Room Temperature Sensor; got to use the 3D printer for real

I used my 3D printer on a real project, the room temperature sensor. I spent a ton of time messing around with various drawing tools and learning a little about slicing software, but didn't actually do anything that could be considered useful. Unless you count a stand for my Star Trek Enterprise that I have on the kitchen window sill:

I got the files for the stand off Thingiverse <link> and just printed it. Thingiverse is wonderful for getting things that other people have worked on, but it isn't the same as coming up with your own idea and stepping through the entire process.

I decided on the basics of an enclosure for the sensor weeks ago, but it took a while to decide on a tool to use to design it, then the actual design process. I tested a couple of enclosure ideas for a few weeks and finally, a couple of days ago, put four more of the sensor electronics together and printed enclosures for all of them. I have five areas of the house being monitored and recorded in my home database that I can chart and examine.

Yes, they follow each other really well over the period charted, but remember, a temperature change of a small amount will make a room feel hot. It's interesting to watch this phenomenon. One room is 75F and the next is 78F, not enough to matter, but it feels really hot in the warmer room. It doesn't help much to put a thermometer in the warmer room so you can see the actual temperature; you just think the thermometer is wrong. Strange.

Here's what the entire assembly looks like on the wall:

It consists of a base plate that you mount to the wall and a cover that slides over the base to protect the wiring. Here's the two pieces before mounting:

Most of the engineers out there could have drawn this up in about 10 minutes, but frankly, this process was tough. Every little item had to be measured and thought out carefully. I had many, many test prints of sections to see if various things actually fit together. I kept running into little unexpected problems. The plastic would be a little too rough in some spots and pieces wouldn't fit together. The model would glue itself to the base of the printer and it flatly refused to come off. I didn't put holes in the sensor PC board and had to come up with a mounting solution for it. How tall EXACTLY should a stand off be for the various pieces?

Most of this stuff was trial and error. Lots of errors. Just look at the pile of various attempts that will go in the trash after I post this:

One thing I discovered towards the end of the project was that I could cut the model and only print part of it. That was a real time saver. I could print the section I was trying to settle on and test it without waiting hours for the printer to finish the entire thing. I didn't read about using this capability anywhere on the web. Lots of folk talked about cutting the model, but didn't mention how it could be used to test a section of it. I guess most folk use 3D printers to make single piece items.

It's surprising how many little things creep up and have to be solved. The battery holder:

was initially a bit of a problem. I didn't want to mount it flat to the base plate because the mounting screws would go through the plastic, so I put in a couple of standoffs for the two screws that go through the holes in the middle of the assembly. Of course, that made it rock on the screws when I messed with the batteries. I solved that by putting in four more standoffs that kept it from moving.

Th circuit board with the electronics on it didn't have holes for mounting screws (oops).

So I had to work out special standoffs that could hold the board in place. Of course I didn't have washers that would fit the standoffs, so I just printed them as part of the base. Here's what the base wound up looking like after I solved these problems:

The battery holder has six standoffs two of them for actual mounting and the others to keep it steady. The standoffs for the PC board have a cutout for the board and I use little washers to hold it down securely:

The little dark spot in the center-left is a moth that was attracted to the heat and became part of one of the sensors. Note to self: close the printer door stupid.

I slid the lid up on one of the sensors so I could show off the construction of the assemblies:

If you zoom in on the picture you'll be able to see the mounting arrangement more closely. They were simple to install since they only take two #6 drywall screws and the entire device doesn't weigh much.

This project has been a lot of fun over the last couple of years. Yes, YEARS. And it isn't finished yet. I still want to use the sensors to replace the two house thermostats by taking the readings of each room and turning on heat, cool, or just fans to control the internal environment. I want to chart the outside temperature along with the rooms to see how it affects things as the day passes from season to season (notice, I only have two seasons here: nice and too-darn-hot). I may even create outdoor sensors to measure the wall temperature on some of the outside walls. That will prove the insulation effectiveness.

Who knows what I'll do with this in the future.

Tractor Hydraulics - Cylinders

I'm still waiting on my valve assembly, but I thought I would write a bit about repairing the cylinders on the tractor. The cylinders are what do the real work of lifting and moving things. The engine turns a hydraulic pump and that shoves high pressure oil down a series of tubes and hoses to extend or retract a rod. The rod is attached to whatever thing on the tractor you want to manipulate.

This is a picture of the tractor with red arrows pointing to the various cylinders visible on one side. These things are awesome; using the power of fluid at high pressure, they can manipulate the various parts around pivots to lift, dig, dump, etc.

