Navitron Renewable Energy and Sustainability Forum

SOLAR THERMAL => Solar Thermal Systems => Topic started by: wookey on July 17, 2011, 10:08:20 PM



Title: Corrosion in plate heat exchanger
Post by: wookey on July 17, 2011, 10:08:20 PM
So, my solar system sprang a leak last week (or so, that's when I noticed), and had nearly filled a 2l ice cream tub in a week of dripping, so clearly (as I'm about to go away for three weeks), it needed to be fixed.

My setup (for those who haven't been reading here for years) is not the usual bottom-coil but a 'side-arm' plate heat exchanger fitted to a standard 114l coper cylinder between the cold feed and the vent pipe. So the PHE is hydraulically part of the cyclinder (mounted level with the bottom of the tank). When the solar heats the PHE, that sets up a thermosiphon in the PHE loop which puts hot water in at the top of the cylinder. The thermosiphon piping is plastic (JG speedfit), so the PHE is electrically isolated from the cyclinder (other than by the water).

So, on draining the cylinder and the solar loop and taking it to bits I found that the PHE threaded mounts on the DHW side are quite badly corroded, whilst the ones on the solar side remain shiny and new. The PHE has been fitted since approx May 2008 - i.e a little over 3 years. It's a standard combi PHE (intended for a vaillant boiler).

So, the question is - why has it gone all rusty? PHEs are normally used for CH water on one side and DHW on the other. DHW is agressive as it's full of oxygen and not full of inhibitor, but as this is the normal use case I don't understand why mine's rotting at a fair rate of knots. It's not the PHE body, which is stainless, but the threaded mount. Is it possible that the mounts on one side are made of different stuff to the other, because the manufacturers know that only one side needs to be agressive-water proof?  The rotted pair is 3/4"BSP, the OK ones are 1/2"BSP. Which size is normally the DHW? Pity they aren't all the same size - I could have just turned it round and got another 3 years out of it :-)

This is the PHE: http://www.ekmpowershop22.com/ekmps/shops/euroboilerparts/he110-vaillant-turbomax-06-4946-064946-289-p.asp

Below is what it looked like once unbolted, and after I cleaned it up - you can see it's got quite pitted. I had to take a good mm off to get a surface the O-ring would seal against. Both those spigots looked the same when it was originally bought and assembled - some kind of coating on the steel, presumably zinc?

Clues welcome. I can keep buying new PHEs every 3 years, but this shouldn't be happening, so I want to understand what's going on first. I also took the opportunity to service my solar loop by replacing the fluid. It was fairly brown (I'm not very surprised as I've had quite a few stagnation days, mostly due to 1-wire unreliability and my dodgy software (maybe 10 days in total?). Pic at the bottom. I need to get some pH paper to test it properly.


Title: Re: Corrosion in plate heat exchanger
Post by: desperate on July 17, 2011, 10:20:58 PM
Hallo Wookey,

That's a fine spot of rust you have there, I have a feeling that a clue may be in the fact that normally there is mains water in that exchanger which hasn't sat around in a cold cistern exposed to the atmosphere giving it the chance to absorb oxygen. I wonder if there is a test you could do that would measure the difference in the oxygen content of fresh mains water and water drawn from your hot tap?

I have done a similar thing with my neighbours hot cylinder as my heat dump, but I made it out of 4 parallel lengths of 10mm inside a length of 28mm, it seems OK so far..........fingers crossed.

Desperate


Title: Re: Corrosion in plate heat exchanger
Post by: EccentricAnomaly on July 18, 2011, 09:56:00 AM
I wonder if there is a test you could do that would measure the difference in the oxygen content of fresh mains water and water drawn from your hot tap?

Put a goldfish in it? Or, perhaps:

How to Measure the Amount of Oxygen in Water (http://www.ehow.com/how_5183453_measure-amount-oxygen-water.html)

The relative complexity of that does indicate there's not a really simple method. Maybe people who keep fish have a method.

