Rover V8 cooling changes--EWP?

[neillr]

Hi all,

I have a 2004 Land Rover Discovery with the 4.6L V8, and I've been doing some experimenting with the cooling system. I'd like some opinions on what to do next. Please note: "Buy a Toyota" or "put in an LS2" are not on the menu.

I live in the Western United States and tow a utility trailer fairly often on trips. Our weather can range from damp and cold to hot and dry, and altitudes can range from sea level to nosebleed. That places a strain on any cooling system, and is a real challenge in the Rover. My goal is to keep the engine between 180*F and 190*F (88C), as it seems to starve the lifters and cams of oil at higher temperatures.

The current engine is a rebuild with flanged liners and improved camshaft bearings. I've ditched the remote bypass thermostat, and I'm using an 82C (180F) inline thermostat. New viscous clutch and mechanical fan, new OEM radiator, new OEM water pump. I've added an oversized engine oil cooler (which was not provided from the factory in the 2004 model year).

Everything works as well as it CAN work. It maintains a pretty constant coolant and oil temperature as long as the temperature is 70F (22C) or less, and works even better if there is some humidity. I'm actually considering adding an oil thermostat to speed up the warm-up on cold days. But on a hot, dry day going up the mountain with a trailer, or idling in heavy traffic with the air conditioning on full blast, the cooling system doesn't keep up. The biggest problem seems to be the flow rate of the pump under 2000RPM--when I back off on the throttle, it doesn't cool down much because the flow rate goes down too. I need a long downhill at speed to get the temperature back under 93C, much less 88C, although with the added oil cooler it will eventually cool off on a level road with the A/C off.

However, I have a dream in which I'm blasting down the freeway at 80mph, up a hill, carrying 7 passengers with the A/C going full bore, while towing a trailer--serenely ignoring the engine temperature which never goes above 190F.

I'm considering three possible modifications to keep things under control in all conditions:

1) Aluminum 3-row radiator. This will cost me about $1000, but I'm not sure it would do anything for me at idle, since the OEM pump is so bad.
2) Custom high-flow water pump, to replace the OEM pump. This would be a custom Flowkooler impeller fitted to the OEM pump housing and shaft, which would create higher flow at low RPM but still not cavitate at higher RPM. This would cost upward of $1000.
3) A Davies-Craig EWP115 as a booster pump, in addition to the mechanical pump. I'm pretty sure this would solve overheating at idle, and I would be able to have it run-on after shutdown which should solve heat soak issues. But does it do anything for me in the 1500RPM to 2000RPM range? This would cost me about $300 all told, so it would be the most cost-effective option if it will solve my problems.

Any opinions?

[neillr]

Thanks for the good replies, everyone. All in all, it sounds like a bigger radiator may be the way to go. There's a couple of details I'm wondering about, though. Kia_diesel, the cooling system performance has remained constant through 3 Rover engines and two Discoveries, with various OEM components replaced. Currently all components are less than 18 months old and have less than 10K miles on them. I suspect Max_Torque will agree that this is a design flaw with the Discovery 2--my suspicion is that it was mostly for the purpose of pleasing the EPA and getting an import license. Only changing the thermostat type and adding the oil cooler has made things manageable.

GreenV8s: I've looked at the pump curve on the D-C units, and they seem to have plenty of flow at low pressure differential. If it is set up to run constantly once the engine is warmed up, would the D-C unit really be a negative to the system flow at higher RPM? All it has to do in that regime is not put suction on the mechanical pump intake.

Max_Torque's point about the waste heat at idle being small is a good one. If the heat rejection is sufficient to keep the block and head temperature under control at throttle, then there should be no problem at idle even if flow is reduced. My experience with adding the oil cooler confirms that--when there is some humidity in the air and there's low load on the A/C, the idle temperatures are stable now that the oil cooler is getting more heat out of the engine.

Still, higher flow should reject more heat, yes? I don't think I'm having trouble with pressure drop across the engine--the coolant temperature runs about 2C to 5C warmer than the head material when everything is warmed up, which makes me think that the heads are getting plenty of coolant. The worst temperature rise comes at about 2000 to 2600 RPM, at load, as Kia_diesel surmised. If I could get higher flow at the same system pressure drop or higher in that RPM range, that would increase heat rejection as I understand things.

I can confirm this somewhat--the engine cools off very nicely going downhill at about 2600RPM. There's very little throttle in that condition, and plenty of flow. Also, as a matter of duty cycle, it's a truck not a race car. The engine works hard in short bursts, and if it can cool off in between no matter what the ambient conditions, then that might be sufficient.

Still, on a hot day it's possible to watch the engine temperature go up about 2C at idle just from turning on the A/C, which increases the effective ambient temperature at the engine radiator. So a bigger radiator is likely the best bang for the buck.