They're also the thing that many people won't even attempt to fix themselves. While they will eagerly take on rebuilding an engine, they'll send a cylinder to the shop for someone else to disassemble and fix. This option could be justified if the rod is bent or something actually broke inside, but most of the time it can be repaired on a table in the garage using tools some of us already have, or can borrow. So far, I've rebuilt seven of these things and all of them work fine. No leaks.

First, let's talk about hydraulic shops. There are some really good ones out there that have excellent people doing great work, then there are the shops that I seem to find. There's one guy that is really good and a bunch that can't buckle their own belt. Guess which one usually deals with me? And, these places are expensive. A friend of mine had three cylinders done by Case; it cost him just a little less than $3000. This is under the price of replacement Case cylinders, so it was a better deal than buying new, but sheesh, that would make many folk abandon the tractor. What they did to me was say $300 to $400, but there may be problems, and that will raise the cost. Additionally, I have to take the cylinder off and get it to their shop.

So, they get the cylinder, take it apart and find a 'problem' that cost more than I can afford. I'm left with a cylinder that has most of the parts in a box, and a disabled tractor. No, I'll pull it apart and see what needs to be done. If it needs major work, I'll find a machine shop to do the work, or hunt for a replacement cylinder somewhere.

I'm not going to go into great detail describing how to work on one of these things, there's videos on youtube that cover it for the most part. Additionally, there's the perfect thread for someone attempting this for the first time over at tractorbynet.com <link>. Instead I'm going to tell you the things that you don't get from the videos and various web sites. There are just some things that they don't mention that I had to find out the hard way.

The single biggest complaint that folk have is that the cylinder leaks. It will leak inside, which causes the cylinder to change positions over time, and it can leak externally which also allows the cylinder to change position, but this time it gets hydraulic fluid all over the place. Both of these problems are fixed the same way, replace the various seals and o-rings in it and put it back in service.

If it's that simple, why don't more people do it themselves and save money? Some of these things are big, heavy, and all the fittings are stuck. You can't just grab a crescent wrench off the table and turn a couple of bolts to get them apart. And, to make it a little more complex, most of them require special wrenches. Here's the ones I had to buy to work on mine:

The two black ones, not the open end next to them.

It takes a LOT of force to open one of these cylinders, so a piece of pipe to use as a cheater bar is necessary to break some things loose. I used a three foot pipe that fits over the various wrenches I used. Let's take a look at what's inside one of these:

I know, it's a bit hard to see, but I wasn't thinking about posting it when I took this picture. Here's a parts diagram of one of my cylinders; they are all basically the same:

Part number 1 is the long silver rod in the picture above. 4 is the external oil seal. 6 - 8 are what's called the 'gland' and they go on the rod, but at the end near the rod mounting pin 3. 9-13 are the piston.

So, assembled the pieces look like the photo above. This is what you'll see when you finally get it out. But first you have to overcome the gland and external oil seal.

My tractor had two styles of glands. The one in the diagram above that is simply tapped into place and another that is threaded and you screw into place. Let's talk about disassembling the tap in one first.

It's really easy to put these things in, but not so when you try to take it out. First you have to unscrew the external oil seal. I talk about this process below. Then the press-in gland removal instructions I found said to drain the cylinder of fluid, then repeatedly tap the piston against the gland by shoving the rod in and then pulling it out, slamming it against the gland until it comes out of the cylinder. That didn't work at all. First, you simply can't get all the fluid out and each time you pull the rod, it hits the fluid inside and stops before hitting the gland. Also, the fluid sprays out one of the fittings and makes a mess.

How did I get it out? First, I unscrewed the external oil seal on the end before disconnecting the end nearest the bucket (more detail below). Then I set it up like this:

Yep, I pulled the Jeep in front of the tractor and hooked the wench up to it. This worked pretty well on all of them but one. That one was really stuck, and I actually pulled the jeep across the ground without getting the piston to come out. To make that sucker give up, I had to get inside the jeep and stand on the brakes so the front wheels weren't turning and run the wench. The two rear wheels weren't enough friction to make it happen. This is the gland, rod and piston from that cylinder:

There's two things that are important. The yellow and rust thing on the right is the external oil seal that holds stuff in, and it is hard to get loose. I used the wrench shown above and a three foot cheater pipe, and still had to hit it with a two pound hammer a couple of times to get it free. Some people use a pipe wrench, and that will work too, but it's really hard to keep the pipe wrench on while applying pressure. You have to get a neighbor to hold the pipe wrench in place while you stand on it and jump up and down. No, a longer wrench won't help; it's just harder to control. If you go the pipe wrench path, be careful not to crush the oil seal. This bugger is easier to break than you might think having just pounded on it with a hammer, but read on.

