Hacker News new | past | comments | ask | show | jobs | submit login
Drip Irrigation Design Guidelines (irrigationtutorials.com)
190 points by bryanrasmussen on July 16, 2017 | hide | past | web | favorite | 56 comments

I have a small irrigation setup that works somewhat well for my needs. I have a small pond/waterfall pump inside a rain barrel that pushes a decent flow to my emitters that water my lettuce and cucumber plants. The pump inside the barrel is connected to an outdoor Z-Wave switch that is ultimately controlled by my SmartThings hub.

The main downside is it's currently manual and water is limited in the barrel. The rain barrel can only hold so much water and at times, I find my ZWave switch will turn off, then back on, which eventually drains the barrel if I'm away. I usually turn it on/off on hot nights when I'm home so I can monitor it.

Next summer I'll take a different approach with some of the ideas from this article. Perhaps use the rain barrel only if it's full (using sonar sensors that ping off the water level), but use the main house line as the primary.

That hacker/tinkerer in me wants to ultimately build a system that waters based on weather patterns/demands. I'll save that for next year.

This seems way, way too complicated.

Just last weekend I visited a 30 acre University farm - where they teach agroecology and have graduates worldwide who are running farms. I noticed that they were just using normal water spigots at different locations on the farm, and they had timers of different sorts on those spigots to what looked like 1" poly tubing. When I asked about how they adjusted for all their different crops, they said they just adjusted the timer for a set of rows and used a sharp point to poke different sized holes in the tubing at the base of plants. No big deal. They said it sometimes takes a few adjustments, but it works fine. The have been doing this successfully for decades. Very easy and very low cost.

So one issue with what you posted is that it's a recipe for contaminating your drinking water. You need to have some type of anti-siphon device to prevent the back flow of water from your irrigation system into the house water supply. My guess is that somewhere the farm you are talking about had a giant and very expensive double check back flow preventor. Having designed and installed several irrigation systems I can tell you that pressure drop is a thing. To get the correct CFM, I had to install 1 1/2" mains so that the rear of my lot would have enough pressure to deliver the cfm I needed. Also, water hammer is a thing if you have lots of pressure. Your consumer grade fixtures will be destroyed by water hammer and your system will need lots of maintenance. Also, if you live around the system, water hammer is noisy and when your system trips on at 5am, it's going to wake you up. So while I don't doubt the farm did this, I'm not sure it's an appropriate installation for a home owner that just wants the system to work reliably for years with little to no maintenance.

Your water hammer reference reminded me of a setup I once saw. It used a hydraulic ram to get water from a small waterfall (maybe 5m fall) up to a stock trough about 50m up and 200m horizontal. Those rams are awesome - and loud.

Did you have a way to release excess air from the hydraulic ram? I set one up with similar dimensions to what you described but it would always quit after about 15-30 minutes, presumably due to air in the line. If I evacuated the line and got it going again then it would run for 15-30 mins again and then stop. Never figured out a solution :(

It wasn't a homebrew solution and came as a kit. I'll try and find out where it was from. Presumably it had a way of releasing the air as it would go fine for weeks (until a flood took it out). The thing was awesome and could have gone way higher (with reduced flow). It had so much water running into the trough that the main issue was trying to work out how to reduce the flow, which is surprisingly hard.

In California every outdoor spigot has to have an anti-siphons installed directly on it, so it’s not too hard to imagine that being the case here.

I've owned several houses in California and not one has had an anti siphon valve on it. A quick check around my neighborhood shows that non of my neighbors do either (at least on their front yard spigots).

I wanted to run a tap with flow backwards and found that all outside taps in both big box hardware stores near me had a valve in them. There was no way to tell this externally.

Sorry but this is not correct. The anti-siphon valve is almost always apparent. These spigot valves are not very good and fail pretty regularly so I would not trust my drinking water to them.

Also an anti-siphon valve only works if it's above the water lines. If you store a hose above the spigot like many people do, you've just defeated the valve.

Every outside tap here in New Zealand is required to have a back flow preventer (is this a different thing?). They all passed the recent test they had which was required by the local council and they are all pretty old. The 'check valve' is required to be replaced every two years as it can't be tested according to the documentation here. I don't known what this is and don't have one.

And they adjusted for low water pressure by poking a few more holes for each plant nearer the end of the the poly tube. By avoiding "nozzles" on their tubing - which would require pressure to operate -- they greatly simplified the design of their irrigation system.

Again anything that moves liquid from point A to point B requires water pressure or the liquid will not move. So the system you describe still requires water pressure to operate. I really don't see how this simplifies because you have to empirically determine the number of holes at each plant. If you have thousands of plants, I just don't see how this is efficient or scales. Maybe I'm missing something. For small numbers of plants maybe it can make sense. Driving through Watsonville, Ca today on my way home from the beach I saw hundreds of rows of crops growing with traditional drip lines.

