Lishtot - Test If ANY Water Source Is Safe To Drink In 2 Seconds

[EEVblog]

My bullshit meter just went off the scale 

https://www.lishtot.com/TDP1.html

[Urs42]

I'm only guessing what this thing is doing. The TestDrop Pro Thing is only a PCB with Bluetooth, some LEDs and buttons. A  app is needed on a phone, and if you press a button on the TestDrop, the app will look up online if the water is safe. They do have some data on the website: https://www.lishtot.com/global-water-map.html. It looks like most water outside of the US is safe. 

[wilfred]

This video shows how to use the device.


For those who want to investigate a little deeper here are some links to my first 5 minutes of looking.

Patents by Alan Joseph Bauer https://patents.justia.com/inventor/alan-joseph-bauer
This one immediately caught my eye although admittedly I am no patent search expert. https://patents.google.com/patent/US9453810B2/en
the ninth embodiment seems to me to be relevant.

https://www.lishtot.com/LishtotAnalysis_Nepal_Experimental_Plan-Lishtot_TestDropPro_Jan2019.pdf
http://mitwater.org/events/lunchandlearnsusanmurcott

[EEVblog]

This video shows how to use the device.

 

I want to buy one of these and bust this shit right open, but I don't wan to give this scammer any money, what to do? 
Also, it seems to be over 2 years old.

[magic]

Doesn't have to touch the water to get a reading
Tests for 20+ different harmful contaminants

For all we know it could be a Geiger counter.

Erm, wait it's not.

E. Coli

They really detect it contactlessly? Did I miss those little buggers evolving RFID tags or something?

But say what you want, this lady totally convinced me


edit

The device first went on general sale to the public in July, and CEO Netanel Raisch tells TIME that Lishtot is preparing to launch “testing as a service” in India in order to help the technology reach the people who need it most. “The way that we see it, it’s not just about making money,” Raisch says. “We built the company to help people.”

Lishtot has more water purity-related technology on the way, but for now the TestDrop is its main product. The devices cost $50 each, or $35 with the CES discount — but my guess is they’re more likely to spread when bought in bulk by NGOs, utilities and other organizations, which will then distribute them where they’re most needed.

So incompetent charities is the market they are after?

[PlainName]

I want to buy one of these and bust this shit right open, but I don't wan to give this scammer any money, what to do?

You pay for your principles, is what. Or you wanna remain an armchair warrior?

I think you would make far more than the cost via your video income, and even if you broke even or made a slight loss, you would still benefit from actually following your conviction and doing it.

As to not giving them money, $42 isn't going to get them much further along any road they are on, and you could look at it as preventing them getting much more money when you successfully and properly debunk it. Or, maybe, admit that it works 

[Ice-Tea]

It measures 'the field associated with the cup'.

I'm pretty sure this must break a few laws. And sooner or later, people will die and there will be lawsuits.

[magic]

No, because they want to sell it to third world countries But things could get interesting if some American goes hiking with it and gets diarrhea.

I have a hard time believing that just e-field signature of hovering the liquid around can be used to reliably detect everything from mercury ions through halogenocarbons up to bacteria while also not detecting safe stuff like small amounts of salt, calcium ions or whatnot. And that's what the patent found by wilfred appears to boil down to.

The invention claimed is:

1. A method for determining the quality of a liquid sample, including the following:
causing a portion of said liquid sample to flow in a predetermined direction in a liquid delivery element;
positioning a conductive element such that it is not in direct contact with either said liquid sample or said liquid delivery element, wherein said conductive element is adapted to be in electrical communication via a single electrical connection with a passive electrical metering device;
recording electrical readings with said electrical metering device;
comparing said electrical readings with predetermined electrical reading values associated with high quality and low quality samples of said liquid;

They didn't even include any AI buzzwords here

[BravoV]

It measures 'the field associated with the cup'.

I'm pretty sure this must break a few laws. And sooner or later, people will die and there will be lawsuits.

Wonder, if there are few "greedy" people, say a lawyer + doctor and a willing "victim", deliberately make a liquid that strictly will only cause the victim sick, but not seriously though, say just mild diarrhea  , and then used this fabricated case to sue the hell out of them.

[Cerebus]

Erm, wait it's not.

E. Coli

They really detect it contactlessly? Did I miss those little buggers evolving RFID tags or something?

Oh yeah, those E. Coli have really kept up with the times. They even have their own anti-social network now, "The Virulence Signal Path"^*.

^* A joke that will only be understood by someone into electronics and biology. So obscure it'd make a hipster shed tears of joy.

[PlainName]

say a lawyer + doctor and a willing "victim", deliberately make a liquid that

I had an invention that upset some incumbents that used a very basic tool to reap loadsamoney from their clients, and had exactly that threatened. I was told it didn't work and couldn't possibly work (despite it actually working and one of their kind even being asked, unwittingly, to test it), and that they would buy one and deliberately hurt themselves with it in order to sue the arse off me.

I assume you were kidding, but this really does happen. The insurance 'smash for cash' scam was all the rage around here until it got stomped on a year or two ago.

[fcb]

I want to buy one of these and bust this shit right open, but I don't wan to give this scammer any money, what to do? 
Also, it seems to be over 2 years old.

For $64.90 ($49.95+$14.95 shipping) you could get 2 or 3 videos out of it.

You might even beat Thunderfoot.

[Cerebus]

This one immediately caught my eye although admittedly I am no patent search expert. https://patents.google.com/patent/US9453810B2/en
the ninth embodiment seems to me to be relevant.

That 'patent' is an almost pure condemnation of the current patent system. Basically it seems to list every method of detection that has ever been used to detect something vaguely chemical or biological in a liquid using electronics. If it were actually valid it would cover blood glucose monitors, dissolved solids meters, dissolved oxygen meters and so on, and so on (i.e. 'prior art' devices that exist in their millions). It's the most bullshit patent that a US patent 'examiner' has ever pushed through on the nod.

The 'embodiment' that they are selling is based on the weakest method of all those listed in the patent. Testing for contaminates, some of which can be dangerous in parts per billion, by a remote electric field test? Really?

Anybody relying on this device is going to either

• Die of thirst because this will react to anything that is not pure distilled water, or
• Die from contamination with something highly dangerous that doesn't render the water sufficiently conductive to be distinguishable from pure H₂O by a remote electric field test.

If I were in a position to I'd love to call their bluff very publicly with a succession of deadly drinks: Polonium Nitrate, Aflatoxin, Puffer Fish liver, a Thallium salt and so on: "OK, test these and drink the ones that you deem potable".

These people are actively dangerous, they're going to kill people.

[ebastler]

I want to buy one of these and bust this shit right open, but I don't wan to give this scammer any money, what to do? 

Test the crap out of it, the return it for a refund. 

[magic]

Test the crap out of it, the return it for a refund. 

But it would also be interesting to tear it down and see what's inside, how it works and what it really reacts too.

Bonus if Dave can sacrifice a bit of his poo for science and see if the gadget will detect it

[BravoV]

Dave has a baby/toddler, he and his spouse dispose diapers everyday, should be easy. 

[SparkyFX]

My bullshit meter just went off the scale 

From the description it might be just a very narrow use case. Only very pure water, only in sterile plastic container, only for contaminants that change the dielectric properties of anything in the sensors range. It´s probably just a capacitive sensor with a reading of the dielectric properties at a fixed distance, but at the same time covering river and stream water, mineral water, tap water.

So what do you do when all sources are showing red, you can only rely on bottled water as long as the bottle is sterile (especially on the outside) or decontaminate what you find. The information provided would be a help in the ideal case where you find a clean water source that you don't have to decontaminate, but going into the wilderness and not planning to decontaminate the water would be irresponsible in the first place? It kind of pinpoints a hen and egg problem. Anyone who needs to rely on potentially contaminated water sources has methods for those cases.

