Lelit Bianca, particles forming in boiler (rpavlis water)

Equipment doesn't work? Troubleshooting? If you're handy, members can help.
millmountain

#1: Post by millmountain »

Just this past weekend, I was telling myself how very satisfied I have been with the Lelit Bianca since March. She is indeed a great machine, although one small issue did appear that I haven't yet dealt with. If I am interpreting it right, that issue now requires some attention. Today I opened a ticket with Lelit.

The issue?
The hot-water wand emits water tainted with particles. The machine exclusively sees distilled water treated with the rpavlis recipe (addition of 100 mg/L of potassium bicarbonate). Although we use the steam wand every day, we rarely use the hot-water wand. After the Bianca was a few months old (maybe four months), the water from the hot-water wand began to come out loaded with dark particles. I think these particles are a problem, but I am not sure what is causing them to form.

I cannot detect them coming from the steam wand, although it admittedly is hard to draw enough water to be sure. I suspect that the particles are forming on the walls of the boiler. The particles settle quickly in water, so if the hot-water wand draws water from near the bottom of the boiler and the steam wand draws steam from the top of the boiler, it would make sense that the particles only exit the boiler through the hot-water wand. Hopefully no particle-laden lattes.

The alternative, that they could be forming in the hot-water wand, seems unlikely. The particles thin out during extraction, but after a short interval they then again come out intensively. It thus seems they come from the boiler.

Here is a video (warning, turn down the volume), initially the particles are intense, then absent. A second, immediate release is clear. After a short delay, it will again be contaminated.
[Edit: Uploaded video to YouTube on request, but not embedded so you still need to click the link]
https://streamable.com/h5k6zz

Symptom No. 2
This morning when starting the machine showed error 03, the "steam boiler filling alarm." I let the machine sit for a couple of hours, then simply turned her off and back on. This time, she started up and behaved normally without reporting errors. If the particles are forming in the boiler, perhaps they are building up and starting to block the water intake.

My observations:
  • If the rpavlis water were leeching metal from the boiler walls, I suppose it should be in solution and not visible as particles that settle.
  • The brew boiler shows no signs of containing particles. If the water is causing the issue, I suppose there should be signs of particles coming from the grouphead as well. One possibility is that the internal walls of the brew boiler are different than those of the steam boiler, another would be there is no way for the particles to exit the brew boiler.
My questions:
  • What could be causing the particles to form? Could they metallic or otherwise potentially harmful to health?
  • Where might the particles be forming? If in the boiler, it seems likely this will eventually cause damage and lead to a need to replace the boiler.
  • What steps could I take to help answer these questions? I would prefer to understand the issue rather than just bring the machine in for service.
I am happy to produce further evidence and to consider what trials to perform. I might wait on the latter until receiving feedback from Lelit. I took photos under a microscope of the particles filtered in a paper towel, but it's not an ideal surface. I could try getting some on a film substrate. Basically, they're quite small.

BPlus: turning your coffee spirit
Sponsored by BPlus
nahau

#2: Post by nahau »

Is it possible the boiler is picking up water from your driptray? I'm uncertain how your machine is built, but there was a post here in the not too distant past where the poster had the hose of the vacuum release outlet to the driptray below the water line... thus pulling all the driptray muck into his boiler when he shut down his machine and a vacuum formed as it cooled. Any hoses sitting down low in the driptray?

catmug

#3: Post by catmug »

This is a guess, but I would say given your usage habits (steam only, no water tap), you've probably got scale in your service boiler from the same water sitting in there and essentially distilling every day, concentrating what small amount of calcium / minerals is in your water. This could also explain the sensor malfunction and the lack of an issue with the brew boiler, where water passes through regularly.

Fatdonk54

#4: Post by Fatdonk54 »

I have a Bianca as well, and my hot water spout is producing bad tasting water. I've compared it to the water that comes out of the group head, and they taste completely different.

I am guilty of not using my hot water spout often, so I'm thinking that it's just leftover water and build up that could be causing the water to taste weird.

Maybe you are having a similar issue? I've reached out to the distributor I purchased my Bianca from, hoping for some guidance.

millmountain (original poster)

#5: Post by millmountain (original poster) »

