Poor man's VS pump in action

[Nofossil]

I've installed my 'poor mans variable speed pump' (described on another thread) and gotten the control software more or less dialed in. The graph below tells the story.

The three bottom lines are the three pump control relays. The bottom-most line (the light blue one) is the 'assert control' relay, and it shows when control of the pump is taken away from the EKO, starting at about 6:22.

The next line up (the pinkish one) indicates low speed, while the black line indicates medium speed. Neither line indicates high speed.

The four top digital lines show heat loads. From top to bottom: Baseboards, DHW, storage, and hot tub.

The controller is switching heat loads and varying pump speed to try and keep the boiler outlet temp in the 165-175 degree range.

With the 3 speed pump, I'm getting much better control and I'm able to avoid idling much more successfully. The higher speed of the Grundfos has also flushed a lot of air bubbles out of the system.

 

[Eric Johnson]

Knew we could count on you, nofossil. That's great. I'm especially encouraged by your success in regulating the boiler temp to avoid idling. I think you might be onto something there, my man.

 

[antknee2]

Very cool , my father and most of his friends would really appreciate all your electrics there into R&d;. I think a variable speed pump would help my solar heating system on sunny days around midday the output could be over 210 degrees , very hard for me to believe when it's 20 degree's out side . Really enjoy reading about yous invention's and discoveries . Anthony

 

[wdc1160]

Nofo, I know this has been covered, but what were you goals with the variable speed pump.
Energy/electric convservation and added flow during the highest wood heat output?
regardless controllable flow will no doubt have a ton of applications.

I think a variable speed pump would help my solar heating system on sunny days around midday the output could be over 210 degrees , very hard for me to believe when it’s 20 degree’s out side

anthony, I looked all over for a description of your solar setup. If you make one, can you give me a link to it. Thanks Bill

 

[Nofossil]

Nofo, I know this has been covered, but what were you goals with the variable speed pump.
Energy/electric convservation and added flow during the highest wood heat output?
regardless controllable flow will no doubt have a ton of applications.

In this instance, the Taco 007 that I had been using didn't have a high enough flow rate to keep the boiler from idling when the storage tank started to get hot. I also wanted to use a slower flow rate when the boiler was cooler to help it get back 'into the zone'. The Grundfos medium speed is almost exactly the same flow rate as the Taco, so this gives me a nice range.

 

[BrownianHeatingTech]

That's great. We definitely need to be getting variable-speed technology here in the US.

Even the "multi-speed" system you have would be a major improvement over single-speed pumps, as you just demonstrated.

Joe

 

[wdc1160]

Can you calculate how much less in wattage you'll use ? You have a point at which the circ and controller take over then they relinquish control, whats the trick there?
How much faster you can produce btu's (by stopping the idle)? You must actually shorten burn time right? Top speed was engaged in that burn for quite some time. Was it overwhelmed? Do you think it will be overwhelmed on your hottest burns?

 

[88rxn/a]

ok, i am sorry....im sorta lost on the idling? maybe ill go and read your thrad about how you made this to better understand it.

 

[wdc1160]

Idling in this context = All of the heat cannot be carried away from the boiler. It causes the boiler to go into "idle mode" It starts activating mechanisms to stop it from buring as hot. In the EKO's (Nofo's model) it reduces the amount of air that goes into the intake. W/o air gasifing slows down.

 

[88rxn/a]

I see, so it would work for all types of OWB boilers. so im assuming in the other thread "computerized" controlled is the thermostat?
this sounds to be a great idea!!

 

[BrownianHeatingTech]

I see, so it would work for all types of OWB boilers.

OWB's don't tend to idle the way that gasifiers do, given the large water volume of most OWB's.

so im assuming in the other thread "computerized" controlled is the thermostat?

No, nofossil has a computer that he built to control his system. Check out the website in his signature - the system is pretty impressive.

this sounds to be a great idea!!