The other thing is the silver nut on the left. These things are put on with several hundred pounds of torque and are over an inch across. I used an 18 inch crescent wrench with the cheater pipe to free it. But, how did I hold it in place while applying that much pressure? That was easy, just put the rod back on the tractor but not in the cylinder. Then, using the entire tractor as your vise, go for it using the cheater bar to break it loose.

Be careful if you decide to use heat on this nut. It's one of these:

It would be really easy to melt the material inside the nut that helps keep it in place during use, and these nuts are expensive.

That brings up the question of how to torque this thing back on. What I did was to use the cheater bar and crank it down as tight as I could after putting a couple of  SMALL spots of RED Loctite on the threads. They've held well for quite a while now. Same thing with the external oil seal, but don't use a cheater bar on this. Just put the spots of Loctite on the threads then snug it down tight using a wrench. I'll tell you why not to use a cheater bar on it below. Lots of folk will disagree with me on the red Loctite because it's meant to be permanent. You normally have to heat it up to around 400F to get it to let go. However, with the invention of impact drivers and wrenches, mechanics found out that sharp impacts will fracture the bond and let you remove it. Since you'll probably have to hit it with a hammer anyway, that should be enough. Also, nitromethane solvent will soften the Loctite, but that might take several hours. Yes, I had to go through this process and I'll write about that below.

Now for the other kind of cylinder assembly where the gland is screwed in. These are usually larger cylinders that are much heavier and will take some leverage to move around. Sorry, I don't have a good picture of this, the camera wouldn't focus well enough. Here's the rod, gland and piston from one of those cylinders.

Here's what the outside end of it looks like:

The end is actually two pieces; one that screws directly into the cylinder and the other that screws into the first piece. The external oil seal is the bottom piece, it has a seal in the end to keep dust collected on the rod from getting inside as well as a couple of o-rings inside to hold back whatever oil may get to it. The upper one is the actual gland and has packing inside that the oil seal helps keep in place. Both of them are hard to remove, but the external seal is the easiest. The gland was really hard. If you look closely you'll see burn marks where I heated it up to get it loose. I took an acetylene torch and ran it around the outside until I was able to turn it with a wrench and the cheater bar.

The end nut that holds the piston together was not as hard as I expected, I only needed the cheater bar to get it to move.

The reassembly after replacing all the o-rings and seals was pretty easy, but I managed to mess it up. When I put the external seal in place I over-tightened it and broke it:

This was the first cylinder I rebuilt, so I have something of an excuse. I thought that since it was so hard to get off that it should be torqued down and it just crumbled.

Turns out that these are often made from cast iron, not steel. The cast iron just couldn't handle it. Man, was it hard to remove. It was as tight as it could be and had red locktite on it. When I went to replace it I used a chisel and pounded it pretty hard for a while; it finally gave up and turned a little bit. I got a replacement for just a shade under $100. Total rip off, but I couldn't find a used one anywhere.

Now, let's consider the replacement seals. If you go to a dealer, they'll sell you a kit that has the necessary seals and such in it to replace everything. I actually found seal kits for this tractor, but ... $200 for some o-rings?? There had to be a better way. Calling around the various places that serviced hydraulic systems was worthless, they all wanted to sell me the expensive kits or do the job for me. I finally talked to a guy that suggested going to a store that specialized in this kind of thing. I got a hint from the parts guy at a local AutoZone, Martin Fluid Power.

I found Martin Fluid Power on the web and they had a store in town, so I loaded up the first cylinder and went there prepared to show them exactly what I needed. They were great. The guy measured each spot where a seal was supposed to go and gathered up the parts right there in front of me. I took my bag of stuff that cost $25 and went home with it. I just saved myself $175 a cylinder.

What was there originally for packing was called a Vee Pack. This is a specially shaped seal that you combine a number of to fill the space. The orientation of these is important. There are two kinds of basic cylinders: single action and double action. Single action only push one way, they are returned by springs or the weight of what they are moving. Common examples are a floor jacks or the jacks under motor homes. Double action get fluid pressure on both ends to move it in or out. Most cylinders on tractors are double action. The only exception I know of are the swing cylinders on some backhoes, they use two single action cylinder in opposition to swing that big boom around. The Vee Packs are arranged differently in the two types. Here's a picture of mine to supplement the ones in the thread I recommended up top:

See the shape? If you were to cross-section them they would look like a 'V', hence the name Vee Pack. You basically point the open side toward the fluid pressure so they expand a tiny bit and hold back the fluid. What makes it a 'pack' is that it consists of several of these with a special spacer in the middle; some of them point one way and the rest point the other. Out there on the web are some illustrations of this and how they hold back the fluid and move the piston back and forth in the cylinder.

