Thread: regarding engine temp...

OK, so I'm about to put my motor back in, and I'll be using a Meziere electric water pump. My groovy Zeitronix wideband unit can control it based on sensor voltage. So I need to be able to accurately measure water temp, collect voltages off the sensor to program in and run. So I have a couple questions. First, what temps do I want to run at? I know that too cool is bad, and that these EWP's can be a little too efficient. So what is the right temp window to keep it at? At what temp should I have it come on and then at what temp should it go back off?

Also, I am planning on getting a set of STRi gauges, but I'm still unsure as to how precise the numeric reading can be off them. Where is a good, reliable place to measure temps off? Thermostat housing? Lower end of the rad w/fans off? You help is appreciated!

Mine ran at 180 while cruising. The stockers fans kick on around 212 but some of the later B13s had the fans kick on lower like 208 iirc.

K, good to know. So I guess I'd have the pump come on at 210 and then have it let off whenever the car gets below like 190?

What do you think about where to measure temps from? Rad w/fans off?

Do you know what the temp of the stock thermostat is?

Stock USDM SR20DE thermostat opens at 76.5°C and has that number stamped on the unit itself.
The engine is considered completely warmed up at 71-75°C and 70-80°C is the ideal operating temperature. The water pump should be on at all times. It is not used to control the water temperature at all. It is only used to circulate the water. Even when cold, the water pump should be running to prevent hot/cold spots in the engine during warmup.

The radiator fans are a different story. I like to turn them on anywhere from 82-86°C depending on preference. The engine shouldn't ever get near 100°C in an ideal situation.

Are you thinking of running without fans, and using the water pump to control temps instead?

No, I'm definitely using fans, but I'm changing my mind due to what you just said.... The problem, if you can call it that, is that these EWP's are too efficient, and they'll actually get the car back into cold/startup temps. There's a company called Davies Craig that makes an EWP controller that cycles the pump at 50% voltage for a couple of minutes at a time once it reaches the desired temp. When I think about it, and tell me if I'm wrong on this, the water doesn't truly circulate until the tstat opens, except for the small amount that can travel thru the pintle in the tstat. So in my mind, it's no big deal at startup, but it is a problem if the motor develops hotspots during hard running. so, it seems to me that the in-between answer is lower run speed to prevent overcooling, and I'll only run one fan for the same reason.

I'll explain some theory, and you'll be able to make some educated decisions based on that.

At cold startup you'd like water flowing through the engine to keep the engine a uniform temperature while it heats. The water only circulates through the engine, bypassing the radiator during warm-up. This is a good time to have water flowing, and too much water seems like it shouldn't change anything.

Then you reach operating temperature and the thermostat opens up allowing water to flow through the radiator to keep the water cool. If the water cools too much, the thermostat will close again until it warms up again, so this acts as a closed system to keep the temperature around 76.5°C. This system works very well with a stock water pump. I don't know how a bit more flow from an EWP would cause this system to lean too far towards the cold side of things, but it might be possible.

The radiator fans are there only for when you're idling at a stop light or similar. On a hot day with no air moving across the radiator, even sending water through it won't keep temps under control. The radiator needs air going across it to keep the water cool, so as the temps rise at a stop the fans will kick in to bring temps under control. So elimnating one or both fans won't help keep things warm on a regular basis, and will only remove your ability (or hinder if you leave one of two fans) to keep temps under control while at rest.


I do not have any experience with electric water pumps and they very well could make engines run too cold if operated too quickly, but I have no experience with that. As for the ideal control of the water pump... Something similar to stock would be nice, with a speed limiter to stop it from over spinning causing cavitation. Barring that, a single speed set at a happy medium could work, or two speeds (one for low rpm use, one for high) that adjusted based on rpm. The cycling of the pump at 50% duty cylce if temps are too low seems like a viable solution IF you do experience over-cooling.

Maybe I've had too much to drink, but this is my thinking:

When you have a higher capacity water pump that rotates much faster than stock pumps, you run the risk of water not being able to transfer enough heat, efficiently.

Someone correct me if I'm wrong

For instance, if you throw water down a metal slide, and you have a heat source like a propane torch underneath it, the faster the water moves, the less the heat actually gets transferred from the metal, into the water and dissipated.

Make sense? Or should I just go back to my beer?

I have a site on this, you run the pump all the time but not at full duty. i just can't find the site right now.

Tekkie, you're right, except in the case of the engine, the water keeps recirculating. It would be like the water in your slide getting to the bottom and then being carried up to the top. It wouldn't matter how fast the water went by the torch.

Keo, are you thinking of the Davies Craig piece? It has a 3 phase operation... they are as follows:

Phase A: The EWP pulses at 6V-- 10 sec.ON, 30 sec.OFF until coolant temperature is approx. 20 degC below the 'Set' temperature.

Phase B: The EWP pulses at 6V -- 10 sec.ON, 10 sec.OFF until coolant temperature is approx. 5 degC below the 'Set' temperature.

Phase C: The EWP is ON and supply voltage increases linearly with temperature from approx 5 degC below to 'Set' temperature. At 'Set' temperature the EWP is at full output.

So this means that it is going to keep water moving across the motor to some extent and increase its flow according to temps. The problem that I have in my mind is that since this pump mounts on the lower rad hose, before the coolant enters the rad, it seems that it will have a hard time circulating water due to the distance from the pump to the motor. Therefore, it seems to make sense to me that I should run without a t-stat in order to have continuous flow.

I know it's not totally in line with what you're saying Ben, but I trust these guys and their expertise with this stuff, and I think it's the most logical and comprehensive way of addressing it. What's your opinion on that?? I've heard for a long time that this is a good way to go for a track car, which I want my car to be next summer, but now that I'm doing it, I can't find any resources on putting it into action. Plus the blockoff plate looks so slick I can't possibly go back!