It is. We need to get manufacturers selling these things. Should be dramatically cheaper than the true variable-speed pumps and, realistically, the efficiency difference between multi-speed and variable speed is going to be pretty small.

I'm imagining a pump wound for six speeds, using a simple control that maintains a pressure or temperature setpoint by cycling up and down the speeds. No where near the cost of a "proper" variable speed system, but would anyone be able to notice a performance difference?

Joe

 

[Telco]

I think a variable speed pump would help my solar heating system on sunny days around midday the output could be over 210 degrees , very hard for me to believe when it's 20 degree's out side.

Excellent, this calms yet another concern of mine about going solar. How well does it REALLY work in the winter. Love this board!

 

[wdc1160]

How well does it REALLY work in the winter. Love this board!

I am trying to get him to do a separate write up. I was hoping I wouldn't spur another hijack of some ones thread with discussion of solar or Anthony's setup

I see, so it would work for all types of OWB boilers. so im assuming in the other thread “computerized” controlled is the thermostat?
this sounds to be a great idea!!

88rxn/a, I was really talking about idling in the context of nofo's setup. Joe is correct if you take what I said

Idling in this context = All of the heat cannot be carried away from the boiler. It causes the boiler to go into “idle mode” It starts activating mechanisms to stop it from buring as hot. In the EKO’s (Nofo’s model) it reduces the amount of air that goes into the intake. W/o air gasifing slows down.

and extrapolate it into an unrelated tech, it may give you the wrong impression.

I don't know if you understand the language we use when we are talking about OWB's, but OWB's and gasifiers don't use the same strategies to provide heat. Consquently, using like stategies to distribute the heat and regulate output may not be compatible.

In this case, I dont' think what nofo is doing applies to OWB's

 

[Nofossil]

Feels like an interrogation - "Where were you on the night of January 17th?"

I'll do my best....

Can you calculate how much less in wattage you'll use ?

I'd have to measure wattage. It's not my primary concern. Any additional electricity ends up as electric heat, which is cheaper than oil these days. I'd like to use less, but my focus is on optimizing the wood boiler right now.

You have a point at which the circ and controller take over then they relinquish control, whats the trick there?

That's my 'fail safe' logic. If my controller does nothing, then the EKO has full control of the circulator. I take away that control only when I want to manage the circ. If my controller dies or has a Microsoft Moment, the EKO will have control and life will go on.

How much faster you can produce btu's (by stopping the idle)?

I don't really know. Probably not a lot faster. It's more about keeping an efficient burn going.

You must actually shorten burn time right?

Burn time might be a bit shorter. I've also restricted the fan inlet a week or two ago to reduce the peak output and stretch the burns a bit.

Top speed was engaged in that burn for quite some time. Was it overwhelmed?

To be clear - top speed is when neither of the two A&B;relays is active. It wasn't at top speed for all that long, and it wasn't overwhelmed - it never idled, so life is good. I have the relay logic inverted (both relays energized = stop, neither energized = full speed). I did that so that if my controller goes away, the pump is able to be controlled by the EKO.

Do you think it will be overwhelmed on your hottest burns?

I have some ability to juggle heat loads. The hot tub, storage, and superheating of the DHW can all be used to match the heat load to the boiler's output.

 

[Eric Johnson]

I was wondering, nofossil, if it would be possible to run your modified pump with a triple aquastat (or some other relatively simple device) instead of a computer.

 

[Nofossil]

I was wondering, nofossil, if it would be possible to run your modified pump with a triple aquastat (or some other relatively simple device) instead of a computer.

I've thought about that. It would be simple to set it up with two aquastats as a two speed pump, but three speeds would be harder. I deliberately designed it to be easy to control with a computer - takes only two outputs to provide four speeds (including 'off').

I expect a little cleverness could solve the problem, though.

 

[wdc1160]

Feels like an interrogation - “Where were you on the night of January 17th?”

