Coolant leak

[nitroman]

This design works great and is highly durable. I've had one which has lasted 25 or 30 years now. The "long-finger" type of extractor won't work when high-torque is called for.

Good news:
The new thermostat comes with a suitable gasket.

DSC02101 (Small).JPG

Bad news:
Another sheared screw! Arrgh!

DSC02102 (Small).JPG

Any advice? I've tried penetrating oil, heat, mole grips and the rest of the usual options. Next stop is to get a local workshop to weld a nut onto the remaining stud, perhaps?

I've lost track of how many screws and bolts have sheared since I got the car in April...

[ashm]

Hmm, that looks like it might work. I've ordered one, so we'll see how it goes.

I can see one of the bolts securing the thermostat housing, but I can't figure out how to get at the other one - I can't even see it, thanks to all the metal pipes just below the thermostat. I think I can feel it with a finger, just about. I'll try all the options before having to tear out all that plumbing...

Since I'm now in the market, does anyone know where I can get a couple of short M8 flanged hex head HT bolts? My usual place doesn't seem to have them on the shelf. I can get screws, but I'd rather use a bolt if possible, and don't want to buy a pack of 50.

Adam

[bobbrown]

As the bolts for the stat housing do not need to be high tensile as the load is small you could use these
http://www.a2a4.com/acatalog/M8_FLANGE_BOLTS_A_2.html
Stainless steel so they will not go rusty

Bob

[ashm]

Is a stainless bolt likely to get siezed in the alloy housing due to galvanic corrosion? I believe the theory says small SS bolts in an Al lump should be ok, while small Al bolts in a SS lump would corrode quickly. However, many things are ok in theory...

I was going to go 8.8 or 10.9 as an attempt to reduce the likelihood of heads shearing off in the future. This is also why I prefer to use bolts than screws... Now that I think about it, DIN 6921 flanged hex head bolts in the correct length are probably threaded to the head anyway. Still hard to find in 10.9, but funnily enough they're widely available in titanium! For about £6 a pop...

Incidentally, I've just read that 12.9 is the standard grade for socket head cap screws. Seems high, I guess because it's so easy to develop high torque on a socket screw?

Adam

[Nickydee]

Is a stainless bolt likely to get siezed in the alloy housing due to galvanic corrosion? I believe the theory says small SS bolts in an Al lump should be ok, while small Al bolts in a SS lump would corrode quickly. However, many things are ok in theory...

I was going to go 8.8 or 10.9 as an attempt to reduce the likelihood of heads shearing off in the future. This is also why I prefer to use bolts than screws...

Adam

I note you are in Woking . Have you tried

MARGNOR (FASTENERS) LTD
36 STRINGERS AVENUE
JACOB'S WELL
SURREY
GU4 7NW

Never failed me yet for any car related or for that matter house related project and in fact supply components to Kit car trailer and commercial body manufacturers/ garages etc. Always really helpful and happy to sell individual items. Only open M-F business hours but if you stuck PM me and I be happy pop in and get what ever you need as I live in the same viliage.

Regards

Nick

[bobbrown]

Is a stainless bolt likely to get siezed in the alloy housing due to galvanic corrosion? I believe the theory says small SS bolts in an Al lump should be ok, while small Al bolts in a SS lump would corrode quickly. However, many things are ok in theory...

Adam

As the head of the bolt will be in direct contact with the aluminium you should not get any galvanic corrosion, plus you need an electolyte so unless you intend to take the car for a dip in the sea.... I would have though SS bolts with some anti seize compound (Copaslip) would do the trick.

Bob

[ashm]

I note you are in Woking . Have you tried

MARGNOR (FASTENERS) LTD

Ah, thanks! That's really good to know, I hadn't discovered them yet. Just drove past yesterday afternoon on the way back from Dorking.

I often go to AHC Camberley (http://www.ahc-camberley.co.uk), they tend to be pretty good for random screws, bolts and things, as well as cheapish workshop tooling. They're only a small detour on my way to work in Bracknell as well.

Adam

[ashm]

As the head of the bolt will be in direct contact with the aluminium you should not get any galvanic corrosion, plus you need an electolyte so unless you intend to take the car for a dip in the sea.... I would have though SS bolts with some anti seize compound (Copaslip) would do the trick.Bob

Ok, I was worried that the coolant might contain something that'd make it electrolytic, I don't know what chemicals they put in antifreeze these days along with the glycol, and I dropped out of A-level chemistry anyway! I tend to be cautious about these things, it's good to know SS shouldn't be an issue.

Thanks!

Adam

[BFG in the BRG]

Paul P. has a header tank for sale - see Elan parts for sale - hope that helps

[ashm]

Well, the knurled-cam type stud extractor worked (thanks nitroman!) and I replaced the thermostat.
I also took the header tank off again, took a close look at it and gave it a rinse out. Couldn't see any splits in it at all.
Put everything back together, now I've had a couple of days driving and no more leaking! Which means I have very little idea what was wrong, but anyway all is well again apparently.
Thanks for the help!
Adam

[ashm]

Ok, further developments:

After replacing the thermostat the temp gauge sat at about 1/8 during normal running conditions, whereas before it had shown 1/4. This didn't bother me particularly.

This morning on the run into work I noticed the gauge start to rise quite quickly. Eyes glued to the dash I saw it go from 1/8 to 7/8 in about 20 seconds... Pulled over, popped the bonnet and had a look. Coolant didn't seem particularly hot. I got back in and started the car, the fan wasn't on but the temp gauge was now at the top. I switched off, sat there for a while to let the car cool a bit and started it again, gauge was still topping out.

Then I realised I had an elanscan cable and laptop in the car, plugged in and saw the temp was about 100C. I ran the car and watched as it rose to about 111, then heard the fan kick in. Temp dropped to 105 or so and the fan stopped.
I pulled away and got up to speed for a while, the temp dropped to the low 90s... the gauge dropped a little, but only to about 3/4. I finished my drive, and elanscan showed the temp varying between about 95 when moving and 108 when stopped in traffic.

I'm guessing my gauge, gauge sensor or their connections or earths are faulty. Can anyone confirm that the temps I'm getting from elanscan are fairly normal?

Thanks,

Adam

[John_W]

Sounds about right to me.
This ElanScan plot shows how the temperature should rise when sitting at idle for a long period. You can see the thermostat opening at 85 degC then the fans cutting in at 110 degC and going off at 102 degC over several cycles.

John

[dapinky]

Adam,

There are 3 temperature senders in the thermostat housing (and immediate environs) - one feeds the ECU, one feeds the gauge, and one operates the fans......


...from what you have done, you can rule out a problem with the CTS (the one which feeds the ECU) and the fans - Elanscan is looking good for those two.

That tends to suggest an issue with the sender for the gauge - likely to be a bad wiring connection, or possibly a small air-lock inside the housing.

I'd check the wiring first, then if it appears okay, loosen the sender to ensure that any trapped air bleeds out, then re-fit.

Unfortunately, I don't have the resistance values for the sender at various temperatures....... Geoff?

Edit:-

I've found a picture, and from the wiring diagram, the single wire to the sensor should be Yellow with a Black tracer (still no test resistance values, though )

[ashm]

Hmm... seems to be a dodgy connection. I wiggled the wires and it's behaving as before.
Still settles at 1/8 when driving though.

In other news, I've finally found a small leak in the expansion tank. Not at the back or underneath, but around where the cap screws on. Guess I'm in the market again!

Ad

[GeoffSmith]

Unfortunately, I don't have the resistance values for the sender at various temperatures....... Geoff?

No need to poke, dear boy for 'tis all on t'Wiki!