As expected, there is a lot of high-end theory involved with the design of Pelton Wheel hydro installations, and mine is only haphazardly poking at the arithmetic.

Had a look at the Wikipedia page on Pelton Wheels:

http://en.wikipedia.org/wiki/Pelton_wheel

...and learned that the velocity of the stream of water is best calculated to be twice the speed of the rim of the wheel. Since the speed of the runner is regulated by the output voltage of the generator, that is pretty much a constant in my installation. Varying the speed of the water entering the runner is probably more important than varying the flow.

Friday, I stopped by the hardware store to see what they had that I could use for changing the nozzle in the small plant that I built. They had a small brass "driveway washer nozzle" (high pressure fitting) and a plastic "sidewalk tunneling kit" that looked like I could cut it down and cobble something together. I bought them both.

The easiest change was to simply unscrew the nozzle I was using and screw on the small brass replacement. This new nozzle has a 5/32" hole instead of a 3/16" that the original had. Turning on the water produced .78 amps of current, or 11 watts, a 10% improvement. The water pressure at the nozzle ran at about 25 PSI, an improvement for the rest of the domestic supply.

At first, I was rather confused as to why a smaller orifice produced more power, but considering the stream velocity being the most significant factor, I guess this makes sense. Since Pelton runners work by diverting the flow of water, rather than simply by the impulse of the water hitting the runner cups, this makes sense. More water does not necessarily mean more power. Using the water entering the runner cup more efficiently does mean more power.

I did look around at the hardware store for Rain Bird nozzles, but they didn't carry them as replacement parts. I may have to try the bigger stores next time I go to the city.

I'm thinking that multiple small(er) nozzles would produce more power than one big one. Matching the water velocity to the runner speed will give best output for the least water used. If a smaller nozzle produces this desired speed, then several smaller nozzles would add additional power output. It's all about torque, not speed, the speed is fixed by the battery voltage. More nozzles = more torque = more power.

Now I have to figure out how to support the generator with a box or structure that allows me to mount and align four nozzles. Oy, more projects.

I guess it's time to be surprised, as the tiny $7.50 hydro plant is still cranking out power. Most days I have to shut it down by mid morning as the batteries are fully charged from the previous night's consumption.

Since this leaves me with a surplus of power, I have run the extension cords across the yard a couple of days to the 24 volt grid-tied system and run it during the daylight hours. The plant puts out .45 amperes at 26.7 volts nominal, about 13 watts. This makes me think about an automatic change-over that switches the hydro plant to the 24 volt system when the 12 volt system comes up full. I still have the custom designed and built charge control that I used to use to divert excess PV power into the RV refrigerator, it was designed to run an external relay and operate on hysteresis so the relay doesn't chatter. Just the thing to integrate the two systems to charge from one source.

Went up to the spring today and there's a load of water coming off the hill that I can't capture or use. A larger penstock pipe is a must, as is a more industrial pelton and alternator.

Just came out the warm end of a full week of freezing temperatures. Today was the first that the temperature was above freezing for 24 hours.

The neighbors lost their water last Monday. Apparently, running my little electric plant and keeping the water flowing at full volume kept me from having freeze issues. Every year until this one, I had some kind of a failure in cold weather, in spite of running the water at the taps in the house. Nothing like wide-open delivery to keep things liquid.

Of course, this means that there is a lot of water to get rid of, and a lot of ice that can build up. Witness the iceburg that the hydro plant manufactured during the freeze:

The spray of water exiting the wheel built a kind of ice dome around the exterior of the housing, while the inside was completely open. I almost busted my butt a couple of times approaching the area to shut the plant down for showering, it was pretty slick all around.

Of course, having a proper enclosure for the plant and a spillway to conduct the water away from the vicinity would make sense, but I keep thinking I'll move it down to the lower garden and add a nozzle or two. Besides, old bricks and 2x4's are much less effort to construct.

I've been thinking about how to better mount and support the generator I have now. If I add nozzles to it, I need more and better attachments, and it would be nice to have something more stable than an old cinder block as a foundation. A better enclosure would make the operation quieter as well, not that I have a big objection to the white noise of water. A square bucket or aluminum enclosure of some sort with the generator mounted on top of the lid would allow the machine to be opened up for inspection and make exchanging nozzles a lot easier.

Eventually, I'll get my big tax return, and I've already decided that a small part of it will get invested in a commercial hydro plant. There are quite a few choices, I have 180 feet of head, with about 125 gpm in the winter. Thinking of a pair of two-inch penstocks. Return line to the house would be about 600 feet.

Today was cloudy and drizzly, and my 12 volt PV array would never have been able to recharge the batteries from last night's consumption. It's almost 4PM and the NiCd's are just now coming up to full after the hydro plant ran all night and all day, perfect timing for tonight's consumption.

