sharpener Posted February 19 Share Posted February 19 No, it seems not. Of those widely stocked by plumbers' merchants Fernox HP-EG is based on ethylene glycol (CH2OH)2 which is highly toxic and has a pH of 7.7 (slightly alkaline). Minimum concentration is 25% for proper inhibition, which protects to -10C Sentinel R600 is based on propylene glycol CH3CH(OH)CH2OH which is much less toxic but more viscous so more difficult to pump. It has a pH of 6.5 (slightly acid, arguably less desirable). 20% concentration will protect down to -10C. In addition specialist companies like Hydratech make several heat transfer fluids based on the above, plus other low viscosity and bio-based liquids. They also make Thermox DTX which is available retail and is an ethylene glycol based fluid with detoxification additive. 22% is needed for protection to -10C. So what do people actually have in their systems and in what concentration? Has anyone got the chart which distinguishes between the temperature a solution becomes impossible to pump and the (lower) actual freezing point for a given concentration? What would be the recommendation for a coastal site in Devon where the lowest temp in the last ten yrs was -4.9C (on 1/3/18)? Link to comment Share on other sites More sharing options...
JohnMo Posted February 19 Share Posted February 19 (edited) Or just go to Screwfix, Flomasta concentrated antifreeze, it's Monopropylene Glycol based anti-freeze with additional anti-corrosion & anti-limescale components. I am doing between 15 and 20% Dose Rate 25% = Freezing Point - 11°C Dose Rate 30% = Freezing Point -15°C Dose Rate 35% = Freezing Point -18°C Dose Rate 40% = Freezing Point -22°C Edited February 19 by JohnMo Link to comment Share on other sites More sharing options...
sharpener Posted February 19 Author Share Posted February 19 6 minutes ago, JohnMo said: Or just go to Screwfix, Flomasta concentrated antifreeze, it's Monopropylene Glycol based anti-freeze with additional anti-corrosion & anti-limescale components Thanks for that, hadn't thought to check Screwfix. For completeness it would seem this has a pH in the range 7.5 to 8.5 and I eventually found the following dosage table in the Q & A There are other tables here. It says that low concentrations of ethylene glycol (<25%) are at risk of bacterial contamination. I wonder if the biocides you can add to marine diesel would be effective? But they are pretty expensive too. On the whole I would prefer an ethylene glycol based fluid because as you know my application is close to the limits on heat transfer. But I don't want to use a lot of it because I have a relatively mild climate but a large thermal store to fill. Link to comment Share on other sites More sharing options...
SteamyTea Posted February 20 Share Posted February 20 6 hours ago, sharpener said: pH of 7.7 (slightly alkaline) pH is an odd thing as it changes with temperature. Ideally you want the pH to be the same as what it is in contact with, raise the temperature and the pH lowers i.e. more H+ than OH-. It is also a base ten logarithmic scale, so a small change makes a big difference. Link to comment Share on other sites More sharing options...
RobLe Posted February 20 Share Posted February 20 8 hours ago, sharpener said: On the whole I would prefer an ethylene glycol based fluid because as you know my application is close to the limits on heat transfer. But I don't want to use a lot of it because I have a relatively mild climate but a large thermal store to fill. We use Thermox DTX in our gshp groundloops at 25%. I found we needed it as the cold heat exchanger iced up once - I was lucky not to kill it😅. The header tank (diy, old school) ended up with a lower percentage and went a bit funny with floating manky stuff in it. I didn’t have any more DTX, so I sieved the header contents to get rid of the worst, then added half a bottle of fernox F7 biocide. That was a year ago, no more floaters, haven’t noticed any issues 🙂. Link to comment Share on other sites More sharing options...
sharpener Posted February 20 Author Share Posted February 20 1 hour ago, RobLe said: then added half a bottle of fernox F7 biocide Maybe that is the answer then. The min recommended conc is 22%, I presume on account of the possibility of fouling. As upthread I don't really need it pumpable down to -10C; something like -7 with burst protection to -10 will be more than sufficient. The tables don't go down that low (high) but something like 16% looks plausible if the fouling issue can be cracked by some other means. Link to comment Share on other sites More sharing options...
JohnMo Posted February 20 Share Posted February 20 17 minutes ago, sharpener said: don't really need it pumpable down to -10C; something like -7 with burst protection to -10 will be more than sufficient. I really just looked at the risks. Coldest day AND many hours of power cut is required to get insulated pipe contents down to atmospheric temperature. At any other time the heat pump freeze protection program will look after you anyway. How many external oil boilers have antifreeze? Do you ever here of issues with frozen pipes? Adding some extra biocide I didn't concider so perhaps I should get some fernox F7 biocide. Link to comment Share on other sites More sharing options...
