Another puzzling cooling issue

Further to my recent questions about the cooling system, I need to ask for help with the ongoing problem whose investigation led me to ask those first questions.

Apologies for the long post but I'm trying to provide as much info as possible.

I recently picked up a 97 CB with 13k on it. It had been used for short journeys and though in excellent condition needed a bit of maintenance.

Among other things I've changed the coolant and flushed the cooling system. Soon after doing this, my water pump developed a leak, indicating that the seals were shot. Whilst it was in this state, I had to do a 170 mile journey (Oxford to Chester) but stopped to top the coolant up every few miles via the overflow tank.

Since replacing the water pump (and putting another new lot of coolant in) the system has been behaving strangely. Having put 120k on my previous CB I know how they should run, temperature-wise. It hasn't been overheating, and the needle stays just above C at constant high-ish speeds. However, as soon as I stop or slow down the gauge rises, and rises further and more quickly than I know it should. The fan comes on fine and it never goes above about 3/4...but stick with me, I know it's not right. Once I'm up to speed again (above about 40) the gauge moves back down quickly enough.

I've noted that every time I take the pressure cap off the coolant level has fallen. Generally it's only by a couple of inches and coolant is visible down in the thermostat housing. When I top it up to the neck a couple of bubbles always come up when I squeeze the hoses leading to the thermostat (the pair I was asking about in my previous post), and if I start the engine with the cap off a few more bubbles come up. Once these have, no more come up even if I rev the engine a bit.

This is what I have done so far:

–Fitted new pressure cap
–Checked thermostat – fine
–Snipped off each end of the overflow hose to provide 'new' rubber for the seal, and checked clips.
–Checked all other hose fittings

The new pressure cap seemed to help things a bit (maybe) but the problem is still there.

Today, when out on the bike and stopped in a layby with the needle about 1/3 up the temp. gauge, I decided to look in the overflow tank and rev the engine. Seemed to be bad news: there was swirling and little bubbles appearing in the coolant in there, albeit only at high revs. Although no smell of exhaust coming out at all. Head gasket failure? Well I've just done a compression test and both cylinders show 196psi which as far as I know is fine. There's no evidence whatsoever of oil in the coolant or vice versa.

I'm stumped. Basically air is getting into the cooling system somehow and I can't see where from. Should I replace the overflow hose? Could air be getting in at the thermostat housing o-ring? You're *meant* to replace this when opening the housing but I didn't. There's no evidence of coolant leakage there.

The closest I can get to an answer is that the air only appears to get in after the bike has got properly hot. Am I correct in thinking that as the cooling system does its job and the water temperature goes down from reaching a maximum, a vacuum develops in the cooling system to some extent, or is this only once the engine's switched off and the whole thing's cooling down? This is the only situation in which I can see air being sucked in somewhere. 

I'm really hoping I'm just overlooking something stupid. I'm considering swapping the radiator with the (quite new) one I have sitting on the old bike, in case the current one is rather blocked, but I can't see how that would allow this air to be getting in.

It was at this stage after replacing everything to do with the cooling system that I swapped the engine out. I still haven't pulled the original one apart to check the HG but it seemed the only thing left to be causing it was a blow between a cylinder and the water jacket. It did cure the problem btw, I'd bought the spare lump previously for silly money "just in case"

Thanks, well, FINGERS CROSSED I think I've sorted it and it's not the HG. Just been out for a ride with the new (old) rad and the temp. gauge is back how it should be - barely moving. Revved it whilst looking in the overflow tank - swirling but no bubbles.

I'm guessing the rad that was on there was full of shit (it didn't look great when I peered inside) and was trapping air bubbles. I'm guessing that as soon as there's some air in the system the coolant is able to boil, hence the bubbles I saw in the overflow tank?

Well, I hope someone might read this post one day and gain some use from it...