Looks like my predictions about the longevity of this project were based on sound reasoning. I had to take the hydro plant off line a day or two ago due to wear and breakage in the Pelton assembly. The plant had been getting noisy very fast, making kind of a rattling/clattering racket when it was running. It was still making full power, but it sounded like a hollow gourd filled with beans being rolled around.

Inspection of the Pelton revealed that the attachments for little buckets had become badly worn due to the motion of the buckets while running. This Pelton wheel is assembled with the small buckets held in place by a locking tab attached to the stem of each bucket. The buckets stems sit in a groove in the top wheel and are secured by a mating bottom wheel cover. There was a tiny amount of motion in each bucket when I put the plant into service, but after 1,200 hours of running, the mating parts of the buckets and wheels are showing some wear. Two of the buckets had broken part of their retainers on the wheel and were kind of flopping around when the water jet hit them.

I removed the Pelton runner and after removing the bottom wheel cover, I had a handful of worn parts laying in a pile.

I wanted to get started on this project today, but guess what, without the hydro plant keeping the water lines cleared, I had to go up to the spring and do the cleaning and draining thing on the head end of the system.

After I figure out what kind of plastic the wheels and buckets are made of, I'll either cement or solvent weld the buckets back into place on the wheel and probably pot the whole mess with more cement to make it a solid block of plastic. No going back after that step, the next time it needs repair, it's into the trash can with the whole thing.

As for the power I'm not making with the plant down, I miss it already!

The hydro plant is back on line. I couldn't find a solvent-based cement that would fuse both the wheel and buckets. ABS plumbing cement reacted with the wheel, so I used it to "pot" the buckets into the lower wheel, filling all of the voids between the bucket mounting tabs and the wheel assembly. Before putting the top cover back on the assembly, I drizzled cement over the tops of the buckets, then pressed the cover into the layer, latching it into place.

The result is that the wheel and buckets are now one solid piece. Although the cement did not burn into the plastic of the buckets, it does adhere well to them, so they are unable to move.

Putting the plant back into operation, it makes full power (10 whole watts) and now runs much more quietly, with the sound of the water spray and a tiny background hum as the buckets rotate in and out of the stream.

Done for now, next time this needs attention, it will be the end of life for this project and time for a new plant.

Running up to the three month mark, with a week off for rebuilding...

Yesterday, I decided to do something about all the tail water coming off the plant. I've been a bit concerned about it from the beginning, soaking the ground right next to the house on a hillside didn't seem real smart, and that ice buildup couldn't have been good either.

Spent a few minutes with a second plastic water jug and a razor knife, and made a slip-on bottom cover to catch the water and direct it into an old piece of gutter to carry the water away from the area.

The gauge is reading a bit low these days, I have a leak somewhere in the pipe coming down from the spring. This afternoon, I went up to the spring to clean the screens, etc, and after removing the supply from the barrel, I watched the water level drop rapidly. Something is amiss with the pipe somewhere down the hill. If the next couple of days are rain-free, it looks like I'll be trudging up the ravine to see if I can find the problem.

A year and counting. This little project is still paying off, as I'm using it once again to keep the batteries topped off as the light disappears towards winter. To be fair, it didn't get run much over the summer, but it still seems to be holding together.

No improvements since the last post, I haven't even had an opportunity to take it down to the lower garden and try it at higher pressure. I do think about trying to put together a "linear current booster", essentially a buck/boost converter so that I can find the optimal running voltage for the plant and then convert that (whatever it turns out to be) into a charging current for the batteries. I do know that the plant puts out more current at voltages above 12 (13.5 - 15 V), but less when connected to the 24 volt system, and even less when the 24 volt system is full (~28 V). Somewhere between those two potentials is a "sweet spot" where the voltage and current curves intersect to produce the maximum number of watts.

Now that I've written it down, it sounds like a lot of trouble for a device that will still produce less power than a couple of night lights can consume...

Today, I did something that I've wanted to do since the first day of this project, I took the hydro down to the lower garden and connected it to the water supply there. Being as the garden is about 40 feet lower in elevation, there was more pressure available. I was able to put the larger orifice nozzle on the plant and maintain about a 27 PSI running pressure.

The power output of the plant into a 12 volt battery was doubled with this increase in pressure and nozzle combination. I was able to put out a scorching 20 watts!

Of course, there are some drawbacks to this configuration. One problem is that I am about 200 feet away from the house, meaning that there's no easy way to get the power back to the batteries.. The other problem is that with the running pressure at 27 PSI at the plant, there was nearly zero pressure and no flow of water up at the house. Both of these problems are surmountable, I can string wire all up the hill to return the power, and some manner of servo-operated valve will need to be fitted to the plant so that it can be shut down automatically when water is drawn from the domestic supply. I just need to decide if it's a priority project or not. Having 480 watt-hours available is attractive, but that's less than 3¢ worth of electricity a day. Lots of other important project around the homestead waiting ahead of this in the line of to-do's.

For now, the plant is back on it's cinder block humming along and making 10 watts of power to keep the lights on in the Housetruck. Maybe that's enough for now.