HughF Posted March 1 Share Posted March 1 Water in my system, and a single, fake, anti-freeze valve (£30, alibaba)…. Link to comment Share on other sites More sharing options...
sharpener Posted March 1 Author Share Posted March 1 (edited) 1 hour ago, HughF said: Water in my system, and a single, fake, anti-freeze valve (£30, alibaba)…. What is the point of the fake valve? Who is it intended to kid? Some further digging shows that Fernox HP-5C may be more economical, it can be used at a dosage as low as 10%, this protects down to -4C and has a specific heat of 4.15 at 55C. It is based on propylene glcol, but at a low conc. maybe the viscosity is not too bad. From the graphs it seems to be less viscous than 25% of Fernox HP-EG, which is the lowest allowable conc. of their ethylene glycol product. Cheapest I have found is here, still costs about £8/l though and I will need ~40 litres. Edited March 1 by sharpener Link to comment Share on other sites More sharing options...
TonyT Posted March 1 Share Posted March 1 Trace heating on external pipework would be cheaper and not affect efficiency like glycol does. Link to comment Share on other sites More sharing options...
TonyT Posted March 1 Share Posted March 1 BES pluming supplies sell stuff under £5 litre Link to comment Share on other sites More sharing options...
sharpener Posted March 1 Author Share Posted March 1 5 minutes ago, TonyT said: Trace heating on external pipework would be cheaper and not affect efficiency like glycol does. My concern is what happens in a power cut. Can't rely on the house battery to be fully charged. 10 minutes ago, TonyT said: BES pluming supplies sell stuff under £5 litre If you mean this stuff then the min conc is 25% so more expensive overall. That was the attraction of the Fernox HP-5c. Link to comment Share on other sites More sharing options...
TonyT Posted March 1 Share Posted March 1 And do you suffer from power cuts? I would rather go trace heating and anti freeze balance to get a better efficiency all year round Link to comment Share on other sites More sharing options...
SteamyTea Posted March 1 Share Posted March 1 I am not sure how much the lower specific heat capacity of a water/anti-freeze mix really affects efficiency. There may be a small penalty to pay for extra pumping, but probably not measurable in reality. Link to comment Share on other sites More sharing options...
PhilT Posted March 1 Share Posted March 1 4 hours ago, SteamyTea said: I am not sure how much the lower specific heat capacity of a water/anti-freeze mix really affects efficiency. There may be a small penalty to pay for extra pumping, but probably not measurable in reality. Interesting point, the formula for delivered energy being Q = M x Cp x delta T, so water Cp reduces from 4.18 to around 3.8 for a 20% glycol mix, but increased flow rate M would cause a decrease in delta T, which seems to suggest the conventional wisdom may be correct Link to comment Share on other sites More sharing options...
sharpener Posted March 1 Author Share Posted March 1 26 minutes ago, PhilT said: Cp reduces from 4.18 to around 3.8 for a 20% glycol mix I think that's for ethylene glycol though these tables have a value of 3.95 which is not quite as bad. For propylene glycol there is an increase in density which is several times the (small) reduction in specific heat, so on a heat capacity per litre it actually gets better with increasing concentration. The tables in the links use a variety of intervals for both concentration and temperature (in both C and F) so are not easy to interpret but that is my tentative conclusion. I think the only material effect is the increased viscosity, for 25% EG this is about 2x and for 10% PG is <1.5x at 80F so the latter will be easier to pump, and provided there is sufficient head available it will merely result in an increase in pumping power, which will mostly end up inside the house anyway. Seems odd that PG can be used down to 10% whereas EG cannot be used at less than 25% bc of fouling. Perhaps it is because EG is a better substrate for biological growth. I will pursue this with Hydratech next week to see if they offer a direct equivalent to the Fernox product (as those are notoriously expensive). Link to comment Share on other sites More sharing options...
PhilT Posted March 2 Share Posted March 2 (edited) 9 hours ago, sharpener said: I think that's for ethylene glycol The specific heat capacity of ethylene glycol is 2.36, propylene glycol 2.5, at 25degC, so a 20% mixture of either would have Cp of around 3.8 at 25degC I wish it wasn't so as my system has 20% antifreeze source = https://waterbaths.net/blogs/blog/thermal-conductivity-and-specific-heat-of-bath-and-chiller-fluids Edited March 2 by PhilT Link to comment Share on other sites More sharing options...