nahau wrote:Is it possible the boiler is picking up water from your driptray? I'm uncertain how your machine is built, but there was a post here in the not too distant past where the poster had the hose of the vacuum release outlet to the driptray below the water line... thus pulling all the driptray muck into his boiler when he shut down his machine and a vacuum formed as it cooled. Any hoses sitting down low in the driptray?
Now this is very interesting. The answer, however, is no. The driptray outlet is closed off, has no hoses attached and typically has to be emptied once or twice a day. The driptray water is also brownish, since the single-shot basket leaves a bits of grounds on the screen, which we frequently rinse and brush off into the driptray. If this were getting into the boiler, even diluted with fresh water it would not be alternatively gray and then clear, as I think is visible in the video. Good thought, though.
catmug wrote:This is a guess, but I would say given your usage habits (steam only, no water tap), you've probably got scale in your service boiler from the same water sitting in there and essentially distilling every day, concentrating what small amount of calcium / minerals is in your water. This could also explain the sensor malfunction and the lack of an issue with the brew boiler, where water passes through regularly.
I hadn't thought about buildup from something coming back out of solution (precipitate), as opposed to leeching material from the boiler. The main reason is that there should be essentially zero calcium in the water. I distill the water myself using this beast, which has activated carbon to catch VOCs in the output. My only additive is potassium bicarbonate, and the granules are tiny gray particles, I would estimate on the order of 10 microns. From their color and form, the particles are not calcium carbonate (limescale), which is white or beige.

There is probably not a sensor malfunction. This morning my wife shared that the pump drawing water into the boiler has been running much longer in recent weeks during starts (I am usually not around), noticeably longer than short pump run times in the past. I assume it has to work harder due to some clogging from the particles, in which case there is now quite a buildup in the boiler and it will have to be serviced soon.
Fatdonk54 wrote:I have a Bianca as well, and my hot water spout is producing bad tasting water. I've compared it to the water that comes out of the group head, and they taste completely different.

I am guilty of not using my hot water spout often, so I'm thinking that it's just leftover water and build up that could be causing the water to taste weird.

Maybe you are having a similar issue? I've reached out to the distributor I purchased my Bianca from, hoping for some guidance.
I noticed this difference in taste from the very beginning. I sort of ascribed it to the very soft water (neither distilled nor rpavlis water taste great to me), and the difference to the water mixing with some residue from the grouphead. This bad taste is why I decided against using the hot-water wand for americanos, and why I don't use the wand very often. Now I am wondering whether the different taste is due to the water pulling some material from the boiler walls, or possibly from somewhere in the steam circuit piping.

A shame about the hot-water wand taste really, which I decided to accept because of the special rpavlis water. It seems to me rather ironic IF the rpavlis water is causing any damage/leeching from the steam boiler, because the whole reason was to increase machine longevity by avoiding descaling, damage from sulfates, etc.I am not so quick to believe the rpavlis water is the issue, because it is much discussed and Prof. Pavlis knew his topic well; however, I can't rule it out, as it could be that the Bianca has some components that are not common to other machines with which rpavlis water has been tested.

I would like to know:
Are there any other Bianca users who have been using rpavlis water?

Thanks,
Mike

millmountain (original poster)

#6: Post by millmountain (original poster) »

For anyone following the water angle, I think we have to assume that I could have been miscalculating and not actually mixing the rpavlis water recipe correctly. So as an aside, here are my calculations and how I mix, so we can either correct my mistake or check it off as being correct.

As described above, I distill approximately 4 L of water at a time using a distiller, which is less than a year old.

The dry ingredient is pharmacy-grade potassium carbonate, KHCO₃. I will double check the label that it is pure KHCO₃.

I prepare a concentrate by adding 10.0 g to 0.25 liters of distilled water. This is of course a concentration of 40 g/liter.

When I mix, I use a volume of approximately 2 liters of distilled water. I then calculate how much concentrate by weight (assuming 1 g ~ 1 ml) I need to add to achieve the desired concentration. The desired concentration is 100 mg/l.

A: 100 mg/l, desired rpavlis concentration
B: 10 g / 0.25 l = 40 g/l, KHCO₃ concentrate
D: 2.0 l, volume of mixture
C: don't ask, don't tell

Added volume in liters of concentrate needed for the mixture volume:
= A / B * D
= 100 [mg/l] / 40 [g/l] * 2.0 [l]
= 0.1 [g/l] / 40 [g/l] * 2.0 [l]
= 0.005 l = 5 ml, or about 5 g

In case this is hard to follow, here is another way to get to the same result. Start by asking what weight of KHCO₃ we need per 2 l to get 100 mg/l:

weight of KHCO₃ = 100 mg/l * 2 l = 200 mg = 0.2 g

Then ask what volume of concentrate we need to get 0.2 g of KHCO₃:

0.2 [g KHCO₃] / 40 [g/l] = 0.005 l = 5 ml

In other words, since there are 0.2 g KHCO₃ in 5 ml of concentrate, 5 ml is the right amount of concentrate to add to 2l of water in order to get 100 mg/l concentration. (While this neglects the volume of 5 ml vs. 2000 ml, we can happily live with this 0.25% error.)

I therefore add 5 g (about 5 ml) of the concentrate to the 2 liters of distilled water to get rpavlis water.

Using the average molar masses of potassium, hydrogen, carbon and oxygen, for a concentration of 100 mg/l I calculate 60.9 mg/l of HCO₃⁻ and 39.1 mg/l K⁺. This results in a calculation of 49.99 KH in mg/l eq. CaCO₃ (from HCO₃⁻) and 75.6 TDS in mg/l. (GH is theoretically zero.) I plan to buy a TDS meter today and measure the distilled, rpavlis and hot-water wand values. Hopefully the fish shop will also have some kits that will allow me to verify GH and KH.