Time for some unscheduled maintenance. Last week, I went out and turned on the plant at dusk in preparation for the evening's lighting. About nine o'clock, I noticed that the plant wasn't putting out any power, as indicated by the dedicated ammeter inside the Housetruck. "Great" I thought, "the spring has quit again, I'll have to trudge up the hill again tomorrow and get it put back into shape", and got on a coat to go out and switch the house and truck over to the backup water supply.

When I got outside, I could hear that the water was running full on at the plant, but it didn't sound quite "right". Looking at the plant revealed that although the water was blasting the pelton cups, they weren't rotating. I reached underneath the plant and gave them a twist. It was stiff, then started spinning of it's own accord. I just assumed that since the weather was below freezing, I just had some ice buildup somewhere that prevented the plant from starting up, and I just hadn't noticed.

This morning, I went out to shut the plant down when the batteries looked like they would get filled by the PV array, and found the plant stalled again. This time I know it was running, because I had checked for charge current when I got up. Obviously, something was binding inside the generator, and putting enough load on the shaft that the pelton had been unable to supply enough power to keep it turning. Time to investigate.

Just at dark tonight, I went out to turn the plant on, and decided it was the right time to find out what was going on. I removed the plant using the pipe unions, unplugged the electrical wiring, and took it into the garage. I was afraid that it would be a rusted-solid block of corrosion, but amazingly, it all came apart easily. I attribute this to all the lithium CV joint grease that I slathered on all the exposed surfaces, motor shaft, etc before I assembled it last time.

Inside the motor/generator, I found that the brushes were worn almost all the way down (no surprise), and that the motor case was well filled with carbon from the brushes. The front (shaft) end of the motor seemed to be very stiff in the sleeve bearing, so I completely disassembled the armature and end plates, cleaned everything up, and relubricated the bearings. I also chucked the armature up in the drill press and used some emery cloth to polish up the commutator.

Nothing to do about the worn brushes, they still had a few hours left on them, so I reassembled the motor, put it all back into the housing and reassembled the pelton runner, etc. After it was done, the motor shaft turned freely and had no sticky spots.

Putting it back online brought it up to full output, and it runs quieter than ever. Next time I go to the city, I'll have to stop by the motor shop and see if they have any brushes that can fit this motor, or something that can be modified to fit. If I can replace the brushes, I think I can get another year and a half out of this thing.

Catching up on some of the background on this project.

Since last time, when I was concerned about how long the little permanent magnet motor and castoff pelton wheel were going to last, the system has evolved, and continues to change, however slowly.

Every time the motor/generator would show signs of wear or have a failure, I would manage to get it up and running once again. I cobbled carbon brushes purchased from the hardware store to replace the originals when they wore too small to be usable. The bottom bearing, which supported the armature thrust, was wearing steadily. I would disassemble the motor regularly and lubricate the bearings, clean out the carbon dust and insert shim washers to compensate for the bearing wear, which would cause the armature to ride lower in the casing that it was designed to. This also caused the commutator to wear in a cone-shaped pattern. At one point the armature was running so low in the motor casing that the windings on the armature wore away on the end plate. I dutifully picked apart the damaged windings and soldered in jumper wires to reconnect the windings, making the plant function again, potting that end of the armature in epoxy to hod it together.

It was increasingly obvious that I was going to have to replace this motor at some point, and it would make sense to use something more heavy duty. I purchased a "blue spoon" turgo runner from Hartvigsen Hydro with the intention of getting an upgrade underway. I also purchased a variety of small nozzles to feed water into the runner and a steel arbor that the runner would mount on to allow it to be connected to the shaft of an automotive alternator.

The new runner was going to require a turbine box to mount the generator and nozzle. Turgo runners require that the stream of water be applied at an angle, coming down from the top of the runner, rather than the side, as a pelton runner does. I decided to throw together a small box made of cedar and redwood for a trial, using brass wood screws to hold the parts and pieces in place. I figured that it would be temporary, and I'd build something more sturdy and rot-resistant after I earned more about the runner and the process.

For a dynamo, I had never intended on using a car alternator for such a small amount of power that I could make on the amount of water I had to work with. Instead, I pulled the permanent magnet alternator off of a Southwest Wind Power AIR403 wind generator that I had. It had never proven to be useful in the climates that I had lived in, so putting it to experimental use made some sense.

This is one of the early trials at the house. I'm running the output of the alternator through a random transformer in an attempt to step the voltage up or down to see the effect on the total power output. By the look of the ammeter, I'm putting around 10 Watts into the 24 volt battery bank, a bit less than the PM brush motor, but not bad for a first attempt. A dual Schottky diode assembly is rectifying the output to feed the batteries.

One attractive feature of an AC alternator was the possibility of using transformers or even variable transformers to"tune" the speed of the runner for best efficiency.