Wookey, you mention that the PHE is electrically isolated from the cylinder (apart from through the water) but don't take the thought any further. Might it not be better to connect the two to short out any galvanic potential or would the resulting current just make things worse? That's something I'm a bit confused about.


Title: Re: Corrosion in plate heat exchanger
Post by: Brandon on July 18, 2011, 10:30:43 AM
wookey, in a combi I would expect to find the DHW in the 1/2" tappings, and the system in the 3/4".

Had never thought about it before though


Title: Re: Corrosion in plate heat exchanger
Post by: djh on July 18, 2011, 11:10:14 AM
Just speculation here ...

What material screws onto those terminals? Is it setting up a galvanic couple?

Presumably your solar piping is copper and is earthed somewhere and presumably your cylinder is earthed. If you measure the voltage between your cylinder and the PHE terminal (when installed), is it really zero?

I believe it is possible to fit anodes to potable water systems. That might be worth exploring if no other solution appears.


Title: Re: Corrosion in plate heat exchanger
Post by: dhaslam on July 18, 2011, 11:34:41 AM
Although it doesn't look like it  part of the  build up could be limescale with  a little rust adding the colour.   

Connecting the DHW side to the 3/4 " connection seems logical since in the original use  the heat exchanger the other circuit is pumped  and very short.


Title: Re: Corrosion in plate heat exchanger
Post by: dimengineer on July 18, 2011, 01:47:12 PM
Isn't it the chlorides in the DHW doing the corrosion? You will have warm/hot water, with chlorides in there. Very corrosive.
Its a thought.

TIm


Title: Re: Corrosion in plate heat exchanger
Post by: baker on July 18, 2011, 09:58:38 PM
Hi
it looks like electrolyse from common brass bsp fittings to stainless steel  corrosion
this can be sorted by using a dzr  type brass fittings designed for your application
can be got from / BES
john


Title: Re: Corrosion in plate heat exchanger
Post by: Philip R on July 18, 2011, 10:52:52 PM
High chlorides would destoy the copper cylinder, the water would run verdigris stains everywhere.

Heat exhanger possibly contains bosses off different grades of stainless, ferritic v austenitic.

Have you got earth currents running through your hot water system. May need to check earth bonding around the cylinder.

Philip R


Title: Re: Corrosion in plate heat exchanger
Post by: wookey on July 19, 2011, 12:20:57 AM
Some useful thoughts here. The solar piping is not earthed explicitly, and indeed wrapped entirely in insulation, so only earthing is through expansion vessel attached to wall and panels attached to roof - both pretty shoddy earths.

The connections to the PHE at the corroded spigots are JG speedfit straight tap connectors (7088 on
http://www.bes.co.uk/products/154.asp ). (I had to cut down the flanges slightly to make them fit). So are those using DZR brass or not? As they are intended to connect to taps, I'd expect them to be DZR. So that's a plastic bodied fitting with brass, just for the nuts. On the non-corroded side it's 15mm angled tap fittings. Those are chrome-coated brass.

Yes there could well be some limescale in those deposits, as the PHE is often hot enough to scale up. However the inside of the PHE is very clean in fact, with no signs of limescale.

I'm not convinced by the 'water that's been via a roof tank is much more oxidised' theory. If that was really a problem then everyone who lived near a water tower would have terribly corroded combis wouldn't they?

I'll see if any voltages can be measured.


Title: Re: Corrosion in plate heat exchanger
Post by: tony. on July 19, 2011, 08:15:28 AM
Could it be bacterial growth?


Wookey,

Whats your current email as i have solat thermal installers guide which may be of use.
12MB email ok for you?

