Recently I obtained an HX espresso machine and want to be certain I never get scale buildup in the boiler. Steaming milk is the only way I routinely draw from the boiler and that isn't that often. The boiler is a whopping 14 liters, so I really don't want to drain and refill unless it needs it.

I was thinking of buying a HM TDM-3 water TDS meter and check the water periodically. When TDS goes over a certain reading, I would drain and refill.

Does anyone else use this method? What would be a good reading on this meter? It is a NaCL cal meter and according to many posts is not likely going to indicate hardness that would match a calculation.

I use Rpavlis water(100%) recipe going in. My thoughts are to check the meter reading with this, then check boiler out. If readings double, I likely am due to flush.
Thoughts?

Hi Ed, I may not be well equipped for it, but am willing to think about this with you. Here is some input on reasons to drain a boiler:
Draining your boiler periodically

With pure rpavlis water, you will effectively avoid scale deposits. If I follow your thoughts on the TDS measurement, you are concerned about carbonate build up. This would be the result of extracting steam over time, leaving behind mineral content. Since you don't do this very often, and the boiler is enormous (café sized, likely quite inefficient for home use), this will only happen extremely slowly. You would need to sample water from the hot-water wand, and then if carbonate buildup is the concern I personally would prefer a measure of alkalinity over TDS, to make interpretation easier. There are inexpensive test kits for this sold for aquariums.

If you measure TDS, it could be useful to know the material the boiler is made of, in order to consider whether any leaching could change the TDS content. I don't know this, but I wonder whether it could even be a good idea to purposely form a thin layer of scale---by introducing calcium for a limited time---in order to reduce leaching.

Does the user manual give any input on and reasons for the frequency of draining the boiler? It might be possible to do regular small drains rather than a big one periodically.

EddyQ wrote:I was thinking of buying a HM TDM-3 water TDS meter and check the water periodically. When TDS goes over a certain reading, I would drain and refill.

That would work, and a simple NaCl calibrated meter is fine here -- just be sure to measure the reservoir water and the boiler water at the same temp.

For some informative example numbers see David Buchanan's recent post: On the fence about descaling your espresso machine? See here... He also uses rpavlis water, but steams quite a bit and doesn't have a 14 liter boiler.

If you have a water tap that pulls from the steam boiler you can make a practice of drawing some hot water from the tap before each steaming session. I do that and my steam boiler water is generally around 30% higher in conductivity than my input water. (It hits a rough equilibrium because you are removing concentrated water, removing pure water, and adding unconcentrated water each time you steam.) On my small home machine (Rocket Evo) I always draw enough to at least trigger an autofill - that is one way to assure that it doesn't do an autofill while I'm trying to steam. (On my machine a cold water autofill will cause my steam pressure to plummet.)

Thanks Pat. I read though DaveB's thread at about the same time you replied here. Very informative!

Yesterday my TDS pocket meter arrived and I went ahead and calibrated it in 1382ppm cal solution. Then measured my distilled water(measured 1ppm). My Rpav water measured 62ppm (20C) which I was happy to see how closely it matched your 60ppm. I drew some water from my boiler with hot water spigot and waited til this morning to measure (left in closed jar overnight to cool). It measured 83ppm. It has been roughly 3 months since I filled the boiler and I did not do too many steamed drinks. Furthermore, my wife was using my water spigot quite a bit for the first month or so. I estimate, she pulled though five or more gallons. I am wondering if my water is dissolving some possible old scale that came to me from previous owner. If so, perhaps I should flush a bit earlier.
If not, it sounds like this TDS reading could double to 150ppm without risks.

I like this solution.

So I thought about what I had done with this method and realized there is a serious flaw. Measuring boiler water from the water spigot has to be corrected for the fact that it is mixed with incoming cool water!

So today, I filled an 8oz cup with water from the spigot and covered til cool. I also measured the temperature just after pouring. Since my boiler is 257F and incoming water (Rpav recipe), I computed the mix. Now this is where I may not be doing the math correctly. The water in was 75F, boiler is 257F and my cup with mix was 190F. So, (190-75)/(257-75)=63.18% of Rpav water and the rest(36.8%) is from boiler. Since Rpav tds measured 64ppm and mix measured 108ppm. 63.18% x 64ppm + 36.8% x X = 108ppm. Solve for X which is boiler tds. The result is 183ppm, which is getting high and likely needing a drain/refill. Seem right?

EddyQ wrote:Measuring boiler water from the water spigot has to be corrected for the fact that it is mixed with incoming cool water!

So today, I filled an 8oz cup with water from the spigot and covered til cool. I also measured the temperature just after pouring. Since my boiler is 257F and incoming water (Rpav recipe), I computed the mix. Now this is where I may not be doing the math correctly. The water in was 75F, boiler is 257F and my cup with mix was 190F. So, (190-75)/(257-75)=63.18% of Rpav water and the rest(36.8%) is from boiler. Since Rpav tds measured 64ppm and mix measured 108ppm. 63.18% x 64ppm + 36.8% x X = 108ppm. Solve for X which is boiler tds. The result is 183ppm, which is getting high and likely needing a drain/refill. Seem right?

You have a fancier hot water tap than most of us.

Your math looks fine to me. Of course is idealized a little - ignores the cooling of the water between mixing and collecting and measuring. But ballpark is fine here and you can assume that 30% or more of that water is from the reservoir and not the boiler, which means your observed increase in conductivity of water out of the the hot water tap (about 1.7x) probably means about a 3x concentration of the water inside the steam boiler. Fortunately since you're using rpavlis water that's no reason for alarm - won't scale or harm the boiler even if it does have 300 mg/L of potassium bicarb in there.

homeburrero wrote:You have a fancier hot water tap than most of us.

Perhaps, but the auto fill controller flipped out yesterday and now the machine sits cold. My La Pavoni to the rescue!

Thanks for checking my math! Yes I agree the mixed temperature is likely cooler due to other contacts. But if it was off 5 degrees, that would only mean my boiler TDS is lower than computed by some small amount. I think I will drain and refill since it is now cold and easy to do.

Another data point. I drained the boiler and measured the tds at 140ppm. Not the 183 that I calculated, but we know the temperature drop likely would have increased the calculated result. So I am not too far off, but worth noting the ratio from spigot to boiler.