I’ll do my best....

Sorry its just my busiest time of the year. I had been trying to throw some of these out to you since you posted. I have to get right right to the facts. And, their are a number of design and backend implications and even some things you can infer. Thanks for all the help.l

Top speed was engaged in that burn for quite some time. Was it overwhelmed?

To be clear - top speed is when neither of the two A&B;relays is active. It wasn’t at top speed for all that long, and it wasn’t overwhelmed - it never idled, so life is good. I have the relay logic inverted (both relays energized = stop, neither energized = full speed). I did that so that if my controller goes away, the pump is able to be controlled by the EKO.

Do you think it will be overwhelmed on your hottest burns?

I have some ability to juggle heat loads. The hot tub, storage, and superheating of the DHW can all be used to match the heat load to the boiler’s output.

I used too much shorthand in these question. ESP would be much faster. It appeard from 8-9 pm some of the longest periods of high speed pumping were done. Did it ever get close to a point where the load could not be stored quickly enough? Or, that the boiler just makes more heat than could be carried away -- even if their was more room for the heat.

 

[Nofossil]

Sorry its just my busiest time of the year. I had been trying to throw some of these out to you since you posted. I have to get right right to the facts. And, their are a number of design and backend implications and even some things you can infer. Thanks for all the help.

No apologies needed - they're great questions. I think it's great that you and others are thinking about the implications. You raise questions I haven't thought of.

I used too much shorthand in these question. ESP would be much faster. It appeard from 8-9 pm some of the longest periods of high speed pumping were done. Did it ever get close to a point where the load could not be stored quickly enough? Or, that the boiler just makes more heat than could be carried away -- even if their was more room for the heat.

To be very specific, the longest period of high speed was about 12 minutes starting at about 9:22 (Black and Pink lines missing = high speed). During that time you can see the output temp gradually dropping. There was an earlier period starting around 8:26 that lasted about 8 minutes. During that time the temp stayed about even. The output never reached the point where the EKO goes into idle, though it came close rigt about 9:00. The pump didn't go to high speed then because the inlet temp was too close to 140, the result of the a baseboard zone coming on and introducing cold water a bit earlier.

 

[Eric Johnson]

I was wondering, nofossil, if it would be possible to run your modified pump with a triple aquastat (or some other relatively simple device) instead of a computer.

I've thought about that. It would be simple to set it up with two aquastats as a two speed pump, but three speeds would be harder. I deliberately designed it to be easy to control with a computer - takes only two outputs to provide four speeds (including 'off').

I expect a little cleverness could solve the problem, though.

My mind starts to lock up when I try to think about this stuff, but what about the middle speed being the default, with the high-limit stat kicking it into high gear and the low-limit stat engaging the slow speed? Would that work? You'd have to set the differentials to zero so as not to encroach on the other speed's domain.

 

[Nofossil]

My mind is melted at the moment (long day with client), but here's the problem:

My approach needs two signals - relays, aquastats, switch closures, whatever. Each signal is on or off, giving four possible combinations.

I call the signals A and B, The results of the different combinations are as follows:

A on, B on: OFF
A off, B on: SLOW
A on, B off: MEDIUM
A off, B off: FAST

I think it's tough to make the transition from slow to medium. Both signals have to change at the same time. Again, I think some cleverness could solve it. I have none left at the moment.

 

[raybonz]

My mind is melted at the moment (long day with client), but here's the problem:

My approach needs two signals - relays, aquastats, switch closures, whatever. Each signal is on or off, giving four possible combinations.

I call the signals A and B, The results of the different combinations are as follows:

A on, B on: OFF
A off, B on: SLOW
A on, B off: MEDIUM
A off, B off: FAST

I think it's tough to make the transition from slow to medium. Both signals have to change at the same time. Again, I think some cleverness could solve it. I have none left at the moment.

