All posts by David Strip

Dealing With Snow

One of the challenges of living where we do is dealing with snow in the winter. A snowfall of 6″-8″ is relatively common and with about 500′ of driveway on a hill, shoveling is not a viable option. It’s even less an option if the snow is 18″ deep in a single storm. We have a variety of tools to deal with snow, starting with the humble snow shovel, which is used for tight areas close to the house. Because you have to walk outside to get between the house and the bedroom, we keep a snow shovel outside the bedroom door on nights when snow is in the forecast.

For more serious work clearing sidewalks and patios, we have a 26″ walk-behind Craftsman snowblower. For mid-sized areas this is the right tool, though for some reason the wind is always in whatever direction blows the snow back on the operator.

Next up on the hierarchy is the tractor-mounted snow-blower. It clears an 84″ path and can (at least in theory) deal with snow 30″ deep, throwing it 30-50′ to the side. This is the weapon of choice on the driveway as it throws the snow well clear of the driveway, which reduces the length of the mud season in the spring since the snow is not just pushed to the side, just to melt down the driveway for weeks in the spring.

Snowblower in action

The final option is to use the plow on the truck. The plow will get closer to the ground than the snowblower (which will throw dirt from the driveway if set too low) and is easier to operate for long distances, as would be the case if I had to plow all 8 miles back to the paved road, which has happened once, and 4 miles to the junction, which has happened a couple of times. I tend not to use it on the driveway except for the last inch or two left by the snow-blower so as not to build up a wall of snow on the side of the driveway.

The plow is made by SnowDogg and has a 96″ wide stainless steel blade, which means no painting and no rust and the ability to clear a path the full width of the truck even angled all the way to the side. It’s pretty tall, too. Tall enough that with wet snow it can build up more snow than the truck can push – another reason to clear the path first with the snow-blower.

Clearing the road or driveway eventually runs out of steam – you’re trying to get someplace and you can’t plow the whole way, or at least you don’t want to. For that, there’s THE BEAST – a 2013 Toyota FJ Cruiser with 3″ of lift and 35″ tires. There’s also a locking rear differential and A-TRAC traction control. The 35″ tires give about 15″ clearance at the lowest point, quite a bit more under the rest of the vehicle, meaning it takes quite a bit of snow to pack under the car and stop it. (Unlike the RAV4 which can hang up in 9-12″ of snow).

Windmill Woes

What’s wrong with this picture? Yes, the blades of the windmill are missing. The fan assembly came off about three weeks ago, apparently due to a failed cotter pin. We’ve been trying to get it fixed but the well service firm that erected the windmill has decided they don’t want to work with us anymore (in spite of the fact that we have of course paid every bill promptly and in full, no haggling, no delays.)

Fan on the ground.

The fan is 10′ in diameter and weighs about 75 pounds. Together this means it’s no easy feat to get it up the tower and installed. My first couple of attempts to reach tech support at Aermotor, the firm that made the windmill, were futile. Some of the sales guys gave me information on the required cotter pin but didn’t say anything about what it would take to restore it to it’s proper place.

Thinking it was just a matter of getting the fan up the tower and sliding the shaft in place, I worked out a way to rig an upright, a boom, and a pulley system. We bought a 24′ extension ladder, which gets me up to the platform. I pulled a length of 2.5″ water pipe into a vertical position and anchored it to the platform on the tower.

Tower rig

Not visible in the picture is the boom which slides into the upright pipe and provides a 4′ arm to support a pulley. I got this all done before lunch today. I came back to the house for a lunch break, which is when I finally got hold of someone at tech support. He explained it wasn’t just a matter of slipping the shaft back into the gearbox and reinstalling the cotter pin. Apparently some other gears are in the way and have to be disassembled first. He was pretty emphatic that there was no safe way to do this with the windmill on the tower.

After lunch our friend Ramey gave me the name of a well-service firm in Española. I called immediately and got a friendly voice on the other end. He’s backed up about two weeks, but seems to be very happy to do the work. To top it off, he’s familiar with Aermotor windmills, so he knows what he’s doing. With a little luck I’ll have another post in 2-3 weeks with a repaired windmill to show.