I think you're spot on re the rad clogged up with shite, I'm tackling my bro's old Daytona on Saturday with exactly the same issue and I'm sure it's because there's hardly any flow through the rad; going to give Holts rad flush a go!
I know my previous post reads like a horror story but I worked my way through every component in the cooling system (new rad, stat, fan/switch, water pump blah blah) and it still blew the coolant out so there was nothing left to try.

Well, although the new rad has improved cooling matters, the original problem is still there. Once cooled down, the coolant level is consistently about an each or so lower then the top of the filler neck, and it's still not running quite as cool as I'd like (although if I didn't know these bikes well I probably wouldn't think anything was wrong – shows how effective their cooling system is).

So am I now considering HG failure? Well I was, but then I did something I should have done much earlier: ran the engine up to operating temperature whilst stationary but with the fuel tank off so I could get a good look at the filler cap area. What do I see almost immediately but a slow leak around the filler neck to thermostat housing joint. I think the big hose to the radiator might be leaking a tiny bit as well, although I can't tell as the leak from the filler neck trickles down to that pipe's junction.

I feel a bit foolish for not noticing this earlier, although it's a very slow leak so once the bike's up to temperature there's hardly any evidence of it. 

So I'm off to Lings to order a full set of new hoses (being 17 years old I think the bike could do with this anyway) and o-rings. Will post back once I've put them all on and tested the system.

I also replaced the o ring at the filler neck  (as it was slow leaking), replaced almost all hoses, but still have the same issue...overheating from time to time,and spilling coolant....the only part not replaced yet is the radiator and one hose connecting to the cooling inlet of the engine...

Well for the time being I swapped the filler neck for the (non-leaking) one off my retired CB.

Is it normal for the coolant level to be an inch or so below the filler neck once cooled down? When I remove the cap there is a slight 'glug' and the bottom of the cap (the smaller seal etc.) is wet (dripping). Is it possible that there can be a bit of residual vacuum in the system which is released when I remove the cap, causing the level to fall a bit? If I top up the coolant at this point to the top of the filler neck it only takes about 10 or 20 ml. This seems to happen consistently – it doesn't drop further.

The cooling system is now, I think, acting normally – when warmed up and on the move the needle sits pretty much over the boundary between the white 'C' block and the rectangle on the scale (even in this hot weather), moves about a quarter or third of the way up if I go into a 30mph limit but goes down again after a couple of minutes of riding at 30, or more quickly when speed is picked up again, and now doesn't really go above halfway and activate the fan unless I'm sitting still for maybe five minutes. Does this all sound OK? I know it's clearly doing its job properly but I still want to know that it's all completely working as it should. I just don't know (and can't remember noticing either way on my old bike) whether it's normal for the level to be where it is when the pressure cap is removed when cold. My instinct/knowledge says it should be right up to the top if that's where it was filled to when cold, but I may not be right there.

This sounds exactly like my bike after fitting a new radiator. Maybe the coolant is going to the overflow bottle? I'm not too concerned as it is running great so far.

I keep checking it after just one ride but I think actually cooling systems like these need a few heat cycles to stabilise. My hypothetical explanation for my slight residual vacuum after cooling down is that there is still some air in the overflow hose which is acting like a spring for the vacuum process that returns coolant from the reservoir to the engine. Does that make any sense? 

I'll stop opening the rad cap, enjoy riding and maybe look at the level again in five hundred miles' time or so.

jchesshyre wrote: 
I'll stop opening the rad cap, enjoy riding and maybe look at the level again in five hundred miles' time or so.

That sounds like a good idea - from what you said about the temperature gauge earlier it sounds like it is fixed.

It might be wise to keep an eye on the temperature gauge too. Just don't look at it too often...

Problem still not solved.

I top the system up and go for a ride – everything fine.

Check the level under the pressure cap after a ride (when fully cooled down) and it's gone down by about 1.5 inches max. If I don't do this and go for another ride after it's cooled, it runs a bit hot indicating that there's air in the system.

If this were the head gasket I would imagine the level in the reservoir tank would go down as the system is designed to do this. Besides, my compression test showed 196 psi in both cylinders.

