Tankless Water Heater

Here is more than you ever wanted to know about fixing my tankless water heater.

A few years ago I bought and installed a Poloma model RMTG-53DVP tankless water heater in my home. This exact same model is also sold under the brand names of Rheem, Ruud and Richmond. After operating for about a year it shutdown and started flashing error codes C7, 13 on the remote control. If I unplugged it for a while it would work again but would only run for about 3 minutes before shutting down again. Some digging around on the WEB indicated this was a oxygen sensor fault. In other words the oxygen sensor was telling the controller there was low oxygen. Never having worked on a tankless water heater before I called a plumber that advertized fixing tankless water heaters. He came out and we had a nice conversation and basically recommended I fix it myself. For this I paid for a service call. Apparently he was not familiar with that brand of heater and did not want to figure it out.

Now a tankless is way more complicated than a conventional water heater with a microcontroller and lots of sensors and valves.


Here is my heater opened up. At the top is the copper coil assembly where the water gets heated. The plastic with the black lines on that is an electric heater that keeps the water from freezing if the temperature gets too low. Below that is where the burner assembly goes. The burner assembly and manifold is removed in this picture. Below the burner assembly is the controller, valve and fan. There is a whole bunch of screws that have to be removed to get thing apart.

The oxygen sensor work by monitoring a flame burning in a special chamber. There is a precision orifice that controls the amount of gas burning and a thermocouple to monitor the flame temperature. The controller waits for 3 minutes after the heater starts to check the flame temperature so that everything is stabilized. If the temperature is too low the controller shuts down the heater and sends an error code to the remote control.


This is the burner assembly. On the left is a round brass hole. That is where the sensor flame burns.

So what could cause the problem? Obviously there is plenty of oxygen and nothing blocking the air flow into the heater. The thermocouple could be bad but those thing are very reliable. The controller board could be bad (horrors). After a bit of poking around and inspection I discovered that there was some crud deposits around the precision orifice.


It did not seem like the orifice was blocked that much but since I could not find any other problems I cleaned the orifice and reassembled the heater. That solved the problem.

Since then I have had to clean the orifice about once a year. Something in the propane  is forming the crud.  They also recommend cleaning the water lines by running vinegar through the heater with a pump for about an hour. I have done that once. It did not seem like there was very much lime build up though.

Posted in appliance repair, DIY | Tagged | Leave a comment

Soapstone Stove Restoration

Many years ago we bought a soapstone stove at a garage sale for $25. It was not in the best shape with numerous stones broken. I brought it home and dismantled it so I could move it. The complete stove weighs around 700 pounds. I stored the pieces in the barn. Now that I am retired, with a little more time on my hands I decided to fix it up and install it in the house. I have a lot of wood and it would be nice to have a backup for the propane heat we have.

The stove is a Hearthstone I made by the Hearthstone stove company of Vermont. I believe that the Hearthstone I is the first model made by the company and is no longer in production. I did learn from an online dealer that the Equinox 8000 stove replaced the H1. It looks similar but the innards have been modernized.

Replacing the stones

The first task was to see if it was even possible to replace the broken stones. A few years ago I visited the local Hearthstone dealer and asked about getting replacement stones. They said I should talk to the factory. I contacted the factory and they were kind enough to send me an owners manual for the stove but said I had to go through the dealer to get part. The old run around. The factory would probably have had to make the stones and that would be and expensive way to get the stones. I did find an online dealer that had some of the stones but they are quite expensive. One valuable thing I got from the online dealer was a chart showing the layout of the stones in the stove. That leaves me with getting some soapstone and making the stones myself. It turns out the stones are 30 mm thick which is the same thickness as the soapstone slabs used for kitchen counter tops and the like. There does not seem to be any soapstone counter top dealers nearby so I have been contacting soapstone suppliers. I got samples from three suppliers, Finnish soapstone, Brazilian soapstone and Virginia soap stone. I wanted  to match the look of the old stones. Shipping turned out to be quite expensive for the stone so I finally found a dealer a few hours drive away in Wisconsin. The slab I purchased there does not match the stones of the stove but arranging the new stones symmetrically it will look OK.


The picture above shows the broken stones. I used this to estimate the amount of new stone I needed to purchase. After I got the slab of new stone I realized that I actually needed more stone as I wanted to move the flue outlet from the back of the stove to the top. This requires a few different stones, which I have to make. To compensate for this I decided to glue the stones with simple breaks and no missing parts. This is a stove and the stones get quite hot so you can’t use just any glue. I have a gallon of sodium silicate which can be used as a glue. Sodium silicate when dry is essentially glass and can withstand high temperatures. This worked quite well and after refinishing the stones the breaks are hardly noticeable.


