hot water flow sensor

[jessey]

Hello Everyone,

I'm looking to build a tank-less on demand electric water heater for myself. I just replaced my 40 gallon hot water tank about 5 months ago and it just started leaking yesterday. The hot water tank is still under warrantee and will be replaced at no cost (but not the installation) by the manufacturer but I thought this would be a good time to build my own on demand one and selling the replacement one would more than pay for this project.

I've already built a programmable thermostat circuit to control an electric space heater using a DS1820 chip so I should be able to utilize one for this project. I'm thinking that I'll have to use a water flow sensor of some sort to sense when the hot water tap is turned on so the microprocessor will know when to turn on/off the heating elements and also a DS1820 or similar chip to monitor the temperature and if it gets too high to shut off the elements. I've never had any experience in using a flow sensor and there are so many different ones on the market. Would anyone know of a good water flow sensor that would suit my application and would be easy to interface to a pic chip?

Also if anyone has done this sort of project I would be really interested in any do's or don'ts or comments about this type of project. I was a welder/heavy duty mechanic for years before I got into appliance repair (my current vocation) and have a big shop in my back yard with all the equipment that I'll need to accomplish this project on my own so I should be able to build one at a very low cost.

Can anyone help me to get started with this project?

Thanks
jessey

[Normnet]

PID
Proportional Integral Derivative Control
page 145

http://www.parallax.com/dl/docs/books/edu/ic.pdf

Norm

[jessey]

Hi Normnet,

Thanks for the information but I don't think I'll have any problems controlling the temperature of the water and was thinking of just using the temperature sensor as a sort of high temperature limit switch to shut off the heater elements in the event that the temperature exceeds a preset high temp level then turn them back on when the temp reaches a preset low temp level. The elements will be either on or off using a relay or a triac. I feel confident and shouldn't have any problems writing code for that part. Actually I think I'll be ok with the flow sensor as well once I find one that would be suitable for this application. I was also thinking of having the micro control a Proportioning Valve to control the amount of water passing through the heaters so that if I use a solar pre-heater for the in coming water supply then the water temperature would control the flow rate. The warmer the incoming water then the more flow rate I would have and when the incoming water was cold then the micro could limit the flow rate some to maintain a preset temperature for the water. I'll just use a manual valve to experiment with to see how it works and when I get it to work good then I'll use a proportioning valve. Do you think this would be a good approach?

Thanks
jessey

[Dave]

jessey, Use an inexpensive diferential pressure transducer across a fixed orifice. There are many different company's out there that supply small diferential pressure transducers. Motorola MPX2100 first comes to mind but there are many more... LOL

Dave Purola,
N8NTA,

[Darrel Taylor]

Jessey,

If you can accurately control the temperature of the water, then you don't need a flow sensor. Just maintain the water at a slightly elevated temperature, under no flow conditions. Once the tap is turned on, the water temperature will suddenly decrease. Detect the change to turn on the heaters. When the tap is closed the heaters will turn off, due to normal temp control. If they are off for a certain amount of time, revert to standby mode, which allows it to slowly cool off again.

That seems to be how our's work. It has a relay that clicks on for about 10-15 seconds every 20-30 minutes, just to keep it warmed up.

.

[jessey]

Hi Darrel,

That sounds really good as it would certainly make it more affordable and simpler to build. I'm not sure if I'd really need it but what do you think of having the micro control a Proportioning Valve on the inlet side of the heating unit which could control the amount of water passing through the heaters so that if I use a solar pre-heater for the in coming water supply then the incoming water temperature could control the flow rate. The warmer the incoming water then the more flow rate I could have and when the incoming water temperature was cold then the micro could limit the flow rate some if needed to maintain a minimum preset output temperature for the water? Something to think about.

I managed to fix my hot water tank yesterday with an epoxy glue for now so I won't be rushed to get this done and can now take my time. It sure is nice to have hot water again!

I was thinking of just welding a rectangular box together just big enough to house the heating elements with an inlet and outlet for the water and of course a pressure relief valve and possibly a sacrificial anode for corrosion protection. I'll do a search for some kind of anti rust protective coating for the box but that can be later. Thanks to all for the comments about doing this.

Thanks
jessey

[Darrel Taylor]

Personaly, I think I would rather sacrifice temperature, than flow. And the main heater will take up the slack if it's not hot enough anyway. The Solar Pre-heater is there to save money, which it will do by not making the main heater work so hard, but the end result should be the same... Full-Flow, hot water.