Closeup of the turgo runner. The side of the turbine box has been removed to get a close-in look and to help align the nozzle in relation to the runner spoons.

I did end up running a similar plant configuration using this alternator and runner combination for a time down at the lower garden, where the water pressure is better, making around 30 Watts and running the power back to the house through a long wire.

The problem associated with having the plant at the garden was that the water pressure at the house was very low due to the use lower on the line. Gathering some cast-off parts from my storage locker, I built a motorized valve, using the windshield wiper motor from a VW Rabbit (what else?). A controller regulated the speed of the motor, and limit switches told the controller when the valve was either fully open or closed. The intention was to be able to remotely control the valve, so I didn't have to keep running up and down the hill to have enough water pressure for laundry and showering.

In the end I didn't get around to installing the valve before a conservation project to restore salmon habitat on the creek relocated my gravel driveway between the old garden and the house. Without an easy way to get wiring between the two locations, and the fact that the tail water from the hydro plant would make the ground around my new driveway soggy, I shut the plant down and went back to just using the small, original hydro apparatus at the house again.

If this project was going to get off the ground in a serious way, I needed more water to be delivered down off the hill.

One day while cruising through an auto wrecking yard down in Reedsport, I spotted a big stack of 1¼" yellow polyethylene pipe. There were about twenty coils of the stuff, each 500 feet in length, still secured (mostly) by the original strapping. I inquired about this, and negotiated to purchase four coils (2,000 feet) for a reasonable price. I would have to come back with my trailer to pick it up, which I did a day or two later.

The yard monkeys who brought the pipe out of the yard did so by arriving with all four coils held up by a forklift. I was pretty sure that I wasn't going to be able to lay the pipe down, so I told them to stand it up in the trailer bed. After I got a couple of ratchet straps on the pipe, they let the forks down and beat it back into the yard.

The pipe wasn't very cooperative, and two straps barely kept it from falling over, so I ended up spending quite a bit of time trying to manage the bundle while putting on the remaining straps. Eventually, I had it secured well enough that I thought I could get underway. After a few miles, I stopped alongside the road and checked the load, it seemed to be holding.

At home, I used my backhoe to lift the whole stack off the trailer and laid it on the ground. One by one, I picked up the coils, moving them to an area behind my storage locker for eventual installation up to the water source on the hill.

Having secured the penstock, I happened to be at the recycle yard in town one day and came across some commercial dishwashing equipment, something that had been removed from the casino, I think. There were three motors in the equipment, each three-phase, 240/460 volt. Two of the motors were 2 horsepower, while one was ¼ hp. All of them had stainless steel shafts, which is never a bad thing when working with water-driven systems. These would make usable generators for my upgraded hydro project, so I removed and purchased them. The 2 hp motors cost me $6 each, while the smaller motor was $3.

I put the larger motors into storage, they were very dirty, and had noisy bearings, but they were rebuildable. The smaller motor, I intended on experimenting with sooner. I purchased an arbor that would fit the motor output shaft and allow me to attach the new turgo runner.

I had to do some quick self-education, as induction motors are a bit different from using a permenant magnet DC generator. In order to make an induction motor operate as a generator, it's necessary to cause self-excitiation by placing capacitors in the correct configuration and values across the field windings. I scratched around and found some old caps in air conditioning units at the yard and cobbled together a test of the motor-as-generator.

Because I has very little water to work with, I found I had to set up the field windings of the motor for 460 volt operation to get it to put out approximately 200 volts AC. The value of the capacitance seemed to make a lot of difference to the voltage, and also to the speed of the runner. I tried various configurations of the field windings, Wye and Delta at each voltage, as well as connecting incandescent lamps as loads. Changing the nozzle diameter changed things too, the power the plant put out and the net head pressure on the water line. What I found was that the plant operated best with the smallest nozzle.

This was all a lot to process mentally, and I needed to go "back to the books" to help me understand. I had also downloaded some spreadsheets that allowed calculations to be run taking into account net head pressure, runner diameter and number of spoons, nozzle diameter, motor efficiency, etc. This was becoming a very complex lesson.

One thing that was obvious right from the first was that with so many electrical connections, and so many possible configurations of wiring, that clip leads wren't going to cut it. Aside from the fact that I didn't have a whole trunk full of leads, they tend to fall off, have exposed ends that can cause shorts, and when you have a big tangle of them, trying to keep straight which lead was connected where and to what becomes a brain buster. What I needed was a test board...

...So I built one, literally on a board.

Each of the windings in the motor were brought out individually to a long-ish cable with a multi-conductor plug and socket which was terminated on a terminal strip. Wires with push-on connectors and labeled with adhesive numbers allowed me to easily and quickly identify windings. All of my hand-drawn diagrams had the same numbering system. Three 10μf capacitors with quick connect terminals were mounted, as were three medium-base edison sockets for incandescent lamps. There were a couple of "unassigned" terminal strips that could be used to connect quick-connect leads together, allowing me to make series or parallel connections.

