Maker Pro
Maker Pro

How to control the speed of a 24vdc pump from circulating water temperature?

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Stu, that cap will work fine but I'm accustomed to paying that much with free shipping. One other big difference is I'm addicted the Ebay's China vendors. The same item from China would cost about as much but it would be a pack of 50 or 100! :)

BTW, these are very common caps and can be cannibalized from junked appliances. In the old days we didn't have PC Boards so cannibalized components came out with reasonably long leads. PC Boards changed that significantly but is not a problem if the component is being re-purposed on another PCB.

Chris
 

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
Stu, that cap will work fine but I'm accustomed to paying that much with free shipping. One other big difference is I'm addicted the Ebay's China vendors. The same item from China would cost about as much but it would be a pack of 50 or 100! :)

BTW, these are very common caps and can be cannibalized from junked appliances. In the old days we didn't have PC Boards so cannibalized components came out with reasonably long leads. PC Boards changed that significantly but is not a problem if the component is being re-purposed on another PCB.

Chris

Hi Chris

This is all looking good. For now (at least) there is just the question of what appears to be a connection between the red and white wires at the pot. The two connections are very close together so I want to confirm that the 'apparent' connection shown in your circuit (post #46) is indeed a real connection.

You've all been incredibly generous with your advice - so I'd hate to accidentally screw it up!

Stu
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
In this image the wiper (center terminal) of the Pot is shorted to the bottom terminal of the Pot with a jumper wire. Is that what you're asking?

Also, the Cap can be soldered to the Pot terminals or to the bottom side of the PCB directly between the GND and the 0 - 5V connector pads. It's actually a better placement but I wanted to keep your soldering iron away from the PCB because I don't know what your soldering skills are.

Chris
upload_2016-7-1_0-47-55-png.27653
 

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
In this image the wiper (center terminal) of the Pot is shorted to the bottom terminal of the Pot with a jumper wire. Is that what you're asking?

Also, the Cap can be soldered to the Pot terminals or to the bottom side of the PCB directly between the GND and the 0 - 5V connector pads. It's actually a better placement but I wanted to keep your soldering iron away from the PCB because I don't know what your soldering skills are.

Chris
upload_2016-7-1_0-47-55-png.27653


Chris - You got me! I'm probably safer away from the PCB and will use a jumper between wires rather than risk roasting the board. I am ready but there will now be an intermission..... We're making good progress with the plumbing, but on a boat nothing is EVER straight forward.

We've now got a local guy with a plasma torch to cut the holes in the stove for the back boiler we're going to use.

It's been a long time since I actually looked forward to an English winter, but I'll be making an exception this year if only to see how my experiment works!
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Stu, here it is again with some notes.

Chris
upload_2016-7-17_0-38-0.png
 

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
Hi Chris

Many thanks for your time and patience. Absolutely chrystal clear.

I'll be back (here) during next week at some point with intermediate results using hot water and a 15 watt load. In a few weeks time we should have the heating up and running and I'll be able to report back again with the whole thing in situ

Stu
 

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
No matter what thermistor you choose. the range may be 100 'C for 0-5V but you need a control range of 1% of this for comfort so an Op Amp with gain of 100x and offset control to set the temperature for regulation point.

The best controllers have a PID loop to anticipate change (D) or derivative and integrate (I) temp error. This design is proportional (P) only, so high gain reduces error and offset adjust controls the temp.

If you add a pot to the thermistor, thus can offset the control but also reduces gain and thus wide temperature error and small control range from a smal change in thermistor resistance.

Maybe Amazon or Ebay has such a simple amplifier with offset control.
 

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
Another option which makes it simple is a digital home thermostat with integral thermistor. then you control the motor PWM board with an on off relay between two speeds set by a pot and series R for continuous flow at slowest speed for off to reduce corrosion or boiling and highest necessary for On with quiet operation. not max speed .

This gives you a timer, setback temp contols, and the flexibility with a Pot to adjust min, max speeds. Not as good as a proportional speed control for noise, but then maybe this will limit the power better when the wood stove burns out and the home thermostat can shut off.
 
Last edited:

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
Another option which makes it simple is a digital home thermostat with integral thermistor. then you control the motor PWM board with an on off relay between two speeds set by a pot and series R for continuous flow at slowest speed for off to reduce corrosion or boiling and highest necessary for On with quiet operation. not max speed .

This gives you a timer, setback temp contols, and the flexibility with a Pot to adjust min, max speeds. Not as good as a proportional speed control for noise, but then maybe this will limit the power better when the wood stove burns out and the home thermostat can shut off.

Thank you for this. Having now received all the bits and pieces to follow advice given (gratefully rec'd) I'm going to test that approach first. Hopefully posting results soon.
If there is not enough control I may have to come back and explore other avenues.

Stu
 

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
Stu, here it is again with some notes.

Chris
View attachment 27934

Hi Chris

I've carefully soldered the wires and the cap. I think some kind of adjustment may be required to make this work properly. The pot itself appears to be working normally - as it was before the cap and the thermistor splice. At first I couldn't really tell if the thermistor was working. To check this I turned the pot down to a point where there was only the vaguest glow on the 15 watt bulb. I then dipped the thermistor into water heated to around 70C and the glow on the bulb brightened very quickly. Before putting the thermistor into hot water I checked the output voltage at the PWM and got a reading of about 0.6 volts. With the thermistor in the hot water it went up to about 1.6 volts.
I have the PWM connected to the boats 24VDC circuit so there should be no problem with input voltage.