Somehow air is getting in during the cooling down phase and the coolant that is expelled to the reservoir tank is not being drawn back.

I've replaced the hose from the filler neck to the reservoir, including fitting small jubilee clips instead of the standard spring clips.

The pressure cap is new. Could it still somehow not be sealing properly?

I've replaced all large hoses in the system, and o-rings.

Any ideas as to how air's getting in? Crack in the reservoir tank? I haven't removed that to inspect it yet – does anyone know how to do this? The Haynes manual tells you to remove the 'tool box' and rear mudguard but I can't see how to do that anywhere.

I plan to once again run the engine from cold to fully hot with the fuel tank off and watching the whole thing like a hawk, and then watch it cool down (so much time!).

This is starting to really piss me off now... >:-(

It's a pain in the arse, isn't it? It takes aaages to cool down.

I guess you are right - there must be air getting in somwhere or else the coolant from the reservouir tank would be re-filling the system as it cools.

By gum, I think I may have found the problem.

I recently invested in a cooling system tester, which I had used but only when the system was hot.

Just pumped it up, when cold, to 1.1 bar and got a small leak from the hose connecting the two carbs. Took the hose off and...



It doesn't leak when the bike's hot, as far as I can see, but I'm guessing that when cooling down and a vacuum starts to form in the system it's enough to suck a little bit of air in. This would explain why each time I topped the system up and then started the engine a few bubbles rose to the surface (initially I thought that meant it was the HG but they didn't carry on forming, it was only a couple).

Off to buy some hose from town and will post back...

That's a very small hole, but it could well be the problem.

Is a cooling system tester just a bicycle pump suitably connected so you can pressurise the system? or is it more complicated than that?

Yes a cooling system tester is basically a bicycle pump with pressure gauge attached to a modified pressure cap.

*sigh*

I check the level this morning and the problem is still there.

It always falls to exactly the same level – to the bottom of the level of the tube leading to the thermostat. I don't know if this is a clue – well it is but I don't know what to make of it!

The other clue is that there are always a couple of bubbles in the system when I then top the level up and squeeze the hose leading to the thermostat. Although these could just be in the thermostat housing actually.

I feel like I've tried everything now. I don't know what to do! All I know is that the problem occurs during the cooling down phase, as when I go for a ride after topping it up the cooling system behaves perfectly however long for or however hard I ride. Then when the bike's cooled down, the level is low.

All I know is this: some of the coolant that overflows to the reservoir is not returning and is somehow being replaced with air. The amount is consistent each time.

Perhaps it's time to remove the reservoir and check that for cracks.

I've just run the bike up to full temp (four fan cycles or so) but with the overflow hose disconnected from the reservoir tank and instead immersed in a bottle of coolant, to eliminate the tank and the hose's fitting to it as a source of the leak.

I'm currently waiting for the bike to cool with this bottle in place. If the coolant returns as it should then I'll know it's a problem with the reservoir; if it doesn't then I'll be looking at the pressure cap or elsewhere.

Despite waves of despair, I'm generally keeping a cool head as I know that ultimately there is a rational explanation for the problem which will only be found through experimentation and not through anger! Bike maintenance is good for the soul ; )

jchesshyre wrote:I've just run the bike up to full temp (four fan cycles or so) but with the overflow hose disconnected from the reservoir tank and instead immersed in a bottle of coolant, to eliminate the tank and the hose's fitting to it as a source of the leak.

I'm currently waiting for the bike to cool with this bottle in place. If the coolant returns as it should then I'll know it's a problem with the reservoir; if it doesn't then I'll be looking at the pressure cap or elsewhere.

Despite waves of despair, I'm generally keeping a cool head as I know that ultimately there is a rational explanation for the problem which will only be found through experimentation and not through anger! Bike maintenance is good for the soul ; )

Having same problems as you. Head gasket is kaput is the diagnosis for mine. New parts will be £50, labout will be 5 hours!