I never really understood the Solar Idea too well. You end up spending Thousands of dollars, just to save a few hundred bucks in electricity. Kind of like buying one of those Hybrid cars. Pay an extra $10,000 to save $500 a year in gas. Doesn't add up.

If you can, use 316 or at least 304 stainless steel for your box. 316 will last forever, 304 will still rust a little, but it's cheaper. If you have Hard Water, it will soon be coated with calcium which helps with the corrosion too.

However if you use the anode ... I hear that sacrificial Zinc's can help prevent Cold's. :)

.

[Ioannis]

... If you have Hard Water, it will soon be coated with calcium which helps with the corrosion too.

However if you use the anode ... I hear that sacrificial Zinc's can help prevent Cold's. :)

.

How about Water Descalers using Electic Field? Has anyone used them? What were the results?

Ioannis

[jessey]

Hi Darrel,

Thanks for all your suggestions and for giving me an approach to follow for doing this. Well I finally got my test code working good, at least on the bench for now anyways. I thought I'd post it for you and others to have a look at for any possible comments. Once I build the unit to see how it works then I can cut the size of the program down considerably.

It's a fairly short program. I have (probably not needed but good for testing) created variables where as I can programme the temperature of the water coming out of the tap, an adjustable high limit temperature point for safety and also to know when the tap is off. And instead of a timer in the program & relay to control the ambient temperature when the tap is off I'll have it temperature controlled. I'm also thinking that I should have a differential for the ambient temperature of a few degrees so the relay or triac (relay for now) doesn't chatter or cycle too often.

I'm just going to use mild steel for the prototype to see how it works then when I have a good design then I'll switch to stainless steel construction. My first design will be a unit just big enough to house the heating elements with a very minimum water capacity. What kind of water capacity does your commercial unit hold?

For the DS1820 sensor, I was thinking of getting a suitable sized stainless steel tubing and epoxy the sensor into it then with the appropriate size fitting installed into the tank use a furrow and a nut to secure it into the tank or just epoxy the sensor to the outside of the tank. Any suggestions for accomplishing this?

Thanks for your help and I look forward to any further comments that you or anyone else here might have.

Thanks
jessey

[jessey]

Hi Darrel,

The code I posted in the last post is not correct, I caught it after I posted it. The new revised code here has been tested and is working good.

I'm now thinking of putting the DS1820 on the bottom of the inside of the tank directly in the path of the incoming cold water (epoxyed into a SS tube) so my code will catch the sudden temperature change.

I was also thinking of placing a mechanical reset-able high temperature limit switch on the side of the tank for added insurance. I can use the same one that was on my old hot water tank.

Thanks
jessey

[Darrel Taylor]

Looks like it's safe to post again. Yahoo!

>> What kind of water capacity does your commercial unit hold?

I have an under sink model, which looks like it could hold maybe a half-gallon. Yours sounds bigger.

It might be tricky getting a fast enough response from the temp sensor. The tube and epoxy might slow things down with the extra mass. But ya gotta isolate it somehow I guess.

... mechanical back-up ... good idea.

Can't wait to see what happens.

DT

[SteveB]

Hey Jessy,
Here is a link to the Tankless system I've been using for the last year: http://www.seisco.com/pages/internal-dia4.html. May help with the conceptual ideas. Of note, notice there are individual cambers for each heating element. Allow for future easy replacement of the heating elements, as this is the most likely component that will require replacement. Can't really speak to the corrosion issue (The one I have is actually made out of a non metal polymer). If you have hard water, you sould allow for easy cleaning out of the cambers to get rid out the scale build up. Also keep in mind, these can draw a lot of current. So make sure your electric service to your house is up to the task. The sight above has some good info on that as well. If you look around a bit on the web, you may find some temp probe products that would work well. If you construct your own, here is an epoxy designed for heat transfer (http://www.arcticsilver.com/products.htm) that would work well.

As for using the solar heater as a preheater, you need to "protect" the backup heater (and people) from the excessive temps possible from a solar system. Simply put, use a passive temperature controlled mixing valve between the solar water heater and the backup water heater. This will allow water temp into the backup heater to be at, or below, a preset temp, which would be 5-10 degrees above the max output temp of the backup heater. This way, the backup heater only comes on when the solar has run out. And the hot water from solar water heater, when it is really hot, will be used at a slower rate, increasing its effeciency.

