Brass and lead toxicity in midrange espresso machines | Hacker News Reader

Brass and lead toxicity in midrange espresso machines

coffeegeek.com

197 pointshedora57 comments

One thing I noticed was that the guy who tested the shots from a Silvia was very proud of his scientific rigor, but never bothered to test (or at least he didn't mention it) whether there was lead in the water he put into the machine to begin with. Which for me renders his results meaningless.

Agreed - without a control it's definitely not super reliable, with that said he does mention he's using spring water to test it (I'm assuming shop bought - if that had lead it'll be a whole other scandal).

On the other hand, maybe he's just in Flint..

Disclaimer: I am not a subject matter expert. But I would like to point out that the collection section begins with: "With spring water in the machine". One would presume the "spring water" means lead-free as it has not been subjected to lead piping.

I, too, am a bit suspicious as the numbers seems to vary a lot and, with the exception of first 24h collection, the lead-content decreases while the time of exposure increases. But I wouldn't call the results meaningless.

Spring water isn't necessarily lead free. There are maximum allowable limits(at least in the US), but it could be anything less than that.

Assuming the spring water company is actually complying with the limits. That would also be worth testing. I wouldn't be particularly surprised if a spring water company was lax with ongoing testing and making sure to test the water once it has been through their whole process and is in bottles.

There was a case with cranberry juice in Australia (I think it was) where school kids tested store bought cranberry juice and found it contained zero vitamin C. Something about the way the juice was processed and bottled destroyed all the vitamin C but no one had ever bothered to test the juice after processing and bottling.

was it difficult to get distilled / purified water? I could go sprinkle some lead into Nestlē’s springs. They took the time to note that it’s spring water, why didn’t they just go “oh maybe we should use distilled” for the sake of easily ruling out a questionable impurity? Have we not learned anything from the polywater debacle?

> One would presume the "spring water" means lead-free as it has not been subjected to lead piping.

You would presume. Still, the whole analysis now hinges on that one thing - what really was in that "spring water". Maybe some lead got in there in some way they didn't expect. Or maybe they're just lying and that spring water was really tap water. Or maybe it really was lead-free. Point is, this is a core thing that should've been tested, and not doing so implies either malice or laziness, neither of which instill confidence in the research.

Hanlon's razor applies here.

"Never attribute to malice that which is adequately explained by stupidity."

-- https://en.m.wikipedia.org/wiki/Hanlon%27s_razor

Hanlon's razor would only apply if the person doing the test was being accused of deliberately ignoring the issue of initial water quality for malicious purposes. Besides that, it is a simple, judgement-free fact that if you are measuring how much a device changes something, you have to compare the output to the input, independently of any hypotheses about how the input got to be the way it is.

Hanlon's Razor applies because the GP literally accused the person of malice. Direct quote:

> Point is, this is a core thing that should've been tested, and not doing so implies either malice or laziness...

My apologies, you are right.

he literally accused the person of either malice or laziness, not of malice. The either is pretty important.

He's accusing him of malice or he's accusing him of laziness. Hanlon's razor applies to the former and not the latter. You're splitting hairs.

There might be lead in the plastic in the the water in the machine.

> in the least, sample the water from your silvia and sample the water from your water source if your source isn't already testing their water and send the sample(s) to an accredited lab

It's at the end of his post.

Where? I can’t seem to find it.

Search for "water source" on the page.

All I see is that the geologist says you need to test both the water from the machine and from the source, but he or she still never posts what the source results were nor directly claims that their results were zero.

Also no mention about the purity of the nitric acid used. Did they include nitric acid in the control sample, if they even sent one?

There's no indication that he actually sampled his water, though. If he did, he should/would have given us figures for the lead level in said water. It's unlikely that it was absolutely zero. I think he just trusted his water source.

Yeah without a tare, it's useless

The German Federal Institute for Risk Assessment also published some results from their research on lead migrating from espresso machines into the used water (http://mobil.bfr.bund.de/cm/343/freisetzung-von-blei-aus-kaf... and http://www.bfr.bund.de/de/fragen_und_antworten_zur_freisetzu...). Based on that, there is a huge thread in a German coffee board (https://www.kaffee-netz.de/threads/bfr-studie-zu-blei-in-sie...) where you can find several test results for different machines.

