53 comments

[ 1.8 ms ] story [ 132 ms ] thread
This was fun to read. Anyone know where I might find the relevant equations/math?
Any undergraduate intro to fluid mechanics textbook will cover this material thoroughly. Looking at my bookshelf I have "Fluid Mechanics" by Frank M White from third year fluids back in the early 00s which treats this material in a chapter called Turbomachinery, building upon the preceding 10 chapters or so.

If you don't want to deal with calculus you can get most of the way there with the Bernoulli equation and friends.

Edit: this looks pretty good after a cursory glance: http://brennen.caltech.edu/fluidbook/

If you only get 26 feet of head, how do they manage to pump water from (say) the water tunnels that are hundreds of feet under Manhattan? Is it just that the pressure in the tunnels is well above 1atm?
The pressure in NYC's water tunnels brings the water up to about the 6th storey[1]. Beyond that additional pumps and tanks are needed, which is why NYC's skyline is covered in wooden water tanks (except for the newer buildings, which often have internal metal ones).

[1]: https://untappedcities.com/2021/01/22/how-does-water-tower-w...

That's not what I'm referring to. Yes, the water system is pressurized (as are all), but the tunnels themselves are ~500 feet below ground [1][2]. Water is pumped to valve chambers which are themselves far (~250 feet) below ground, then again to the surface.

The head pressure on a 500 foot column of water would be over 200psi. This article [3] says that the first leg of water tunnel three runs at over 300 psi, though, so maybe that's the answer. It just seems crazy.

[1] https://en.wikipedia.org/wiki/New_York_City_Water_Tunnel_No.....

[2] https://www.theverge.com/2013/10/19/4853636/underground-with...

[3] https://www.baltimoresun.com/news/bs-xpm-1998-08-21-19982330...

Pumps might only be able to suck water up 26 inches, but there is no such limit on how high they can push water.
I believe the suck-height limit is more than 26 inches. From memory (it's been a long time) the theoretical maximim for 1 atmosphere is 10m. Practically, because of inefficiencies a lot less.

Annecdotally I have a well-point (pump at top) that lifts water 2 to 3 metres. And I have an agricultural pump that sucks about 4m,and then pushes water up a hill the next 50 vertical metres or so. And its a pretty small pump.

>there is no such limit on how high they can push water.

Of course there's a limit. If there were no limit we'd have infinite free power generation.

You’re missing the fundamental difference in the power vs distance curve. In the case of sucking water, it flatlines and no amount of additional power can move the water another inch once you’ve exceeded the pulling power of a perfect vacuum.

Without looking up the actual equation, I would guess that this is never true (or at least practically never) when pushing water, because you’re not limited by atmospheric pressure.

When pushing you have to provide enough energy to move all the water in front of you, but if you do provide that energy the water will move. At some point I suppose the pressure will get so high the water will solidify in the pipe though?

>You’re missing the fundamental difference in the power vs distance curve.

You're missing the fundamentals of physics.

>At some point I suppose the pressure will get so high the water will solidify in the pipe though?

Before (way way way before) the water solidifies you will reach the maximum pressure that a given pump can deliver.

If you think the little 1/2 horse sump pump you can buy at home Depot will push a column of water up to the stratosphere then I have a bridge to sell you.

Who said anything about 1/2 a horsepower? I’m talking about the most powerful pump in the universe, both (known and unknown). How far up an unbreakable column can that pump push the water? And then, what if you make it just a little more powerful, can the water hypothetically be pushed higher?
If there's a limit on how much positive pressure a pump can generate to push water above itself, then what approximately is the value of that limit, and how would a pump capable of exceeding that limit allow you to construct a perpetual motion machine?
The limit depends on the pump. If you're pushing water uphill the weight of the water will eventually equal the maximum pressure that the pump can deliver and you won't be able to push any further.

This is basic high school physics.

You've completely misunderstood this this conversation. The limit on the height from which a pump can suck water is a limit that applies to all pumps, no matter how powerful. There's no comparable limit for pumps pushing water upward. Yes, each particular pump will have its own limits, but there's no comparable overarching fundamental limit.
Thank you, you’ve explained this better than I could
Isn't pushing water up doing some work? And then it coming down can do same or less work? Just because pump can push without limit, it doesn't mean it's free.
I think you are out by a factor of 12, it's around 26 feet. Or maybe you are thinking of a column of mercury in a vacuum which is about 26 inches.
Put the pump at the bottom and push the water up, don't suck it up from the top. Same way with deep wells.
Fundamentals of fluid mechanics by Munson et al is also good.
Tangent: seeing the title and the Honda url, I was certain this was going to be related to the engine air pump model[1]. Since engines are limited by the amount of air they can push through, you then think of the engine as an air pump (and try to optimize around that). Although this model is fairly commonly used in the automotive industry, details online are surpassingly sparse. Taylor[2] has a bunch of details on it.