Go visit the UCSC agroecology center farm. They have been using variations of this technique for years. I was surprised when I saw it too - I'd thought that it would need to be more complicated.

Wouldn't you have to replace sections of tube each season with this approach?

That's true - I asked about that and they said that they could sometimes reuse the tube if they were planting in a similar configuration - otherwise they just discarded the tubing (it is very cheap - another advantage of their system.)

My experience:

Use valves that are suitable for low flow/pressure and work off 24 VAC so you can use them with a cheap line-powered electronic timer.

If you need to filter or pressure regulate before the valves (probably at least filtering), be sure to use parts rated for continuous pressure at above typical household pressure. (100 psi ratings are probably enough)

Minimize the types of different tubing and emitters you use, it's easier to have spares that way, so if you weed with sharp implements and cut something it's not too annoying.

Don't go significant distances with less than the 1/2" black polyethylene tubing.

Drip-tape is super-cheap but mostly for farm-scale row crops.

The emitters I've had the best experiences with are the Netafim Woodpecker emitters and the take-apart-to-clean-out "flag" drippers sold by dripworks.com

As always, anyone trying to work with gravity fed systems that sit in the range between ~2 metres head (the usual maximum of very low-head valves) and standard / city pressure, are somewhat stuffed.

I'm working with about 10 metres of head - a decent standing pressure, despite coming through 400 metres of 50mm polypipe - but it's too much for the specialised 'water barrel' type valves mentioned here. Combined with no power (if i had power I'd have a pump :) means it's also not feasible to use always-closed solenoids, or have sufficient pressure to reliably get residential grade water computer valves to open / close.

Have you tried something like this:


Since it's a ball valve operated by a motor it works with zero pressure. It's a relatively small ball valve but should be enough for almost all applications. I have one of those controlling about 100 meters of drip tubing from a source that's not more than 2 meters high (less if the tank is empty) and it works fine. For better flow I'll be trying one of these ball valves:


It's much less convenient as they don't have an integrated controller. I'll be trying it with this controller:


and a 12V battery together in a waterproof casing of some sort. I've been looking for a plastic case similar to the water timer one that's a screw on with an o-ring but haven't found one yet.

I actually picked up a few of the first ones there via ebay a while ago, to trial off a smaller tank with around 2 metre head, but haven't plugged them in yet. Winter means frosts, and no great need for automatic watering. I note someone else has had bad experience with them - I've used the normal pressure versions of those with some success - a couple of years, with 9-monthly battery changes, seems reasonable for the price - just frustrating when they go wrong (usually just before a fortnight of 40+ Celsius days).

The actuator though, that looks very relevant to my interests. I've just done a google spiral on CR02 (and similar) wiring, to control via an arduino, say. With 3s to close / open, and only 12V required, this seems quite feasible to have a cheap solar charged battery sitting and running a half dozen of these of a manifold. The benefits of a switching solenoid in terms of power consumption, with a simpler control mechanism?

The last link you posted seems to imply it will talk to a solenoid (NC or NO) only - would be very curious how that goes.

I have several dozen old water timers with the o-ring seal ... where the o-ring seal has failed, and the insides rusted out. I now cover them all in thick plastic bags, and keep them under pit covers / galv steel lean-tos. Probably explains why they tend to last more than one Australian summer, too.

>The last link you posted seems to imply it will talk to a solenoid (NC or NO) only - would be very curious how that goes.

The way CR02 wiring works is that you have a common ground and then you feed the 12V to one of two wires. The valve takes care of stopping the motor when fully open and fully closed so it should minimize power consumption. So my plan is take that normal NC/NO relay and feed the 12V to the common wire and wire NC and NO to each of the two valve wires. Since the programmable relay itself is also powered by 12V I get away with a single power source. My hope is a simple pack of low self-discharge 8xAA NiMH or similar will last a season. But a 12V lead battery and a solar panel could work as well.

Aliexpress has a very big selection of different kinds of actuated ball valves. If they actually work reliably it seems like a great solution as it's a normal valve operated by a motor. See this for example for a selection of voltages and the 5 wiring types:


That one I linked was picked quite carefully to be stainless steel and CR02 wiring which seems to be the easiest to operate with just a single relay. I've received the relay but not the valve yet so I haven't tested everything yet.

>I have several dozen old water timers with the o-ring seal ... where the o-ring seal has failed, and the insides rusted out. I now cover them all in thick plastic bags, and keep them under pit covers / galv steel lean-tos. Probably explains why they tend to last more than one Australian summer, too.