JRising

#7: Post by JRising »

Does it smell bad?
My first thought was "Oh, that's burnt milk that's been sucked up the steam wand into a cooling boiler, I should write about checking the anti-vacuum valve", but I would have thought that the smell would be something that you'd mention, not just the discoloration, so I'm curious, does it have a weird odor?

Aida Battle: Indigo Reserve from world renowned Finca Kilimanjaro in El Salvador
Sponsored by Aida Battle
millmountain (original poster)

#8: Post by millmountain (original poster) » replying to JRising »

I didn't think of this. I knew of the possibility, but we've been very careful about cleansing the steam wand after every use. There were a very few instances over the last 10 months where it didn't get flushed for a couple of hours, so it's not impossible that there could be milk contamination.

I had all family members smell it and distilled water at dinner; the son thought the distilled water smelled stronger. Hard to describe any difference in smell, maybe there is a slight difference. I just smelled cool vs. fresh out of the wand, there is nothing clearly funky. Glad you thought of that.

millmountain (original poster)

#9: Post by millmountain (original poster) »

Got a TDS meter on the way home, so now we have some data.

The water samples are:
  1. Distilled. After distillation, sat in a plastic container a couple of days.
  2. Rpavlis. As described above, freshly made. The concentrate solution and added amount were accurate to about 0.01 g.
  3. Grouphead. Did a backflush, then ran water for ten seconds. Did another backflush and ran another 10 seconds, then took the sample.
  4. HWW, first gush. From the hot-water wand, the first bit that came out. Some water was let out yesterday, and some was let out 4.5 hours before this, but not more recently.
  5. HWW, constant. After first gush ran some more through, and took the sample from a constant stream.
  6. Tap water. From the tap using the cold setting. Allowed to warm to room temperature.
The TDS meter says it accounts for temperature variation, but not to use with hot water. The samples cooled or warmed during dinner and I measured the temperatures by handheld IR from directly above the samples. Here they are, in order from left to right starting on top, although after reading this thread you can visually tell which ones came from the steam boiler. The difference is even starker in person.






Here's the data:
                   | Temp, °C | TDS, ppm
                   |----------|----------
1. distilled       |   21.5   |  000
2. rpavlis         |   22.0   |   55
3. grouphead       |   22.5   |   64
4. HWW, first gush |   24.0   |  224
5. HWW, constant   |   23.5   |  191
6. tap water       |   20.6   |  134
7. fish tank       |   21.9   |  141
My preliminary conclusions from the data:
  • The distilled water is pretty clean. I used sample 1 after measuring it to clean the TDS meter after subsequent uses, and even that small contamination raised the value to 9 ppm.
  • The rpavlis water is not exactly at its theoeretical value, but it is close enough for me to not be concerned.
  • The grouphead does not seem to be specially contaminated. There is sure to be enough residual oils to account for +9 ppm; nevertheless, it is used frequently.
  • The steam boiler very clearly has some new additions. Since it only gets used for steam, most of the water may have been in there for months.
  • Our tap water is drinkable and "okay" for immersion methods, although I don't love the taste.
  • I changed the aquarium water this past weekend.

User avatar
homeburrero
Team HB

#10: Post by homeburrero »

millmountain wrote:For anyone following the water angle ...
I'm following the post, but am pretty convinced that it's not related to your water. I think you have somehow gotten contaminants into the steam boiler. I think nahau's suggestion above is certainly worth looking into.

Re your recipe math for the R Pavlis water; it looks fine. Pretty straightforward given your 40g/L concentrate, which is 40 mg/ml. Putting 5 ml of that in distilled to make 2 liters is 5 ml * 40 mg/ml ÷ 2 L = 100 mg/L KHCO3, which is the full strength rpavlis recipe . The total alkalinity is 50 mg/L CaCO3 equivalent*, the total hardness is zero, and the pH nicely neutral. There is nothing in that water that might produce a dark precipitate that you're seeing, and nothing that should corrode metals or strip oxides from the boiler, pipes, or fittings.

The actual TDS would be 100 mg/L because you have dissolved that much stuff into the water. But the conductivity of KHCO3 is expected to be low compared to the typical mix of stuff in natural water, so when you check it with a typical inexpensive NaCl calibrated conductivity 'TDS meter' at 25℃ it will read somewhere around 60 ppm.



* KHCO3 happens to have almost exactly the same molar mass as CaCO3 (100.12 g/mol for KHCO3 vs 100.09 g/mol for CaCO3) , but only neutralizes half as much acid, so that's why the alkalinity, in CaCO3 equivalents, is 50 mg/L rather than 100 mg/L.
Pat
nínádiishʼnahgo gohwééh náshdlį́į́h