The picture above shows the method I used to cut the new soapstone slab into the sizes for the stove. I put a diamond tile saw blade in my circular saw and cut the stone wet with a clamped on guide. This produced a nice smooth cut. I was careful not to get the electrical part of the saw wet and it is plugged into a GFI outlet.

I planned on cutting the grooves  and bevels for the stones on my table saw with just an ordinary carbide wood cutting blade. I tried cutting some of the broken stones to see if this would work and they cut quite nicely. When I went to cut the new stones they promptly ruined the carbide blade. The new stone was much harder than the original soap stone in the stove.  A 10″ diamond blade for the table saw solved that problem.

Each stone has a decorative beveled edge and grooves in the edges. Flat metal strips fit into the grooves to hold the stones in place.

Metal Parts

After sitting in the barn for years it was impossible to remember how all the pieces of the stove fit together. It was a jigsaw puzzle type of challenge to sort out the stones and metal parts. The metal parts had become quite rusted.


As can be seen in this picture of one of the stove doors. All the metal parts got sandblasted and the cast iron parts repainted with flat black high temperature stove paint. Some parts like the screen for the doors had to be replaced. I replaced the screen and all the nuts and bolts with stainless steel. I also had to replace some of the steel strips that hold the stones in place because I changed the vent from the back to the top of the stove. This requires some different length strips. The ugly bolts in the door handles were replace with some stainless steel rod and some nice brass acorn nuts.


Because the complete stove weighs somewhere around 700 pounds there was no way I was going to assemble it and then move it into place.


In the picture above the base of the stove is placed along with some of the internal workings. There are cast iron plates and a channel around the bottom that supplies air around the base of the fire. The incoming air is regulated by a thermostatically controlled damper on the back of the stove. The 400 pound water filled cast iron radiator behind the stove provides and excellent radiation shield to protect the wall behind it from the stove heat. Not seen in this picture is another metal radiation shield on the right wall.


Above is another view of the stove being assembled. The three stones in the back with the fiberglass screen on them are some of the replacement stones. The original stones were shorter to allow for the vent in the back. You can also see the ends of the steel strips that hold the stones in place. Because of the way the temperature of the stove is regulated by controlling the incoming air it is important that the stove be airtight. The rope gaskets for the doors were replaced. New ceramic fiber paper gaskets between the door frames and the stone were installed. When assembling the stones into the stove each stone mortared with furnace cement to seal all the joints between the stones and between the stones and the cast iron frame.



Above is a picture of the completed stove with a nice fire going. Normally I use the side door for ash removal, loading wood and starting. This winter I have been using the stove daily. It will keep the downstairs part of the house warm without any additional input from the propane heat even on below zero days. When the temperature is in the 40’s it is easy to get the house a bit overheated. Because of the way the air is regulated I believe you are supposed to be able to load the stove up with wood and have it burn most of the day. I prefer to keep only a small amount of wood in the stove and add a piece every hour or so during the day. This way if it gets too warm I can just stop feeding it wood.


Since the stove is in the dining room it can be used to keep food warm. Here it is keeping thanksgiving dinner warm.

The dining room gets a bit warmer than the rest of the downstairs so I have a fan to help circulate the air to the other rooms.


Hearthstone stove manufacturer http://www.hearthstonestoves.com/

Soapstone supplier http://www.elegantstoneproducts.com/

Posted in Alternate Energy, DIY | Tagged , , , | 11 Comments

Solar Power

I have wanted to put up a solar power system for some time and last month I finally got around to getting one installed. When I first started looking at solar power the panels were small and expensive. The inverters were fairly primitive and not too reliable. All that has changed in the past few years. Now when I drive around, particularly in rural areas I frequently see solar panels. It is still hard to find installers so when I saw a sign for a business along the road in the next county I called then and asked for a quote. In the past I would have probably done the installation myself but I am getting a bit old to be scampering around on a roof carrying big solar panels. It would have taken me all summer to get the job done. They got the hardware all installed in two days. After that it was a two week wait for all the paperwork to get done and approved.