Tony


Title: Re: Corrosion in plate heat exchanger
Post by: JohnH on July 19, 2011, 12:49:40 PM
Do you have any soldered joints anywhere?  The stuff around the inside of the PHE flange looks like rosin flux.
Flux can be corrosive and the flange itself could be acting as a cold spot.
JohnH


Title: Re: Corrosion in plate heat exchanger
Post by: ecogen on July 19, 2011, 01:22:27 PM
Two forms of corrosion at work here. Firstly Crevice corrosion which is a type of pitting corrosion, the cause of which is often overlooked as nothing appears to be visible that could cause such an attack. In this case it is likely to have started around the rubber O ring where the fitting face and boss face come together, helped along by crud producing a stagnant crevice. The stainless steel becomes less noble due to low oxygen weakening the passive film. Chlorine in the tap water is only a few ppm at most, even in warm water this is not a problem. However, permeation through crud of such corrosive material into the stagnant crevice can lead to concentrations where pitting will occur.
Cleaning the boss face with a steel wire brush prior to assembly is certain to initiate pitting. Lower cost lower chrome nickel steels are more susceptible.
Secondly, a dissimilar metal nut on the outside of the joint in a dry environment will only start corrosion pitting once electrolyte seeps past the O ring. The three conditions needed for galvanic corrosion now present.
Partial solution is to remove all pitting from the boss face and reassemble using a new O ring and paste both mating parts, threads too with an anaerobic sealant eliminating all crevices. I find Loctite 572 enables many a good bodge. Screwfix 10. Corrosion will continue inside the pitted boss bore, not helped by low flow rates, periods of no flow, crud and biofilm.


Title: Re: Corrosion in plate heat exchanger
Post by: wookey on July 20, 2011, 02:02:20 AM
That sound like you know what you are talking about :-) I didn't clean the spigots with a wire brush. The only surface dmaage could have come from the extra bit of plastic flange on the fitting that I had to chop off. But I can't imagine speedfit plastic doing much harm to steel.

Also we seem to have corrosion on both spigots. The pic is the top one - I couldn't get the bottom one off at all so just left it as it wasn't leaking. Why would both have the necessary scratches to induce corrosion. And I can imagine a bit of crud collecting at bottom fitting, but top as well?

I did wonder about something to re-seal the surface, but couldn't think what would last. I'll try the loctite - cheers.


Title: Re: Corrosion in plate heat exchanger
Post by: biff on July 20, 2011, 09:06:21 AM
hi wookey,
          i used to fit a salamder to the mains inlet in the house,it was supposed to polarise the iron ore contents in the water which meant that the crud passed straight through the system and would not stick to anything.
    that gunge in the 3/4 boss looks suspiciously like ore,horrible stuff, the salamder might be a good bet,it will not get rid of it but it will keep it in suspension, the febrox in the c/h does a similar job for the c/h but in a different way.
                                                                                                               biff


Title: Re: Corrosion in plate heat exchanger
Post by: ecogen on July 20, 2011, 01:03:14 PM
The reason I mentioned the wire brush was not because of surface damage but because it forms anodic iron on the surface of the stainless steel and will initiate immediate surface corrosion. This is why stainless steel wire brushes must be used.
Surface roughness and corrosion resistance are related, but of minor relevance in this case.
Place a magnet on the boss. If the attraction is strong similar to that of steel then its not annealed austenitic stainless. Compare the magnetic property of the boss to the . Does the boss show any difference in any magnetic attraction to that of the plate material? This is not definitive, as cold working changes the magnetic properties. Assuming the plates to be 316, a lower grade of stainless will be less noble / more anodic in contact with a higher grade and more prone to corrosion. This may be comparing apples to oranges, but I have put a rare earth magnet to a Bowman brazed plate 316 H/Ex. Both plates and Boss non magnetic.
If you were to show me the before and after corrosion photos, with the S/S H/Ex blanked out. Top of my material list would be cast iron.


Title: Re: Corrosion in plate heat exchanger
Post by: TimSmall on August 20, 2011, 08:44:28 PM
FYI, I suspect this pic is of the same (or a very similar) model of exchanger - it's never been used, and was sitting in my (fairly dry) basement for a couple of years, til I noticed that it looked like this.  I suspect from the look of your one Wookey that I'll now not bother doing much with it!

It was originally destined for a project almost identical to Wookey's, but the relative which it was bought for keeps umming and ahhing about the idea - fortunately by the look of it.  Time to find a different make/model I think.