Those are some rough conditions if the o-ring failing is enough to rust everything. I'd expect the threading itself to provide quite a bit of protection already. Can't imagine how you'd get much moisture through all those threads. One thing I have noticed is that you shouldn't torque the lid hard so as to not deform the o-ring too much.

UPDATE: finally got the valve and it works fine with the controller. The combo uses 8mA when closed, ~80mA while opening (for ~5secs) and ~40mA while open (half is the relay in the controller half is the valve). This makes it outside the range of a simple battery but should work fine if you have a solar setup with a 12V battery or an AC mains and a simple 12V power source. So far I'm happy with the domestic all-in-one I linked so I'll probably not pursue this more. If a simple NiMH pack worked I'd probably trial a full setup. But if I was going further I'd use a DIN 12V programmable relay like this:


and a DIN 12V power source like this:


all enclosed in an outdoor distribution box like this:


That way everything gets properly packaged using DIN elements. If you have a 12V panel/battery you don't need the AC->12V power source of course.

I used to have the first one, it was garbage - didn't even last an Australian summer, I think it overheated.

That sucks. I've had good luck with the solenoid version of those. But I can see how the fully enclosed design can cause overheating if it's left in the australian summer sun.

Have you evaluated how much it would cost to use solar power?

Or would the possibility of the system not running be a disqualifier?

Yeah, solar for charging batteries to then run solenoids, absolutely feasible - but most solenoids are either Normally Open or Normally Closed, and need constant power when holding them in the other position. For relatively low-flow situations these are non-starters, as you usually want to have a circuit open for an hour or more.

There are switching solenoids that only draw power as you change state. Downside, they don't fail safe. They're also more expensive, and AFAICT you can't use standard (NC / NO) controllers with them. To complicate matters, switching solenoids usually want a certain amount of pressure (about twice what I'm working with).

It's an area that's slowly getting better served by cheap fab places out of China. You can already get A$20 'residential' grade 13mm single line water controllers that are mostly reliable, but their valve comes down to about a 1mm hole, so big impact on flow rate, and highly susceptible to debris.

What about stuff like holding open a larger solenoid to charge a gravity tank?

> but it's too much for the specialised 'water barrel' type valves mentioned here

what are you using instead?

Why not use something like a Tropf-Blumat system? It doesn't need electricity and is self regulating. You only need a water source (it can be a barrel of water at some height or a regular water source) - it stops feeding water when the soil is moist enough. It uses ceramic cones to regulate the amount of water.

This is the first I've heard about Tropf-Blumat systems, sound s pretty great if it really works as promised.

I wonder why it isnt popular in US, price too high?

I use such a thing for a pot on the garden table but a larger drip system for the garden itself. The Blumat doesnt require running a pipe from under the deck onto the table and is movable but does require filling in water every few days.

I've been planning an automated (raspberry pi controlled) watering system for my garden lately. This page is a great source of information about irrigation. It took quite a few web searches for me to find this kind of information before I found this site. Great to see it on HN

I had 24 zones running with opensprinkler on a pi and loved it. About half the zones were drip. I'm in a new place now and I'm about to put in the first zone. I'll use the same hardware. It was stable and easy to use.

Best advise I saw in the tutorial was to make sure tubing and fitting sizes match. It will take a few days, but they will blow apart when you aren't looking.

Show HN: GreenPiThumb – A Raspberry Pi Gardening Bot | https://news.ycombinator.com/item?id=14644539

> Drip irrigation (sometimes called trickle irrigation) works by applying water slowly, directly to the soil, bloop, bleep, bloop, bleep.

Such a satisfying description

Anyone know of some kind of in line valve for each dripper that senses moisture and will stop flow once adequately watered?

I container garden in pots and have to over water to ensure everything gets enough, but this causes all sorts of other problems of pooling water and leaked water onto the balcony, leaching of nutrients.

"Anyone know of some kind of in line valve for each dripper that senses moisture and will stop flow once adequately watered?"

There are different emitters for different rates ... I have 1gph as well as .5gph emitters for various purposes and I think there are other choices as well.

I would think you could run the water the same duration for all of them, but vary their irrigation with different rate emitters.

I recommend graduating from the hardware store tooling to the professional Netafim hoses/emitters/couplings.

Agree with other posters - it's either high cost, or flaky, or both. Even the expensive ones require replacements a bit too often, and you'd have to be worried about some false measurements firing off excessive over-watering.

The 'water well' pots may be a simpler, cheaper, reliable option -- if you can hook up multiple inverted water bottles, or some ballcock arrangement into a cistern and share out from there with 13mm pipe -- if the bottoms of all your pots are the same elevation.

There's a small period of time while the roots head down to find water, that you need to top-water, but, again, convenience .vs. cost .vs. resilience tradeoff.

