Project Korean Cortina - WTF is going on with my cooling system?

[Bren]

11 hours ago, mat_the_cat said:

Let the measurements commence!

And the results were most strange. I double checked my arrangement of thermocouples, and am certain I've got top and bottom correct. But this is what I get (note, I took the thermostat out a short while ago, just to see if that had any effect. None, apart from a slower warm-up).

So the radiator *outlet* is actually hotter most of the time, apart from at idle (the two peaks near the beginning, and the one at the end). The smaller peak at around 2000 seconds is where I drove a bit more briskly, and didn't really notice anything on the gauge - to be expected, as that is taking its reading from the engine outlet, i.e. the top hose which doesn't rise as much as the bottom does.

From this to me it suggests that the coolant is actually flowing the opposite way round, i.e. into the bottom of the radiator, cooled a bit (in normal running), then back into the top of the engine! A quick bit of Googling, it it dawned on me (or so I thought at this stage...)

My belt layout is as the Series 3 diagram on the right, spinning the pump clockwise, but I think what I've done is use a pump as fitted to a Discovery serpentine belt model, which drives the pump from the back of the belt - hence designed to be spun anticlockwise. So it looked like I'd be experiencing reduced pump efficiency. When the engine is spinning fast, it just about manages to pump some coolant in the reverse direction through the rad. At idle, there's nothing, and it even looks as though the thermo-siphon effect is dominating slightly, as the top hose gets slightly hotter and the temperature rises swiftly.

 Then I had a look at the picture I'd taken of the currently fitted pump.

Remember this is spinning clockwise as viewed from the front of the engine, hence ACW from the impeller side. And opposite to its design rotation - or is it?! To the best of my knowledge, those impeller vanes are correct for spinning ACW as viewed, drawing coolant through the impeller from the hose stub at the top of the photo, flinging it centrifugally outwards and into the block. In which case why does mine appear to be working in reverse?

Think I'll sleep on it, and maybe try pinning the pump with a drill in both directions, with the hose disconnected so I can see what's happening.

Lastly, this was probably a bad time to see this comparison table...

Yep. The rover V8 is a boat anchor these days - previously you could justify having one as they were cheap secondhand. Not any more.....

[bunglebus]

11 hours ago, mat_the_cat said:

Think I'll sleep on it

A quick Google suggest the pumps you have would rotate ACW as we look at your pictures, so clockwise as you look at them fitted to the engine 

[mat_the_cat]

12 hours ago, mat_the_cat said:

I think what I've done is use a pump as fitted to a Discovery serpentine belt model, which drives the pump from the back of the belt - hence designed to be spun anticlockwise.

 

1 hour ago, bunglebus said:

A quick Google suggest the pumps you have would rotate ACW as we look at your pictures, so clockwise as you look at them fitted to the engine.

Yep, you're right - I'm a numpty. All I had to do was look at one of the two serpentine belt Rover V8s I have here, and the pump is clearly different. So that's a relief in a way that I haven't been completely stupid, but still doesn't explain my apparent reversed circulation.

[Talbot]

Are you sure you haven't just got rather more air cooling while driving forwards on the top hose thermocouple than you have on the bottom one, hence getting a false low reading on the top hose? Which would of course then not be an effect when the car is idling, and hence you get the top hose showing hotter.

Either that, or is the bottom hose thinner than the top hose, and hence you get better heat transfer through it?  Where are you attaching the thermocouples?

[mat_the_cat]

7 hours ago, Talbot said:

Are you sure you haven't just got rather more air cooling while driving forwards on the top hose thermocouple than you have on the bottom one, hence getting a false low reading on the top hose?

  It's possible - certainly there doesn't *look* like there'd be a visible difference but there may be something funky going on with the airflow which isn't obvious.

7 hours ago, Talbot said:

Either that, or is the bottom hose thinner than the top hose, and hence you get better heat transfer through it?  Where are you attaching the thermocouples?

I'm attaching it to the hose directly at the radiator ports. I was thinking of repeating the experiment with some lagging, as obviously the thermocouple is only measuring its own temperature. It looks like there is a significant cooling effect, as the overall temperatures are a lot lower than I'd expect, and certainly lower than the gauge/radiator fan switch would suggest. I may make up some very fine wire TCs, which I can thread into the coolant directly. That has to be the best option.

[Zelandeth]

Seems an obvious one...but you have checked both sides of the impeller casting of that pump are pointing the same direction haven't you?

With modern part quality control I'd not put it past some factory churning out pumps with 50/50 flipped impeller geometry...

[Talbot]

Have you got an IR thermometer gun?  It would be interesting to see what the TCs are reading, and then blat the area with an IR gun.  Whilst I wouldn't expect them to be identical, there shouldn't be more than a few degrees difference.  It would confirm (or not) if you are getting real coolant temperatures or not.

[mat_the_cat]

26 minutes ago, Zelandeth said:

Seems an obvious one...but you have checked both sides of the impeller casting of that pump are pointing the same direction haven't you?

With modern part quality control I'd not put it past some factory churning out pumps with 50/50 flipped impeller geometry...

The reverse side of the impeller has no vanes, if that's what you were wondering? It's been quick to rise with two different pumps now, so unfortunately I don't think I can blame a dodgy pump!

2 minutes ago, Talbot said:

Have you got an IR thermometer gun?

We must have inherited one, as I borrowed it previously. I've yet to find it though! It suggested coolant temperatures ~110°C though from memory, although the question was what emissivity value to enter for the surface I was pointing it at. The hose should approximate a black body source fairly well though, so close to 1  I'd have thought.

[mat_the_cat]

On 6/11/2021 at 12:00 PM, Talbot said:

Do you get air locks in it though, being a high-spot with no vent?

This has been on my mind since you mentioned it - could my problems be down to something as simple as an airlock? Even if not, it can't hurt to include a bleed screw at the high point, so I purchased a short length of joiner tube with a threaded boss in it.

But I still wanted to have the appearance of a factory fit, one-piece hose. So I notched the ends of the hoses to fit around the screw:

I also had to replace a hose, as it had been partly eaten by the alternator fan. There's clearance when the engine is running, but obviously things must shift around under load. A new hose was cut down to fit:

Using a hose clip to try and get as straight an edge as possible, seeing as I don't own a hose cutter.

I then shrunk all the hose 'clamps', and scuffed them up to remove the in your face writing.

The alternator fan I had previously trimmed down so it's just smaller than the body. This was hastily done before the trip to the FoD, using the slightly dodgy method of an angle grinder flap disc, and an idling engine lathe.

Apart from having a fun day meeting up with folk down there, the ~400 mile trip was an absolute joy. It's my first long trip in it for a while, and for once there was no traffic jams on the M6 at Birmingham! I particularly enjoyed the return journey, with fewer cars on the road, convoying with @AnthonyG (for a short while 😉). The fuel economy is so much better getting out of the Welsh hills - I normally average low 20s!

[mat_the_cat]

I am a real-life Gary King, and I don't know whether to be pleased or not.

https://youtu.be/dshmyllg2HE?t=55