[SiliconWizard]

It possibly can't be reliable as a testing tool. There can be many different kind of contaminants in water, some of which would be very hard to detect without proper analysis.

So apart from the bullshit meter, what I'm wondering is what exactly could possibly happen if this device ever became a product? If someone uses it, it falsely detect some water as "safe" to drink, and the person still gets sick or dies? Is the founder comfortable going to jail?

Looks like some people don't realize what they're getting into fiddling with health-related devices.

[PlainName]

There can be many different kind of contaminants in water

Looks like it's explicit in what it can (allegedly) detect, so many different kinds that aren't on that list aren't relevant.

Is the founder comfortable going to jail?

Vendor - the founder is safe behind the sales-front, I think. Let's suppose it can robustly detect the things it is alleged to detect, and someone falls ill through something it doesn't detect (and isn't said to detect or even meant to detect). Is there still an issue? Could be quite lucrative for the lawyers arguing duty of care and similar.

[magic]

That 'patent' is an almost pure condemnation of the current patent system. Basically it seems to list every method of detection that has ever been used to detect something vaguely chemical or biological in a liquid using electronics.

I don't think so. Sure, it rambles a lot about different things, but a some point you get tired, scroll down to the list of claims and it appears that they only claim contactless sensing, and electric field in particular.

Vendor - the founder is safe behind the sales-front, I think. Let's suppose it can robustly detect the things it is alleged to detect, and someone falls ill through something it doesn't detect (and isn't said to detect or even meant to detect). Is there still an issue?

The banner at the top of the website posted by Dave says: "Test If ANY Water Source Is Safe To Drink In 2 Seconds". That's a rather strong claim and I think lawyers will be salivating themselves if this product ever takes off in the West.

They also say it has been independently tested, but both test reports contain disclaimers to the effect of "we only report about the results of particular tests that have been performed under laboratory conditions, nothing is implied about suitability for any purpose". The vendor has nothing to cover their ass.

[Cerebus]

That 'patent' is an almost pure condemnation of the current patent system. Basically it seems to list every method of detection that has ever been used to detect something vaguely chemical or biological in a liquid using electronics.

I don't think so. Sure, it rambles a lot about different things, but a some point you get tired, scroll down to the list of claims and it appears that they only claim contactless sensing, and electric field in particular.

We could quickly go down the rabbit hole trying to dissect/decode something written in patent-ese  but I think the claims are potentially broader than that and the only limitation is 'not in direct electrical contact', so potentially including all the colorimetric methods etc. Definitely one where one would want to ask a patent lawyer "Has X kind of interpretation been allowed in the past?" - I know that this is the kind of area where common sense and the law, as interpreted, don't mesh and I'm most definitely not going to try to figure it out using mere logic.

[PlainName]

The banner at the top of the website posted by Dave says:

Fair enough. I note that they also say 20+ rather than explicity 20. Mea culpa.

[daqq]

(when the first person dies of drinking improperly analyzed water)
Release the lawyers!

[Daixiwen]

Where is the blockchain?
You can't sell a gadget in 2020 without a blockchain!

[EEVblog]

It possibly can't be reliable as a testing tool. There can be many different kind of contaminants in water, some of which would be very hard to detect without proper analysis.

There own website is filled with info about why it might display and inconsistent result, how you have to use clean cups, have a specific height above the water etc.
It clearly doesn't work properly and they know it.

Mrs EEVblog who is a water quality scientist and published in the field of water contamination just looked at the web page stared in disbelief and walked away.

[PlainName]

how you have to use clean cups, have a specific height above the water etc.

So much like, say, EMC probes or that I-Prober thing (got an Ebay search on those but they're not getting any cheaper). Needing a particular environment or support isn't in and of itself a fail. And if it does what they say given those circumstances, it should be hard to argue that it doesn't work.

The issue is whether or not those things are appropriate - for the suggested usage (i.e. the girl on the river bank taking a sip) they clearly aren't, but saying it doesn't work when they can show that it does - albeit given a laboratory setting - is asking for a fail fail.

I saw this the other day on BBC Click! and whilst they tend to be non-judgemental I'm fairly sure they wouldn't show it working if it didn't. You could be the man to show them up

[KaneTW]

About the only thing I can think of for water quality that would kind of work in that manner is UV spectroscopy, and even that is super broad (and rather expensive).

What they're probably doing is just measuring the conductivity of the sample, which is utterly useless.



E: "Doesn't have to touch the water to get a reading" OK, this is total bullshit. I guess you could do some sort of reflected light measurement with a UV source but I don't see how that'll give any decent results.

[magic]

Journalists will cover everything these days for a bribe from PR droids.

The chick at the river is of course as stock photo, as you would expect given the filename which they didn't even bother to change
https://www.dreamstime.com/stock-photo-hiker-woman-drinking-water-river-creek-hiking-zion-national-park-happy-female-taking-break-fresh-image44158426

The testimonials are laughable

since our new neighbors started to use our water source, we have suffered from unexplained pain. The TestDrop Pro has consistently shown that our incoming water is very bad

I'm only surprised that they managed to score those successful tests of protein detection at borderline homeopathic levels. I wonder if all their devices are exactly identical or perhaps they are "a dynamic startup" and "constantly iterating", read: "every device we show is tuned for the particular tests it will be subjected to".

Theranos 2.0?
At least it's not a fucking "Woman In Tech vs Hateful Misogynist Crusaders" this time around

OTOH, they are based in Israel so good luck going after them.

[Domagoj T]

Mrs EEVblog who is a water quality scientist and published in the field of water contamination just looked at the web page stared in disbelief and walked away.

Would she be up to joining you in front of a camera for this one debunk? I'd think it would make a great video to show how this thing works (or doesn't) from electronics perspective and then what it takes to actually analyze water, assuming she has access to the equipment and can arrange a bit of filming in her lab.

[ebastler]

About the only thing I can think of for water quality that would kind of work in that manner is UV spectroscopy, and even that is super broad (and rather expensive).

What they're probably doing is just measuring the conductivity of the sample, which is utterly useless.

E: "Doesn't have to touch the water to get a reading" OK, this is total bullshit. I guess you could do some sort of reflected light measurement with a UV source but I don't see how that'll give any decent results.

No, that's not how it works (or claims to work). It responds to charge build-up on the plastic container. You might consider reading this thread, the web site, and/or the patent cited earlier in this thread. 

Can this work reliably, and provide a measure of water quality? I don't think so. Even if we assume that they can get a charge reading via capacitive sensing, and limit the discussion to those substances which do have an effect on charge build-up -- why would the "critical" charge be the same for all kinds of contaminants? It's a simple scalar measurement, after all, while different contaminants will likely vary a lot in toxicity.

[magic]

It takes a few seconds so it may not be just a single scalar measurement.

But I also doubt that they are the first to study electric fields around chemicals so somebody else would already be building such detectors if it were so simple 

And they didn't patent the measurement process itself, only some general mumbo-jumbo about testing water quality with electric field, so perhaps there is nothing there to patent

[KaneTW]

I'm not claiming that's how it works, I'm claiming the way I outlined is the only way I can think of to maybe measure something remotely relevant.

What they're actually claiming to be doing is hopeless and can maybe detect distilled water unreliably at best.

[daqq]

I saw this the other day on BBC Click! and whilst they tend to be non-judgemental I'm fairly sure they wouldn't show it working if it didn't.

Oh dear, you havent been on the internet for long, have you?

[PlainName]

BBC Click! is not the internet.