I'm not too sure about that diagnosis with mine, as no bubbles emerged from the overflow hose once the bike was warmed up (other than those that emerged as the air in the hose was pushed out) and I had it running for a good 10 minutes once hot. Surely if it was the HG in my case then at least one bubble would have come out?

I have no smell of exhaust anywhere in the cooling system. Have you checked for this?

Also, I would think that if the HG was releasing gas into the system then rather than the level in the filler neck dropping, the level in the reservoir would drop as extra coolant would get sucked back in as the bike cooled down to replace the bubbles that had entered the system. Surely if the level is dropping in the system as the bike cools down that suggests air entering during cooling down rather than whilst running?

Yep, he said he could see the exhaust gasses in the coolant and initially it was low, even thought it was refilled a few weeks ago!

Ah, right that does sound more like the HG.

With my bike, the coolant is flowing to the reservoir but then not returning as the system cools, so the level in the reservoir is gradually rising and the level in the main system falling a little. This indicates a vacuum leak somewhere.

Hope you get yours sorted easily.

jchesshyre wrote:
Despite waves of despair, I'm generally keeping a cool head as I know that ultimately there is a rational explanation for the problem which will only be found through experimentation and not through anger! Bike maintenance is good for the soul ; )

Yes, I fear that 'a process of elimination' is now the only thing left... you just have to convince yourself that it is going to work and you *will* find the answer (eventually!).

Can I just check, eternally_troubled or others, would you say it's normal, in this warm weather, once the bike's properly warmed up (having been riding on national speed limit roads for an hour or so) for the gauge to rise to 1/3 or occasionally 1/2 way up the scale once back in town and sitting at various traffic lights, not getting above 30mph, etc.? Once I get back to 40 or above it very quickly moves back down to the bottom of the scale, just above the white C mark. In all this fuss I've sort of forgotten what the bike's normal summer behaviour is!

If that sounds OK, then I'd say it's fixed. I think my running hot issues have been down to a number of combined factors rather than one thing. Clogged rad, a few bubbles in the system, and possibly a tiny leak from the carb heat hoses. I think one of those hoses was actually blocked as well, as before I replaced them all and blew through the 'heater' units on the carbs, the hose from the LH carb to the water pump didn't seem to have any coolant in...that might explain why I appeared to have air in the system.

What I haven't yet worked out is why the level kept appearing low, with some residual vacuum, after cool down, but I'm going to ignore this for the moment and leave the pressure cap alone if I decide that the cooling system is operating normally. I *think* it is but if you are able to put my mind at rest by confirming that the behaviour I describe above is OK then that'd be wonderful!

jchesshyre wrote:Can I just check, eternally_troubled or others, would you say it's normal, in this warm weather, once the bike's properly warmed up (having been riding on national speed limit roads for an hour or so) for the gauge to rise to 1/3 or occasionally 1/2 way up the scale once back in town and sitting at various traffic lights, not getting above 30mph, etc.? Once I get back to 40 or above it very quickly moves back down to the bottom of the scale, just above the white C mark. In all this fuss I've sort of forgotten what the bike's normal summer behaviour is!

That's the way mine behaves and I haven't had any problems with the coolant system so it sounds "normal" to me

jchesshyre wrote: having been riding on national speed limit roads for an hour or so, [is it normal] for the gauge to rise to 1/3 or occasionally 1/2 way up the scale once back in town and sitting at various traffic lights, not getting above 30mph, etc.? Once I get back to 40 or above it very quickly moves back down to the bottom of the scale, just above the white C mark.

Yes, that sounds like what my bikes does and (touch wood) it is working fine in the cooling department.

I think your 'combined factors' and/or 'multiple problems' idea is probably the right answer: it was just a matter of finding all the problems and fixing them.

Completely normal.
Our CB500 is liquid cooled, but also air cooled for any matter.
On summer days mainly, if i stop for 2 min with engine working, the needle goes up very noticeable.