Wildlife at the Windmill

The next step after getting the windmill up and running was to lay some pipe and place a stock tank where wildlife would be hard to spot from the road. (Sure, it would be illegal to hunt on our property, but that doesn’t stop everyone around here.) The stock tank is about 75′ west of the windmill, slightly downhill along the spine of the hogback. Animal tracks started appearing around the tank the very night we first filled it. I placed a motion-activated game camera on a post by the tank two days ago and here are some shots of our visitors. The black and white shots are infra-red images taken after dark. The camera has an infra-red flash, which the animals can’t see, but reflects from the eyes, giving a glare in the image.

The Windmill (Part 2)

In late September the well service company (a one-man operation with an occasional extra hand) showed up with the windmill on the trailer, along with some pipe and “sucker-rod”. They accessed the site by following the road to the edge of the property where it has dropped to the valley level, then back-tracking across the valley and more gradually up the hill. Even this was a bit much for the load, which wasn’t well anchored. In the process of crossing the valley they bounced several lengths of pipe off the trailer. With the trailer on the site it was pretty straightforward to use the truck-mounted crane to lift the windmill off the trailer and lay it on the ground. By the time all this was done it was early afternoon and the thunderheads were rolling in. That was it for the day – too dangerous to be working with the crane in an electrical storm

Lifting the windmill from the trailer.

The delivery of the windmill was followed by a three-week hiatus while we waited for Mike (the well-servicer) to return and get to work on the well itself. The first step in refurbishing the well involved puling out the old 3″ pipe and sucker-rod. This is a pretty straight-forward operation in which the crane is used to lift the entire “string” of pipe until a section clears the opening of the casing. A metal plate with a slot cut in it is slid around the pipe belong the coupling to support the pipe. (The plate is called a “skillet” because of it’s similarity in appearance to a cooking skillet.) Large pipe wrenches are used to unscrew the upper pipe from the coupling and that now-free length of pipe is laid on the ground, and the process starts over until all the pipe is out of the well.

The second step is to “shoot” the well, which involves exactly what the name says – a .357 revolver is fired repeatedly into the well. The impact of the bullet in the water is believed to create a pressure wave that emanates outward and clears debris from the filter screens at the bottom of the well. Most of what I’ve read says this is a myth, which seems the likely explanation.

After shooting the well, the next step is bailing. Just like the name suggests, this is like bailing a boat – you’re pulling up the water and dumping it. Unlike a boat, you’re not going to reach down with a bucket. Instead, the process uses a 15′ length of 3″ diameter pipe with a valve on the end. The valve is basically a metal plate that closes off the end of the pipe. The plate has a bar attached to it, pointing straight down. When the bailer reaches the bottom of the well the bar pushes up the plate so water can flow in. (Of course, if the water is deep enough it just flows into the bailer from the top as it sinks into the water. When the bailer is lifted the valve plate closes on the bottom of the pipe and holds the water. The bailer is then pulled up and out of the casing and the crane swings it around to some place where it can be drained by lowering it until the bar on the valve plate is pushed up and the water flows out. Since the well had not been used for probably 50 years or more, the water was pretty nasty – filled with all sorts of slimy plants, sludge, and generally smelly crap.

Once the well has been bailed, you reverse the process and start dropping a string of pipe into the well. The bottom of the string is the working tube, which is a brass cylinder with a polished interior surface. This tube is about two feet long and is screwed to a length of water pipe, which is typically 21′ long. You lower the pipe until the coupling sits just above the casing and slide a “skillet” around the pipe so you can detach the pipe string from the crane. You lift the next piece of pipe with the crane, screw it to the coupling, remove the skillet and lower the new length of pipe until its coupling sits on the skillet at the top of the casing.

Attaching the last length of pipe. The skillet is sitting on top of the casing supporting the coupling.

Before the last length of pipe is attached, a disk with a hole big enough for the pipe but smaller than the tee at the top is slid onto the pipe to support the pipe string on the top of the casing. (Or in our case, you lower the last length of pipe, realize you forgot the disk, lift the last length, remove it, insert the disk, then screw it back on and lower it again.)

(There actually was a two-week break while the pipe was lowered into the well. We unfortunately didn’t have the right combination of lengths to reach the bottom of the well, which meant breaking off work until a 10′ length was purchased and brought back to the site.)