[1] https://www.roadandtrack.com/car-culture/news/a29588/how-int...

[2] https://direct.mit.edu/books/book/4840/Internal-Combustion-E...

That's a very common occurrence outside of software.

I know a guy who is famous within his field for certain important discoveries. Later, he "discovered" something else, only to find out that it had been published decades earlier by people in an adjacent subject field, and that parts of industry had been using it for a long time. Years later, there are still no new resources about it.

If you know, you know. If you don't, you don't.

> This is similar to sucking on a straw

Totally random and off topic, but since it took me 40 years to learn the only 100% effective way to get rid of hiccups, this is it.

You need to create an extended period of negative pressure in your oesophagus, to de-activate the vagus nerve, or whatever it's called. The reliable way to do this is to have a bottle of water and a straw. Plug your ears if possible - helps but it's not strictly necessary. Crimp the straw somewhat with your teeth. Now spend 20 or 30 seconds slowly, but with a high pressure, sucking water through that straw and swallowing it. Remember, your goal is negative pressure in the oesophagus for an extended period.

It is the only reliable method and for some reason is almost completely unknown. Well, now you know!

</tangent>

> Crimp the straw somewhat with your teeth

Damnit. Can't do that with the glass and metal straws we have now...

> It is the only reliable method and for some reason is almost completely unknown. Well, now you know!

I'm going to patent a straw with a tiny hole on one end and sell it as an anti-hiccup-device.

Do you also know of a way to trigger hiccups so I can test my anti-hiccup-device?

Pumpkin seeds do it for me. There's something about the fiber in the shell that irritates the throat after a while. Eat enough and I'll get hiccups for hours.
Or, just try holding your breath for 30-45 seconds. I'm betting you've discovered a round-about way to achieve the same mechanism of action (raising level of CO2 in your blood). https://www.bbc.com/future/article/20140127-do-hiccup-remedi...
For that method, I have to clench my abs as hard as I physically can, and usually a hiccup will still force its way through.

It does work, but it's not easy and usually take 5+ attempt to stifle the one that forces through.

My (somewhat weird) roundabout way is to take a deep breath in and exhale slowly while imagining myself at a time (like say 5 minutes ago) when I did not have hiccups, but it requires concentration (usually 5-10 seconds).
Not sure where i read this one, but it's worked out great for me (someone who seems to get hiccups a lot).

    - breath in through your nose till you fill with air
    - hold your breath for 10 seconds
    - try to breath in again (will be a very small breath)
    - hold for 10s
    - try to breath in again (will be even less)
    - hold for 10s
    - try to breath in again (almost nothing)
    - hold for 10s
    - breath out
I solve hiccups by drinking water. It is an instantaneous solution for me.

I also have to say that sucking on a straw is a terrible analogy for pumps because people think they are pulling the liquid when they are pushing it up the straw.

> sucking on a straw is a terrible analogy for pumps because people think they are pulling the liquid when they are pushing it up the straw

How is lowering pressure by sucking on a straw a bad analogy for lowering pressure inside a mechanical pump?

I just start holding my breath a bit before a new hiccup arrives and swallow a couple of times, it usually does the trick.
Yup this, and even without swallowing. Hold that breath, press down hard.
The method I use has a medical name, but I can't remember it unfortunately (it's actually used for heart palpitations, but I found that it worked just as great for hiccups). It's basically lying on your back while holding your breath and applying pressure down on your lungs from both your chest/throat.

In my 20's I got a lot of hiccups, usually because I was drinking and talking at the same time; but due to the placebo effect eventually I would only need to lie on my back for 1-3 seconds, pretend to hold my breath and then I could get back up and it'd be cured.

Drinking a full glass of water very quickly has never failed for me, perhaps it does something similar to your method?
Same ultimate solution, but no need for any apparatus - lung packing. I learned this incidentally while I was into freediving as a teen.

Just take the deepest breath that you can, and then use your tongue and throat to pack more air into your lungs until you feel fit to burst - it may take 20-30 packs. Hold it for ten seconds, then slowly exhale - if you’ve done it right you’ll feel like a deflating air mattress. Hiccups will be gone. If they aren’t, you didn’t pack enough.

having small amount of solid sugar, let it mix with saliva, let it go down on throat, and then drink some water -- this has never ever ever failed me!
I'm in the process of researching to replace my swimming pool pump. My current one nominally runs at 2850rpm and has a nominal power consumption of 1650W, but I think I measured it at around 1100W. I imagine the maximum load is when starting and also if it has to "suck" with a large head. My pump is about 0.5m above the pool level. It is a 14 year old pump that is obviously just a single phase (240V 50Hz) AC induction motor.

The replacements are permanent magnet DC motor driven, with a variable speed (presumably frequency and voltage) drive can control speed from 600 to 2850rpm.