And, in answer to your patronising put-down, I am an intertubes n00bie of only 28 years or so. How is that relevant?

[daqq]

BBC Click! is not the internet.

A lot of, if not most of the current science and gadget 'reporting' is just some 'reporter' regurgitating the marketing presentation from cool sounding startups without any critical thinking applied to the process. Just look at the coverage Solar Roadways, Triton Gill, Fontus or any number of obviously awful designs got - all of the media coverage was high praise for an innovative design that would save the world! The fact that they could be disproven or shown absolutely impractical by a few quick calculations by any reasonable engineer did not matter.

And, in answer to your patronising put-down, I am an intertubes n00bie of only 28 years or so. How is that relevant?

That it's really an extremely optimistic assumption that tech reporting "wouldn't show it working if it didn't". You need to work on your cynicism.

[KaneTW]

I've looked at their site again.



This is such a humongous array of different compounds that you'd need a well equipped analytical lab to determine all of these with any degree of selectivity.

You can't detect all those different metals (at safety-relevant levels) without specific reagents and a whole bunch of assays or a mass spectrometer.
Somehow Hg is different from HgCl2, implying that "Mercury" doesn't implicitly stand for mercury salts.

The entire bottom section is just "scary sounding IUPAC names" (and somehow 4-chlorophenol is a small organic compound while 2,4,6-trichlorophenol deserves a separate bullet point).



Basically the only thing I'm seeing is that they do a capacitance measurement and any deviation from some threshold is "bad". They're not testing for any of the above things, they're just measuring capacitance (which depends on conductivity, which depends on ion concentration).
I bet if you do a test against DI water and DI water with anything that doesn't dissociate in water, it won't even notice it. Calling BS on trihalomethanes especially. There's 0 chance it can detect chloroform in relevant concentrations.
Totally useless as a test, as water with high harmless ion concentration will register as bad, while low-conductivity but toxic water will register as "fine".

[PlainName]

optimistic assumption that tech reporting "wouldn't show it working if it didn't"

I specifically mentioned BBC Click! because it isn't common or garden 'tech reporting'. Yes, it is fairly superficial due to the format, but they do tend to point out faults where appropriate.

[PlainName]

high harmless ion concentration will register as bad

"Better safe than sorry". No worse than sell-by dates.

low-conductivity but toxic water will register as "fine"

But that would be very bad indeed.

[KaneTW]

https://youtu.be/Gvh_Fkg9VJw?t=904

"chance of contamination: 40%" for safe-to-drink water.
"90%" for water with some unspecified additive.

And then the host goes "does seem to work" after the most unscientific test you could possibly do.
Not looking good for BBC Click here.

[magic]

BBC Click
A guide to all the latest gadgets, websites, games and computer industry news.

Sounds like startups to me. If you have any familiarity with that culture, you would know that it's all about bullshit, hype, young naive millennials trying to change the world, and then some more lies and defrauding the investors, and the investors putting up with all that for a chance of attempting to establish a monopoly by dumping the product for the first few years.

Those journalists can't bee too critical or they wouldn't specialize in reporting on that industry in the first place.

In similar vein, TechCrunch

It’s that simple to operate; I did it myself successfully after one or two tries. It’s so simple, in fact, that I was suspicious. I thought it might be some kind of spectroscopy, but where was the emitter? Why did you have to move it, if not to create some kind of doppler effect?

Turns out that the whole thing is based on the electromagnetic fields that surround everything. Water creates its own local field, which is measured by moving the TestDrop through it, and it turns out that clean water emits a slightly different field than water with lead or chlorine in it, water with E. coli, water with dissolved animal matter and so on.

The device has been subjected to third-party testing, two reports from which I read; the thing really works. It detected even tiny amounts of lead and protein instantly and with 100 percent accuracy and no false positives or negatives.

[PlainName]

And then the host goes "does seem to work"

What's wrong with that? He didn't say it deffo works, he correctly stated it seemed to work. And he told us he already knew which beaker was the fixed one so we're well aware of the failings in the demo. Nevertheless, I would trust him 100% to have not faked that segment. If he'd had to do it several times before getting the 'good' clip, he would have said so (they have done that previously).

[KaneTW]

I trust that it works as he'd shown. But it's just nothing more than smoke and mirrors due to how the test was structured.

That TechCrunch quote though, wow. I'd love to see those reports, especially on selectivity. "Yeah, we added lead acetate to DI water and it detected something"

[magic]

I suppose it's the same third party tests that are linked on the company's front page.

[KaneTW]

Ah, yet another website that doesn't load properly without all third party scripts enabled. I found them.

https://www.lishtot.com/reports/180108%20Lishtot%20TestDrop%20Pro%20-%20PFOS%20-%20FINAL.pdf
https://www.lishtot.com/reports/180105%20Lishtot%20TestDrop%20Pro%20-%20Protein%20-%20FINAL%20-%20AB%20AK%20Signed.pdf

As expected: HPLC grade (ultrapure) water as negative control, sensitivity on "high", no selectivity tests. And a nice legal disclaimer at the bottom.

[PlainName]

I trust that it works as he'd shown.

Exactly! That's not saying it works as advertised, but it does say to me not to write it off as impossible without looking any further.

[langwadt]

high harmless ion concentration will register as bad

"Better safe than sorry". No worse than sell-by dates.

here they have been changed to "use before" for stuff that might make you sick after the date and "best before" for stuff that will probably be fine but maybe not as good after the date

[PlainName]

They've moved to that here too, at least in places. But 'sell-by' is well-known, covers stuff that doesn't really need to be sold by then as well as stuff that does, everyone knows what I mean by it, and in full flow one tends not to keeps pausing to dot Ts and cross Is.

[Daixiwen]

I think the ultimate test would be to record a video with the CEO, give him two glasses of water, tell him that one of them is infected with E. Coli and ask him to drink the good one 

[EEVblog]

I think the ultimate test would be to record a video with the CEO, give him two glasses of water, tell him that one of them is infected with E. Coli and ask him to drink the good one 

https://youtu.be/BGX4nMrnxg0?t=72

[EEVblog]

The CEO has emailed me, I've invited him to join here and defend his product. This will be fun 

His two best proofs are:
1) The usual blah-blah about customers in 70 countries being the "proof" 
2) They have an Australian company that tested it and are selling it, so therefore it must work!
https://www.optimosgroup.com/products/testdroppro-personal-water-tester/

[amyk]

"richlorophenol" - that's rich!

The fact that they don't even bother proofreading their site already says volumes about how much they should be trusted

[magic]

https://www.lishtot.com/coming_soon.html

They also have an improved version which can list the exact contaminants it found and their concentration level. Coming soon in 2018

Of course you must sell your soul to The Cloud and they monitor your every move to sell the data you collected to third parties

edit
On the upside, at least they have some marketing team which monitors the Internet for threads like this. Savvy

 

[Cyberdragon]

On the upside, at least they have some marketing team which monitors the Internet for threads like this. Savvy

 

Perhaps they saw what happened to the "digital electricity" guy, saw that this is 100 times more BS, and realized immediately what would happen to them.

Come on scammers...show yourselves...we have our logic cannons armed and ready.

[Xenoamor]

I'm not a chemist but wouldn't this at least need to be placed in the water to even stand a chance at doing something?
It's Thanos 2 for water, with wireless tech!

That gif on their site looks sketchy as well. Looks like they hold the button down and it blinks and then when they release it it shows the "results", very easy to fake but if they're selling something it should be a laugh to have a teardown of it

[taydin]

Question to the folks here that know water quality measurement: What kind of unsafe contamination CAN BE detected using an electronic device, and without water contact? (not this device, I'm talking generally).

