[mark e]

Well, it's like this... I've had no end of grief recently with my EWP installationon my 9000CS, not cooling properly, me blowing up the controller, getting too hot etc etc (EWP= Electric Water Pump)

I've also had some very good advice from Abbott about how to make things run cooler (I might do a brief write up in while...)

Meanwhile though, I've still got the EWP and no mechanical water pump- and I have to drive the car and the weather is due to be warm for the next few days. Tonight I was trying to track down the cause of the air that always seemed to be drawn in to the pump, and indeed which I reckon could be the cause of cavitation at higher rpms in the mechanical pump... and found it!!

With the EWP running at max, air is drawn in through the small bore pipe that runs from the radiator to the expansion bottle. If I clamp it tight, no air is drawn in and the pump therefore flows much more water.

It appears the vacuum presented by the pump does a better job of sucking down the pipe leading to it from the header tank than the radiator! I'm also quite coinvinced that if it happens with the EWP flat out, it will also happen with the mechanical pump at higher engine speeds, hence the cavitation reported by Abbott...

So, the question is, what exactly is the small bore pipe for? If I clamp it off, will I get any problems? I can see that it might be used for some sort of pressure equalisation in a standard system fitted with a stat but as I don't have a stat any more this shouldn't be a problem. I can also see that it is a route for any air that develops in the system to migrate out of the radiator.

 

[sgould]

I'll dip in here and guess. I don't know the 9000 but water and pumps and things are the day job (when I'm not watching people dig holes in roads

)

I would imagine that the small pipe is just a bleed pipe so that when filling the system through the header tank all the air comes out of the top of the radiator. If it's a problem (and it works as I think) you could clamp it. If you do, try opening clamp from time to time to release any air. If none after a few goes leave it clamped. Trial and error really.

One thought though. I'm assuming that the small pipe goes to the 'cold' side of the radiator?

If so, Sucking air means that you have a large pressure drop across the radiator (actually between the point at which the downpipe from the header meets the system and the point at which the small pipe meets the system). This could mean that the cores are becoming blocked. You've already removed the other restriction in the system, the thermostat. Since it is a closed system, the air must be returning to the header tank via the main downpipe, unless it's being trapped in the engine/radiator and the header tank level is rising.

If you know that the cores aren't blocked. Is the EWP rated too highly for the system? Do you know the temp drop you are getting across the rad? It presumably isn't enough if you've got overheating. It may improve if no air comes in to lower the level. You may need a bigger radiator

 

[mark e]

Good points, there, Will- thanks.

The small hose is the same side of the rad (Cold) as the inlet to the pump, so I reckon there might be a problem even if the rad is in perfect condition because there will inevitably be a pressure loss across the cores.

The air is indeed becoming trapped elsewhere in the system as the header tank level rises by about 30mm, and thus impeding the coolant flow.

It was cool on the way in to work this morning, so I left things as "normal". However later when it has warmed up, I will clamp the hose and see what happens...

The pump is meant to be thermostatically controlled (but isn't at the moment 'cos I blew the controller up- don't ask!

) and varies the flow according to temperature. Typically, an engine at idle will flow 300-400 gallons per hour. The EWP flat out flows about 1300 GPH, which is the sort of rate you would expect to see round about max rev range (the mechanical pump loses efficiency at higher revs, so it isn't a linear relationship)

 

[Mark B]

The thin pipe is to allow any air in the system to migrate out to the relative peace and quiet of the expansion tank. In most cars that I've looked at, once the air is gone, a small but steady stream of coolant circulates down the hose, into the expansion tank, and back via the tank's bottom hose. I am puzzled as to why this flow should have reversed, but then I'm not familiar with your new set up.

A short term fix would be to use some kind of one way or float valve in the small hose, but that would cure the symptoms rather than the basic problem.

 

[mark e]

Mark,

I think you've just made me realise the problem...

What I've done is ground the impeller off the mechanical water pump and put the EWP in the bottom radiator hose. (as recommended in the installation instructions

)

The pipe from the expansion tank feeds in to the pump housing, and I'm guessing that there was probably a slight vacuum on it as a result, which would indeed result in a small flow through the small hose in to the top of the expansion tank.

Having the EWP in the hose will of course now put the entire mechanical pump chamber under positive pressure, and thus also the feed hose from the expansion tank.

Oh p00ey sticks!

Righty ho then, I now know my course of action. I'll put the standard mech pump back in, use some other heat reducing measures (I'll write them up now) and use the EWP in the top hose as a booster.

And write a snotty note to someone about the EWP not doing the job as it should...

 

[Mark B]

Mark E,

I am now very curious as to what you've been up to! Can you please write a short description of the setup, and what it was intended to achieve, especially if you have no thermostat.

Thanks.

 

[sgould]

Don't be too hasty. I'll try and think a bit more. A few random thoughts while I'm here.

You have a closed system once the pressure cap is on. So any pressure put on the header tank will be transferred via the small hose back to the system.

You need to look at it as the "header tank, small hose" running in parallel with the radiator. You will get flow through both. If the small hose is submerged in the header tank you will draw water through. As long as it's not significant flow compared to the flow through the radiator, I would ignore it.

Your problem is the air being drawn into the pump through the small hose and possibly reducing the efficiency.

Can you relocate the small hose connection to the other side of the rad? Or plumb ithe top end into the downpipe of the header tank where it will be submerged all the time?

Before you put the pump in the top hose check that it can operate with the higher temperature. A pump on the hot side usually shortens the life. Well it does on central heating systems

 

[mark e]

I've posted a general write up on the installation here.

Meanwhile to answer the rest of the questions...

If I try to run it with the cap off the tank, it's worse and will overflow out of the top.

Whilst I think that relocation of the hose would work, I'm not sure it's viable owing to the difficulty of getting reliable connections under temperature and pressure. I'll have a thmub through some catalogues to see if there's anything suitable to tee it in.

Yes, their will be a reduce life expectancy from installing the pump in the top hose, but if, as I plan, I'm only going to run it when necessary, this may not be a problem.