Once the pipe string is in place you drop a check (foot) valve down the pipe where it falls to the bottom of the working tube. (Why don’t you just place it in the working tube at the start? Probably because dropping it is more fun.) The next phase is to install the pump. The pump piston is a machined brass assembly with “leathers” (now silicone) that seal to the sides of the working tube. A floating ball valve allows the water to pass the piston when the piston is lowered and supports the water when the piston is lifted. The piston is connected to lengths of sucker rod that screw together. These are lowered in a manner similar to lowering the pipe string, with each length supported on top of the well casing as the next is screwed in place. In spite of the fact that the bottom of the string of sucker rod will live under water, the sucker rod is wood, possibly ash, maybe oak.

Attaching a length of sucker rod

At this point the well is complete and it’s time to put the windmill over it. The windmill is first moved so that it’s immediately adjacent to the well pad. It’s then lifted from an attachment point near the top so that it’s basically pivoting into an upright position. In the picture below Tony is holding back the top of the tower so that it doesn’t swing too far and too fast once the center of gravity shifts to inside the contact point.

With the tower upright over the well pad, it’s all hands on deck to rotate the tower to the right orientation (the holes in the base are not symmetrical, of course) and then to guide the tower over the anchor bolts cast into the concrete pilings. The anchors are 5/8″ and the holes are 3/4″, so while we have leeway, we don’t have much room for error. We get the tower onto the bolts on the first try. No “persuasion” was needed to make things align. That’s two successes in a row on my concrete casting. With the tower in place we adjust the nuts under the base to level the base. Another set of nuts is threaded onto the achors above the base to lock it all down.

In addition to 12″ diameter, 3′ long pilings, the tower is anchored with guy wires from the corners. Although they had a ladder on the truck, it was deemed easier to just climb the tower. Clearly not an OSHA-approved procedure.

Tony climbing the tower

All that’s left at this point is to attach the sucker rod from the well to the mechanism in the windmill itself. That involved cutting the last length of rod to match the position of windmill mechanism and we were off and running.

Two weeks later Elaine and I are back with a 350 gallon stock tank and a bundle of PVC pipe and parts to provide a watering tank for wildlife. We dig a shallow trench from the windmill to the tank, about 75′ away. An adapter, a union, a couples of elbows, and we’re attached to the well head. Then it’s just a straight run of pipe to the tank. Pretty easy work.

The well head piped to the stock tank

Post script: While this all worked pretty well, water was flowing out of the top of the pipe on the up-stroke of the pump because the flow down the 1″ pipe to the stock tank was too slow. This was solved by adding a length of 2″ PVC pipe above the 2″ nipple above the tee in the photo above. This required detaching the sucker rod from the windmill. I followed the same procedures the service crew did to keep from dropping the sucker rod down the hole. In reality this was not much of an issue because even with the piston sitting at the bottom of the working tube, the last length of sucker rod still sits above the piping (or at least I believe it is.)

The elk found the stock tank the first night it was there.


Kohler 15-RYG Coolant Replacement

The Kohler manual tells you that you can drain the unit by opening the drain valve at the bottom of the radiator. This will drain about two gallons from the radiator (if full), but will leave about one gallon in the block. There appears to be no drain plug in the block, nor is there any way that I’ve found to snake a hose into the block to siphon it out.

According to the tech support people I talked to, if you are just doing routine maintenance changing the coolant and it’s still in pretty good shape, just add two gallons of fresh, diluted anti-freeze to the radiator and be done with it. If, like me, you ran radiator flush through the system before reading this, you will need to purge that from the system. You can do this by removing the lower hose from the radiator and running clean water through the block. A female hose end just fits inside the coolant hose to the block. You can even tighten the hose clamp for a better fit. Run about 10-15 gallons of water through the block to purge whatever is in there. Tap water is OK for this purpose. But now you will need to purge that with clear (de-ionized or distilled) water. I used a submersible pump in a bucket with clear water. You might be able to get away with lifting the hose end above the generator and pouring water into the hose using a funnel. You should run at least 3-4 gallons of clear water through the block at this stage.

Installing a Shelix Jointer Head

Spiral cutter heads for jointers and planers have been gaining popularity over the past several years. These heads utilize a large number of carbide inserts rather than the typical two or three high-speed steel knives found in traditional cutter heads. In addition to the greater durability of carbide compared to HSS, each insert has four usable edges, so if you nick one, rather than having to resharpen a whole knife, you just rotate that one insert (or eventually replace it.) No more trips to the sharpening shop, an important consideration when it’s 2+ hours away. Below are the steps required to replace the head in my 8″ Woodtek jointer. This is a pretty generic Chinese import beneath the green paint, so the method probably applies to a wide range of import models. Byrd Engineering, the maker of Shelix heads, offers an 8″ head for these generic Chinese jointers. The main issue is whether you need the metric variant or not. The Woodtek does use the metric version.