Because several operating conditions can cause an induction motor to lose its residual magnetism (short circuits, running down with a load connected, sudden overload, etc), it is necessary often to "flash" the windings with a small battery. I installed a small knife switch that would allow me to disconnect one winding and momentarily connect it to a 9 volt battery to do the job with the push of a button.

This test rig would make experimenting with the induction motors much easier, as many of the needed connections could be made on the board quickly with the quick-connect wires, and for the most part, there was some order to the wiring, as many of the connections had an established place on the board.

Because capacitance plays such a significant role in making an induction mtor run as a generator, and the value of the capacitance needs to be varied to "tune" the output, I needed a lot of capacitors on hand. These aren't the fifty-cent small components from radio shack, but $5 - $7 and up components. Fortunately, I ran into a large number of high-intensity discharge light fixtures that had been scrapped. Each contained a 15 microfarad, 440 volt AC capacitor with quick-connect terminals and a bleeder resistor attached. In exchange for separating the various components (cast aluminum, steel, copper wire, etc), the recycle yard operator allowed me to keep the capacitors, which had little scrap value. Now I had 24 capacitors that could be connected in series or parallel combinations to create nearly any value that I needed.

A few minutes with a cable cutter brought me the wiring guts from a couple of commercial air conditioning units. With this I made a big batch of test leads with quick-connect terminals on each end to allow me to make connections between the various capacitors, windings, loads, etc. In all, I was slowly building an entire hydro laboratory.

The results of continued experimentation with the ¼ hp motor showed that when optimized, I could make not quite the same amount of power as the DC permanent magnet motor when running up at the house. It was time to go down to the lower garden where the line pressure is greater and try again.

Hence, the tailgate lab setup.

More water pressure meant more power. It also meant that I needed a different value of capacitance. In the image, I'm using a 3 phase variac transformer to allow me to add or subtract an additional bank of capacitors to 'tune' the output of the generator. I'm running the resulting 240 volts AC through a 3Φ transformer and rectifier. The resulting direct current is charging a 24 volt battery in the bed of the truck. Pity that the analog meter face is washed out, I'd like to remember what the total amount of power was...

Not all of this project was sweat-free intellectual exercise. The next phase or two involves hard physical labor, not something of which I am overly fond.

My reliable and energetic friend Mark came down from Portland twice in Fall of 2015 to help route and install the yellow poly penstock pipes that I had purchased a year earlier. Mark is a veritable human dynamo, and he's seemingly unstoppable on any task.

The first trip to Minerva, we spent two days clearing the old logging road that leads up the hill to my water source, and clearing a path down off the hill for the penstock pipes to bring water to the "powerhouse", an old, dilapidated milking shed at the entrance to my property.

Because we would be making numerous trips up the hill hauling materials and equipment, I arranged to borrow my neighbor's 4x4 quad ATV, to which we coupled a small garden trailer I use around the property.

Using chain saws and gasoline powered brush cutters, we hacked and cut our way through the brush that had grown up in the road over the years. When I would walk up the hill to work on the water collection system, I would always carry a machete, keeping a path cleared. I would usually leave small fallen trees, stepping over them. This time, we basically scarified the entire quarter-mile roadway right down to the bare grass.

Once at the "spring" (I call it this, although the source of the water is higher up on the hill), we the began descending, clearing a path in the ravine that is the drainage for that waterway. This would be the route of the penstock, not the road we had just cleared.

About halfway down the mountain, we started leveling off, coming out of the ravine and traversing laterally on the hill side. The idea was that I wanted to come off the hill in a different drainage, crossing between the two at a kind of "saddle" in the topography. It was here that we ran into some of the most rugged terrain and obstacles of the clearing project. There were decades of thick brush growth, and many, many layers of small downed trees that had been growing, dying and falling to decay. In one place, which I have come to call "the bear woods", we had to crawl on our hands and knees in a tunnel of brush. This was obviously the haunt of woodsland creatures, and the entire area surrounding our work area was riven with bear trails. I was more then a little nervous, although our pussy black bears here would be no match for a running chainsaw.

I'm not really sure how, but in the end, we managed to scout and clear a path that was almost perfectly all downhill from the spring site. This is important, as moving water efficiently over distance depends on not having "hills and valleys" in the elevation of the pipe for air to collect in.

At the far downhill end of our journey, we exited the woods just above the country road, right across from the powerhouse. It was a steep 100 feet or so off the last of the hill and down to the road.

Two weeks later, in October of '15, Mark returned, bringing along his adult son, Chet. Mark and I are of modest physical build, but Chet is more of a 'dozer of a guy. Guess who was going to be harnessed to pull the pipes up the hill?