[neillr]

Oh, one other thing. GreenV8s, the Discovery is an interesting beast. Some things are truly excellent on it--the radiator ducting is one of them. I don't think there is any significant leakage of air back to the front of the radiator. The viscous clutch fan is quite good, too, and it's nearly new, about 2000 miles on it. I've thought about putting in an electric fan from a Mustang or Lincoln, but I think I would need a bigger alternator for it, and I'm not convinced that there will be much benefit.

[neillr]

Radiator, mechanical fan, electric fan, and ducting are all factory spec.

The engine oil cooler is the same as the transmission oil cooler that is fitted to the Discovery from the factory. On the older Discovery 2, they were equipped with a 7-tube engine oil cooler, I believe. This was deleted in 2003 when they changed from 4.0L to 4.6L engines, but there are still mounting points for it in front of the radiator and the block still has plugs for the piping. I mounted an 11-tube transmission cooler in the factory location and piped it in to the existing ports. It helped a great deal.

Coolant is the basic green stuff, which is what Land Rover recommends (after having corrosion issues with orange coolant).

[neillr]

Ha! Yes, this is a known problem. The "solution" is to get a rebuilt engine with flanged cylinder liners to make the engine more tolerant to overheating. Which I have done. But they are still likely to have wear issues with cams, camshaft, and lifters.

The best improvements that I've found are the ones I've done--inline 180*F thermostat and the oil cooler. Yes, the oil cooler is not in the best location, but there's not really a great place for it, because of the extensive ducting behind the grill to ensure that the radiator receives maximum air flow. An additional fan would be difficult--there is already a pretty good-sized electric pusher fan, which comes on with the A/C (under 40 miles per hour) or when engine temperature hits 212. The only real improvement I could make to the fan would be to replace the belt fan with the Mustang GT electric fan I mentioned. That would not solve the issue by itself, but might do the trick in combination with other improvements.

Did you have any further comments on the D-C electric pump being restrictive? I had figured that adding one couldn't hurt, but your input was quite alarming. Did you find it restrictive even when it's running at full voltage?

There's easily enough room to add a third or even fourth row to the radiator. This one would probably work quite well, but would come to about $750 shipped:

http://www.allisport.com/product/discovery-2-v8-al...

[neillr]

I do not currently have gauges for oil temperature or pressure. I am able to monitor changes oil temperature, at least, by using an infrared thermometer to measure the outside temperature of the piping to the oil cooler. I have a pretty good idea of the delta-T across the pipe, so I can get pretty close that way.

But I never have to make that measurement when the oil is too hot except as confirmation--I can just listen to the lifters clattering at the rear cylinders! The Rover V8 has a known issue with starving the rear heads of oil when the oil gets thinned out, or at least that's my understanding. Hard experience and listening have taught me that if the oil temperature gets above 210F and stays there, it will soon go above 220F, and bad things start to happen if you do not give it a chance to cool off. The engine runs quietly and smoothly when the oil temperature is less than about 200F. I use 10W40 as per mfg spec. I have known people who use 10W50, but I don't think it really helps much when things get hot.

It is also a "feature" of the Rover V8 that you are likely to crack the block if coolant temperatures get into the 220F range. All in all, it's really best to keep the darn thing cool.

I have the remote bypass thermostat properly deleted. With a good 180F thermostat and the oil cooler, the coolant temperature runs between 187 and 194, as long as there's some humidity in the air and the temperature is 70*F or less. As I said, hot and dry conditions increase the coolant temperature required to reject the same amount of heat. I only know two ways to change that--increase surface area or increase coolant flow.

[neillr]

KiaDiseasel said:

Engines can live quite happily with high metal temperatures and coolant above 90c but what you must avoid is localised, and especially general, boiling.

I think you mean *modern* engines can live quite happily with high metal temperatures. Which the aluminum Rover V8 is not. It was not designed with high temperatures in mind.

[neillr]

Yes, oil pump was redone. It's only got 5000 miles on it. I've gotten this same behavior across three different engines, though. 100C won't do it, it's just that once the oil gets to that temperature something changes, and it continues to get hotter. The oil cooler has helped quite a bit, in that the oil will eventually cool off if I reduce the load--which it simply would not do with the factory remote bypass thermostat and no oil cooler.

How did you get your V8 to 125C without cracking the block? It's pretty well-understood in the Discovery crowd that coolant temperatures of 105C will crack the block.

[neillr]

If you check out a U.S.-centric Land Rover forum (we were't allowed to import the TD5 diesels), you will read an awful lot about blown head gaskets, cracked blocks, slipped liners, and clattering lifters on the 4.0L and 4.6L engines. I can't say for sure what is happening inside the coolant passages, but I've used a few different measurement techniques (including a borrowed infrared camera) to get the temperatures I've mentioned here.