Title: Re: Corrosion in plate heat exchanger
Post by: wookey on August 20, 2011, 11:36:01 PM
That is indeed extremely similar , if not identical (I'd have to check the number). It does strongly suggest that  only the 1/2" side should be used for DHW.  I could do that.

Alternatively does anyone know of reasonably-priced PHEs with nice threaded bosses like this? I actually found it quite hard to find such a thing. Most combi PHE do not have threaded boses. Most PHE that do have threaded bosses are quite expensive, or excessively large for the max 2kW solar trasnfer.


Title: Re: Corrosion in plate heat exchanger
Post by: climber on August 21, 2011, 09:49:36 AM
I'm still searching for a reasonably priced PHE with threaded connections, but am considering this one:

http://www.ebay.co.uk/itm/Stainless-Steel-Heat-Exchanger-10-Plates-ca-22-kW-/270793678539?pt=UK_Home_Garden_Hearing_Cooling_Air&hash=item3f0c8f9ecb

I think the photo is generic for all sizes they sell. This particular one is 10 plates, 29.4mm high - How does this compare dimensionally to the one you have been using Wookey?

What I am unsure of, is the best method/type of fitting to use to connect to the threaded connectors........


Title: Re: Corrosion in plate heat exchanger
Post by: TimSmall on August 21, 2011, 11:02:31 AM
That exchanger looks reasonable, I think.

I think the best sort of connection to go for is the style used by washing machine hoses - i.e. a brass fitting with female parallel BSP thread, which compresses an EPDM washer between a flat "shoulder" in the fitting and the heat exchanger boss when tightened.  I think (but I'm not entirely sure) that BES part 17613 is what I have in mind.  I think screwfix may also stock them as "female couplers", but they don't really give enough detail.

Maybe best to contact the ebay seller and ask if they would supply fittings suitable to connect their exchanger to 15mm copper tube (I'm assuming that 15mm copper is used in germany, I may be wrong).

Tim.


Title: Re: Corrosion in plate heat exchanger
Post by: al_uk on August 21, 2011, 11:03:04 AM
That PHE looks very similar to the one I've got. I ended up using flexible hoses from screwfix to connect, as I couldnt find anything else that would fit. The problem with the hoses is that they are not full bore, and don't look great. I'd be interested to know if you find any better fittings in either copper or plastic.

I've had mine fitted for a few months now. I suppose I'd better check it for corrosion.

My thread is here

http://www.navitron.org.uk/forum/index.php/topic,13123.0.html


Title: Re: Corrosion in plate heat exchanger
Post by: climber on August 21, 2011, 11:48:30 AM
That PHE looks very similar to the one I've got. I ended up using flexible hoses from screwfix to connect, as I couldnt find anything else that would fit. The problem with the hoses is that they are not full bore, and don't look great. I'd be interested to know if you find any better fittings in either copper or plastic.


I would be concerned using these flexible connectors at high temperatures i.e. solar side >100 degrees C - not sure how long they would last before the hose failed.

The search continues - have the solar panel sitting in my garage, don't want to start the installation until I have the PHE problem solved!




Title: Re: Corrosion in plate heat exchanger
Post by: wookey on August 21, 2011, 05:14:27 PM
Tim, I don't think there's a shoulder inside 17613. I reckon you you want 17643 (4/5/6/7).
But I found that the awkward part is that tap conncetors expect a particular ID and the PHE has a smaller ID than normal for the 3/4" thread so 3/4" BSP tap connectors don't fit without cutting off the bit that goes inside the pipe (so that the washer can actually be compressed. (I realise a picture would help here, but hopefully you can work out what I mean).


Title: Re: Corrosion in plate heat exchanger
Post by: Solal on August 21, 2011, 05:40:03 PM
Are the plates corroded  or just  the connection?


The earlier  ebay link phe is from Wiltec. I've used a couple of their heat exchangers.

You can purchase  big 160kw alfa laval units  here.....
http://www.eco-equipped.com/alfa-laval-brazed-plate-heat-exchanger.html
Includes the insulation  casing  and reasonably priced also.



