain temperature, decreasing warm-up time to optimal temperature.
6. Allows the water pump to be configured to run for some time after the engine has shut off.
9. Makes water pump and/or thermostat replacement trivial later down the line.

#1: VOD access in 10 minutes on a well set up EFI system, even with the original intake.
All of the other could be computer controlled
#5 might create hotspots around the liners and I wouldn't do it
#6 would just put you back in to warm-up mode faster and decrease efficiency. You want the engine up to temp as quickly as possible.


3. Eliminates long heater return line from front to back of car (reducing front-to-rear coolant lines from four to three).
It's aluminum tube and can't weigh more than a few pounds.


7. Eliminates failure mode of engine overheating with a shredded/malfunctioning belt.
If the belt goes, the dash lights up due to alternator failure. What will be the comparable "tell" with an electric pump?

[David T]

It adds additional load to an already overloaded electrical system. For all of the work and expense, IMHO it isn't worth it. You would also have to figure out how to run the alternator without the water pump. Could be as simple as just getting a different size belt. The water pump is very simple and reliable as it is. If you are doing it because it is an interesting project, go for it.

[Timeless]

I would take all that money you're going to spend on AN fittings, hoses, electric pump, wiring, etc., and put it toward EFI instead
+2. This is a FANTASTIC idea.

[dn010]

See post #3.


It adds additional load to an already overloaded electrical system. For all of the work and expense, IMHO it isn't worth it. You would also have to figure out how to run the alternator without the water pump. Could be as simple as just getting a different size belt. The water pump is very simple and reliable as it is. If you are doing it because it is an interesting project, go for it.

[gongloo]

Thanks all for your great responses thus far. Very interesting discussion!


You will still need a belt to run the alternator so you'd need to figure that part out as well. Might be simple as getting a smaller belt for just the crank to alternator unless the angle of the tensioner doesn't allow it to work.

Yes, I'd want to try to find a belt that is the right size to run the alternator directly off the crank without any other idlers or pulleys. I assume that the angle of the tensioner would not be a problem, but perhaps this is a poor assumption.



You'd probably want coolant to circulate when the engine is running so that you don't get hot spots in the engine and all sensors get the same temp coolant.


Good thinking, I hadn't considered hotspotting! I'll update the original post with a note to this effect.



You'd probably not want to run the pump when the engine is off unless you also plan to have it run cooling fans at the same time [battery strain].


I think a few amps for a minute or so shouldn't really strain the battery too much, but of course that depends on the battery system.



How often do you plan to access your VOD for maintenance that justifies removing the pump and redesigning the cooling system?


Yearly, because I'm anal. Monthly if it's trivial to reach right in!



Just a preference, but I would rather have a mechanical pump that slowly indicates it's need for replacement [by weeping] over an electric pump that either works, or it doesn't, and usually the doesn't part comes at the worst times.


Yeah, I'm not really familiar with the failure modes of electric pumps, so that's totally worth considering further. Lots of cars now exclusively use electric water pumps stock, and it seems for extra insurance some of them include a flow meter to throw a CEL when they malfunction. Hooking something like that up to the fan fail circuit would be fairly trivial, and possibly wouldn't even require a flow sensor (sensing voltage/amps across the pump might be sufficient, depending on failure modes). Lots of food for thought here!

I'm so sick of coolant leaks at this point, though, and totally wouldn't mind carrying an extra small, light electric pump in the trunk as a backup (instead of the emergency coolant bottle I currently carry), especially if it's fairly trivial to replace. Wouldn't be the first time I've had to replace a component like that on the car at the side of the highway (last time was the alternator).


I would take all that money you're going to spend on AN fittings, hoses, electric pump, wiring, etc., and put it toward EFI instead

Why not both? :-P



#6 would just put you back in to warm-up mode faster and decrease efficiency. You want the engine up to temp as quickly as possible.


I'm not talking about bringing the engine way down in temperature; I'm just talking about reducing hotspotting after shutdown. Definitely less of an issue for naturally aspirated engines (which, to be clear, mine is), but nevertheless seemed like a neat trick to help a little bit with longevity. Maybe just a minute or so of circulating coolant post shutdown.



It's aluminum tube and can't weigh more than a few pounds.


I'm concerned less with the tubing weight and more with the water pump weight and location. That unit is surprisingly heavy! Honestly, though, I'm way less concerned with the weight than the other thoughts I had above. In particular, I hate that the system has that many hose connections in it. My aluminum tubes are also a bit pitted and worn, so getting a great seal on them is a pain. Could I replace the aluminum lines? Sure, but where's the fun in that?!


It adds additional load to an already overloaded electrical system.