Darrel:

I never really understood the Solar Idea too well. You end up spending Thousands of dollars, just to save a few hundred bucks in electricity. Kind of like buying one of those Hybrid cars. Pay an extra $10,000 to save $500 a year in gas. Doesn't add up.

I've seen some set-ups for between $1000-1500 (here's one (http://shop.solardirect.com/product_info.php?cPath=22_74_78_108&products_id=188)). With this setup (where I live) I could probably recovery my costs in three years. Even if the thing only lasts 10 years (likely more), I am money ahead, especially if energy costs continue to climb. I am actually planning to go this route in the fall (at the same time my roof is replaced). We have so many good solar heating days here that I my not really need a backup heater for 10-11 months out of the year.

Good luck jessy. Keep us updated on how things work out.

Steve


BTW, if you would like a closer look at one of these let me know. I could take some close-up pics.

[jessey]

Hi Darrel & SteveB,

Thanks for all the info and the links. I'm now waiting for a friend to return my plasma cutter and as soon as I get it back I'll be starting my project. I think I'll use a copper tubing (for better thermal conductivity) for the DS1820. I enquired to the link that you sent me SteveB for a good thermo substance, I described my application and just got an email back today suggesting to use Artic Silver Adhesive and also got a phone number for arctic silver so I'll call them tomorrow to see what they have to say.

Could anyone here suggest an alternative heat transfer compound (maybe not the best but something that will work to try) that is widely available and easy to get just in case I have to wait for the artic silver adhesive as I live in Canada and sometimes I have to wait a few weeks for some stuff to clear customs. If Active Electronics carry it then I shouldn't have to wait too long.

Its hard to tell in a lot of the pictures of the electric on demand water heaters in the sites that I visited but I would assume that the enclosures for the heater elements are round with convex ends (like a hot water holding tank) so I think my flat design will have to be beefed up somewhat to withstand the extra stresses especially if the water pressure is high. A friend suggested drilling multiple holes in between the length of the elements and using solid round stalk inserted into the holes and welded on each side of the tank to beef it up, a good idea that I think I'll employ. What is the shape of your tank for the heating elements Darrel? Looking at the pictorial drawing of SteveB's non metal polymer tanks, it looks like they are round as well which I think results in less psi stresses overall.

I think I'll put the solar thing on hold and just try and get the heaters working with the code that I posted for now then if I get it working good then I'll consider it. Green is good.

Anyways, thanks for your help and I'll keep you informed on my progress, maybe with some pictures later on.

Thanks
jessey

[Darrel Taylor]

Round here too.

Here's a model that's similar, except ours has a relay, instead of triac.
http://eemax.com/operations%20manuals/New%20SP%20instr.ppt

Like I said, it's small.

DT

[air1kdf]

Cool project.

DIY solar setups can be very inexpensive, check out ebay.

I am setting up a solar panel system to heat my 30,000 gallon pool. There are systems like Goldline that make and sell differential temp systems to optimize the heating. They also use 10k thermisistors that sense temperature. There are many sensors that can be placed inside or outside of your setup.

If it was me, I would use 2 sensors. 1 going in, 1 going out. Make sure to add code for circuit failure(like no temp reading, and invalid readings).

I would also forget about limiting flow. It would prove to be more complex than it needs to be, expensive for the control valve and actuator, and a PITA for the end user.

The thermal grease that you mentioned is great for putting a fan on a CPU. But it is a grease and not an epoxy.

FYI a 4x10 ft panel made 20 years ago can boil water in 5 minutes.

[BobP]

Hi,

This may be a simple and cheap solution. When I installed my power shower I was surprised to notice that the flow sensor on the outlets of the pump were no more than a magnet and reed switch positioned on opposite sides of each of the 2 plastic outlet pipes.
I can only assume as the water flow starts the magnetic field changes releasing the reed switch.

It is simple and cheap enough to try?

It has worked faultlessly on the pump for over 2 years now.

Hope this helps,
BobP

[Ken Howell]

It might be tricky getting a fast enough response from the temp sensor. The tube and epoxy might slow things down with the extra mass. But ya gotta isolate it somehow I guess.


DT

I've put these temperature sensors in small copper tubes (about 3/16...just big enough to encompass the housing) and epoxied them for environmental sealing.....they seem to respond almost instantly to increased temperature from holding them between fingers.....so I don't think the lag on temperature measurement is going to be a problem. You will have to have one hell of a heater unit to match temperature rise with water flow. This is especially true if your water temperature changes with the seasons as ours does here in Boise, Idaho.....