This building looked to be the ideal place to put the panels. The house is not suitable due to trees and not facing south. The building next to this one has a shallower roof angle and is shaded more by some trees. The solar array I wanted consists of 21 270 watt panels. When we first looked at putting the panels on this building it looked like 21 panels would not fit. The panels are normally installed in portrait because that uses the least amount of racking. In that orientation they would not fit. I decided to spend a bit more on the extra racking to get the panels on this building.  As can be seen in the picture above the 21 panels just fit in the landscape orientation. The tilt angle of the roof on this building is 35 degrees which appears to be a good compromise. The panels are wired in three strings of seven panels each.


The DC wiring is run underground to the adjacent building because it has power and is a better place overall to mount the inverter. This inverter is an impressive bit of technology. It is rated at 6 Kw and is capable of  MPPT on two independent  circuits. One of the seven panel string is wired to one of the MPPT inputs and the other two strings are wired in parallel to the other MPPT input.  The inverter is fully weather proof and capable of being mounted outside. Although I am much happier that it is inside a nice building. When you turn the inverter on it boots up and goes through a list of self checks before it connects to the grid.  A variety of information is available on the LCD and internal settings can be changed after entering a password. It keeps track of things like total energy produced, peak power for the day and overall and even the value of the energy produced if you enter the cost per Kwh. For some reason the default is in euros rather than dollars. The inverter is capable of being remotely monitored and controlled. This is done by RS485 signalling . This is a good choice overall but I would rather not bury a cable all the way back to the house for monitoring. Also I plan to have my shop in this building and want internet for the shop. So I am arranging to bring internet to the building via a wireless link. the open box on the left in the picture above is a place to mount the RS485 to Ethernet converter and the power supply for the wireless link. When that is all hooked up I will be able to monitor and log data from the inverter in the house and even put the data on the web.


I took the above picture during testing around noon on a sunny December day. The power output is about 1 Kw less than the rated power of the array. I don’t really expect to ever see the full rated power but it will be more in the summer when the sun is higher and more square with panels.  I got permission from the utility company to go online and switched the system on. So far the weather has been foggy and cloudy. It still makes 400 or so watts at noon on a foggy overcast day. The utility company came out and installed a digital watt hour meter that keeps track of coming and going power. This size array is expected to provide about 40% of my annual energy consumption. Time will tell.

Technical Details:

The solar panels are Suniva OPT270-60-4-100. These panels have 60 6″ mono crystalline cells. The 21 panels are wired in three strings of 7 panels. This gives a maximum open circuit voltage of 269.5 Volts for each string.

The inverter is an ABB PVI-6000-OUTD-US-A. This is a transformer-less inverter rated at 6Kw. It provides 240 volts to the grid when operating. The inverter manufacturer was formerly Power One. This caused me some difficulty  when looking for the users manual and monitoring software online as many of the web pages are still under the Power One name.


I was planning on comparing the monthly electricity usage with the solar panels with the month from the previous year. The electric company has nice online access to the information. Only one small problem, you get an new account number when you set up for solar and the information from the old account is not available. I should have saved the usage information before the account switched. The first two months on the solar system December and January have the shortest days and are generally dark and cloudy. Despite this the overall bill from January was less than half of what I remember for a typical bill for this time of year. Based on this I suspect there will be months this spring and fall, when I am not using the AC that the solar system will make more than I use for that month.


Green Transitions http://greentransitionsllc.com/

Posted in Alternate Energy | Tagged , , , | Leave a comment

PC Problems

A while back the computer that runs my instrument platform died. I don’t know what happened but I ended up replacing the internal power supply, motherboard and memory stick to get it working again. The front panel USB board was also dead but I don’t use that so I did not replace it. After repairing the computer everything worked except the computer would not communicate with the Arduino that runs the weather instruments.  At first I thought the Arduino was dead also but it would work if plugged into a different computer.  A different Arduino not programmed with the weather instrument software would work on the computer. Finally after a lot of digging around in the windows registry and playing with the device manager I got it to work. Here is what I think happened.

When I replaced the motherboard and started the computer windows booted up and found that there were two new serial ports (the ones on the new motherboard) since COM1 and COM2 were assigned to the serial ports on the old motherboard windows assigned COM3 and COM4 to the new ports. Now with the old motherboard the weather Arduino was assigned to COM4 and windows remembering that left the Arduino assigned to COM4 even though COM4 was also assigned to a motherboard port. So when I tried to communicate with the Arduino I was actually trying to talk to the motherboard COM port even though the device manager said COM4 was assigned to the Arduino. Using the device manager application assigning the Arduino to an unused COM port made thing work. I still have windows saying there are 4 motherboard COM ports. There are probably only two. All sorts of deleting registry keys and playing with the device manager have not been able to get rid of the extra ports. I probably should have reinstalled windows when I replaced the motherboard but that requires so much putting thing back together that I will leave it the way it is unless some other problem shows up.