I notice you used the word 'splice' when joining the thermistor wires to the PWM's red wire. I've tinned and soldered the wires, does this count as splicing?. Can you suggest what may be required next, what readings I can take etc to ensure I've not messed it up somehow?

Stu
 
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Stu, the PWM output can't be accurately measured with a voltmeter. An Oscilloscope is required to actually see PWM voltage but you don't need one to know it's working. Your test lamp is telling you what you needed to know and seems to be working as expected.

Yes, soldering the Thermistor wires to the Pot wires constitutes a 'soldered splice'. If you didn't slide heat shrink tubing over the wires prior to splicing them then you will have to wrap each splice with electrical tape to keep them from shorting.

Chris
 

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
The circuit you have , has insuffient gain or sensitivity , thus excessive regulation error.
Get a digital hermostat.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Stu, a photo of your hookup will aid in determining if you made any wiring errors but as I've previously stated, it seems to be working as expected. ;)

Chris
 

Minder

Apr 24, 2015
3,479
Joined
Apr 24, 2015
Messages
3,479
A bit late to the party, but I would tend to agree with Sunnysky that you are going to need PID from such as a digital thermostat etc.
They are much higher in the way of cost, but I suspect with the system so far the hysteresis could be all over the map?
An added bonus is that most now incorporate remote control and monitor from a cell phone.
M.
 

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
If you dippped in room temp water and could see light turn off and on with 2 deg shift, then it works. 70'C indicates far too insensitive.

Remember you are trying to detect air temp. But if you also want to detect water temp to disable pump, then that is a good idea to add.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
No, Stu is not trying to detect room temp. He has a thermostat for that. He's detecting water temp and slowing down the circulation pump to insure the heater has sufficient time to heat the water.

Chris
 

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
I must have missed where he indicated there was a hot water valve thermostatic regulator.
If the intent is to prevent boiling and/or excessive pump noise/speed then when the thermostat indicates off, the excess water heat must be dumped to prevent boiling with a recirc. valve and/or flame controlled .

so it is not clear to me how heat, pump, air temp and flame are controlled.

Normally design specs would indicate setpoints for safety and comfort with upper/lower values for water, pump and air temp. with sensor fault detection.
 

cygnusv

Oct 7, 2014
79
Joined
Oct 7, 2014
Messages
79
I must have missed where he indicated there was a hot water valve thermostatic regulator.
If the intent is to prevent boiling and/or excessive pump noise/speed then when the thermostat indicates off, the excess water heat must be dumped to prevent boiling with a recirc. valve and/or flame controlled .

so it is not clear to me how heat, pump, air temp and flame are controlled.

Normally design specs would indicate setpoints for safety and comfort with upper/lower values for water, pump and air temp. with sensor fault detection.
,

Hey Guys!

I'm not even thinking about room temperature. The boat you see in the top left hand corner of this post really is our home - myself and Kath;s (Gord bless her) - and has been for the last 15 years. We love our life aboard but the control aspect that you would associate with household heating goes out of the window. Our boat is 30' x 9'6" and seperating us from an English winter is 3/4" of mahogany. We've had (in 2010) temps down to -17C, very unusual but does happen here from time to time.

Our aim is to redistribute the heat from our solid fuel stove, from the salon to both the bow, where we sleep and the stern where we need to keep the electronics in our washer /dryer above freezing. We can control 'room temperature' with the rotating air wheel on the front of the stove. If we do get another hard winter we'll top up the heating shortfall with marine electricity.

The main point of this experiment is for the PWM to adjust the circulating pump's RPM to a minimum to do the job, without allowing the system to boil and do it at the fewest decibles possible. The only possible place onboard, for the pump, is directly below our bed and the pump noise, at 14 litres a minute (flat out) would be intrusive, If the fire goes out - the pump should stop.

Thoughts? (no, getting a house will never happen!!!)
Thanks to all, Stu

Edited twice to add
Thanks Chris, apart from the room stat you were the closest I think.

To avoid any possible confusion, at best we can rely on between 5 and 6kws from the stove anything more would come from electricity.
 
Last edited:

Sunnysky

Jul 15, 2016
541
Joined
Jul 15, 2016
Messages
541
Thanks so water control only.

An analog temp Voltage and PWM voltage controlled power to pump will perform as well as you define it which depends alot on water radiator efficiency and flow rate.

Tcut-off for pump. .. I suggest 20'C
Tcut-in for pump at slow speed... perhaps 25'C
Tmax for water temp.. assuming water nearest flame is 20'C hotter, perhaps 70'C
Vmax for acceptable noise on say 0-5Vin control to PWM.
Then 70'C should provide full speed on PWM and 25'C near 25%PWM becuase motor may stall and overheat if not enough torque.

If one has a PWM controller spec and pump. and thermistor sensor mounted near flame on pipe, it may need some adjustment for gain and offset to achieve quiet continuous operation until flame goes out or bell rings to add more fuel.

Thanks for details on boat
 
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Then 70'C should provide full speed on PWM and 25'C near 25%PWM becuase motor may stall and overheat if not enough torque.
One of the benefits of PWM motor control vs older variable (analog) voltage control is that it nearly eliminates loss of torque at low RPM.

Chris
 
Top