Naturally, being me, I didn't replace them with a normal Vee Pack. I used what they called a Poly Pack. These are high tech materials that work the same way, but survive much much longer. You almost never see Poly Packs on tractors, but for a tiny bit more money, they should last until the tractor wears me out. They look like this:

 

They have the same action of expanding one edge a tiny bit under pressure to fully seal the cylinder, but the different shape and extra spacer piece in the middle work to decrease friction and wear. The material also helps with friction and keeps the temperature down during use. They work the same, buy enough of them that fit the particular cylinder and fill the space; put half of them facing one way and the other half facing the other. When you lay them out remember the wide end faces the incoming fluid.

I used these on the pistons as well as the interior of the external oil seals. They just work with no hassle at all. Getting the right size may be a tiny hassle, but a good parts man can help with that.

Lastly, a couple of suggestions for doing this. Take a LOT of pictures; this isn't the good old days when you had to have film developed. I didn't take enough, and when I put back one of the cylinders, I messed it up. Here's a picture of the installation showing the hydraulic hose:

Don't see the problem? Let's look a little closer:

Most of you see the problem, but for the ones that haven't messed with a tractor, the answer is, I got the cylinder in upside down. When the cylinder pivots up it will crush the connector. Guess what:

It did. If I had taken a before picture and then actually looked at it, I would not have had to order a new hose. That brings up another point hydraulic hoses.

Hydraulic hoses cost way too much. Well, that is if you go to the places that various folk recommend. There's a place here in Phoenix that everyone touts as being THE place to go for hydraulic hoses. I went there for the first couple of hoses and paid out the nose for them. The next set I went to AutoZone and got them. Frankly, if you don't live in a farming community, AutoZone is a pain because they don't understand pressure and flow well enough to put the hoses together. I talked the parts guy through it and got what I needed, but there's an even better way: go online. There's several places that sell hydraulic hoses online that have great customer service. You call them up describe what you need, maybe send them a picture or two, and they will put together exactly what you need. And, the price is fair.

I don't know if you noticed or not, every single place that I went to wanted to charge me several times what the stuff was actually worth. I fully expect to pay retail, but two hundred dollar seals, undefined labor rates that are under their control, triple the price hoses (or more) just isn't what I want to do. Look online for the stuff, the wait of a few days may save you hundreds of dollars and make an impossible project possible.

And,

When you pry the various things apart and scrape the crud out of the various grooves, be careful with the darn screwdriver:

The rods are heavy and covered with a film of hydraulic fluid. When you handle them, wear gloves that increase the friction between your hands and the rods. This will offer some protection and lessen the chance of dropping it on something soft:

Have fun.

Speaking of Tractor Hydraulics ...

A while back I bought an old, big tractor. I already had a small landscape tractor, but I wanted to be able to dig in this rocky soil and the little one just wouldn't cut it <link>. This big guy had been poorly maintained for many years out here in the Arizona sun and has been a constant pleasure for me to mess with.

But, wait, how can a poorly maintained tractor be a pleasure? In case you haven't noticed, I like to fix things; especially things that I know little or nothing about. I get to dive in, take it apart, figure out how it works, and eventually, actually use it for whatever purpose it was intended. This tractor has been a great project machine for the last couple of years and has been running (well mostly) the entire time. It's dug holes for me, the neighbors and a friend down the road; it also moves dirt for folk that managed to get a pile of it delivered to start a garden. Fun machine.

The valves for the front loader have been leaking the entire time, but hydraulic fluid is much less expensive than the $1800 valve assembly, so I just lived with the leak until about two weeks ago. I had several choices, replace the valve, fix the valve or install a different valve. Each choice, naturally, had its own shortcomings:

If I replace it, I'd have to shell out $1800 plus shipping for a new one. There are a few used ones that show up from time to time on ebay, but who knows their history? I'd have a new valve and it wouldn't leak.

I could send it to a hydraulics shop. I called a couple of hydraulic shops and the estimates ran from $700 to $1000 with the caveat that they might have to custom machine some parts that weren't available. That, of course would drive the price up. The up side is that it wouldn't leak.

Replacing it with a different kind of valve would require some modifications to the tractor and special fittings would have to be put in place. This was the cheapest solution in raw money, but would take some plumbing and fabrication time.

