Dribbling Hose

[mudguard]

My fisher-price chocolate-teapot part-plastic radiator gave up bothering to contain the coolant at Dunsfold. Thanks Marshall Big Mike for the moral support and sorry to anyone whose TGT session I delayed! I did manage to get home fine by driving gently, but this, along with the weather forecast, put to bed any thoughts of doing the Taffia run again this year.

Anyway, I've now swapped it for a new all-ally one and it seems to be working great - almost. Bit of an air of concerned anticipation doing my first K-bleed but fortunately I have a secondary filling point in the heater hose which makes it pretty easy to get the airlocks out. The car is now controlling its temperature better than I've ever seen - it's almost like its (mostly) glued at 82. Only thing is, when I drive enthusiastically, I get a bit of dribbling from the bottom rad hose followed by the inevitable airlock and the needle creeping up towards the danger zone.

Of course this didn't happen when I ran it up to temp on the drive and patiently checked everything over multiple fan cycles, oh no of course it was fine then. Nor did it happen cruising at 30/40, but only when I was several miles away from home where the nearest NSL exists. It went straight away at that point, about 1-2 drips per second, so I tightened up the jubilee clip as much as I dared, re-bled, and that seems to have mostly cured it. Now I reckon it would go all day at moderate levels.

But it's still weeping a bit when I give it the beans, now about 1 drip per 10 seconds, and it soon stops once idling. Still a big problem as even this is enough to cause an airlock and a roadside re-bleed. Scratching my head a bit as to how I could reproduce the "stress" situation back on the drive, I disconnected the fan, temporarily, to run it up nice and hot. Temperature climbed towards 100, coolant level in expansion bottle rose steadily, top and bottom hoses went rock-hard. I assume that the weeping is due mostly to the pressure in the bottom hose, no? Anyway, not a drop lost. But I had just repositioned the jubilee clip so that it was closer to the flange in the radiator spiggot; previously it closer to the lip of the hose. Maybe that's fixed it? I did of course then reconnect the fan and check that temperatures, expansion bottle level and hose pressures returned to normal. But haven't had chance to re-road-test the car yet with the repositioned jubilee clip.

I found a brief mention from nickh7 in the useless search facility of roughing up something with emory cloth (but what?) in this situation. Will only this stop my dribbling?

[garybee]

The weep from the bottom hose won't be causing your airlock.  The pressure is greater in the coolant system than where it's leaking to, so the flow will only go one way.  If your system is airlocking itself I would suggest a chemical test for combustion gasses in the coolant.

[Jonathan Kay]

Are you sure that it's coming from that junction and not dripping down from somewhere else? (Guess why I ask!)

What type of hose?

Is that a genuine Jubilee clip or something else? Some experts recommend a correctly-sized Mikalor clamp.

Do you lubricate the hose as it goes on the spigot and the clip on the hose?

Could you add a photo of the spigot, and are you sure that it's smooth all round?

Jonathan

 

[mudguard]

That's what I feared someone was going to say :(

There is no mayo in the oil or in the coolant, but I'm aware of how there doesn't have to be (fire ring problem) :(

[mudguard]

Thanks JK, I am quite sure it's coming from there - verified by touching the hose lip.

It's the same jubilee clip that was on the same hose (silicone) previously with no leaks from this point. With the old radiator the leak was from the crimping at the junction of the plastic top tank and the ally core, near the top hose, which I believe is a well-know failure point. I don't know whether it's a genuine clip. I did not lubricate but I did check that the spigot was very smooth - it is.

[CageyH]

So do you still have the problem?

If it runs up to temperature, I would be tempted to go for a drive (with spare coolant and tools to bleed if necessary) and see if everything is ok. Then monitor the coant level to make sure everything is ok.

 

I am not a fan of the Mikalor clips, as I could not get the system to seal as easily as using normal jubilee clips.

[ScottR400D]

I had a persistent, though very slow, leak from the bottom hose/rad connection. 

Tried a new clip, reseating the hose to no avail. 

Cut a 35/40mm length from an old cycle inner tube and fitted it over the rad inlet. Replaced the hose and clamp. 

No more leaks. 

 

[mudguard]

Thanks CageyH I did exactly that - and yes it's still leaking, but slightly less so since I re-seated the clip. I think I will try the inner tube trick next, thanks ScottR400D.