I had spent quite a lot of time trying to figure out the best method of uncoiling the yellow 1¼" poly pipe without arriving at a solid decision. It was thick walled and very stiff, and had been coiled up for years, getting it laid out flat was going to be a chore. The coils weighed 150 pounds each, so they weren't impossibly heavy, but they weren't light, either. Several thoughts came and passed, laying it out along the country road, unspooling it in my pasture in the sun, etc. Mark wanted to just cut the binding straps and play it out. I was concerned that it was kind of like a coiled spring, and if released improperly, would tangle itself in a giant mess when the tension was allowed to get loose.

In the end, I built a hanging, revolving turntable with caged sides out of scrap steel, supported by the loader on my backhoe.

We set up a rope and pulley, hanging a block from a big tree at the top of the first rise from the road. This area was so steep that it was impossible to climb it without using hands and feet, and getting the pipe up this first part would be the most difficult. Using the rope to haul the end of the pipe up first was a big help.

The pipe spool worked great, Mark and Chet were able to pull the first 500 feet up with a moderate amount of effort while I managed the unspooling. We needed to do this twice, once for each penstock. When each section of pipe had cleared the spool, I climbed the hill and the three of us would pull that section all the rest of the way to the top of the hill to the water source, about another 250 feet, before pulling up the next run.

The next two runs didn't have as far to go to meet the ends of the fist pipes up the hill, it went pretty quickly. We also put in a new run of 1" NSF certified poly pipe that would be the new supply for the domestic water for my residence. The old ¾" pipe was not "drinking water safe", and had been patched with sections of rusty metal pipe, etc.

It was necessary to get the new pipes on the opposite side of the county road. In these parts, that's what drainage culverts are for, and the county basically turns a blind eye towards the practice. In my case, the culvert I needed to go through was a special "salmon passage" culvert that had been installed to facilitate spawning fish accessing upstream reaches of the creek tributaries. The culvert is huge, and has had the bottom third or so covered in rock and native soil to emulate a natural streambed. It is possible to walk in the culvert in a hunched-over stance.

Because of the significance of this, I chose to not just place my pipes on the bed of the stream through the culvert, but instead, suspend the pipes from the roof of the metal shaft, so as to not draw any ire from the department of fish and wildlife, etc. So that the county couldn't get pissy about any atachments to their culvert that I might make, I used forged steel hooks and a turnbuckle to suspend the 3/8" EHF steel cable that served as the carrier wire for the pipes. Poly rope was used to lace the pipes to the wire.

At the ends of the culvert, where the pipes entered and exited, I disguised the pipes under native vegetation and by enclosing them in black drain pipe covered with a wired-in-place layer of tree moss. From a short distance, it's impossible to tell that anything is different in the culvert without actually getting down in the stream bed and looking in.

All that was left to do this season was to trench 50 feet or so from the culvert to my power house, a project that took less than a day thanks to my Case Model 60 trenching machine.

In the typical fashion of my projects, progress on the water system and penstock pipes in 2016 was exactly none for the first seven months.

One morning in the first week of August, I got up to find that there was no water at the house. This was not an unusual occurrence, the water collection system up on the hill was a slightly modified leftover from when the previous hillbilly owners lived here. It was always in need of attention, and I would have to hike up the hill every couple of weeks when it was working well to clear screens, arrange the rocks that diverted water from the stream bed into the collection basin and reset the pipe that conducted the water into a rusty 55 gallon barrel to be sent down to the house. When we had storms, or in the fall when there were leaves and the rain washed the summer's dust and soil into the stream bed, visiting the site was an every-other-day chore, usually at the the worst possible times.

As usual, I got my gear on, machete, gloves, towel (I always end up getting wet doing the spring work), a bottle of water, etc, and hiked the quarter mile up the hillside, hacking away any brush and branches from the path on the way.

At the water collection site, I walked into a confusing scene. It looked like a tree had fallen, obscuring the upper parts of the site. I fact, a large red alder tree had dislodged from the hillside above the area and the root wad had fallen directly on top of my catchment basin. A small landslide had followed, completely burying the whole area in a hundred tons of dirt, rock, and tree. This wasn't going to be a simple arrange-the-rocks repair, my water system up there was destroyed.

Digging the area out was out of the question. Even if I had wanted such a job, there is nowhere in the steep ravine to get rid of the spoils of that work. Anything that you cut, dig or dislodge tumbles to the bottom of the channel and becomes an obstacle to further work. It would be necessary to hand-carry the tree, soil and rocks away from the site. Not going to happen.

The first order of business was to get some water going back down to the house. Amazingly, the rusty barrel that served as a reservoir was intact, the tree had fallen only inches away from it, but it was still sitting on its precarious stack of stones, connected to the now dry pipe running down the hill. The usual small trickle of late-summer water was running down the rock face above the slide site, so I made a trip back down to the house and grabbed three ten-foot sections of plastic rain gutter, some 16d framing nails, a hatchet and bailing twine. With this, I was able to construct a 25 foot flume to catch some of the water and direct it into the barrel, restoring service to the house.