After priming at full speed for a few minutes (to purge air) it looks like you can then run it at say 1300 to 1800 rpm and save a coal truck worth of power. (I do have solar PV panels on my roof, but I am hoping maybe 1000kWh or more a year saving ).

Strange that Honda would put this out. Automotive water pumps are usually not vacuum pumps. They're down at the bottom of the cooling system pushing coolant, not up top trying to make a vacuum.

This is, inevitably, covered in one of those Jam Handy movies made for Chevrolet. "Water Boy", 1936.[1]

[1] https://archive.org/details/0799_Water_Boy_08_01_01_00

Uh, this is not about cars (I thought that as well). It's about water pumps for stuff like watering plants and stuff.
Honda has a popular line of engine powered water pumps, like you may use to empty a pool or a flooded basement.
Why there is no unit conversion with a flick of a button yet? It clearly looks that it has been translated from SI unit (e.g 1 bar or 14.7psi pressure at sea level)
Well (as opposed to all the HN posts about Rust and GoLang), this is a topic I actually know a smidgen about while running a solar water pump venture out of East Africa. AMA.
Nobody took advantage of the opportunity! I will now.

I am somewhat involved in a restoration project of a building in Italy (pet project, not my job, I'm a tech guy).

We are considering installing geothermal power to warm/cool the building. Anything of what you know applies to that?

There is a water canal nearby, and therefore there's a plan B that involves exchanging heat with the water in the canal.

I wish I could speak to geothermal, but it’s just a different industry & technology (maybe it’s like the the difference between architecting a chat app vs a ride-hailing app..? :) ). Yet +1 for the sustainability approach there, always a fan of those intentions =) Things I know revolve around the content of this video I produced: https://www.youtube.com/watch?v=1ZLkmrAvAzE

Relevance to original article, for example, are that there are ventures around here that focus on water suction, but their scope is practically limited to surface water. We’ve focused exclusively on submersible pumps so that farmers with wells can tap their groundwater for agriculture with us.

What do you use for energy storage for dark hours?
We avoid electrical energy storage currently. Couple of reasons: 1. Product has to be simple to be adopted effectively. Batteries or overhead tanks imply additional management overhead on the part of the farmer, and that management has to be instructed. Difficult to do in this market context. 2. Support - we’ve talked with customers of competitors who do feature a battery storage element, and they’ve complained about getting support when the battery was failing after some time. 3. Cost. We try to get customers onto the first rung of the income-generation ladder. Higher-income rungs can be achieved with incremental investment, and energy storage is a good example of this. 4. Agronomy: Water-logging plant roots consistently at night is no bueno: fungus, restricting nighttime gas transfer between the roots and soil.

That said, the case for storing energy during dark hours can still be a good one, provided 1-4 are satisfied/mitigated. Early morning hours are an excellent interval to irrigate over, and the solar irradiance at those hours is often still too low to utilize the pump’s full delivery & depth (head) capability.

I'm a bit late, but what's the technical setup? I'm curious about the panel, controller and pump. What's the wattage and parts etc. What's the final cost?

Finally, what problem are you solving for?

>Finally, what problem are you solving for? Smallhold farmer livelihoods that are underachieving relative to their capacity.

Solar water pump (SWP) systems are expensive to our target market of smallhold farmers when purchased with up-front cash. Since a SWP is an income-generating appliance for its user (apply water toward grow crops -> sell crops -> $$) we focus on how the farmer can sustainably make a down-payment and pay off the SWP on a rent-to-own basis. The market and industry space is widely distributed, so we target equipping distributors to accomplish the above with farmers in their respective geographies.

>what's the technical setup? > the panel Typically 120-200W collapsible solar panels that the farmer deploys daily.

> controller Our IoT controller that collects local environmental & usage data that equips distributors to monitor their productivity (and so their anticipated loan performance). > and pump. DC-submersible, capable of delivering shallow groundwater accumulated over recent rainy seasons. (Power level matches the panels minus losses). Difficult to go into more detail than that without exposing more about our specific go-to-market strategy rather than anything that helps bring more farmers and distributors to the table. At the end of the day, pump models & panel sizes run the gamut in any given country's market - we help distributors select among these in their local market rather than exporting one SKU to everybody.

>What's the wattage and parts etc. As above, with hosing & cabling sufficient for the groundwater depths involved. Accessories beyond that like sprinklers and a bag for carrying are generally appreciated by farmers.

>What's the final cost? We have offerings between $600-$1300 paid out over a year. Of course a distributor can choose whatever payout plan they feel their local farmer market will bear.

All of the priming and vacuum issues they describe is why I swear by submersible pumps. I use a pair of deep well pumps to move water up 70m, each doing a 35m lift, with a relay tank halfway.
having small amount of solid sugar, let it mix with saliva, let it go down on throat, and then drink some water -- this has never ever ever failed me!