Title: Re: Corrosion in plate heat exchanger
Post by: TimSmall on August 21, 2011, 07:22:02 PM
Tim, I don't think there's a shoulder inside 17613. I reckon you you want 17643 (4/5/6/7).

Sounds plausible.  I'll see if I can find the one that I think I have here to illustrate - originally from Screwfix I think.

Tim.


Title: Re: Corrosion in plate heat exchanger
Post by: wookey on June 03, 2017, 07:27:47 PM
So, 6 years later I come back to this issue. My solar setup seemed to be being rather inefficient this spring, with not as much hot water as I was expecting. I noticed that the pump was working in the morning, but was cutting out when the system got properly hot. So I put in a new pump - it's done 9 years. But the tank still wasn't getting hot.  At this point I realised that the thermosiphon wasn't circulating, and also that I had a leak from the upper PHE spigot.

So, looks like it lasted almost another 6 years before springing a leak. And has probably been slowly bunging up for quite some time, (we had noticed degraded performance for a year or so at least).

I now can't get either speedfit connection off the PHE - the nuts seems quite solidly corroded on. But the blockage turns out not to be in the PHE, but in the short horizontal bit of pipe, and is made of what looks like blue calcium carbonate, so that's hard water deposits, but with coppery bits in?

I'm probably going to have to get a new PHE as I can't fix the leak without getting the leaking spigot off and that may well destroy the PHE, and then I'll probably  find there isn't enough material left to get a seal against anyway.

I was hoping to have got round to fitting a whole new tank and setup by now, but what with one thing and another don't seem to have got round to it.


Title: Re: Corrosion in plate heat exchanger
Post by: martin W on June 10, 2017, 09:35:12 PM
Perhaps not relevant as we are in a relatively soft water here in wales, but I have used several of the following make of plate heat exchangers in industrial settings with no issues. They are usually the same boss size for both sides

http://www.ebay.co.uk/itm/Stainless-steel-PLATE-HEAT-EXCHANGER-NORDIC-TEC-25-65kW-Ba-12-INSULATION-BOX-/252813982017?var=&hash=item3adce33941:m:m2HP_EZMaDw4nkEJhoSan-Q

I usually buy them direct from the manufacturer (LEDNET) in Poland, but ebay might be your best bet.

As they are used in and industrial environment, they have a LOT of water going through them... Hot side is around 70 Dec C, with cold mains on the exchanger side. around 7000 Litre per day of fresh water going thought them. The hot side is permanently pumped 24/7, with a flow of around 4-8000 litre per hour IIRC.

I've also used with chilled water (5 Deg C) to mains water.

They are connected with brass fittings and copper pipe. I tend to used CXFI (compression female connectors) with fibre washers and fernox water hawk, or similar jointing compound.

I took a replaced two heat exchangers last year for much bigger ones they had seen around 2 years of industrial usage, and no signs of any sort of corrosion or anything.


I've noticed over the years' that any leaks on the hot water side does product a nice build of crust around the leak quite quickly. I've always assumed that's it due to the large volume of 'fresh' hot water we get travelling around the system every day (about 30,000 litres).


Hope this might be of help


Title: Re: Corrosion in plate heat exchanger
Post by: MR GUS on June 11, 2017, 01:38:23 PM
Wookie, I know folk laugh at my calcium carbonate slurry magnet (low voltage set up) I bought years ago ...but it works.

We have same (ish) water as you being in close proximity, did you fit one on the incoming supply ever?

When you have resolved your current issue I'd recommend one still as a preventative bandaid.

Mine has always worked a treat, & whilst I could not tell if there had ever been a sacrificial anode in my he tank (looks plausible) there is negligible calcification in my pipes nor hot water tank, the crust it slowly removed over a two year period & ejected was considerable,  & my shower head only requires a thumb nail on it every 18 months or so (had 2x proper cleansings in its life just for fun in a decade).

If calcium is a problem compounding other issues worth adding maybe!? (I seem to recall you tried one some years ago though)!?