Look at this dutch system, mentioned by another commenter, pretty interesting: http://www.blumat.com/en/startseite.html

It’s the holy grail really. Moisture sensors are either flakey or super expensive. The best compromise I’ve found is to time the flow. Sensing adequate hydration is tricky particularly at the home garden level.

Well for one, define 'adequately watered'. Secondly, a reliable continuous sensor will be a tdr sensor, the cheapest of which start around usd 200 - and then you have to do a bunch of hacking yourself still. It's not so easy - there's a reason real indoor growers don't use pots with soil. Look into using other mediums, or maybe use a cyclic flooding system, but both require adapting the environment the plants are in, they're not so nice to look at with the tubing and overflows etc.

I have the same problem, I would like to develope a circuit to recycle dripping water to a tank and reuse it to allow overwatering eithout leaching the nutrients and wasting water.

This is how my grandmothers does it. Put a glass / bucket of water next to the plants, take a string of wool, hang it in the bucket and push the other end in the soil. Perfect moist for the plants all the time.

how high is the water in the glass ? capillary traction can pull water a lot ?

won't the constantly moist wool spoil quickly?

I'm pretty skeptical about drip's performance compared to flood method when it comes to resource intense crops like sugarcane.

Is anybody qualified enough on this topic here to counter me on this?

I want to landscape my yard and switch from sprayers to drip. I have an old system, and honestly am completely overwhelmed by setting up irrigation.

Am I overcomplicating things? Or is it really that hard to get it set up properly with correct pressure such that I'd be better off just hiring someone to install it?

Slow down, take a deep breath it's not hard! It should not be hard particularly if you are on a city water supply. If the system is old you'll want to check that there are appropriate backflow preventors on the system and that they are installed correctly and that they work . Once you have that you install an inline pressure regulator to get the pressure down and simple inline filter on the line at the manifold where your valves are. This helps keep the drip lines from clogging. If you have pop up sprinkler heads you can easily retrofit them with drip. If you have the old style fixed spray heads you can still retrofit them to drip but it may be more work if you want to hide the drip lines. The whole conversion is very easy because your main lines have already been trenched. Trust me I've trenched my own lines and this was hard tiring work even with a trenching machine.

Some good info here: http://www.urbanfarmerstore.com/converting-sprinkler-to-drip...

Thanks for the info. TBH, it seems like describing the process at a high-level is very easy, but each of those steps you listed probably has some education that is needed to set them up properly, know what to look for, research into brands, etc.

Or am I overthinking that part of it?

Are you coming off city pressure water supply? It shouldn't be too hard to experiment and set up yourself, and if you've got an existing system you've got the convenience of being able to experiment slowly as you roll out new equipment.

The pressure-compensating drippers, as mentioned in TFA, may be useful for you, depending on your run lengths, and contours. Bit more expensive, but the sound of lots of emitters going 'click' at the same time can be quite cool.

For most systems I use a dribbler, rather than a dripper, device. I use 25mm (well, 1") LDPE almost everywhere so I can easily puncture and insert an adjustable barbed dribbler. Do not go for the cheap ones - they won't last one summer. Dribblers let you compensate for distance from the tap and contour changes.

You should have a go designing and implementing something - it can actually be a whole lot of fun.

Thanks for the insight. Yes, coming off city water. My problem is that my life is full of lots of other priorities and free time is at a steep premium. So yes, I could probably slog my way through getting educated on this, plan things out, experiment, etc., but I'm trying to figure out how crazy it really gets for a basic setup to inform whether I have the time to invest vs. paying someone as part of other landscaping work I'd like to do.

It should be reasonably easy if you follow the guidelines in this document and google around a bit more. One of the real upsides of drip is that it's really pretty tolerant to pressure issues - just use large enough tubing/short enough runs to keep your pressure loss under control and do some rough mental-math estimates of the zone pressure and flow rate.

The one area where I feel like irrigation goes beyond my comfort zone (as someone who has only laid out irrigation like twice) is if you're working off of a well, since irrigation with a flow rate that matches the well poorly can cause the pump to cycle excessively and burn it out faster. But if you're on city water that's really a zero concern.

Accumulators work well to reduce pump cycling.

Have you looked in to whether a hydraulic accumulator would suit this purpose?

Wells are always equipped with a pressure tank (which is an accumulator), but the capacity is pretty small. To prevent excessive pump cycling with irrigation you would need a rather large tank since you'd want it to run irrigation for at least minutes at a time - I think this is beyond what's practical with a pressure tank, so that you'd have to go to a small water tower. I'm not sure though, it's an interesting idea.

For spot drip irrigation, I take an empty gallon jug, poke a hole in the bottom with a tack, fill with water, and set it next to the plant.

Zero mentioning of limescale

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact