Webasto failed with water logged circuit board?

[wetfoot]

Now on my second webasto thermotop C., both have been fitted on the wall of the engine bay mounted on a plywood backplate The first one failed for a few reasons, burner, using red diesel etc. which I avoided in my second one,  but also because one of the pins in the power socket (X2), second one in, next to the diagnostic lead socket, had completely corroded off due to water getting inside the power socket.

 

Thinking it was due to condensation falling of the roof, when I fitted the new heater I put a curved stainless steel shroud completely over the heater so that any drips from the roof of the engine bay can't fall on the heater.

 

The second one has worked well for two years but has now failed. When I took it out not only was the same power socket wet inside with severe corrosion of the pins where it had been arcing but the circuit board was sitting in a pool of water and had shorted out and burnt.

 

I am at a complete loss as to how so much water could get inside the circuit board housing when it has a plastic cover with a rubber gasket and the only water available is water vapour in the air. Its a lot of water if it's due to condensation, especially as its sitting on top of a hot heat exchanger

 

Has anyone else had this problem?

 

Any ideas what I can do to prevent it happening a third time if I buy yet another heater?  

[Tracy D'arth]

There is huge condensation all around the heater, your shield will probably cause drips rather than catch them!

[Bee]

It certainly looks like condensation, I wonder if having it sitting on top of a hot heat exchanger is the problem?  If there is any moisture in the air inside it will warm up and when you turn the heater off it will condense on the cold plastic box. As the air cools and contracts it will pull in a bit more damp air if there is the slightest gap and so on. It does seem like a a lot of condensation though. It might even be better to take the top off and ventilate the whole assembly but Its a bit of a  mystery as that would mean any electrical stuff in boxes would suffer too.

[wetfoot]

30 minutes ago, Tracy D'arth said:

There is huge condensation all around the heater, your shield will probably cause drips rather than catch them!

Where is this huge amount of water?

The water on the floor is just a small amount that came out of the flow and return pipes when I took them off prior to removing the heater. It was all dry around the heater before I started work.

 

I don't agree that the shield will cause more drips as it is mounted on wood so it is pretty much at the same temperature as the air all the time. But even if drips could form underneath it, the rate of curvature and the size of the thing is designed so that any drips on top, or underneath, will run down the curve and  fall to the floor 5 or 6 cm away from the side of the heater, not on top of it

[system 4-50]

I'd say something below the heater is creating high humidity which is condensing on anything cooler (not necessarily cold) above.  When the system is running, pass a cold dry metal surface around the area and see if you can detect an invisible source.

[wetfoot]

16 minutes ago, Bee said:

It certainly looks like condensation, I wonder if having it sitting on top of a hot heat exchanger is the problem?  If there is any moisture in the air inside it will warm up and when you turn the heater off it will condense on the cold plastic box. As the air cools and contracts it will pull in a bit more damp air if there is the slightest gap and so on. It does seem like a a lot of condensation though. It might even be better to take the top off and ventilate the whole assembly but Its a bit of a  mystery as that would mean any electrical stuff in boxes would suffer too.

That was my feeling too. I have no choice about it being fitted on top of the heat exchanger, that's the way it's made. Partly this is so that one component on the board, a thermistor, can be in contact with the heat exchanger to measure the water temperature.

 

I can't see what two O rings in the power socket didn't stop water getting in. They're more or less the same plug and socket as used on the brake pad wear indicators on my car and they get sprayed with salty water every day in winter without leaking.

 

I would have thought that the rubber gaskets on the cover would have worked too, since these heaters are fitted under the rear wheel arch of Swedish land rovers!

 

Be interesting to see if anyone else has had this issue. If I eventually fork out for another one I think I'll cover the circuit board with silicone grease or even encapsulate it in silicone rubber.

3 minutes ago, system 4-50 said:

 When the system is running,... 

Er... good idea but it looks like that might not be until I've forked out another £800 for a replacement unit!

[blackrose]

I don't suppose it could have been something as simple as a cable running from above that contacted a wet surface and then fed water down to the top of the heater even when it wasn't running?

[wetfoot]

Its possible but unlikely. I deliberately led all cables above the shield so that the shield was the last line of defence so to speak. (in the photo there is a cable under it but that's just because after I removed it I screwed it back up again with one screw just to take the photo. ) Normally its just the heater underneath  so any drips caused by cables would run down the outside of the shield and drip harmlessly to the floor.

 

Bee's suggestion seems the most plausible from a physics point of view (I'm a physicist and an engineer) but it relies upon the O rings and rubber gaskets not working and doing their job in a system that is designed to operate in harsh environments. It supposed to have a 3 yr warranty but I don't suppose that's worth the email it came in.

[GUMPY]

I bet you have binned the old unit

If you still have it maybe you could have swapped the control unit over

My wife asks why I never throw  stuff like that away when it breaks.

[Mad Harold]

Many years ago I owned several Triumph twins.Late fifties and early sixties models were fitted with a distributor with a long rubber "sock"over it running up to under the tank,to make it waterproof.It also made the distributor almost airtight.

In heavy rain the engine would misfire and frequently stop,and taking the distributor cap off there would be condensation inside.

The fix was to insert a sliver of wood between the bottom of the "sock"and the distributor bottom at the back,thereby letting a bit of air into the distributor and still keeping rain water out.

Just a suggestion for your problem that might be worth looking into.

[wetfoot]

@Mad Harold 

Actually, that's a really good idea. There is no real need for the plastic cover and gasket over the electronics in the location it is used. I could just put a loose bit of something, even just a square of Goretex or something like that over it, just to keep out excessive dust and  stuff and leave it open to the environment.

 

The wiring plugs I could do away with as well and solder some wires directly to the board, attaching them with 'chocolate block type' cable connectors.

Little (unnecessary) physical protection but zero places for condensation to collect either. 

 

@Loddon - No I keep everything, never know when it will be useful so I still have the old broken unit with missing pins on the circuit board. My garage is stuffed full of everything that ever broke in my life. My Dad was the same, He used to say "It'll come in handy even if you never use it")

 

If I don't get a freebee unit on warranty then, to quote a 70's series - We can rebuild him!

 

Thanks all.

(but if anyone has had the same issues do let me know.)

[Tracy D'arth]

3 hours ago, wetfoot said:

Where is this huge amount of water?

The water on the floor is just a small amount that came out of the flow and return pipes when I took them off prior to removing the heater. It was all dry around the heater before I started work.

 

I don't agree that the shield will cause more drips as it is mounted on wood so it is pretty much at the same temperature as the air all the time. But even if drips could form underneath it, the rate of curvature and the size of the thing is designed so that any drips on top, or underneath, will run down the curve and  fall to the floor 5 or 6 cm away from the side of the heater, not on top of it

On the contrary, there is lots of uninsulated steel around the heater, all of which will cause condensation in the area. So when the heater fires up and the heat comes on, there will be water vapour around.

 

Gaskets on the electronics chambers will not maintain water integrity, as heating and cooling will cause the air pockets to breath, breathing in the water vapour.

[blackrose]

30 minutes ago, Tracy D'arth said:

On the contrary, there is lots of uninsulated steel around the heater, all of which will cause condensation in the area. So when the heater fires up and the heat comes on, there will be water vapour around.

 

 

But that's also the case in the vast majority of canal boat installations. There must be tens of thousands of the same heater installed in similar uninsulated steel engine spaces. So why aren't we seeing the same issue reported every couple of months on the forum over winter? 

Edited December 5, 2021 by blackrose

[wetfoot]

Ok, I've found the problem.

 

It wasn't condensation at all, Two pin holes had developed in the heat exchanger, both under the circuit board. 

One went through into the water jacket and the other went through into the water intake pipe. I've attached a pic with a fine wire pushed through the holes.

It was corrosion from inside out but I've always tried to maintain 25% Mono Ethylene Glycol antifreeze in the system, as recommended by Webasto.

 

However further research shows that  that Mono Ethylene Glycol solution is corrosive to aluminium when it gets hot and acid formation occurs.

It also causes pitting of aluminium when in the presence of dissolved oxygen and / or chlorine as halide ions are formed. Pitting is what I had, right through into the space where the circuit board is. Aluminium has a self healing protective film on it that is stable when the coolant is at a pH of between 4.0 and 8.5 so the acids could cause it to drop below a pH of 4.0 where the film won't self repair.

 

It also seems important to reduce oxygen and chlorine getting dissolved in the coolant water. If you use a closed system the oxygen gradually gets absorbed and not replaced so that solves one problem but Webasto recommend a twin pipe, open, header tank with the water flowing through the tank continuously (which is what I have). Whilst this might be great at getting rid of large air bubbles, the continuously circulating water surface open to the atmosphere is absorbing oxygen all the time!

 

It also seems likely that I had a lot of chlorine in the water I filled up with as I filled the system from a standpipe in London, where the tap water is heavily chlorinated. 

 

So it looks like a combination of using London water and an open, flow through  header tank is about as bad as you can have!

 

I've now got a new heat exchanger and circuit board but I think when I re-fill the system I might do it using de-ionised water instead of London tap water, put 25% glycol in it and replace the open twin pipe header tank with one of the closed, pressurised ones (I've found one for £65 which is less than the £125 some places are charging for a plastic bottle but still a lot!).

 

I've put all this down in case others who have webasto heaters in their boat want to take steps to minimise corrosion as well.   