I decided to pull the valve loose from the tractor and see how bad it would be working on it. The hydraulics were weird compared to other tractors of the same era and the valve assembly was a lot larger than I initially thought. I just put it back on the tractor thinking I would decide later (after a couple of quarts of hydraulic fluid) which route I would pursue.

But (there's that word again), after putting it back, the front bucket wouldn't hold position. I noticed it throwing away a pile of weeds. I put the bucket about a foot off the ground in the up position and went to get a wagon load of weeds. When I got back, the bucket had rotated down.

After rotating the bucket up again, I put the weeds in the bucket and went to get another wagon full; when I got back, the weeds were on the ground and the bucket was down again. Taking it for a test drive, I found out that I had to rotate it back up every few minutes or it would wind up facing the ground. The arms worked fine, it was just the bucket that was trouble.

My trouble shooting skills kicked in and told me that messing with the valve put it over the edge and it was time to make one of the decisions above, so I decided to rebuild it myself instead of taking it to a hydraulics shop. Yep, that was choice number four that I hadn't thought of before.

Out came the big tools and cheater bar, and about an hour later I had the valve free from the tractor. The idea was to pull it apart, replace the various o-rings and put it back. So much for plans.

Here's a top view of the valve assembly:

The two handles are for the loader lift and the bucket tilt. From the back side, it looks like this:

People familiar with tractors might notice a couple of things. There are the usual 'work' ports, two each for the loader up down and bucket rotation cylinders and an input port for the fluid. What isn't there is the output port to return the fluid back to the system. That can be explained by this picture of the mounting plate after the valve was removed:

See the hole in the middle of the picture? It's the relatively clean black spot. That's the return line; they actually fed the fluid through the mounting plate to a pipe that leads to the frame of the loader where the reservoir is. This has the advantage that it will not rot in the sun and the disadvantage of making the valve almost impossible to replace with something else. Probably wanted to ensure the sale of parts later.

Here's the valve after I got it out:

I carried it inside the garage and started taking it apart. But before I show you the guts of the thing, I want you to notice the fitting on the very lowest part of the valve body in the picture. Doesn't that look like a galvanized pipe fitting? It was. Here's another view of it while it was still on the tractor:

I knew this wasn't supposed to be there, but didn't really understand what the problems might be. Notice the wet fittings from the leak; I have to keep in mind what the actual problem was I was trying to fix.

These are the two 'spools' that are the actual moving parts inside the valve body. The cylindrical cuts in them allow hydraulic fluid to move to the inputs on the cylinders depending on where you have the control levers positioned. The galvanized fitting is right there at the bottom of the spool on the right. I hunted down a parts diagram of the valve assembly to see what was supposed to be there:

I wrapped a red rectangle around parts that were missing. What happened was that the end cap (20 on the drawing above) was taken off and probably lost. The other parts (17-19) probably went with it, so they sealed that part of the valve body with a kludge of galvanized pipe fittings. Naturally, pipe threads didn't match the threads of the valve it was a little bit too small, so someone wrapped the fitting with about three yards of teflon tape and screwed it in that way.

This was the cause of the bucket rotating down on its own. When I took the valve off the first time I probably bumped the fitting and it moved enough to cause a small fluid flow, just enough to cause the bucket to move slowly. When I messed with it the second time the already deteriorating teflon tape turned into slime and just came apart. Its that pile of stringy looking stuff in the spool picture above. Here's a closer look:

I also found my leak:

I used a flashlight to give you a better view. That hole in there is where the seal for the top of the valve should be. That little pile of black debris down below it is what was left of the seal. All the top seals were that way, brittle powder. It's actually a wonder that the leak wasn't much worse.

So, there I was. The valve had over 200 bucks worth of missing parts and a bunch of cheap seals and o-rings that needed to be replaced. Roughly a couple of weeks wait for delivery and a few hours of work.

Naturally, I decided to take a totally different approach. I ordered one of these:

This is a 24 gallon per minute, double spool valve assembly. Brand new, latest technology and a JOY STICK ! For some work in re-plumbing and mounting this, I'll have the latest in loader control. Yes, it'll cost me about $100 more to do it, but this is just another project to have fun with.

Now, I'm pacing the floor waiting for it to arrive.

Edit: I got it in and installed it, then forgot to follow up on the blog. Here's the finish of this project <link>.

I FINALLY Got My 3D Printer

Many, many moons ago I decided that I couldn't afford a 3D printer since I don't want to make 3D printing a hobby, I just want to design and print things to use in my other hobbies.