On the positive side, I found yet more air in the system (after previously convincing myself that I had properly bled it!). With this out, the car can now handle enthusiastic driving without the temp creeping up to 100. After an enthusiastic stint, lifting the bonnet on the idle reveals moderately-firm - not rock-hard - top and bottom hoses and expansion bottle level raised by 1cm over rest, and, very quickly, an activated fan. All as it should be I think? Expansion bottle level then gradually drops back to resting level (the lip halfway up the bottle).

So hopefully the early fire ring failure theory seems less likely? But a combustion gases tester kit is in the mail anyway.

[mudguard]

I did two tests with the "Dr Headgasket" blue fluid tester; once with the tank bung and once with the syringe.

I had to let the engine run for much longer than the instructions suggest, because until the thermostat opens there's no coolant flow anywhere near the expansion tank, and no bubbles through the tester. I was able to rev the engine a bit (tolerant neighbours!) and at high revs see some flow(?) bubbles(?) in the tank apparently coming from the expansion hose to the lower radiator hose. I sampled the tank gases until after this point.

The instructions say the coolant should be at 50 degC or less; well definitely so for the coolant in the expansion tank, but definitely not for the coolant in the circuit - the thermostat had to open!

Both tests negative.

The only thing I can think of now is take the tester on the road, exercise the car thoroughly, and test again after that. But I fear that the expansion tank coolant will be > 50degC at that point.

[nickh7]

A new Ali rad needs to have the inlet and outlet bits roughed up a little with emory cloth to help with the seal especially if you have silicone hoses 

[CageyH]

As well as using a piece of inner tube, you can also use silicone self amalgamating tape to help fix a leaky hose. Thos isnwhat I did recently when I had a persistent leak on my submarine to bottom radiator hose joint. A slight leak under pressure. Having cleaned and roughed up the pipe, it still leaked. A couple of layers of tape, and all is well.

[7 wonders of the world]

A light smear of vaseline in the inside if the hose and on the inside of the clip ensure the hose seats and the clip exerts uniform pressure without wrinkling the hose which leads to small leaks.

 

[mudguard]

A follow-up on this.

I was able to cure the slight leak using self-amalgamating rubber tape on the ally flanges and lubrication on the hose clip.

However, this didn't cure the tendency to climb towards overheating. I had a suggestion from elsewhere that, despite sounding like HGF-driven coolant pressurisation, the symptoms could just be due to a persistent airlock in the thermostat area causing the thermostat to open late. The suggestion was to get a new thermostat, and just before fitting it, as a test, cut the centre out of the old one and the run for a bit to guarantee no airlocks and hence prove or disprove the HGF theory.

I decided to adapt this suggestion, so instead of cutting out the whole centre of the thermostat, I drilled the infamous 2mm hole in the outer flange, orientated to be upright. Thank you Ian SM25T for the picture showing exactly where. I then refilled very slowly, and, lo and behold, the problem (so far) seems cured. Still a bit of high temp tendency with heater running, but I'll get that minor airlock out of the heater matrix when I have time - before winter!

 

[SM25T]

Good news. Fit a 16mm bleed tee in the highest heater hose ... don't cut out any surplus hose when you fit it .... then you can raise the tee to be higher than the heater matrix. Final fill from here with heater valve open ... ie knob pulled out in cockpit.

[SM25T]

 

The mentioned photo. This shows the housing in the installed position.

[mudguard]

Thanks Ian -I do already have the additional plastic bleed tee close to the heater bypass junction, so that should be easy :)

Worth mentioning that my thermostat sits at 45 degrees clockwise rotation relative to the above picture, with the ally raised bar running skyward to groundward (I checked carefully, and it would sit at no other position!) Hence, in my installation, the hole is drilled above where this bar joins the flange, so that the hole is in the same position as above.

[mudguard]

Had hoped there would be no further posts on this thread, but alas not. With the drilled stat the car behaved very well over June/July/August, I wouldn't say perfect temperature control but good enough. We took it to Wales and drove the chip run in each direction but including a Snowdonia extension northbound and a Hampshire extension southbound!

Then, a few weeks ago, the misbehaviour came back more severely. On every drive I had to stop, switch off the engine, let out a lot of pressurised air from the heater bypass junction, and then it would be OK-ish for the rest of the drive. This was even with gentle driving.

Today I'd had enough of it and I finally followed that piece of advice I had a while back - I cut out the guts of the thermostat which means the rad is now permanently in circuit. So the suggestion goes: if the overheating/overpressurisation remains, it has to be a gasket problem, otherwise it's something less worrying/expensive. 