This is definitely one of my bigger concerns. Even the new DMCH alternators are a bit anemic for the car, in my opinion. I've switched several components to reduce electrical load (LEDs all around including headlights and high beams, modern fans, etc.) but still see the electrical system taxed with the AC on at idle. Looking forward to throwing in Dave McKeen's Idle ECU with idle speed bump on AC signal for this reason. If I go down the electric water pump route further, I'll definitely be taking some very specific measurements around electrical load and capacity before committing. Perhaps there are direct fit higher-amp alternators around that I'm not aware of; if so, I'd like to install one regardless.


If you are doing it because it is an interesting project, go for it.

Nailed it. That's what this whole thing is about!

[Ron]

Would love to hear your thoughts...
It looks like you have several bad circuits to me, E.G., the radiator's inlet and outlet are virtually connected to the same place, the thermostat outlet(s). I guess you could add restrictor passages, but that would be a nightmare to figure out...
(Easy to see the flow if you redraw it where the pump is a battery and the rest of the components are light bulbs.)

The otterstat should be immediately downstream and close to the thermostat.
Actually, it's in a bad place already. Ideally, it should be in one of the heads (replacing it with a sensor, as previously suggested).

[dn010]

I'm sure you already know this but higher amp alternators won't fix how they wired up the car, the size of the wires they chose to use and corroded/dirty connections, etc.

[gongloo]

It looks like you have several bad circuits to me, E.G., the radiator's inlet and outlet are virtually connected to the same place, the thermostat outlet(s).

Agreed it's not clear from the diagram, but the intention is that a bypass thermostat is used, with flow going exclusively up to the radiator feed when hot, and flow going exclusively direct to the pump bypassing the radiator when cold. I'll see if I can update the diagram at some point to reflect this.


The otterstat should be immediately downstream and close to the thermostat.
Actually, it's in a bad place already. Ideally, it should be in one of the heads (replacing it with a sensor, as previously suggested).

That's an interesting thought. I'm not sure I understand why placing it in one of the heads would be better. Admittedly I'm no expert in cooling systems, so I have a lot to learn here. Can you elaborate, perhaps with example scenarios and expected behaviors?


I'm sure you already know this but higher amp alternators won't fix how they wired up the car, the size of the wires they chose to use and corroded/dirty connections, etc.

Absolutely. Separately on my list of "someday" projects, I'd love to completely gut the car's wiring and build completely new harnesses throughout. Seems like an awful lot of pain but it would give me peace of mind around longevity of the electrical system. I haven't made time yet to do the whole system, but for certain isolated systems I have gone through the trouble to rebuild the wiring for from scratch (some of the lighting circuits, for example). It's especially tedious to do while matching wire colors and plugs, but in my experience totally worth it and (perhaps masochistically) very satisfying.

[Dangermouse]

FWIW, my only experience with an electric water pump have been on the BMW N52 engine; I'm sure others are available on the market but I will offer the following experiences of that particular unit (other units may have different strengths and weaknesses YMMV, etc):

1. When it fails, there is no warning (no weep hole etc) unless you routinely run BMW software to check it's health, which brings me to
2. It is BUS operated only. No straight 12VDC. No idea if you can get a 12VDC On/Off automotive electric water pomp
3. The BMW one is variable speed according to engine revs, which keeps engine at most efficient cooling and supposedly is good for 0.0001 mpg gain ;)
4. When it fails, the BMW ecu is smart enough to put the engine in limp mode and protect it from overheating. But you are suddenly stuck at the side of the road. The Delorean's isn't and you will be.
5. It is surprisingly heavy, possibly heavier than a comparable mechanical one, as all the electronics are integral to it
6. Keep it at the lowest point in the circuit as the rotor and bearings are cooled and lubricated by the coolant fluid
7. It cost $400.

[David T]

The "O" switch (Otterstadt) is supposed to be in the coolant flow far enough away from the motor so radiated heat won't affect it. Once the thermostat opens and the hot coolant flows over the switch, it senses how hot the coolant is and if hot enough, turns on the fans. Most cars have the switch threaded into the inlet tank of the radiator. The wiring of the whole car could stand a LOT of improvement. From increasing the gauge of the wire to better plugs and connectors, improved fuseblock, maybe break it into two, one in front and one in back, a separate grounding system, a V+ bus and a V- bus, more relays, all kinds of things. There is just so much you can do to an old car that was designed and built long ago.

[gongloo]

The "O" switch (Otterstadt) is supposed to be in the coolant flow far enough away from the motor so radiated heat won't affect it. Once the thermostat opens and the hot coolant flows over the switch, it senses how hot the coolant is and if hot enough, turns on the fans. Most cars have the switch threaded into the inlet tank of the radiator.

I've been thinking about why the otterstat should be on the inlet side of the radiator rather than the outlet side (disregarding the change in otterstat trigger temperature that would need to be accounted for), and realized that likely it's better to actually have it on the inlet side in all cases; really what we want to control for here is a maximum temperature of coolant leaving the engine, without regard for the temperature of coolant entering the engine. In practice, given a correctly tuned system, I suspect it's likely to not make all that much difference anyway.