The yellow penstock pipes had been tied to a small tree above the catchment basin using a length of parachute cord, and left that way since the installation in 2015. There was a bit of a hollow under the tree roots, and I could tell by reading the markings on the pipe that it was only a couple of feet to the end of the pipes. Some digging, and then sawing through the dirt with the machete severed the cord, and with some effort, I was able to pull the two pipes free of the landslide. This meant that I didn't have to shorten the pipes, and that I was able to recover the two $20 brass fittings terminating the pipes.

The rain gutter delivery system wasn't going to survive even the first heavy rain storm, so something better was going to have to be constructed. Because the water system had always collected ground water coming from a spring above the site, I thought that I would just go higher on the hill and find the source of the water and tap it before it "daylights", that is, dig out the spring source and put in a subterranean collection pipe.

I had been above the property line, on National Forest property before, the steep ravine levels out just on the other side and runs back into some small valleys. I was determined to find the source of the spring, and run a pipe the additional distance to tap it properly.

Two and a half hours of stumbling back through some of the toughest brush possible got me in about 2000 feet. I was carrying my gas-powered brush cutter, but even with that, the going was slow. In some places, the stream disappeared below huge tangles of decayed trees and brush, it was impossible to even walk there. I skirted some of the worst areas, rejoining the stream on the uphill side.

What I found as I went in deeper was that the stream was steadily getting smaller and smaller and no one source of water was to be found. There would be side channels of damp, marshy ground, some with a tiny trickle of water leading away. In practical terms, it appeared that the water was a multitude of small seeps, and attempting to tap any one would be fruitless for my purposes. A new plan would have to present itself.

Because of all of the rock and dirt, any collection system I put in would have to be above the landslide site, otherwise I'd be collecting silt and dirt for years to come as rain dissolved the root wad and landslide materials and washed them into the collection intake.

When I first purchased the property, I did some checking into a product that was sometimes used to build hydro plant intake systems manufactured by Hydroscreen. This was a series of manufactured screening systems of various types that used a specialized stainless steel grating to exclude debris while allowing a full flow of water. The design made the screens self-cleaning, requiring little in the way of regular maintenance. This seemed to be the way to go forward, keep the collection site pretty much were it was, but improve the collection and flow of the water, while eliminating the regular trips up the hill to deal with the environmental variables that always caused problems. That, and get rid of the rusty barrel.

After perusing the Hydroscreen web site to get an idea about the types of products that might apply to my situation, I called them up and got Bob Weir (not that Bob Weir), the owner, on the phone. After some discussion, and a few email messages back and forth, I decided on a .5mm pitch screen and collection box with a capacity rated at 50 gallons per minute, more than I would need to feed the house and my hydro plant. Bob was very accommodating, and gladly accepted a few modifications that I proposed. These included three 2" outlet ports instead of the usual one, and stainless steel flanges welded to the bottom of the box for mounting. He said he'd have it ready to ship in a few days, but I think he went right out into the shop and went to work on it, because it appeared to go out the next morning. This was not an inexpensive purchase, but one that would be well worth the cost in the long run.

Because I knew that I'd be making numerous runs up and down the hill with tools and materials for a good, long while, I convinced my neighbor to loan me one of his 4x4 ATV's so I could get up and down more efficiently. We were under Industrial Fire Protection I rules due to the dry summer, so I outfitted the quad with a bucket of water containing a saturated burlap bag, an axe, shovel, and an ABC fire extinguisher. This is required of all vehicles operating off-road in the National Forest, and I wasn't going to be the one to burn down the neighborhood.

The first task was to clear the road again, it had begun to grow brush back after the year before's clearing, and there were the usual downed trees and limbs to deal with. Hauling the chain saw and motorized brush cutter up the hill with a quad was a whole lot easier than trudging along with it over my shoulder. It also had the bonus that I was bringing loads of firewood down off the hill in a small trailer from my clearing efforts. I'm still burning some of the wild cherry that I cut then.

The quad would get to within 300 feet or so of the work area, then it was on foot to the site. Since I'd be handling tools and materials over what was less elegant than your average goat track, I spent some tine with a shovel leveling and widening that part of the path, and cutting away additional woody debris that I'd been stepping over for years.

I knew that part of this gig was going to be dressing and securing the two penstock pipes and one domestic water pipe down the hill, so I set off with the brush cutter and saw to clear and level a foot path the entire length of the pipe run to the knoll overlooking the power house (milking shed).

It was a fair amount of physical activity, taking four or five days to get the most part accomplished. This would have been down time while waiting for the Hydroscreen to be delivered, so I felt I was making use of the days as best I could.

The Hydroscreen arrived in good condition along with some additional pieces that I ordered to facilitate the installation, so my focus turned to designing and building a frame to mount the water collection equipment on the steep rock face up the hill.

