Not your average bleeder...

[JBlake]

I've moved into a house and the bathroom radiator is only getting warm at the bottom. Straightforward bleed job... or so i thought!

Bled it ok, and the rad got hot - top and all. However, within about 3 days it was back to it's all old tricks. Tried bleeding it again and sure enough, the problem continues to persist.

Can anyone hazard a guess as to how air keeps finding it's way back into the radiator... and perhaps more importantly how to fix it.

Thanks in advance.

 

[meldrew's_mate]

Problem is due to poor system design, sucking in air, or corrosion in the system releasing flammable hydrogen/oxygen mix which is collecting in this rad.

If you like pyrotechnics try this:-
Take a small plastic freezer bag or similar.
When bleeding this rad collect the gas in the plastic bag.
Seal the bag and tie it to a garden cane.
Using a long match tied to an even longer cane, set fire to the bag.
If it explodes it's corrosion products, wipe the molten plastic from your face and flush the system....drain, refill, X400, drain, refill, drain, refill, X100, etc..
If it just flops then it's air, look for signs of pumping over / sucking down (look, I didn't make this up, that's just what they're called, OK)

 

[JBlake]

Pyrotechnics seems rather apt, given the time of year!

I'll don the overalls and goggles and have a go with that tomorrow.

If news of the HA3 postcode area being blown off the map reaches you tomorrow, we'll at least know it was corrosion.

After a bit of logical thought...
Assuming it is corrosion, is there not a possibility that this gas could be bubbling up through the pipework and into my loft via the expansion tank? Thus rendering my loft a fire hazard.

Don't mean to sound too dramatic, but it would be very like me to march confidently up the ladder, blow-lamp in hand...

 

[meldrew's_mate]

Tharrr'll be a mighty large lump of corrosion, or a teansie weansie loft for you to worry yer 'ead 'bout thar.

And don't even think about setting fire to the escaping gas from the rad like a small blowtorch! It could burn back and ignite the remainder in the (enclosed space) rad = airbomb repeater + loss of limb/life + new rad required + wall to hang it on etc..

 

[PaulAH]

Now I may be playing with fire here - literally - but do you have to go to all that trouble? Wouldn't simply holding a small plastic beaker upside down above the bleed valve and putting a match to it (away from the rad) show hydrogen without blowing the place to smithereens?

Explosions occur when the expanding products of a chemical reaction - eg combustion of hydrogen and oxygen - are constrained in a tight space. A thimbleful of petrol will burn in an unspectacular way if you set fire to it, but confine a few drops vapourised in a cylinder with oxygen, squeeze it wih a piston, show it a spark, and the explosion is enough to propel a car.

In school chemistry lessons (before health and safety regs decreed you have to dress up like Neil Armstong) we used to test for hydrogen by putting a lighted taper at the mouth of a test tube containing the gas. At worst it would only give off a little pop or squeak. A small plastic cup would offer even less resistance to the expanding gases.

Of course you have to hold the container upside down to stop the hydrogen escaping because it's lighter than air.

I must confess I've actually tried this method on a rad that has been persistently emptying itself without any signs of a leak. Not so much as a snap, crackle or pop. Nothingsville. So either the method doesn't work or something else is causing the water level to drop. Can't for the life of me think what.

Has anyone else tested for hydrogen this way? How would you normally do it?

Regards
Paul

 

[chrishutt]

And don't even think about setting fire to the escaping gas from the rad like a small blowtorch! It could burn back and ignite the remainder in the (enclosed space) rad = air bomb repeater + loss of limb/life + new rad required + wall to hang it on etc..

Much as I respect the good advice of MM, I think in this case he's making it up.

For the hydrogen in the radiator to explode, it would have to be present with a large volume of air (- ideally sufficient to give an Oxygen atom for each two Hydrogen atoms to produce H2O) which is unlikely to occur. Even if it did, the hydrogen would "float" on top of the air, rather than mix with it, so it's hard to see how a flame could possibly pass through the air-vent.

Anyone able to verify this from practical experience?

 

[JBlake]

Cheers for all your comments guys.

Whilst I guess it's unlikely the radiator would explode, I have grown rather attached to my wall and my arms for that matter, so i'll play it safe and take a sample out into the garden.

Picking up on what chris put about hydrogen being lighter than air - the lightest of all the elements in fact - if I trap some gas in something of negligible weight like a freezer bag, it should float much like a helium baloon.

For those that are interested, I'll post the results when i get round to doing it.