First step is to unplug the machine. (You presumably knew that.) The inserts on the cutter are extremely sharp, so you will want to wrap the head with masking tape to protect your fingers when you’re handling it.

Now that the preliminaries are done, remove the fence, the bracket for the fence, and lower the infeed and outfeed tables as far as they go. Remove the belts from the pulley and the pulley from the shaft. Take care not to lose the square key that locks the pulley to the shaft.

The bearing blocks are held in with nuts that are underneath the body of the jointer. On my machine they required a 14mm wrench.

Once you’ve removed the two nuts, you can lift the head and the bearing blocks. You now have to remove the bearings from the shaft. This may be difficult without the right tools. I happen to have a 20 ton arbor press, which made pretty short work of the job. I hung the bearing block on two pieces of angle iron laid across the press and pressed out the shaft, taking care not to drive it so far that the head dropped on the floor. A gear puller is an inexpensive alternative if you don’t have access to a press. If you don’t want to deal with this step, you can buy new bearings when you buy the Shelix head. Compared to the price of the head, they’re pocket change.

With the bearing blocks removed, it’s now just a process of reversing all the steps, starting with pressing the bearings onto the new head.

Now place the assembly in the jointer body, reattach the pulley, the fence bracket and fence, and raise the tables back to their normal positions. There’s no shortage of guides on the web about aligning the outfeed table to the cutters and the infeed to the outfeed.

Wildlife Friendly Fences

Last spring the Forest Service fenced the road that runs through our property. It’s a fairly standard four-strand barbed wire fence. In spite of our requests (and the recommendations of the Natural Resources and Conservation Service, a sister agency to the Forest Service), the fence was not constructed in a manner to facilitate safe passage by wildlife. The easiest way to achieve this is to use smooth-strand (“barbless”) wire on the top strand. The fallback is to cover the wire with PVC pipe. Since the fence was already strung, adding PVC requires slitting the pipe so that it can be slipped over the wire. I expected to see a variety of postings about this on the web, but found none. I came up with a very simple solution for slitting large amounts of pipe on the table saw. A simple box-like structure aligns the pipe to the blade. It’s a simple matter of just pushing the pipe through about as fast as you can. The blade is completely enclosed, making this a very safe solution.

Splitter/riving knife at the back of the fixture keeps the pipe from rotating as it’s fed.

The Windmill (Part 1)

As mentioned in the previous post, Elaine bought Sonya’s old windmill with plans to install it over one of the old wells on the property to pump water for a wildlife stock tank. The windmill will need some sort of footings, so the first step in site prep was to pour 12″ diameter, 3′ deep concrete footings.

The windmill site is about 1/2 mile south of the house, out on the end of a “hogback“, a usually steep-sided ridge formed by erosion. There is of course no electricity or water at the site, so everything is loaded into the pickup – a generator, cement mixer, 1260 pounds of concrete sacks, four 5-gallon buckets of water, a 55 gallon drum containing about 40 more gallons, and assorted tools – shovels, post-hole digger, level, … . Finally, a template for locating the anchor bolts.

A truck-load o’gear

Here’s shot of the site, behind the truck. (Why behind the truck? Because I took the picture after we were done.)

The well site

It’s a pretty bumpy ride, smashing through the sage. The shot below shows what we have to navigate to get back to the road. Directly below the RAV4 the bank slopes at about 40°. Behind the RAV4 the slope is gentler, maybe 20°. I drove mosty across and somewhat down the slope. After doing it a few times it stops feeling like the truck is going to roll over.

I had help on this job from John Rohr, who helps around the ranch, mostly with Elaine’s garden and other nasty/physical jobs. I had the holes half-dug before John came up. He helped finish the holes and helped with all the mixing and pouring. Even so, it took almost five hours on a pretty hot morning.


We used 12″ sonotube and three lengths of rebar in each tube. I tried to tie the triangular arrangement but it was such a mess that I ended up welding the rebar – it was much faster and a lot more rigid. The top of the old well is centered between the footings. The “sucker rod” is free enough to move by hand (with a fair amount of effort – you’re pulling up 40′ or so of rod). A 2″ galvanized pipe runs about 100′ down the hillside (to the right in the photo) to where the stock tanks must have been.