Hi Bill,
Are you trying to control 3 speeds using 2 sets of contacts (A and B)? If so you could easily do this with a $99.00 PLC.. The program would be very simple as you would only have 2 inputs and 3 outputs.. That link I gave you to Automation Direct has the software for free and the logic would not be hard.. FYI it is programmed using ladder logic.. I can write the program for you if you like..

Ray

 

[raybonz]

My mind is melted at the moment (long day with client), but here's the problem:

My approach needs two signals - relays, aquastats, switch closures, whatever. Each signal is on or off, giving four possible combinations.

I call the signals A and B, The results of the different combinations are as follows:

A on, B on: OFF
A off, B on: SLOW
A on, B off: MEDIUM
A off, B off: FAST

I think it's tough to make the transition from slow to medium. Both signals have to change at the same time. Again, I think some cleverness could solve it. I have none left at the moment.

The logic would look something like this..

Ray

PS: I think you'll need 3 contacts and the one you would need to turn the circ. off would be an aquastat that would close is series with the other contacts to give you the off mode other wise you may default to high speed.. Does this make sense?

 

[Nofossil]

If I read this right, then you have it correct but inverted. My interpretation:

- A diagonal line indicates closed contacts
- X0 / X1 are contacts that when closed energize my A and B relays
- You have three 'states' indicated as Y0, Y1, and Y2

The way I have it wired, high speed is neither relay energized, and off is both relays energized.

You could of course swap the NO and NC contacts on my relays and it would work as you have it drawn. To Eric's questiuon, the problem with non-computerized control is that both relays change state going from low to medium speed. I haven't figured out how you could do that with aquastats and the like.

For all you 'C' programmers out there (both of you?), here's the actual code that I use. The newbit variable carries the two bits that drive the A and B relays. Two bits gives you possible values of 0, 1, 2, or 3.

     // ************************************************************************************************ 
    // DIO 5 bit 0 & 1: Wood circ A & B relay control
    // 
    //   A  B
    //   0  0  On / High            Temps above 75 or wood circ control not enabled
    //   0  1  On / Medium          65-75
    //   1  0  On / Low             55 -65
    //   1  1  Off / Low            < 55 or not in woodmode
    //
    // ************************************************************************************************ 
    
    // Create mask, capture old bit value
    rmask = 3;
    oldbit = dio5 & rmask;
    
    // Hysteresis
    if(oldbit){
      outlimit=55;
      inlimit=0; 
   }else{
      outlimit=60;
      inlimit=1;
   }
 
    newbit = 3;             // Off

    // Faster speeds at higher temps
    if ((mode == WOODMODE)){
      if(WOOD_OUT > outlimit){
        newbit = 2;
      }
      if(WOOD_OUT > (65+inlimit)){
        newbit = 1;
      }
     if(WOOD_OUT > (75+inlimit)){
        newbit = 0;
      }
    }

    // Input protection trumps output temp rules
    // If inlet is low and pump is medium or fast, slow pump down
    if (WOOD_IN < 60 && newbit < 2){
      newbit++;
    }

 

[wdc1160]

If I were going to attempt to do what eric is suggesting using no computer power -- I think it could be done with temperature triggered swiches.
In this finite state automata you've got 3 destinct states that shouldn't overlap.

*********************************
Two temp switches maybe even possible?

In this not registering a speed default to low speed(not important, just as easy to reverse the logic).

temp switch A is a complete circuit > 65
temp switch B is a complete circuit > 75

00 if 65 is reched your OFF/LOW
01 if 66 is reched your ON/MED
10 if 76 is reched your ON/HIGH

We don't technically need an on/off for what Eric is talking about. He's using no computer. "ON" in this example is energized relays. So you don't need a controller state ON or OFF. He would intuitivly choose 2 switches and the default to low speed - or vice versa.
I don't know it for a fact, but I bet Taco's senor system could work for this purpose. Some of the other guys here have used'em for thier home jobs.