Thanks again

 

[ChrisR]

I think these are true, based on anecdotal experience. Comments welcome:

- If it fills the rad really quickly it's less likely to be corrosion than air getting in somewhere. The boiler makes air come out of soution so it collects in the first available rad - often the bathroom.
- If it is only happening in one rad, it's unlikely to be corrosion which would happen more or less , everywhere.


So first, get in the loft and have someone turn the heating, the HW, and then both, on and off and see if any water comes out of the feed tank's vent. (Or, unlikely, if the level of water in the feed tank comes up a lot for a short time or while the pump goes on (ie so air could be taken in via the vent)).

What's the boiler, and how old? Can you see the sequence of things connected, eg does it go boiler, vent, feed, pump, motorised valve then to rads?

 

[Softus]

Problem is due to poor system design, sucking in air, or corrosion in the system releasing flammable hydrogen/oxygen mix which is collecting in this rad.

Although the Hydrogen experiments are entertaining, hasn't this gone off topic a bit? MM was going in the right direction straight away.

If a pumped system is sufficiently sludged then it can pump over, drawing in aerated water which will tend to collect in the nearest radiator.

If the vent is connected at the wrong point in a pumped system then air can be drawn down the vent pipe with the same symptoms as above.

If the guilty radiator is corroding, then it will do so with both the valves shut off. If you do this and find that 'gas' still accumulates [in that rad], then it points to the rad; if it doesn't then it's the system, and you should then find the air accumulating in a different rad.

There may be other permutations that I've missed out, but hopefully you'll get the general idea from this kind of analysis.

 

[htgeng]

I think these are true, based on anecdotal experience. Comments welcome:
?

you don't even need to collect the gas to test if it is hydrogen. crack open the bleed valve and hold a cigarette lighter flame to it. You will see a neat blue flame if it is hydrogen

 

[meldrew's_mate]

Much as I respect the good advice of MM, I think in this case he's making it up.

For the hydrogen in the radiator to explode, it would have to be present with a large volume of air (- ideally sufficient to give an Oxygen atom for each two Hydrogen atoms to produce H2O) which is unlikely to occur. Even if it did, the hydrogen would "float" on top of the air, rather than mix with it, so it's hard to see how a flame could possibly pass through the air-vent.

Anyone able to verify this from practical experience?

You are right Chris, I haven't actually blown up a rad, but have set fire to the jet of hydrogen eminating from a bleed nipple. It burns with an invisible flame, so it's easy to burn fingers! My assumption that it has the correct amount of oxygen present for complete combustion inside the rad is based on the gases being liberated due to corrosion, thus external air is not required, the mixture is stoichiometric.

My fears arose when I thought through what may happen if the jet velocity was lower than the flame propogation rate, and remember brownian motion will cause the gases to mix, not seperate. So whilst the dramatic wording may have been a little excessive, I would rather this than encourage someone to risk injury, even though I brought up the method in the first place.

 

[meldrew's_mate]

you don't even need to collect the gas to test if it is hydrogen. crack open the bleed valve and hold a cigarette lighter flame to it. You will see a neat blue flame if it is hydrogen

That's what I'm warning against, entertaining yes, but understand the risks and you may be pursuaded not to do it like that again. The only saving grace may be the cool mass of the rad metal will cool the flame as it tries to burn back and extinguish it, like the old Davie Lamp principle, and explosion proof electrical junction boxes (with wide flanges).

Softus is right, lets stick to the topic, unless someone wants to start a new thread?

 

[chrishutt]

My assumption that it has the correct amount of oxygen present for complete combustion inside the rad is based on the gases being liberated due to corrosion, thus external air is not required, the mixture is stoichiometric.

I take your point about Brownian Motion, MM, but don't see how the above can be true.

The basic corrosion reaction if between water and iron, producing iron oxide (FeO) and hydrogen (not sure of formulae for equation). Any oxygen present (dissolved air) would also react with the iron to form iron hydroxides (e.g. Fe(OH)2). So how would free oxygen be available to mix with the hydrogen in anything approaching explosive proportions?

 

[meldrew's_mate]

(not sure of formulae for equation)

And neither am I so I'll paraphrase my old teacher who would say "let's learn together". Help from those who know would be appreciated (and would save ime too, though I think that, as this corrosion is due to electrolytic action between different metals oxygen and hydrogen are given up in the correct ratio for burning.

It would be nice to know if my fears are unfounded, as I could re-introduce my party piece.

 

[PaulAH]

Help from those who know would be appreciated

http://www.diynot.com/forums/viewtopic.php?t=5644