A few days later I returned to the site to check on everything. I walked up the spine of the hogback from the direction of the main road, a path I hadn’t taken before. About 10 yards short of the footings I stumbled on some relics of the past. No idea what they are other than some sort of long-forgotten machinery. One piece has “McCormick” cast into it. If this is the same McCormick as the inventor of the mechanical thresher, that company was bought out by International Harvester in the early 1900s.

Wrapped Around the Axle

About two years ago Elaine bought Sonya’s old windmill, with plans to place it over one of the old wells on the ranch so that it could pump water for a stock tank for wildlife. Ever since then we’ve been trying to get one of the local well-maintenance people to move the windmill from Sonya’s, repair the pump in the well, and erect the windmill. Finally last week one of the drillers came by and committed to completing the project. The first step is for us to install footings for the tower. These will be cast concrete, three feet deep. I started digging the holes for the Sonotube forms on Monday. The well is about 100 yards off of the side road that crosses the ranch and required some “cross-country” driving to reach the well. Although I detoured around the old fencing, there apparently was a cache of old fence wire buried in the brush that I driving over. The wire wrapped itself quite thoroughly around the right front axle and the drive shaft as well. After trying various approaches to removing the wire I finally grabbed my angle grinder and put a cutting wheel on it. I was able to separate the wire into two large chunks that I could pull out over the hub. (I had already removed the wheel.) Needless to say, future visits to the site involve a more careful search for residual wire.

The cut chunks of wire.

Router Table safety interlock

Safety consciousness in the shop is of paramount importance – a moment’s inattention can cause lifelong damage, especially to your hands. In extreme cases, people die from shop accidents. I don’t think I know anyone who pays more attention to safety than my friend and fellow woodworker Derek Roff. Hands-free switching of the router in a router table means you never have to let go of the workpiece. A footswitch solves this problem, but introduces a new hazard; you could accidentally step on the switch and spin up the router. Several years ago Derek built an interlock using high voltage AC relays that solves this problem. In order to start the router you need to step on the footswitch and press a momentary pushbutton. The router stays on as long as the footswitch is depressed. Inspired by this device, I’ve designed a next generation interlock that takes advantage of cheap microcontrollers and low-cost relays to add features to the interlock while lowering the cost.

New safety features

Like Derek’s original interlock, you must step on the footswitch and press the start button to start the router. Unlike Derek’s, the new device enforces a strict ordering on the switch presses. You must first step on the footswitch, then press the start button to start the router.

The more important addition is a delayed start mode. If you first step on the footswitch, then double-click the start button, the router will start five seconds after the double-click. This gives you time to get both hands on the workpiece before the router spins up.

Convenience features

It is quite common to have some sort of dust collection, often a shop vac, attached to your router table. The interlock provides a second receptacle for powering your shop vac. In order to reduce the likelihood of popping a breaker, the interlock provides a 3 second delay before starting the vac. The vac stays on 3 seconds after the router stops (when you release the footswitch) to catch a few more chips.

The device provides separate jumper-selectable soft-start for the router and the shop vac, mainly to reduce the starting surge load on your breakers. When the soft-start is enabled, the router and/or shop vac are spun up to speed over a 2 second span. Because of how soft-start is implemented it will not work correctly with routers that have built-in soft-start. It also might not work correctly with variable speed routers, like the Porter Cable 7518 that I have in my router table. Soft-start is important on hand-held routers, but less so on table-mounted routers, so the loss of the ability is not a big deal.

The Build

The heart of the device is a custom printed circuit board (PCB) that I designed using KiCAD, a free, open-source electronic CAD program. The boards were fabricated by JLBPCB in China. The smarts are provided by an Arduino Nano, which was selected for its low price (<$2) and ease of use. The finished product, mounted on my router table, is at the top of the post. Here are the guts:

And here it is mounted in a box:

The front panel is 1/4″ acrylic, cut on my CNC router. The lettering is cut on the back while the cover paper is still on the acrylic. The routed lettering is sprayed with white paint. After the paint dries the paper is peeled off the back and the entire back (inside) is sprayed black.

It is beyond the scope of this blog to teach you how to build this. If you already know Arduino programming (or at least how to “burn” an Arduino) and can solder a fairly simple PCB, the files below contain the KiCAD schematic and board layout, the bill of materials, and the Arduino code.

Tip of the hat to David Strafford for tirelessly answering all my questions about the electronics I don’t understand.