I don't have much to add to the discussion about coffee machines. I just want to say, that Britta water filters (and potentially those of other brands) leak silver ions into the filtered water. This is to reduce the green algae in the container.

But I would guess, it also changes the microbiome in the human gut. I would love to see some research regarding this, because I stopped drinking water from the water filter when I was really depressed. My depression faded and there might me a causal link or not, I don't know.

There might be something to that. IIRC, intestinal flora synthesize serotonin [0].

[0] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4393509/

Another good set of studies to compare against are those showing leaching of lead from crystal decanters with hard liquor. Where the amount of lead leached may be gargantuan initially, the level leached out shrinks over time until it reaches a certain low threshold, which at least one study showed was possibly due to a kind of protecting coating developing over the crystal.

If the author found that water shots had a high concentration of lead, it could be that this level would taper off over time, or even that some other contaminant was at play. It could also be the lab gave back inaccurate results. And even with high concentrations of lead, it's possible that interactions with the content of the espresso could reduce the effect on the consumer. A study of actual concentrations of levels of lead in the blood of the espresso drinker should be evaluated.

I appreciated this simple overview: https://www.seattlecoffeegear.com/blog/2011/02/04/keeping-yo...

That overview says plated boilers are safe. Yet apparently they are not according to the forum post above.

I own the Gaggia Classic, and as far as I know it uses aluminum or stainless steel (depending on the year) for the boiler. It's a semi-automatic machine, and personally I love it. Barista grade portafilter, makes a very good espresso. I replaced the steam wand with the Rancilio Silvia wand for better steam action; it's a simple five minute job and worth doing if you go with this machine.

So much for the Silvia I had been planning to get. Relatedly, I've been having trouble finding an under the counter water filtration system which does not involve putting the water into extensive contact with plastic. Maybe there just aren't enough people concerned about such things in order to affect the market. On the other hand, in-demand, easy & low cost to implement features such as built in PIDs have not made their way into mid-level machines as standard features, so maybe (baseless speculation incoming:) competing brands have an informal entente to add in improvements no faster than is absolutely necessary.

I’d love an all glass filter system to replace my plastic one now.

One concern is... what difference would that make in the pre-filter housing if the water goes through plastic pipes anyways to get to you?

One can't ever expect the filtration to be perfect, not to mention any contact that happens "downstream".

Plastic may not be so bad. There’s a research group that has studied various pipe materials. The result generally seems to be that PEX varies widely but leaches detectable and smelly things. PVC does too. CPVC doesn’t seem to affect taste/smell or to leach much. PP-R is also pretty good. Copper, somewhat surprisingly (to me) affects taste. I haven’t seen niche materials like silicone or PTFE tested.

The Sylvia is amazing. It really is. Also consider a second hand La Cimabli Junior. Both leave the opposition for dead, are easy to work on and have readily available parts.

It would be interesting to see the results from NSF and ROHS-approved home machines from manufacturers like Nuova Simonelli, La Marzocco, and Astra.

I also wonder how effective water scale deposits on the boiler are at mitigating lead - doing tests on new-ish machines without much scale buildup could be a factor in these results.

Stainless is becoming more of a popular alternative to brass and copper, but isn't as effective of a thermal conductor and is not as malleable. For steam boilers and heat exchanger systems, being able to expand and deform under high pressure is a desirable feature.

For steam boilers and heat exchanger systems, being able to expand and deform under high pressure is a desirable feature.

I'm curious, why? Presumably it's more than just the fact that deformation allows for less catastrophic overpressure failure?

Correct, there is a lower chance of an explosion.

I have a Saeco Aroma which according to its specs is all stainless steel. Looks like that one is ok.

It may be an all-stainless chassis, but in a $250 dollar machine, my guess is you're getting a brass boiler.

Edit: nope, you're right. SS boiler apparently

A stainless boiler is just 2 stainless steel cups held togheter by a clamp, it shouldn't affect the price much, my 120€ delonghi has one.