 

Edited December 8, 2021 by wetfoot

[GUMPY]

10 minutes ago, wetfoot said:

I've now got a new heat exchanger and circuit board but I think when I re-fill the system I might do it using de-ionised water instead of London tap water, put 25% glycol in it and try to invent some way of stopping or reducing the header tank from absorbing oxygen. Maybe some sort of floating cover inside it.

I always use de-ionised water instead of tap water in both my heating and the engine. As for the heating header tank mine has a sealed  cap and a 4" space above the coolant to allow for expansion 

[Tracy D'arth]

I am amazed that Webasto specify a circulation through the header tank. That is against all principles of preventing oxygen corrosion in heating systems and I have never seen is as a deliberate installation requirement.

 

With an open header tank with only a bottom expansion connection the absorption of oxygen is minimal and I have not found it necessary to resort to a sealed system.

 

I would worry about the state of your radiators too.

[wetfoot]

Here is the relevant page from the installation instructions for a Thermotop C

page re header tank from installation instructions.pdf

[system 4-50]

My webasto circulated through the header tank only for getting the air bubbles out when loading the water for the first time, as per the manufacturer's preferred instructions.  After that a valve was opened which enabled the circulating fluid to bypass the header tank but still allowing air bubbles to escape through that tank.  Unpressurized.

Some plastic piping has an extra layer to prevent oxygen being absorbed into an existing coolant system?

[Chewbacka]

13 hours ago, system 4-50 said:

My webasto circulated through the header tank only for getting the air bubbles out when loading the water for the first time, as per the manufacturer's preferred instructions.  After that a valve was opened which enabled the circulating fluid to bypass the header tank but still allowing air bubbles to escape through that tank.  Unpressurized.

Some plastic piping has an extra layer to prevent oxygen being absorbed into an existing coolant system?

Agree, ex 3 is the best method and the one I used.

Plastic Barrier pipe was developed to minimise oxygen diffusion through the pipe walls and dissolving into the coolant as it will cause corrosion.  So if you haven’t used barrier pipe I would replace it.  

[Tracy D'arth]

13 hours ago, wetfoot said:

Here is the relevant page from the installation instructions for a Thermotop C

page re header tank from installation instructions.pdf 1.6 MB · 6 downloads

Well after 40+ years running a heating engineering company amongst others this is the first time I have ever read such a daft instruction. Never have pump over in an expansion tank, it will continually introduce oxygen into the water and cause accelerated corrosion of any aluminium or steel and dezincification of brass.

Sorry Webasto, its very bad advice.

[Bee]

Thanks for posting that. Would never have guessed that chemistry could have done that or that the machine could be so fussy.

[wetfoot]

21 minutes ago, Chewbacka said:

Agree, ex 3 is the best method and the one I used.

Plastic Barrier pipe was developed to minimise oxygen diffusion through the pipe walls and dissolving into the coolant as it will cause corrosion.  So if you haven’t used barrier pipe I would replace it.  

No its OK, the system is a combination of copper pipe and plastic barrier pipe with the odd bit of rubber pressure hose for connections

[Idle Days]

1 hour ago, Tracy D'arth said:

Well after 40+ years running a heating engineering company amongst others this is the first time I have ever read such a daft instruction. Never have pump over in an expansion tank, it will continually introduce oxygen into the water and cause accelerated corrosion of any aluminium or steel and dezincification of brass.

Sorry Webasto, its very bad advice.

 

I agree. I've just had my system altered by the addition of a short-circuit valve as per option 3 in order to eliminate pump over.  It did seem a very odd installation.  

[wetfoot]

That's how I'm going to modify mine as well when I go back to reinstall the webasto. I'm changing the open twin pipe header for a closed twin pipe one with a PRV cap that opens at 1.2 bar. Haven't had it delivered yet but I'm not entirely sure that there will be room between the two pipes to fit in two tees and a byass valve so that when the valve is open the flow can go in a reasonable straight line. I'd rather not have a complex run with lots of elbows etc. just to fit in the valve as that will just cause more turbulence. Still, I'll see when it gets here.

 

The 'proper' twin pipe closed headers are really expensive aren't they for what is essentially just a thick plastic bottle? Cheapest I found from MellorOnline at £70 by the time you've added VAT, delivery and VAT on the delivery. Most were over £100 ex vat. For the engine header tank I got a small one from a car breakers for a few quid but its hard to find twin pipe ones in car breakers and even if you can, the inlet pipes are very small diameter

[jetzi]

I think it is also generally inadvisable to mix different metals like copper and aluminium components in one system, because of the potential for electrolytic corrosion, so if possible it might be a good idea to replace those copper bits with plastic. Not that I followed that advice with my install, I have an eberspacher D4 hydronic, I think the eber heating chamber is aluminium, but it feeds a coil of my copper calorifier. Not to mention steel radiators.