SteamyTea Posted March 2 Share Posted March 2 1 hour ago, PhilT said: I wish it wasn't so as my system has 20% antifreeze Does mean that it takes less energy to heat up the water/anti-freeze mix in the first place. If heat transfer was purely down to the SHC, then a A2AHP would not be much cop. (not talking about a phase change here as that is a special case) I think the only extra losses are frictional in the pipework that increases the pumping load. Link to comment Share on other sites More sharing options...
PhilT Posted March 2 Share Posted March 2 (edited) 13 minutes ago, SteamyTea said: Does mean that it takes less energy to heat up the water/anti-freeze mix in the first place. I think it means that due to lower conductivity the compressor has to work harder to deliver the same unit of energy to the water/glycol mix. Would love to be proved wrong! Edited March 2 by PhilT Link to comment Share on other sites More sharing options...
RobLe Posted March 2 Share Posted March 2 That thermal conductivity is a big deal - inside the brazed plate heat exchangers there’s a thin layer of glycol that exchanges heat with the refrigerant, but it doesn’t mix or transfer it as effectively as water does to the bulk flow of glycol. Somewhere on the interweb I saw a paper on this - I think it was a factor of two overall. That is for example for a given heat exchanger size and pump, a 3degC drop refrigerant to water, or a 6 degC drop to a glycol mix. That 3C loss equates to a COP decrease. Of course, you can get a bigger heat exchanger to compensate. Link to comment Share on other sites More sharing options...
sharpener Posted March 2 Author Share Posted March 2 (edited) 28 minutes ago, SteamyTea said: Does mean that it takes less energy to heat up the water/anti-freeze mix in the first place. Well yes, fractionally for EG, but the thermal mass of the circulating fluid is small compared to the thermal mass of the house so won't affect the time constant of heating the house. For PG the heat capacity per litre is actually slightly more, see upthread. 15 minutes ago, PhilT said: I think it means that due to lower conductivity the compressor has to work harder I doubt it. Conductivity through the bulk fluid is not important bc (i) the rate of heat transfer in the outdoor unit is limited by the surface area of the HX not the bulk conductivity and (ii) it is being stirred all the time by turbulent flow which the design of the HX is engineered to encourage. Edit: that is what the ribs on the HX plates are for. I am pretty sure they are effective (I have done some HX design for Baxi and others). The plate spacing on the refrigerant side and water side is different bc the properties of the two fluids are different (though how they optimise it for both heating and cooling I do not know!). I would be interested to read the paper that @RobLe cites if it can be found. 2x worse dT strikes me as a lot, if true another reason for using 10% PG which is less viscous. Edited March 2 by sharpener Link to comment Share on other sites More sharing options...
HughF Posted March 2 Share Posted March 2 On 01/03/2024 at 15:07, sharpener said: What is the point of the fake valve? Who is it intended to kid? Some further digging shows that Fernox HP-5C may be more economical, it can be used at a dosage as low as 10%, this protects down to -4C and has a specific heat of 4.15 at 55C. It is based on propylene glcol, but at a low conc. maybe the viscosity is not too bad. From the graphs it seems to be less viscous than 25% of Fernox HP-EG, which is the lowest allowable conc. of their ethylene glycol product. Cheapest I have found is here, still costs about £8/l though and I will need ~40 litres. I objected to paying calefi prices so settled for the Chinese clone. Link to comment Share on other sites More sharing options...
RobLe Posted March 2 Share Posted March 2 A bit of rummaging around the internet and I found this, I think it’s what I’ve seen before, but I’m struggling to get the simple glycol is 2x worse than water from it. Note that at the time, I actually initially measured a 3C drop for our gshp propane-glycol bphe - then in feb or so when it started to freeze a little I added glycol. With the same 40W pump this dropped the flow from 18lpm to 12lpm (memory), and increased temp across the heat exchanger to 6C. Anyway, here’s that paper, maybe it helps… https://onlinelibrary.wiley.com/doi/pdf/10.1002/9783527621583.app2 As corroborating evidence, I notice that commercial gshp units tend to have a higher surface area bphe for the ground (glycol) side than the house (water) side. Link to comment Share on other sites More sharing options...
PhilT Posted March 2 Share Posted March 2 13 minutes ago, RobLe said: I’m struggling to get the simple glycol is 2x worse than water from it. Or "half as good", the pure substance having Cp of 2.3 compared to water 4.184 (joules per something or other). The commercial offerings are diluted plus additives. Fernox Alphi-11 (20% in my system) has Cp of 3.99, so mixed Cp 4.15, so nothing much to worry about. Link to comment Share on other sites More sharing options...
SteamyTea Posted March 3 Share Posted March 3 3 hours ago, PhilT said: joules per something or other kJ.kg-1K-1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now