Well, on a little test drive just now, try as I might, I struggled to even get it above 80deg; being stationary is the only way. On fast stretches, even with the beans, it was at 70, which I suppose is regarded as overcooling. Is this bad for the engine? I can't imagine it's worse than creeping up towards 100, as it was previously. I know about the cardboard-blocking-off-part-of-the-rad trick; so if it still has no stat by winter I'll have to try that.

Unless someone wiser and more experienced knows better, the plan now is to run it like this for at least a few weeks, and if the problem has vanished then get a new stat and see it comes back. I don't really know what I'm talking about, but it feels to me that the symptoms might be explainable by an intermittently failing thermostat, that when it fails to open causes quickly rising temperatures and then the boiling of some coolant, overpressurising the circuit. 

[Roadsport06]

Surely just following the advice from SM25T was what was needed, this would tick off an airlock in the thermostat housing from your list. You'll now be running the engine cold and mask any symptoms

[mudguard]

I thought I had followed all of SM25T's advice: hole drilled, heater fill point in place, final fill from there.

The point is that all of this seems to have helped but only temporarily. So it runs for a while as if it is airlock free, then the symptoms gradually return and get worse. At this point re-bleeding keeps them at bay for a few hours, and the re-bleeding process never ends. Surely at some point all of the air has to have been bled out! Reminder: the bled air does not smell of exhaust. There is no mayo in the coolant. The coolant tested negative for combustion products.

The slight difficulty I have with the "mask any symptoms" theory is that I have seen it "go wrong" - as in climb rapidly from a particular temperature - with stat, from about 78. Now, statless, it can get above 80 if slow/stationary. 

Nevertheless, the theory is easily tested. I just need to block off part of the radiator, make the engine run hotter, and see if the problem returns.

[mudguard]

... so I just did that. I blocked off 70% of the radiator and took it out. I expected to be on the high side of the temps with this, but that was the point of the test.

After warm-up the needle settled at 85 and hardly budged no matter what I did, until eventually when I hit a sustained NSL stretch it climbed gradually to 90+. I interpreted this as meaning I'd made the nosecone aperture too small so pulled over to make the hole bigger. The temp eventually dropped, with the fan running. I checked the hoses - they were all properly warm and not overpressurised, no air to bleed from the heater junction. So no recurrence of "the symptoms" in a scenario which, with the old stat in, would very likely have produced them. Not conclusive, but hopeful, I think.

Now, as it turns out, it has a minor leak, due to my hamfistery, which will need yet another drain and refill to correct. I've decided that at the same time I might as well fit a new stat (with 2mm hole added) and see what happens.

[Roadsport06]

The following SM25T advice was just stick with the 2mm hole drilled in the stat rather than taking it out, as you just want to tick off the airlock by thermostat. If you are getting air in the system and you have filled up from the bleed tee then either combustion gases getting in (test kits for this) or air is getting drawn into the system. I had a pin prick hole in a hose that was putting air into the system. I only spotted this when running the car in the garage, when up to temp and reving a tiny jet of coolant would appear and pull in air when cooling. Other was from the rad having a tiny leak that also drew in air when cooling so a new all aluminium rad fitted.

[Tazio]

I have a dribbling hose, but hopefully my Consultant Urologist will sort it.

[CageyH]

You mention you have a heater. This is another source of a leak, which could pull air into the system. It would be a good idea to check this as well.

[mudguard]

It was HGF, pressurising the coolant and filling the top hoses with air. Not at idle, but with heat and torque, and towards the end, not that much.

And a missing bleed ball valve - randomly affecting thermostat opening. 

And a failed Cat, making much more heat to (struggle to) deal with.

And disintegrating sump foam, not exactly helping either.

No wonder it was so tricky to control the temperature, with that little lot. 

But overall it's very good news. The head is still hard, the HGF surface damage was minor, the liners have not sunk, the oil pickup was not blocked, and there were no metal fragments in the oil. 

But it does explain how, towards the end of its former life, the only way it could control its temperature was with a cut-out thermostat. That removed the stat opening problem and ran the head just cool enough to seal the gasket. 

It's early days in its new incarnation, but there's no overheating now.

[Jonathan Kay]

Thanks for the answers.

And, as always, a reminder to others with foam in there.

Jonathan