I purchased some Azek decking material to build a frame to support it on the rock face where I would be collecting the water up on the hill. This decking material is pure PVC plastic, so rot was never going to be a problem. Assembly was made with stainless steel decking screws, eliminating rust and corrosion there as well. A two-piece stainless steel ramp was fabricated out of 16-gauge material to speed and smooth the flow of water before it reached the Coanda-effect grating. The entire fabrication was completed and assembled in my garage before being taken up the hill.

Securing the collection system to the hill was going to be an interesting challenge. the rock face the water flowed down was at about a 45° angle and very uneven. It was obvious that I'd need to be able to adjust the height of the supports as I built them, so stainless threaded rod was chosen as the material, along with stainless wedge anchors, and all stainless hardware, nuts, washers, couplers, etc.

In order to properly set the anchors in the rock, I needed to grind smooth flat spaces around each anchor. This all necessitated use of power tools, so I hauled a small gasoline generator up the hill, along with an electric angle grinder and 1/2" power drill motor, stacking most of it on a pallet cribbed up on a pile of rocks and tree debris.

The screen frame was temporarily supported on surveyors stakes and a plywood cutout of the Hydroscreen collection box profile held to the frame with c-clamps, and adjusted until the frame was at the right height and angle, and level from side-to-side. then measurements were made, the level spot ground into the rock, and the wedge anchor set in a hole drilled into the rock face at the proper angle. For much of this part of the project, I had to suspend my "everything-straight-and-true" work ethic, because there were some odd angles involved, but in the end, the frame was attached just as I wanted it. Here's a shot of that process getting underway:

See that giant old-growth Douglas fir log laying diagonally at the top of the photo? This will become a problem very soon.

With the frame attached to the rock face and the Hydroscreen and ramp system installed, it was time to figure out how to conduct the water flowing from above the site onto the top ramp. I had figured that some sheet metal laid on the rock, perhaps with a lip that went into a groove cut into the rock might suffice. I needed to boost myself above that old log and see what things looked like from above, a "waters-eye view" as it were. As in the many times before that I went up there, I boosted myself up the slippery wet rock by pulling on nearby shrubbery and the big log as an anchor.

This time, for whatever reason, the huge log came loose, and the end started to shift down the hill, rolling on cobbles and stones that were acting like ball bearings. I watched, horrified as the log slowly moved straight for my just-installed frame and screen. It stopped less than three-quarters of an inch from the top of the frame!

Emergency! I quickly completely disassembled the ramps, screen, and frame, moving it all out of the area while I contemplated to do about this. This was a mostly rotted log that was perhaps forty feet long that had collapsed across the ravine probably 150+ years ago. I was much more that I could ever hope to completely cut up and remove in this setting. What I ended up doing after determining that the log had settled enough to work on, was to cut slices off the downhill end that were light enough for me to pick up and carry away. there were mostly 4" or so wide, as the interior of the log wasn't completely rotted. After the first cut, it seemed that the end that was on the rock face had been holding the rest of the log in suspension, and it was now resting on the soil to the side of the site. All of the removed wood had to be carried away from the ravine, as any loose material tended to roll into the center and create obstacles I would be tripping over.

In the end, this actually simplified the rest of the installation, as now I had more direct access to the water, now that it didn't flow around the butt end of the log on the rock face. Somewhere along the way, someone gifted me a four-foot section of 12" PVC sewer pipe that had been cut in half lengthwise, forming a couple of useful troughs. I was able to make a more natural entry for the water by building up the stream bed with native rock and soil, putting the pipe sections in the middle to catch the stream of water above the collection system where it exited a shallow gully before spilling down the rock face.

This turned out to be a very effective method that didn't require any concrete or other extreme measures to catch nearly all of the water. Only a small, pencil-thin trickle escaped the troughs, which directed the water to the top of the ramp feeding the Hydroscreen.

One small problem presented itself: because I had aligned the ramp system to the water that had been running around the bottom of the now-gone log butt, the flow coming off the second trough was at an angle to the trough, and about 30% of the flow was skipping off the ramp and over the side, effecively being wasted.

A little brain power, some sections of stainless sheet from the front of an old refrigerator and some pop rivets, and I had a hood to redirect the water and straighten it to the direction of the ramp. The inset shows the internal fins that help make the water behave.

With these final adjustments, the water collection system was complete enough to connect the old 3/4" pipe leading to the house, and provide a reliable supply of essential liquid-of-life. In fact, the new system worked so well that it nearly eliminated the too-often trips up the hill to see why the water had quit, usually at the worst possible time. With self-cleaning action, and effective filtering of particles and debris down to .5mm in size, trips to the water collection system became a seasonal occurrence, rather than a two-to-three times a week annoyance.

Here's the completed collection system in action just after the first major storm of the Fall season that year:

That night, I laid in bed listening to the rain hammer the roof, wondering how the rocks and soil that I placed at the head of the troughs and ramp were surviving the onslaught. The next morning, a trip up the hill confirmed that everything was still in place and not damaged by the very large flow of runoff that it had endured all night long. It looked like I had a winner!