[wraper]

Question to the folks here that know water quality measurement: What kind of unsafe contamination CAN BE detected using an electronic device, and without water contact? (not this device, I'm talking generally).

You can use XRF to detect chemical elements (not chemical compounds) without contact. But its problematic detecting elements with low atomic number and very small concentrations for point-and-shoot systems. Also you can detect radiation, though without close contact sensitivity will be crappy.

[fcb]

UV fluorescence can detect stuff that absorbs/re-emits UV  light (oils work well). Absorption spectroscopy (using light) is quite a common technique for water analysis.  You can even do atomic absorption spectroscopy.

[Cerebus]

Question to the folks here that know water quality measurement: What kind of unsafe contamination CAN BE detected using an electronic device, and without water contact? (not this device, I'm talking generally).

I know nothing more than the average man about water quality measurement specifically, but I've a fair grasp of some of the potentially applicable laboratory techniques from Physics, Chemistry and Biochemistry.

You can probably detect more than one might at first think with 'non-contact' methods, but not necessarily economically.

Things I can think of offhand - most of these require you to contain a sample in a flow tube or sample cell so are non-contact only in the strict sense that the sensors aren't in contact with the liquid under test:

• Radioactivity - scintillation, Geiger-Müller tubes, semiconductor sensors
• Turbidity - i.e cloudiness - optical scattering sensors
• Colour changes visible and otherwise - optical spectroscopy basically. Everything from obvious visible 'off' colours like iron contamination to compounds that can be reliably detected via (relatively expensive) IR and UV spectroscopy. IR and UV spectroscopy are standard tools for chemists and can identify potentially a large range of compounds.
• Total dissolved solids - electric field measurements. This seems to be the territory of the device in question. Measurement by the effect of solutes on the field behaviour of a water sample would be incredibly fickle and inaccurate compared to the standard method which is resistance/conductivity testing using contact probes.
• The final non-contact detection techniques I can think of are Electron Spin Resonance and Nucleo-magnetic Resonance. (ESR and NMR). Both quite expensive generally requiring cryo-cooled magnets but ESR is potentially quite a powerful technique for identifying substances. Chemists use ESR to watch reactions in progress.
• If you're going to allow vapour phase sampling as 'non-contact' I think we could include Mass Spectrometry and Gas Chromatography both individually and in tandem as GC/Mass-Spec. Massively powerful technique for identifying mixtures of compounds, expensive, with uncontrolled sampling I suspect this would be prone to contamination from things in the sample stream. Stretching the 'non-contact' idea a bit here.

The problem here is differentiating between harmful and harmless contamination. Turbidity testing will tell you if a sample's cloudy, but not of that cloudiness is from some particles of harmless chalk (calcium carbonate) or harmful white lead oxide. Spectroscopy might reliably detect and differentiate some harmful organic compounds from harmless. None of these techniques could tell the difference at any useful level of sensitivity between harmless soil bacteria and harmful cholera bacteria, ditto tobacco mosaic virus or a T4 bacteriophage* and ebola. 

* My favourite type of virus:


A visualisation of a T4 bacteriophage about to infect a single bacterium.

Edited to add: Also available as a cuddly toy

[EEVblog]

You can probably detect more than one might at first think with 'non-contact' methods, but not necessarily economically.

Or reliably with measurement of charge on a surface, and that's why this product and method is ridiculous.
Even if the contaminants produced the charge as claimed, to try and reliably detect the charge on a cups surface is comical.
This is why their own website and videos are filled with finicky details of how not to screw up the measurement.

[Cerebus]

You can probably detect more than one might at first think with 'non-contact' methods, but not necessarily economically.

Or reliably with measurement of charge on a surface, and that's why this product and method is ridiculous.
Even if the contaminants produced the charge as claimed, to try and reliably detect the charge on a cups surface is comical.
This is why their own website and videos are filled with finicky details of how not to screw up the measurement.

I'd go further and say that detecting ionic changes by measuring surface charge falls into the 'plausible, but not proven' category. If it works it will be, as I said previously, "incredibly fickle and inaccurate".

That their method insists on having a fresh, clean, unused plastic glass handy, but their publicity shots show people kneeling to test river water out in the wilds (presumably in deepest, darkest, adventuresome Africa) is an interesting contrast. I wonder how they would defend that particular juxtaposition.

When I was younger and my knees still worked properly I spent quite a lot of time walking in the wilds away from civilisation and given the choice between a couple of pairs of dry socks with something to eat or a stack of unused, clean plastic beakers, I know which I would choose to cart around with me. Moreover, anyone who's spent any time in the wilds will know that, depending on where you are, the contents of your rucksack* are either permanently damp, or full of dust. Not much hope of keeping your precious stack of virgin beakers clean and/or dry.

Perhaps I'm a poor man to judge - I've spent the majority of my backwoods walking days in either Scotland or Wales, both places where if you need clean fresh drinking water you just point your face upwards and open your mouth. 

* When I was typing this I somehow managed to type "rucksuck" instead of "rucksack". Given the amount of water that some of mine have accumulated inside their bottoms I think the typo might be more appropriate.

[Lishtot]

Some healthy skepticism has been raised on this forum regarding the Lishtot TestDrop Pro and its method of performance.  This post is to help explain how the device works and the science behinds its design.

Water has been shown to be a triboelectric material, namely it can deliver electrons (primarily from hydroxide ion, it would appear) to a wide range of materials.  Below are links to two academic papers and one US patent that are based on the triboelectric behavior of water with plastic materials.  It is noted in the first reference that water would be only less positive than air in a proper ranking of materials in a standard triboelectric series.

https://www.sciencedirect.com/science/article/abs/pii/S030438861630002X
https://www.sciencedirect.com/science/article/abs/pii/S0304388611001100
https://patents.google.com/patent/US4594553A/en

As noted in the above references, water imparts a charge on plastic materials by virtue of contact between water and the same, just as a person gets charged up by walking on a rug in the winter.  The static charge deposited on the plastic has associated with it an electric field.  The prior art work cited above measured the charge delivered to the plastic generally with two electrodes.

One important outcome from the second reference was the inadvertent realization that the size of the field associated with charge delivered would be related to what was present in the water.  The graphs below are from the Ravelo paper:





Note the Y-axis labels.  The distilled water has a range from 0-250 mV, while the "ordinary water" has a scale of 0-40 mV.  The Ravelo paper was the first to give an inkling that the amount of charge transferred from water to a plastic material was directly related to what was present in the water, or in other words, water quality.
The TestDrop Pro currently has three US patents associated with it:

https://patents.google.com/patent/US9453810B2/
https://patents.google.com/patent/US10240182B2/
https://patents.google.com/patent/US10060895B2/


The basic function of the device is to run an unpowered electrode through the electric field generated on a plastic surface by electrons delivered from water.  The device makes 100 measurements per second and by moving one device towards a cup with water, we can get a large number of data points in only a few seconds.  As one would expect, as the device gets closer to a cup with water, the electric field becomes stronger. Below is a canonical graph of "clean water".  Note that as the device moves towards the cup (represented by time in seconds on the X-axis), the field becomes more and more negative.
 




This result is completely consistent with the data from the three prior art publications cited above.  Water is a triboelectric material.  When contacted to a plastic surface, it transfers electrons to the plastic.  A static electric charge will have an electric field associated with it.  An unpowered electrode moving towards a source of static electricity will record stronger and stronger negative electric fields as it approaches the charged-up plastic.

So we can make a measurement that is related to water's interaction with plastic.  This interaction has been seen with a wide range of plastic materials in different formats such as cups, bottles, pipette tips, disposable pipettes, etc. So how does one get discrimination?  Or more accurately, why would water samples give different signals that allow for identification of contaminated samples?

The TestDrop Pro is trained to identify clean water samples, namely those that have profiles similar to the one shown above.  One can program the device for different levels of sensitivity but the general concept is that clean water allows for a lot of charge to be transferred to a plastic container or material.  When water is contaminated, fewer electrons reach the plastic.  Why?  Essentially, there appear to be two different reasons:

1.  Inorganic contaminants such as heavy metals appear to coordinate hydroxide and make electron release to plastic more difficult.
2. Organic/biological contaminants bind to plastic near the waterline and due to their association with water via hydration and/or hydrogen bonding effectively extend the waterline higher up in the cup or container.

The current protocol for the TestDrop Pro has measurement of electric field in the open space over the waterline. The reasoning is that this is where one should fiend lots of electrons transferred by water after swirling, without interference of the now positive water and any contributions from whatever its contents may be.  If fewer electrons are transferred to the plastic, the resulting fields are weaker, as shown in the graph below for a sample that included sodium hypochlorite ("bleach") and ppb levels of lead as free divalent ion:
 


The algorithm associated with TestDrop Pro is designed to look for a certain percentage of a graph below a predetermined threshold.  In the above graph, the lead/chlorine sample does not pass that threshold and the sample would be flagged as contaminated.
As mentioned above, the water, missing electrons left on the plastic, becomes positively charged due to unbalanced hydronium ions (hydroxyl radical is neutral).  Below is a graph of approaching the cup, this time below the waterline:
 



Organic/biological samples tend to wipe out signal until the concentration of the contaminant is very low, ppb or less. Below is a graph for suspected carcinogen PFOS in tap water.
 


The performance of the device as a screen for inorganic and organic contaminants allows for rapid determination of whether a water sample needs further laboratory attention.  Below are links to testing performed outside of Lishtot's labs with the TestDrop Pro. We were not involved in any of the testing reported by Analiza or MIT. Additional companies and research labs have performed research under nondisclosure agreements.

https://www.lishtot.com/reports/180108 Lishtot TestDrop Pro - PFOS - FINAL.pdf

https://www.lishtot.com/reports/180105 Lishtot TestDrop Pro - Protein - FINAL - AB AK Signed.pdf

https://www.lishtot.com/LishtotAnalysis_Nepal_Experimental_Plan-Lishtot_TestDropPro_Jan201


For additional information, please visit https://www.lishtot.com/

[KaneTW]

This still doesn't really answer my questions about selectivity: Can you tell apart harmful from harmless contaminants? How well does this device perform using tap water or similar?

All of these tests were using ultrapure water as control and one single contaminant added. This is utterly unrealistic for any actual scenario. Even tap water contains up to 1.5 mg/l of total organic carbon.

[Cerebus]

You say:

The TestDrop Pro is trained to identify clean water samples, namely those that have profiles similar to the one shown above.  One can program the device for different levels of sensitivity but the general concept is that clean water allows for a lot of charge to be transferred to a plastic container or material.  When water is contaminated, fewer electrons reach the plastic.  Why?  Essentially, there appear to be two different reasons:

1.  Inorganic contaminants such as heavy metals appear to coordinate hydroxide and make electron release to plastic more difficult.
2. Organic/biological contaminants bind to plastic near the waterline and due to their association with water via hydration and/or hydrogen bonding effectively extend the waterline higher up in the cup or container.

Let's just take those as a given. I suspect that few of us who've thought about it have any quibbles that both the above conditions can make a measurable difference to the electric field generated by sloshing water against a dielectric such as the PET or Polypropylene that plastic cups are made of.

At issue is whether this device can differentiate between:

• harmful inorganic solutes such as lead, mercury, cadmium
• harmless inorganic solutes such as calcium, magnesium, chloride.

or differentiate between

• harmful concentrations of organic solutes such as PFOA or protein from pathogens such as cholera
• harmless concentrations of organic solutes such as ethanol or protein from non-pathogenic soil bacteria

If you can't make those differentiations then this is useless as a water testing device. Yes, it may tell you if the water is different from the sample it has been trained on but that isn't useful if the difference indicated has no practical effect as to safety, potability etc. I see no reason why this device, if it is sensitive enough to detect unacceptable lead or cadmium concentrations, isn't going to 'alarm' at hard water if it has been trained on soft water or vice versa.

The clear suggestion from your marketing materials is that this device is suitable for field use for testing water potability. How realistic is this implied claim when it requires a supply of clean, unused plastic cups/beakers to work - things that are difficult to obtain 'in the field'?

So, to sum up the two questions this raises:

• How is this device useful as a water testing device when it appears to have no selectivity?
• How does the need for a carefully controlled testing setup fit against your implied claims for this as an 'outback' field water testing device stack up against the realities of operating in the field?

[Lishtot]

Selectivity.  The device detects classes of contaminants, namely organic/biological and inorganic, generally but not exclusively (it can detect high levels of hypochlorite, fore example) heavy metals.  If the device detects the presence of contamination, it is detecting something that should not be in drinking water.  This does not necessarily imply danger.  For example, take clean tap water.  It will give a clean signal.  Take a sip and swirl again and you will read contamination.  Your saliva contains many organic compounds that would not be found in clean drinking water.  The sample is not dangerous but it is clearly adulterated.  The device has been tested against sodium, potassium, calcium, magnesium and other typical water components and has been found to be blind to them at concentrations in the molar or higher levels.  The device can alert a user to water that is contaminated.  And do so in a few seconds.  Alternative technologies either take much longer or provide a single answer, for example, a $15 lead test--it won't say a word about bacteria.  Dipsticks are excellent for testing water but we believe that we have a broader range and the device is good for 4,000 tests on a single watch battery.

As to the perfect water conditions used by the prior art scientists, we have on our servers over 200,000 tests run by 5,000 users in 67 countries.  This device has been tested with a tap, filtered, RO, bottled, air-generated, and natural water sources.  They tested distilled water; we test only the real thing.

[Lishtot]

At a first cut, there is no need to differentiate what type of heavy metal is present.  This device is a screen, it presents the first line of defense.  And as such, it can warn a user that there is something in the water that should not be there.  We have demonstrated sensitivity to heavy metals tested in 10-20  ppb (except for copper which is 1.3 ppm) and thus is useful for detecting at relevant concentrations. If a water provider or user wishes for more information, numerous professional labs can provide the same.

The issue of detecting dangerous versus innocuous things is important, but why would expect ethanol in your drinking water?  Safe bacteria would probably not release enough protein to give a signal (we have seen the same for certain strains of E. coli).  If you ppb levels of PFOA or enough bacteria-produced protein to give a signal, I think that you would be grateful to know about it.

As to the perfect water conditions used by the prior art scientists, we have on our servers over 200,000 tests run by 5,000 users in 67 countries.  This device has been tested with a tap, filtered, RO, bottled, air-generated, and natural water sources. 

[Lishtot]

In addition:

As to the issue of field testing, if you had a device, you would notice that there are three buttons.  The one-button is set for a lower level of sensitivity exactly for nature testing.  We are aware that naturally sourced water has its own "natural" contaminants such as dead skin, leaves, bugs, and more.  We are doing our best to provide meaningful information without allowing our users to drink potentially dangerous water.  Natural water is a challenge for us but we still believe that we can provide useful information to our customers.

[AVGresponding]

The graphs at the top for distilled and tap water show a pretty clear linear relationship between the flow rate and the measured voltage.

Therefore I think it's reasonable to assume that unless you can accurately control the flow rate, the reading won't be worth anything anyway.

I fail to see how "swirling" the water around the plastic cup can in any way be considered to be controlled, in a metrological sense of the word.

[Lishtot]

Swirling generates the charge. We have put into the parameters of the device a wide berth for differences in swirling. Still, 5-10 seconds is enough to generate a large field on a polypropylene cup. In our system, there is no flow issue, only time if swirling to generate charge. Walking on a rug is not the same as rubbing your feet hard on the same rug. Please test the device and see for yourself.

[EEVblog]

This still doesn't really answer my questions about selectivity: Can you tell apart harmful from harmless contaminants? How well does this device perform using tap water or similar?
All of these tests were using ultrapure water as control and one single contaminant added. This is utterly unrealistic for any actual scenario. Even tap water contains up to 1.5 mg/l of total organic carbon.

Yep. I'm sure that no one doubts the effect itself is real.
The thing that makes this product fall into the impractical bullshit category is the ability to accurately and repeatably measure this under real world conditions. The marketing claims will always exceed the ability of the device to deliver a practical result.

It's the age old trap that inventors of things like this fall into - they take some actual phenomenon and then proceed to think they can engineer a solution that will actually have a practical and marketable results. Doesn't matter if it's Batteriser, Home energy saver capacitors in a box, ultrasonic wireless charging, infrared wireless charging, Fontus self-filling water bottle, peltier device energy harvesting etc.

No need to test this product, it will eventually die it's natural death, guaranteed by the laws of finicky engineering.

But hey, good luck trying.
I won't waste my time testing this, I know exactly how it's going to go:
I use my surface charge voltmeter to try and measure charge build up on the cup, results are entirely inconsistent and finicky. I come to the conclusion that it kinda-sort-maybe works if you hold your tounge at the right angle under a full moon. Manufacturer claims a new model till make it more robust. I retest that, with similar results. They say I'm not doing it right and just point to the thousands of people using as proof it "works".
No thanks.

[wraper]

They say I'm not doing it right and just point to the thousands of people using as proof it "works".

Those who drank some dodgy water after testing it as OK probably cannot leave feedback from their grave 

[wraper]

BTW don't forget about Theranos with their magical blood test and former billionaire bullshiter Elizabeth Holmes now facing criminal charges.

[SiliconWizard]

I also warned them about potential legal issues earlier... I hope they take this seriously. As always, one of the best "tests" you can do is ask the inventors to give the device to their kids and tell them to go camping without bringing any water, and just tell them to use the device to know what they can safely drink. Let's see how confident they are then. (And to paraphrase one famous song, I hope they love their children too. )

Sure the idea may be interesting as Dave said. But the practicality of it all?

As was discussed above, even if the thing sort of "works" under certain conditions, how can the end-user be certain some random water will be safe to drink with this, as 1/ the reliability of the measurement with "real-world" water is still not completely certain and 2/ as was discussed earlier, it doesn't seem to be able to detect some contaminants. Even if you clearly write which contaminants can be detected and which can't, how is it going to be useful in the field? That was one of my main questions here. How is the user going to know what kind of contaminants the water they want to test may contain? That is kind of a non-sense from a usability POV. Unless of course I missed something.

[magic]

Exactly. I wonder if the CEO would drink any clear liquid flagged as safe by the retail version of this gadget. Heck, even any water sample from wild nature flagged as safe by this gadget. Bonus if the exact same physical unit of the gadget can also detect safe surface water so that you don't die of dehydration

Natural water is a challenge for us but we still believe that we can provide useful information to our customers.

Nice that you admit so, but that's not what you advertising suggests. May we have a screenshot in case shit goes viral and the website is pulled down or edited? There we go



Is it just me or does it look like you imply that MIT found this device to be safe with any water found in the wild? Defo looks like false advertising to me

All the independent certifications you link also have disclaimers that no suitability for any purpose is implied by the lab issuing them. You have zero independent testing to back up your marketing claims as far as we know.

Some healthy skepticism has been raised on this forum regarding the Lishtot TestDrop Pro and its method of performance.

Yes, folks are accusing you of fraud and called your product "a scam"

[Lishtot]

The device works and has been shown to work by researchers, companies, groups, and individuals who have no relationship with Lishtot, and for your question, YES - the CEO would drink any clear liquid flagged as safe by the retail version of the TestDrop. (The difference between Theranos and Lishtot is that Theranos hid all their technology and we are the opposite, we have thousands of customers all over the world that tested it by themselves).

The device is not perfect and like all such devices have an error rate of false positives and negatives, which we have found to be less than 10% at the concentrations of contaminants shown on our website (clean water shown as not, less than 5%).  That said, our performance is similar to water testing kits in this market and offers advantages of re-usability, speed, and breadth of contaminants to give a warning to a user. For more on water testing kits, please see this link: https://www.lishtot.com/Water-testing-kit.html
 
The disclaimer on our website is standard - we believe in our product and stand behind it.  Do you know that ALL water filter manuals state the following or thereabouts:
"Do not use with water that is microbiologically unsafe or of unknown quality without adequate disinfection before or after the system." So why should a customer buy a filter if he already knows that the water is safe to drink? The only way for a user to test the water in point of use today is with the TestDrop.
 
In many of your posts, you have expressed concern for the user who relies on our device, drinks and then dies.  We know that people have a serious concern regarding their water quality. The situation today is that they use water filters (which come with the above disclaimer) and other solutions with no information as to the actual water quality. How a user can know when should he replace the filter? The TestDrop is another tool in the arsenal to know if the tap water is safe to drink or if the filter should or shouldn’t be replaced.   

As you are all scientists and engineers, we believe that the best way to look at a new product is to test it rather than to just discuss it.  As such, we suggest that you consider purchasing your personal water tester at www.lishtot.com in order to test the device yourselves.  If you have additional questions, please avail yourself of the contact options on the website.
 

[AVGresponding]

As you are all scientists and engineers, we believe that the best way to look at a new product is to test it rather than to just discuss it.  As such, we suggest that you consider purchasing your personal water tester at www.lishtot.com in order to test the device yourselves.  If you have additional questions, please avail yourself of the contact options on the website.

Sorry, I recently spent all my money on some magic beans.

[EEVblog]

The device works and has been shown to work by researchers, companies, groups, and individuals who have no relationship with Lishtot, and for your question, YES - the CEO would drink any clear liquid flagged as safe by the retail version of the TestDrop. (The difference between Theranos and Lishtot is that Theranos hid all their technology and we are the opposite, we have thousands of customers all over the world that tested it by themselves).

The device is not perfect and like all such devices have an error rate of false positives and negatives, which we have found to be less than 10% at the concentrations of contaminants shown on our website (clean water shown as not, less than 5%).

 

[EEVblog]

Exactly. I wonder if the CEO would drink any clear liquid flagged as safe by the retail version of this gadget. Heck, even any water sample from wild nature flagged as safe by this gadget. Bonus if the exact same physical unit of the gadget can also detect safe surface water so that you don't die of dehydration

Natural water is a challenge for us but we still believe that we can provide useful information to our customers.

Nice that you admit so, but that's not what you advertising suggests. May we have a screenshot in case shit goes viral and the website is pulled down or edited? There we go

(Attachment Link)

Is it just me or does it look like you imply that MIT found this device to be safe with any water found in the wild? Defo looks like false advertising to me

Yep. Australian authorities for example take a very strict view on demonstrably false advertising like this and ban the import of many products each year as a result. Especially when it involved public health. And they have a distributor here, hmm...
They had better be very sure of their claims in this regard.

[magic]

Australian authorities for example take a very strict view on demonstrably false advertising like this and ban the import of many products each year as a result. Especially when it involved public health. And they have a distributor here, hmm...

What are their distributors?

Looking at the website I was under impression they just sell direct.

YES - the CEO would drink any clear liquid flagged as safe by the retail version of the TestDrop.

Well, any liquid is any liquid. How would you estimate the detection threshold of sarin on the lowest sensitivity setting? How many HIV viruses or trichinella larvae extracted and washed from contaminants in clean distilled water before it trips in the "river" mode?

It seems that nobody performed such controlled evaluation, and the "success" stories in E. coli might well be based on accidental detection of something else in the same sample of unpurified water. These bacteria grow in s**t, after all.

[EEVblog]

Australian authorities for example take a very strict view on demonstrably false advertising like this and ban the import of many products each year as a result. Especially when it involved public health. And they have a distributor here, hmm...

What are their distributors?

They bragged about having this Australian distributor who sell it, and they evaluated it so it must be good:
https://www.optimosgroup.com/products/testdroppro-personal-water-tester/

Reminds me of that aussie guy who was stocking the Batteriser 

[magic]

The "TOS" is quite a contrast with the usual propaganda they show on their website and give to the media

Look guys, it's a free energy device but we cannot promise that it will be able to perform work, that's the catch.

[SiliconWizard]

Yeah, anyway... whatever the accuracy of the device, if it's not "perfect" as they say and has a typical miss rate of 10% (and that would have to be proven, but let's suppose it does), I still don't know how they can claim it's usable for the typical "traveler" as is advertised. 10% is already huge.

Now this device may be on par with other water testing devices on the market, I don't know much about this market? But I still don't know how useful that would really be in practice.
Unless maybe you're very unlucky to stumble upon water that contains a lot of heavy metals or something - which should be relatively rare in wild nature? - the safest way of handling unknown water is to boil it. If I had a testing device that was not 100% certain, I would still boil the water. So I'm not sure what the benefit of this would be.

[PlainName]

I don't know much about this market? But I still don't know how useful that would really be in practice.

How can you pass judgement, then?

From my point of view, I would want something that erred on the side of caution. That is, 10% OK water being passed as possibly unsafe is fine, but 10% of unsafe being passed as probably safe is not. As an engineer, you should know there are unavoidable tolerances in everything and it's just a matter of making sure they vary the bit that won't be a problem. Have they managed that with this product? Well, no-one seems to know. Plenty of guessing, though!

As an aside, I read about the bloke that got a Nobel Prize for co-inventing the Li-ion battery. Apparently, when they were developing it there was a lot of resistance (ho-ho) and no-one took them seriously. Maybe if the likes of thunderfoot had been around then we might still be cursing the memory effect in Ni-Cds . I loath Sony as a company, but got to admit they did good on that one.

https://www.theregister.co.uk/2020/03/02/canadian_firm_to_develop_goodenoughs_new_glass_battery/

[EEVblog]

I don't know much about this market? But I still don't know how useful that would really be in practice.

How can you pass judgement, then?
From my point of view, I would want something that erred on the side of caution. That is, 10% OK water being passed as possibly unsafe is fine, but 10% of unsafe being passed as probably safe is not. As an engineer, you should know there are unavoidable tolerances in everything and it's just a matter of making sure they vary the bit that won't be a problem. Have they managed that with this product? Well, no-one seems to know.

Anyone with a clue about measuring surface charge accurately in practical scenarios knows this is a complete folly

[SiliconWizard]

I don't know much about this market? But I still don't know how useful that would really be in practice.

How can you pass judgement, then?
From my point of view, I would want something that erred on the side of caution. That is, 10% OK water being passed as possibly unsafe is fine, but 10% of unsafe being passed as probably safe is not. As an engineer, you should know there are unavoidable tolerances in everything and it's just a matter of making sure they vary the bit that won't be a problem. Have they managed that with this product? Well, no-one seems to know. Plenty of guessing, though!

As Dave said, it's highly dubious from the principle itself that even 10% uncertainty could be achieved from an experimental POV, so what they claim is already dubious. And anyone with a clue about not only engineering, but also about the application, even if they don't know precisely the whole corresponding "market", can indeed question the whole thing. It's not even a "judgement" per se, it's a series of questions.

But my additional point was about usability, I thought I was clear enough about this in a couple posts already: I don't know what kind of customers/use case that would target. If this is not clear, maybe read again what I wrote.

Point is: if you have even 10% uncertainty, would you drink the tested water? Again, when you're in doubt out in the wild, your best bet is to boil the water, not test it with any means (that unless you have a full portable lab will never be safe enough IMO.) The only case I see this (or other similar testers) could have a use would be if you had really NO other choice: out in the wild, completely dehydrated, nothing to start a fire with whatsoever (so no way to boil the water). That sounds more like for a critical survival scenario than for your typical camping/hiking/traveling scenario. Then again, if you thought about bringing this tester with you in advance, you probably thought about taking with you what it takes to start fires and a few small containers to boil water and cook things. And if you have that with you, the only way this tester (if it indeed lives up to its claims) could be useful is to handle cases in which boiling water would not be enough to make it safe to drink, and in which this tester could detect such contaminants (that can't be eradicated with boiling) with a high enough certainty. Is that very common?

So to sum it up, I see two major issues with this project: one of course from a practical engineering POV (as others have stated), and another from a usability POV.

Now if you have clear answers to what I raised above, especially a real-world scenario in which this tester would be better than the usual means - and that wouldn't be a catastrophe scenario in which you're not very likely to have it with you either - I'd be curious.

[Cyberdragon]

We sunk their Scammership (I think they're gone).

Unless they want another logic torpedo fired at their ass.

We're still offering the CEO to "drink any clear liquid", but we get to choose the liquids and one of them will be toxic.

[PlainName]

Point is: if you have even 10% uncertainty, would you drink the tested water?

As I said, it depends which way the uncertainty is. If it might give a false negative (that is, reported safe as unsafe and not the reverse) then why not?

[magic]

It is essentially a single scalar measurement of some complex function of conductivity and ionic content of water.

It can't guarantee that your water is safe unless you set detection threshold to "distilled water". And that's assuming that we can't find some other pure liquid with identical / close enough triboelectric properties.

It may be useful in limited circumstances in the same way that smelling or testing pH may tell you something about the sample.

edit

Now, you'll never have to guess about your water's drinkability again.

Not only fraudster and bulshitters but outright double faced liars too

But whatever, I went to their website for the testimonies. So there we go:

I tested water in 3 new buildings and the water was good and when I tested water in an old building, I received Red. It seems like your device is accurate.

since our new neighbors started to use our water source, we have suffered from unexplained pain. The TestDrop Pro has consistently shown that our incoming water is very bad

I tested tap water and it was red, then I tested my filter’s water and it was blue. The TestDrop Pro gave me the evidence that I needed that my filter was working properly.

I am originally from Germany but now I am traveling in Indonesia. I am in a very rural place and the TestDrop helps me to know which water can I drink.

I compared the TestDrop Pro to a TDS meter, and I found the TestDrop Pro is more accurate and more sensitive than the TDS meter.

I always wanted to know if the stream near my house was safe to drink, now with the TestDrop I know that sometimes the water is good and sometimes not. That makes a lot of sense.

All from ignorant users who simply trust the gadget and are satisfied with whatever general correlation they found between its decisions and the visible or presumed "clarity" of the water. Affirming people's own judgments can get you quite far

[PlainName]

It can't guarantee that your water is safe unless you set detection threshold to "distilled water".

And...?

If that's what it takes to say the water is safe, and it does that, what's the problem?

Suppose someone proposes a system whereby they'd send a signal from 11,000 miles away and you could decode the data on that with the gubbins that fit in a wrist watch... imagine the reception they'd get. Data from a -130dB carrier! Ha! On a watch! Yet not only do we do that easily but it's relied upon in life or death situations.

Here, we've gone from this thing relying on pure magic to grudging acceptance that there is an effect. Now it's the size of that effect. Can't wait for the next installment

JOOI, what would convince anyone here that it works? Or really doesn't work? I mean, suppose SiliconWizard (to pick someone at random, no offence meant!) decided to check it out and got one. Clearly, if he then said it's rubbish you'd be crowing "Told you so" and the like. But what if he said it seems to do the biz? Would you believe him - who you don't know - any more than the possibly fake 'recommendations' on the website? What would he have to do for you to believe him, or isn't that a possible scenario?
 

[Cerebus]

What would he have to do for you to believe him, or isn't that a possible scenario?

What would we want to see? Answers to the selectivity questions with specific concrete substances mentioned - they have mentioned a few harmful things it will detect, they have completely blanked any questions about it registering false positives on harmless substances. Answers to the sensitivity questions with specific concentrations mentioned. Y'know, hard science with numbers instead of arm waving.

All we've seen so far is arm waving and some vague claims. If you want an electronics analogy it's a bit like this kind of conversation:

Them: We've designed a new type of low noise amplifier.
Us: Your published information on the design doesn't seem to make technical sense. How does it work and what's the SNR?
Them: Specially doped transistors. Lots of people have told us it's very quiet.
Us: OK, that's perhaps plausible but a bit unlikely. What's the SNR?
Them: Have we told you that we've lots of positive customer feedback? We can sell you one.
 
Us: WHAT'S THE FREAKING SNR?
Them: -- silence ---

If they come back and say something like "It'll detect lead at x ppm upwards, chromium at y ppm, cadmium at z ppm, won't false alarm on sodium, calcium or magnesium until w ppt or above. Will detect Vibrio cholerae at 3 CFU per litre, won't false alarm on common soil bacteria until 20000 CFU per litre. etc. etc." then we'll start believing them.

Literally the last thing they said to us was:

As you are all scientists and engineers, we believe that the best way to look at a new product is to test it rather than to just discuss it.  As such, we suggest that you consider purchasing your personal water tester at www.lishtot.com in order to test the device yourselves.  If you have additional questions, please avail yourself of the contact options on the website.

That sounds to me very like someone who knows they are skating on thin ice with the claims that they are making. If they won't publicly defend their claims then I have to conclude that is because they can't defend their claims and they believe that further pubic discussion will fail to substantiate their claims. I read their last words as "Shut up and give us the money. If you want any more answers please ask us in private where the answers won't be on public record."

[PlainName]

What would we want to see? Answers to the selectivity questions with specific concrete substances mentioned - they have mentioned a few harmful things it will detect, they have completely blanked any questions about it registering false positives on harmless substances. Answers to the sensitivity questions with specific concentrations mentioned. Y'know, hard science with numbers instead of arm waving.

That's not what I asked in principle. Take an average forum viewer who acquires one of these to check it out. What could he do to show that it works or not? Not what could a boffin with a lab do.

[Cerebus]

What would we want to see? Answers to the selectivity questions with specific concrete substances mentioned - they have mentioned a few harmful things it will detect, they have completely blanked any questions about it registering false positives on harmless substances. Answers to the sensitivity questions with specific concentrations mentioned. Y'know, hard science with numbers instead of arm waving.

That's not what I asked in principle. Take an average forum viewer who acquires one of these to check it out. What could he do to show that it works or not? Not what could a boffin with a lab do.

Exactly the same things. The burden of evidence doesn't change according to who is testing. Changing the evidence required is not objective science. Do that and the answer you get is entirely dependent on who you ask. That would rapidly become "every opinion is equally valid, no matter how uninformed" territory and before you know it Gwyneth Paltrow is the new Surgeon General.

It's not like we're asking for difficult science here. Just the ability to make a few solutions of harmless and harmful substances in water at various known concentrations and wave the magic Lishtot wand at them according to the rituals laid down by the Lishtot high priest. If it can't differentiate harmful from harmless either on substance harmfulness or concentration harmfulness then it's a sham, bunkum, horse-pucky, witch-doctoring.

This isn't 'boffin' territory, it's at a high school chemistry/physics experiment level of difficulty. That's why the lack of hard quantitative evidence from Lishtot is being taken as wilful bad intent - this is stuff that a well informed or tutored 15 year old could do in a high school lab. If you don't mind risking poisoning it's even kitchen sink chemistry. The entire equipment and supplies list is:

• measuring cylinder
• accurate weighing scales
• mixing rods
• clean unused plastic beakers as specified by Lishtot
• Lishtot device
• high purity water
• selection of soluble salts of
  • lead
  • cadmium
  • chromium
  • arsenic
  • thallium (just for fun)
  • copper
  • aluminium
  • iron
  • calcium
  • magnesium
  • potassium
  • sodium
• Optional: selection of harmless and harmful organic chemicals and biochemicals likely to be present in the water supply.

No rocket scientists or boffins required, no equipment more than a school chem lab or well equipped kitchen would already have. The only thing that tips this towards being a school rather than home experiment is the toxicity of some of the materials required and the likely response to you ordering a soluble arsenic  or thallium salt for 'home use'.

[PlainName]

OK, thanks. That's probably doable with a little effort (a quick check suggests I, for instance, don't have many of the salts to hand).

[KaneTW]

Don't need that many (and sourcing might be difficult if you're not a company/etc), but getting some e.g. lead salts is not particularly hard.

[SiliconWizard]

Point is: if you have even 10% uncertainty, would you drink the tested water?

As I said, it depends which way the uncertainty is. If it might give a false negative (that is, reported safe as unsafe and not the reverse) then why not?

Obviously. But what would make you think that this device could magically have a very low false positive rate, and maybe an highish false negative one, which indeed would not be a problem in itself? Better be overly cautious than the opposite.

But nothing from what I read here and from a purely experimental POV tells me that this asymetric error would even be possible, unless the device was so "sensitive" that it would likely detect any water as unsafe unless it was maybe pure distilled water.

To give them the benefit of the doubt, they may have found an innovative way (although we are a few here who doubt that) of getting a low error rate, but is this low enough to really be safely usable? That's what I'm highly doubting. If the device were targeted at professionals (meaning they have the knowledge to use it safely), as long as the claimed specs are not fraudulent, that would be alright. But targeting the basic user - traveler, hiker... that's a different story, and some of us have warned them about the risks they could face with that...

[magic]

The thing is that there are no specs So it's hard to even tell whether it works or not. It certainly does something, perhaps even repeatably so, but that's about it.

They came here and admitted it isn't 100% bulletproof and that natural water is "a challenge for them" but they "hope to provide useful information to the user".
And then they post on their website about knowing if any water is safe to drink, tested by MIT If that's not BS than I don't know what is.

This is not a technological problem like receiving a -130dB signal. The technical side of the device is known and they described it. The only "magic" left is a simple question: to what extent does the thing that they measure actually correlate with safety. The answer has already been decided billions years ago (or 6500 for you young earth fanboys), it's a matter of finding out.

I would buy one myself if it were an AliExpress gadget for $5 with shipping and tax evasion included, but I doubt anyone here is going to pay $50 plus overhead for a glorified electroscope.

[PlainName]

But what would make you think that this device could magically have a very low false positive rate

Nothing at the moment. I am taking task with the insinuation that it has to be 100% perfect in all respects, otherwise it's a massive fail. The same attitude is apparent when dissing all products in this section - if one can find the slightest non-positive, that's enough to shout "BUSTED".

As to whether we could know if this erred on the side of safety, I have no idea. I don't drink water anyway without scouring it with coffee or tea, so the concept of drinking the stuff is alien to me to start with