Things like a coffee cup holder for the little tractor, enclosures of one kind or another for the various little computers around the house, custom light switch plate that has a USB plug in it for one of those double outlets, an outside enclosure that will hold an XBee; you get the idea. I've spent hours scanning the web for ideas about enclosures and there just aren't many choices, and the choices there are, frankly, suck.

I don't plan on being the worlds greatest enclosure designer, but it would be great to hide the wires and switches inside something.

But, since it would be a tool, not the focus, I couldn't see spending over a $1000 for a reasonably good printer that I wouldn't have to babysit and nudge all the time. Since I wouldn't spend the money, I decided to use credit card points to buy the printer. It took years to get enough to buy the printer ! Those points don't add up very fast.

Once I got the points, I was up to my neck in various projects around the house that were causing me to spin in circles accomplishing nothing, so I decided to force myself to finish certain projects before I was allowed to order the printer. The well repair got in the way, I also got some very heavy duty shelves for the garage, some serious repair work on the pool filter, and a big storage box for my toy hauler, but I finally finished them. A couple of days ago, the printer arrived.

I got the Flashforge Dreamer basically because of the reviews. There are certainly better ones and cheaper ones, but at some point, you have to make a decision, and good or bad, this was the choice I made. Needless to say, when the printer arrived I unboxed it right away and hunted down the test print.

I did that! Sure, it was composed by someone else, and all I did was print a file, but after literally years of postponed expectations, I finally did it. Dual extruders and different colors of filament and totally enclosed, oh my. (misquoting Judy Garland in the Wizard of Oz a bit).

There's company visiting, so I don't get to spend much time messing with it, but I did create my first little box.

It's about the size of a quarter; too small to actually use for anything, but I drew it and printed it. No, darn it, it wasn't easy. I started off with FreeCAD and got totally lost in the intricacies and nit-picking of design rules. What the heck is a closed polygon anyway?

OpenSCAD is very compelling because you basically program the print. I could probably do that, but I really wanted to be able to just click on an object on the screen and move, stretch, or spin it until it looked like something I wanted to print. Lots of modelling software is geared towards creating characters for games. I'm not at all interested in the skin texture of some scantily dressed teenagers dream girl for a video game. Plus, that software is really, really complex. I want to print things, not spend weeks designing a facial scar for the burly hero of the latest version of Doom. On second thought, maybe the dream girl could be a project later.

In simple desperation I started using Tinkercad on the web. I'm not happy with it being a cloud based tool, but it will do for now. As I gain experience, I'll probably change to something else, but at least it was simple enough to get me started.

The box and lid above was done with Tinkercad. I got the box on the first try, and I based it on a box a friend of mine made me a couple of years ago. The lid, well that was another story:

The first one was too long, and I forgot a bevel. The second one was too long. the third wouldn't slide in at all, to big. etc. It took five tries to get it good enough. Now do you see why I started with a tiny box? I didn't use much filament (more is on order already) and I learned a heck of a lot. Plus, it didn't take five hours to print the thing. It's really disappointing to spend that much time and find out you missed the screw hole by 5mm.

The slide on lid is sort of the idea I want to use for my temperature sensors. Mount everything on the lid, screw the lid to the wall, and slide the box over it. Simple to mount, simple to open, and mounted to something so I don't keep knocking it off the bedside table. What else could I possibly want?

Since I've got relatives staying with me right now, I can't devote hundreds of hours to working out the details, but I can sneak in a little time now and then to play with it. It's funny they don't complain about the smell of melted ABS at all.

This area has an excessive heat warning for the next week. The local weather predicts temperatures as high as 120F (that's 48C to most of the world), so I guess I'll be inside for the afternoons.

Guess what I'll be doing.

My Well, A Parody of Problems Part 6 (Disinfecting)

Remember back in part 1 <link> I mentioned that the inspection found a bad pressure tank and coliform in the well? I've fixed the pressure tank (and a whole lot of other unexpected stuff), so now it was time to get rid of the bacteria. I wasn't in a big rush to get to this because there were no bad bacteria found, just the usual that wind up in water from time to time. Plus, we all have filters, reverse osmosis devices and water softeners in our houses. There was no real danger and not even any noticeable inconveniences. It did need to be fixed to prevent too large an infection that would clog up the filters too fast, but we weren't close to that point.

One of the things that you need to be sure of is to get chlorinated water into all the pipes that feed the house before the various filters and such. Also, the well itself needed a really high dose of chlorine to kill anything hiding inside. You don't want a high dose of chlorine in the houses, because we didn't want to move while this was happening, so I split the system into two parts.