Yeah, I was pretty much assuming that all the cheap machines would use standard small diameter copper plumbing rather than pay for all stainless fittings all the way through (which would be somewhat specialized from a plumbing perspective), so it wouldn't really matter much whether the boiler itself was or not.

Most likely but thats not 100%. Unfortunately the specs may not be correct if the manufacturers of the individual input components and materials have not been honest.

For example there is a hospital in Australia (http://www.abc.net.au/news/2017-06-30/lead-levels-in-perth-c...) that has been massively delayed due to lead mysteriously appearing where there should be none and asbestos also appearing where there should be none. From what I know, the sources have turned out to be manufacturers in China either knowingly misrepresenting their products or using contaminated materials in their own manufacturing.

Unfortunately testing is the only way to be sure.

This report looks to have examined the same question in the EU and seems less alarmed:

https://www2.mst.dk/Udgiv/publications/2015/09/978-87-93352-...

The report monitored "ready-to-drink espresso coffee from 8 cafes and gas stations... purchased in the Copenhagen area."

The OP's forum is talking about particular (older) mid-range personal espresso machines which use brass boilers, like the Silvia V3: https://www.thecoffeebrewers.com/rasihoofesan1.html Those are not generally found in higher-end-coffee-selling cafes.

The report did note that one of the espresso machines tested in Copenhagen put out levels of lead greater than what was considered disconcerting in the OP's forum's calculations: 65ng/ml vs the 20-40ng/ml found in Rob's testings.

But even so, they still conclude that such levels are not a particular health risk in their mind. From the report:

> "The concentration in ready to drink espresso from cafes was similar to the homebrewed coffee in all but one sample, which had a concentration of 65 ng/ml. Even if a consumer has a daily cup of coffee from this cafe it will not be of health concern. The correlation between tin and lead described earlier indicate that the source of the high value could be a lead containing welding in the espresso machine."

There is no safe amount of lead.

There are acceptable amounts.

“No amount of lead is safe.”

https://www.niehs.nih.gov/health/topics/agents/lead/index.cf...

I had no idea lead was so bad for you. It’s terrible.

Only according to some bureaucracy that had part of its mission not to raise costs too much for manufacturers.

Not according to human physiology/medicine.

Dosage is the poison. Oxygen can become toxic quite easily.

I doubt a single atom of lead will be unsafe. Or two. There is definitely a level at which the intake of lead will not lead to any measurable effects during the lifetime of a human.

It is the same with ionizing radiation.

However, the question is about how much it matters compared to global intake.

We get lead from many sources, some of them hard to avoid while still living a normal life. "Safe" doses are doses that represent only a small fraction of that unavoidable part. The standards may change as the unavoidable become avoidable.

For radiation, the baseline is clear, that's background level, I don't know the equivalent for lead but it's clear that we can't keep away every single lead atom.

They’re not equivalent. Radiation is likely to mutate DNA, but it’s also likely that the body self-repairs such mutations. Lead, however, has no self-defense mechanism. It mimics chemistry in the synapses between neurons causing havoc with nervous communication. It does a host of other horrible things across the body depending on where it ends up.

There is no safe amount. Clearly it exists in nature, as does arsinic and uranium, so we will be exposed to toxins unavoidably. That doesn’t mean it’s safe or good, especially to be consciously added to your daily coffee.

The big concern from the forum post data was that the machines hold water near boiling in a brass boiler. In a coffee shop, the flow through the boiler is much higher than at home, so buying a brass boiler for home use may not be a great idea. I don’t think this contradicts the linked study (buying small, cafe-grade machines for home use is kind of a corner case...).

I think I'd actually be more concerned about coffee makers in offices - any of the multi burner ones with a hot water spigot are obviously maintaining a supply of hot water.

I wonder how this applies to all the instant coffee vending machines around South Korea and Japan.

The ones within subways and outside bus stations feel the most suspect.

And some of those bus drivers and maintenance workers drink from those 10x/day.

The list mentioned in the last posted comment would be helpful to see, if anybody has it available.

Link is from 2006.

Site unaccessible from VPN. Guise I promise I'm not a spammer I just value my privacy :'(

Reply on news.ycombinator.com