[Ron]

That's an interesting thought. I'm not sure I understand why placing it in one of the heads would be better. Admittedly I'm no expert in cooling systems, so I have a lot to learn here. Can you elaborate, perhaps with example scenarios and expected behaviors?
WRT coolant problems, the cylinder heads are much more prone to heat damage than any of the other parts. The best way to prevent damage is for the coolant system to react to high head temperatures as soon as possible. This is the main reason the automakers moved many of their sensors (fan and gauge/light) from the thermostat housing and coolant ports in their intake manifolds to the heads. (DeLoreans will do this if you simply rotate the pipe the otterstat resides in 180°!) Another is, if the thermostat sticks, the coolant will not flow over the sensor and the heads will fry before the gauge/light indicates a problem or the fans come on. Same result if there is air surrounding the thermostat. We want to know how hot the head is without relying on things that can fail and prevent the sensors from doing their job properly -- A single device, as close to the place where a problem is most likely to occur, is the best solution.

Briefly on the flow at the thermostat- When the engine is cold, you don't want a lot of the coolant going through the engine. The thermostat's job is to make the engine warm up quickly so that the oil (and O2 sensors on some vehicles) performs properly.

[Ron]

The "O" switch (Otterstadt) is supposed to be in the coolant flow far enough away from the motor so radiated heat won't affect it. ?? On a DeLorean, you couldn't get it much closer to the exhaust manifold, the hottest thing under the hood.


Most cars have the switch threaded into the inlet tank of the radiator.?? Most cars have no switches threaded into the radiator tank, although some have coolant level sensors.

:cheers:

[Ron]

I've been thinking about why the otterstat should be on the inlet side of the radiator rather than the outlet side (disregarding the change in otterstat trigger temperature that would need to be accounted for), and realized that likely it's better to actually have it on the inlet side in all cases; really what we want to control for here is a maximum temperature of coolant leaving the engine, without regard for the temperature of coolant entering the engine. In practice, given a correctly tuned system, I suspect it's likely to not make all that much difference anyway.YES!!!
...except it does make a difference...

[Bitsyncmaster]

I've been thinking about why the otterstat should be on the inlet side of the radiator rather than the outlet side (disregarding the change in otterstat trigger temperature that would need to be accounted for), and realized that likely it's better to actually have it on the inlet side in all cases; really what we want to control for here is a maximum temperature of coolant leaving the engine, without regard for the temperature of coolant entering the engine. In practice, given a correctly tuned system, I suspect it's likely to not make all that much difference anyway.

I thought about the same thing when I was designing my electronic otterstat. My thought was maybe you could reduce the fan run times a little but I never really saw that it would save anything.

[David T]

The "O" switch is only supposed to turn the fans on and off. It doesn't care how hot or cold the coolant is, it only controls the fans. The temperature sender for the gauge IS in the head where most of the heat originates from. The problem with that though is, if the coolant level gets too low and the sender is no longer immersed in coolant, it won't tell you how hot it really is! I never said the "O" switch is in the ideal spot in a Delorean, I said where it should be. I agree it is close to the exhaust but it doesn't really seem to have that bad an effect on it. A lot of cars, especially imports, have the fan switch in the radiator tank. On newer cars the computer keeps tabs on the performance of the water pump and the fans and if it senses performance outside of the accepted range it throws an error code and lights the check engine light. It will also modify the engine parameters to try to reduce the heat generated to "limp home" mode. Deloreans can't do that but the driver can. For instance, if you see the temps are going up you *can* turn on the heater to try to help cool the coolant.

[DMCMW Dave]

The "O" switch is only supposed to turn the fans on and off. It doesn't care how hot or cold the coolant is, it only controls the fans.

Huh?

[DMC-81]

Huh?

:hihi2: Well, the otterstat IS just a temperature controlled switch. It doesn't care about how hot or cold the the coolant is. :ehh:

60532

[David T]

To clarify, the "O" switch is just a switch. Once you pass it's setpoint to turn the fans on it cannot sense anything else and conversely once it shuts off it can't sense how cold the coolant is. It is not like the temperature gauge that can continue to sense the temperature. The "O" switch also only controls the fans. It has no effect on anything else. On cars with computers they can use an "O" switch to control other functions like timing if the motor gets too hot. Very simple, dumb system.

[DMC-81]

To clarify, the "O" switch is just a switch. Once you pass it's setpoint to turn the fans on it cannot sense anything else and conversely once it shuts off it can't sense how cold the coolant is. It is not like the temperature gauge that can continue to sense the temperature. The "O" switch also only controls the fans. It has no effect on anything else. On cars with computers they can use an "O" switch to control other functions like timing if the motor gets too hot. Very simple, dumb system.

Dumb, but effective. My car just stupidly drives around in these 100+ degree Florida summer days and somehow manages to effectively control the temperature of the engine with the simple on/off otterstat switch.