There is a serpentine belt on the front of the engine, which drives the coolant pump/fan, A/C compressor, and steering pump. The ignition is through two sets of coils driven by the ECM. The coils have less than 5000 miles on them, and the computer is not detecting any misfiring, lean condition, etc. I tend to trust that, since the computer is a hypochondriac.

My reasoning for these modifications is that there's no reason NOT to run the engine as close to 180 as I can get, and plenty of reason to do so. I have personally had a liner slip on one engine after sustained running with coolant temperatures at about 100C. Coolant pressure was good, no leaks, no coolant loss--if there was boiling, it was not because of a pressure leak (at least until after the liner slipped). Also, my impression is that the idle gets rough when oil temperatures are above 200 or so. I didn't bring this up because I don't have data, but it has occurred to me to wonder whether the hydraulic lifters are working properly at those temperatures.

You probably have a good point about getting direct measurement of the oil temperature and pressure before installing more stuff. I'll have to look into one of those sandwich fittings that goes under the oil filter.

I agree, my preferred range of 180F to 190F is difficult to achieve with the available choices of thermostat. I have found that the 180F Stant Superstat opens rapidly enough to do a pretty good job. I would like to try a Robertshaw balanced 180F thermostat which opens even faster, but I haven't found a housing that will work yet.

[neillr]

Sorry, who or what is JE? Which type of block was this?

I'm also sadly familiar with vindictive flora.

[neillr]

Wow, very nice. Perhaps I should have had them do the short block for my rebuild.

[neillr]

Thanks for that info 100SRV. Especially the details on your bottom hose adapter and thermostat wiring. I have a similar three-fan setup on my Jeep, I think I'll use your wiring ideas.

I'm going to get some more data on oil temperatures, and think about everyone's input here. I may very well end up buying an Allisport radiator.

Edited by neillr on Wednesday 26th October 19:22

[neillr]

100SRV said:

Would you like a wiring diagram?

Sure, if it's no bother.

[neillr]

chammyman:

I disagree about the remote bypass thermostat. In theory it's an improvement, and it probably would have been an improvement if they'd put it in the top radiator hose so that the pellet always gets the full flow of hot coolant (like some Cadillacs used to do). In practice, I suspect that at low RPMs, the flow of hot coolant through the bypass and past the pellet is insufficient to keep the thermostat fully open. My engine temperatures are much more consistent with the inline thermostat.

My electric fan turns on with the A/C, at low speeds. I have verified that it does this and is functional.

The timing cover (and integrated oil pump cover) is new, as well as the gears. The top-hat liners came in a completely rebuilt short block, including new camshafts and bearings. All the other items you mention are either new or recently rebuilt.

I do not have any obstructions to the airflow through the radiator, and the shrouding is all in perfect condition.

[neillr]

GreenV8S said:

The Land Rover entry stats are designed to control the pump inlet temperature so they are normally rated slightly cooler than an old-style engine outlet stat. It's a very clever design and very effective in my experience. Personally, I much prefer that over an old-style top stat.

My experience is that it is very effective at warming the engine up to operating temperature and holding it there under light load and benign ambient conditions. But once you've worked it hard for a while, it is not very effective at cooling the engine back down. Of course, I changed out the thermostat before I added the oil cooler, which might make a difference.

[neillr]

GreenV8S said:

Plumbed in correctly, the entry stat should give an unrestricted bypass and closed rad when the engine is cold, and a closed bypass and unrestricted flow to the rad when the engine is hot. It's hard to see what more a stat could do to give you maximum possible cooling.

It should do that. I don't believe that it does. My understanding is that it depends on having some flow through the bypass to heat the thermostat pellet and keep the thermostat open. If the differential pressure across the bypass valve is small (for example, at low revs, when the pump is not working at capacity), perhaps there is not enough flow to keep the thermostat open.

Anyway, why would you want to have ANY bypass flow when you're trying to reduce the engine temperature?

[neillr]

GreenV8S said:

EDIT: Responding to the comment by MAX_Torque, the Land Rover PRT entry stat is pressure/flow sensitive as well as temperature sensitive - it is designed to allow the bypass to open slightly at high revs when the pump is trying to flow too much coolant for the radiator. Doing this maintains a higher flow through the engine for better protection against hot spots while maintaining enough flow through the rad for overall cooling.

Edited by GreenV8S on Saturday 29th October 16:20

That is exactly what I measured with an infrared thermometer. Under heavy load, the temperature difference between the thermostat outlet hose and the bottom radiator hose would actually *increase* over the temperature difference when unloaded. And then never quite cool off. In other words, when the engine is working, the thermostat was mixing more bypass flow into the pump inlet. That's the opposite of what would be ideal, isn't it? Under heavy load, you'd want to get nothing but coolant from the bottom radiator hose and no bypass coolant.

[neillr]

100SRV said:

I can email you a PDF if you wish?

No need, got it. Thanks!