One part was the well itself and the pipes coming up from the ground to the tank as well as the well casing going back down underground. The other part was the holding tank, pressure tank, surface pump and all the piping to the houses. My thinking was that I'd chlorinate the holding tank at a level roughly equal to what the cities use and let normal usage circulate the chlorinated water to the houses. Of course, I told the neighbors to run water through their pipes to get the chlorine up to their houses. This was mostly to lower the tank with chlorinated water so I could fill it to the brim  before I disconnected the well from the tank.

Then I would pour chlorine into the well and hook the plumbing back to the well vent so I could circulate the water to distribute the chlorine and (hopefully) splash around on the well casing itself. Our well casings are PVC, with a steel pipe around the top thirty feet to protect the PVC pipe. This is very common out here; actually I don't know of any well that isn't constructed this way. Having a PVC casing is great because I don't have to worry about rusting a steel casing by pouring chlorine down the well where I can't effectively rinse it off. The plumbing would effectively look like this:

During the process, the holding tank would be all the water we have and the well could be stirred up any time by just starting the bottom pump. There would be a high level of chlorine in the well and a lower level in the tank. Of course, the way I actually did it was a little different.

I mixed a couple of cups of chlorine into the holding tank and told the neighbors to run their water for a little while to get it up to their house. First I topped off the holding tank, then I mixed about a half gallon of swimming pool chlorine with five gallons of water and poured it through a funnel down into the well. You really should water down the raw chlorine. That heavy concentration right out of the jug can mess stuff up. I wanted to check the chlorine level of the well as well as stir up the water down there, so I made a couple of adapters for a water hose:

Yes, I got a little wild with the glue, but it was early, OK. The skinny one went on the (new) well vent:

And the other one went on the very end of the pipe that fills the holding tank:

This way I could connect from the output back to the vent and get the chlorine through all the well piping. So, I connected a hose between them:

And turned on the well bottom pump. After a seemingly very long time, the chlorine started to show up, so I let it run for about twenty minutes more and shut down the well bottom pump. To check it, I just used a pool chlorine test kit. It wasn't accurate because I exceeded its maximum, but it let me know the chlorine was circulated.

Since the holding tank was disconnected from water, the neighbors agreed to hold off on laundry, turn off automatic watering, bypass the water softeners and close the automatic fill valves on the pools. The automatic stuff should not be running during this process because we has a finite amount of water to work with. Some of those water softeners can use 700 gallons of water recharging, and a load of laundry takes a ton of water. I think they over did it a bit though because we only seemed to use about 700 gallons for all the houses and one of them have two little boys still in diapers.

Around 7:30 pm after soaking for twelve hours, I ran the well bottom pump for another 20 minutes. This was partly for fun and partly to stir it up again. Meanwhile, folk were using the water from the tank and the level was slowly dropping. It looked like the tank would be plenty of water for the houses, but I checked again around 10:00 PM to put my mind at rest. There was plenty of water in the tank.

Bright and early on the second day, I disconnected the hose from the vent side and drug it away from the well to a patch of dry ground. I turned the bottom pump back on and started flushing the chlorine out of the well. This is always scary because some wells actually go dry during this process. If you pump more water out than the well can supply, you have to shut things down, wait for the well level to rise and then continue.

That didn't happen, there was plenty of water in there to flush the chlorine level down to something reasonable. What I did was flush the well until it had the same chlorine level as the tank and then put everything back to normal. By leaving a level of chlorine that was roughly equal to city water, I could extend the time the chlorine was in the system to kill off anything hiding in a pipe connection somewhere. The chlorine would all be flushed in a few days with the only problem being that our pool chlorine level might get a little higher.

There's some things that can happen when you disinfect a well that usually aren't mentioned:

The chlorine will kill the bacteria in the pipes and well which leaves their little dead bodies in the water. If the infection is bad enough, this will clog the various filters you have around the house. The primary filter on a reverse osmosis filter may plug, the filter some refrigerators have may die, things like that. If you find you have very low water pressure somewhere, check the filter; you may have plugged it up.

Also, if you have a lot of scale in the pipe, hitting it with chlorine may cause it to release from the pipe. That plugs filters also. It can also expose the raw pipe, and if you have lead pipes, lead will dissolve into the water. This is Arizona, not Michigan, our pipes are copper, PVC or PEX, no lead to worry about. This is what happened in Flint, they got an e-coli growth and over chlorinated to clear the infection. That dissolved the scale and exposed the lead in their water pipes. A reaction between the chlorine and the lead pipes dissolved the lead into the drinking water and a major water crises developed.