The computer is a really nice little machine I built up with parts from MiniBox. It is small about the size of a 2 inch thick book and uses an Intel Atom processor. It only draws about 15 watts, which is real nice for something that runs 24/7.

Posted in DIY, Electronics | Tagged , , , | Leave a comment

CNC lathe

I have a Craftsman Atlas 6″ lathe that has been a workhorse for years. Recently I have a couple of lathe projects that are difficult or impossible with a manually operated lathe. So I have been modifying the lathe to add CNC capability. Parts to convert the lathe include stepper  motors, stepper driver, power supply, computer and modifications to the lathe to attach the motors.

I bought the pro driver board kit and a pair of steppers from Hobby CNC. It is fun to build up boards like this and it saves a bit from buying a complete board. It also needs a case for the board. I decided to keep the power supply and driver board assemblies separate so I can swap things around if needed so the power supply is in its own case.

It turned out that the transformer I bought for the power supply put out a little too much voltage. The maximum input voltage to the stepper driver is 42 VDC and the the transformer was putting out just about that when connected to a rectifier and filter capacitors. I think they rated the transformer output with 110 volts input rather than the 120 volts which it the line voltage around here. Actually it runs a bit higher than that most of the time. I really did not want to get another transformer so I decided to take a few turns off the secondary.


Transformer Modification


This is the transformer opened up with a few turns removed.


Transformer Test


Here the modified transformer is hooked up to check the output voltage.

On a lathe there are two axis that need to be driven, the main lead screw and the cross slide screw. To attach the pulleys  for the stepper motors I added and extension shaft on the ends of the lead screws. To do this I drilled a hole into then end of the screw and turned down the extension shaft for a close fit into the hole. This holds the extension straight while I silver brazed the extension to the screw.


Slide Screw Extension


Here is the cross slide screw and the extension shaft before brazing. Of course the machining work was done on the lathe.


Cross Slide Screw


Here is the cross slide screw just after the extension was brazed on. After I brazed on the extension I discovered it was slightly bigger than 1/4″ in diameter and the bearing and belt pulley would not fit. So I mounted it in the lathe and turned it down with a file and sandpaper.


Lead Screw Stepper Motor

An extension was added to the lead screw in the same was as the cross slide screw. The mount for the lead screw stepper motor is simple, just an angle bracket to bolt it to the table. The pulleys provide a 2 to 1 reduction from the stepper motor.


Lathe Cross Slide Stepper Motor

To mount the cross slide stepper motor two holes were drilled in the carriage. The back mounting plate has a ball bearing to support the lead screw extension.

So far I am pleased with the operation of the lathe. One of the difficult without CNC parts I have made is a brass rod with a long taper. That only took a few minutes to machine. Currently I am working on adding a spindle position encoder to the lathe. When that is complete I will be able to cut threads under CNC. Threading is difficult manually and the lathe does not have any gears to do metric threads. CNC will solve these problems.


Here is a YouTube video of the lathe in action. https://www.youtube.com/watch?v=vzmgTqwl5tA It only took a few lines of G code to do this operation.


Hobby CNC http://www.hobbycnc.com/

Linux CNC http://www.linuxcnc.org/


Posted in DIY, Electronics, Machine Shop, Metal Working | Tagged , , , , | 2 Comments

Wren Trouble

This morning I went out to change the bucket on my digger. It has two buckets, one narrow one for trenching and a wider one for moving dirt. There was a tiny bird sitting on the digger arm cussing me out and it did not fly away until I got quite close. Then I noticed there were two of them and I suspected I was near their nest. Sure enough they had built a nest in the arm of my digger.

wren1 wren2

I will postpone working on the digger until the nest is vacant.  I managed to record the call of this bird.

Looking in my bird book this appears to be a house wren (Troglodytes aedon)

Posted in Amateur Science, Nature, Photography | Leave a comment

Fairy Ring

A fairy ring has shown up in my yard for the past two years. This one is strangely centered on my bird feeder. Articles on fairy rings suggest that sometimes rabbits are the cause of the rings. In this case my theory is that the bird droppings are doing the same thing that the rabbit droppings do.

Fairy Ring

Fairy Ring

There are also several smaller mushrooms behind the feeder that are hard to see in this picture.  Also hard to see is that the grass is greener and taller in the ring. Click on the picture for a larger image.

Posted in Nature | Leave a comment