They had lead pipes because it was the law. Yes, there was a regulation that the pipes from the city supply to the house had to be made from lead. And also, yes, they are replacing all of them now. Stupid rules.

Both of these mean discolored water may happen as well as a chlorine smell that well users aren't used to. It isn't a bad thing if you know it might happen, but it can be a bit annoying. We didn't have any of that happen, things went really well.

Thus, the debacle of the well is over, done, fini. I still want to make some other improvements to the system though. I want a 'too full' indicator that will tell me when the tank get too much water in it and shut off the bottom pump. This can happen if the float fails. I also want an alarm to tell me when the tank is too empty and shuts off the surface pump. This can also happen if the float fails.

It would also be really cool to monitor the pumps for when they run, and develop a feel for what should be happening. That way I might be able to develop some heuristics for predicting problems. I also would like to know the level of the well so I can get a feel for the actual water usage of the three houses. Sounds like I need a waterproof ultrasonic distance sensor for the tank and a power monitor for the incoming power. It won't take long to recognize the power usage of the two pumps, and I'll be able to keep track of the tank level so (maybe) I can predict a float failure.

No, I won't monitor the individual lines to the houses. We're in this together, and I refuse to gather any data on other family's usage. Aggregate data is OK though.

But, for now, I don't even want to see that well.

My Well, A Parody of Problems Part 5 (Discussion)

Part 1 of this project is here <link>.

A reader asked that I put up a diagram of typical well plumbing and describe how it works. The problem is that there are a lot of ways to plumb a well, so I'll describe how my well used to be plumbed and what I changed it into. You can better understand what went on, and maybe, apply it to some problem you may be involved with in the future.

Let's start off with a picture of what the plumbing used to look like:

Not shown is a pump at the bottom of the well a few hundred feet down that raises the water up to ground level. The water came in from the left and hit a tee fitting that would send water to both the well and a booster pump. The booster pump had a pressure switch attached to the output to control the maximum and minimum pressure. Next was a pressure gauge so you can control and monitor the water line pressure followed by three pressure tanks to help maintain the pressure and cut down on the booster pump cycling. Then another pressure switch, hose bib and the manifold that led out to the three houses.

Obviously, this will work since we had been using it for years. However, there are some problems. First, since the well head feeds the tank and the pump, the pump will suck water from both the well and the tank when it is on. If there's any air in there, the air will cause cavitation and give lower pressure. It could, over time, destroy the pump because there isn't enough cooling and the pump will overheat. Additionally, any sand that comes up from the well has a good chance of being picked up by the pump and sent to the houses causing problems.

Second there are two pressure switches. Each one is trying to control the pressure on the lines to the house and you have to switch back and forth between them to get the pressure set right. Since they'll age at different rates and one of them might be shaded while the other is in the sun, you can never rely on the pressure being right. These things have a spring inside to control them and heat will change the tension. In the Arizona sun, this is a problem.

So, I changed it to this:

Now the well feeds into the top of the storage tank so the water has a chance to settle out whatever debris happens to get through. The booster pump can only pump from the bottom of the storage tank which prevents any air from getting into the system. There is only one pressure switch to eliminate one failure point and make it much easier to adjust. This is a configuration that is used a lot in my area. There is another configuration that is used here, but most people change it over time to the way above. The other configuration looks like this:

No storage tank, no booster pump and less parts to fail. Of course the disadvantage is that everything relies on the well bottom pump. The heavy cycling of the pump causes it to fail more often and those things are expensive. But, this configuration will save several thousand dollars initially and the other stuff can be added when money isn't as tight.

I hope this answers any questions the last well post may have left.

The other notable (at least to me) thing is that the well was inspected twice in the last few months. One inspection passed it with flying colors and the second one found the bad pressure tank. I don't attribute the inspection with finding the coliform infection, that was a lab in town that a water sample was sent to. Neither inspection looked at the wiring; they looked at the conduit, and ran the well, but they didn't actually check anything to see if there was a problem brewing. Sure, they have disclosures that say they don't open things to look inside, but would you buy a refrigerator without opening the door? It runs, and doesn't make any weird noises, so it must be good ... right?

Never trust those inspections, go look for yourself. Read a little, look up common problems on the web, decide for yourself. Sure, you may find something wrong, but at least it could be a bargaining point.

I still haven't disinfected the well, but everything is all set up for it. I'm trying right now to coordinate a date for that since it will mean turning off the well bottom pump overnight. One of the folk is pretty ill and the other family has little kids. Tough to coordinate.

I'll describe that part of this project in the next well post.