trying to understand how this water tank pump pressure switch works

[tfnw]

This water pressure switch connects to the output of a water tank pump, and switches the pump motor on until a certain pressure is reached in the diaphragm chamber. The pressure is held in the diaphragm chamber and downstream by a one way valve on the inlet.

The pressure is measured by a diaphragm with a spring pushing against it. When the pressure reaches the desired level the diaphragm will begin to be pushed outwards by the water pressure. When the diaphragm moves outwards, a rod connected to it will also move outwards. This rod is solid plastic as far as I can tell and is not magnetic. This rod slides in a blind hole cut into the inside of a cylinder protruding out of the diaphragm chamber.

The circuit board has a hole in it and sits around this protruding cylinder in the diaphragm chamber (refer to the images if this description is not clear, and i'm sure it isn't, even i'm having trouble reading it ).

The part I am having trouble with is that I do not know how the circuit board is sensing the position of the rod within the cylinder in the diaphragm chamber.

The input power is switched to pass directly to the motor controller by a 24v dc coil input relay, I'm assuming this 24v dc is generated by a diode bridge and voltage divider because the only chip on the board is a quad schmitt trigger which i assume is used for hysteresis on the pressure sensing.

I tried replacing the relay as it had burn marks on the contacts but that didn't work. I have a whole replacement unit now so I'm only trying to understand the sensing mechanism now.

The only suspects I can see are the two raised components (marked RD? on the circuit board, I couldn't find any info on that marking) sitting on what looks to be custom plastic mounts, in the same light blue as the other custom plastic. I'm having trouble identifying what these components are (temp sensors?), they appear to be coated in a clear material. But I don't see how temp sensors could sense the rod position.

The only wild theory I can think of is that one of these components is acting as a transmitter and the other as a receiver and the position of the rod changes the transmission efficiency in a detectable way.

I'm sure it is something simpler that I just don't know about and cannot easily search for. I'm also sure someone who does know would know what's going on straight away so thanks in advance for any help.

[Seekonk]

Well I see a magnetic reed switch at RD2.

[IanB]

The only suspects I can see are the two raised components (marked RD? on the circuit board, I couldn't find any info on that marking) sitting on what looks to be custom plastic mounts, in the same light blue as the other custom plastic. I'm having trouble identifying what these components are (temp sensors?), they appear to be coated in a clear material. But I don't see how temp sensors could sense the rod position.

The only wild theory I can think of is that one of these components is acting as a transmitter and the other as a receiver and the position of the rod changes the transmission efficiency in a detectable way.

These are reed switches. The contacts inside them close the circuit when a magnet is brought close to them.

Does the blue thing have a magnet inside it? It seems most likely that the blue rod activates one or both of the reed switches when it moves closer to them.

[tfnw]

My first guess was that it had to be magnetic, I did a quick test and the rod seemed unmagnetic.

I just checked again, and found that the rod is indeed *slightly* magnetic, I guess these magnetic reed switches are fairly sensitive.

I knew it had to be something simple I didn't know about. Thanks for the help

Is there a good source for circuit board reference designators that would have had RD? in it?

Is there any better way to identify unknown components?

[IanB]

Is there any better way to identify unknown components?

Components mostly are identified by appearance. This video gives a good introduction: https://youtu.be/6Maq5IyHSuc

In this case a thin glass tube with metal contacts inside is very characteristic of a magnetic reed switch.

[tfnw]

Cool, I guess I've just never come across a reed switch before, thanks again

[Mechatrommer]

i have exactly this unit in operation for about 4 years now... thanks for tear down now i know whats inside... since you didn't mention the damage symptom i assume the relay will always in off position. this is similar situation when the rod is far away as in "over pressure limit" condition but its actually not. my suspicion is that the magnetic power of the rod is diminished hence reed switches are not detecting any magnet presence. you may test this even in circuit off condition, put magnet near the reed switches, put your ear closer and you'll hear a faint clicking sound, that is reed switch activating. but you cannot simply put the magnet on the glass, it must be at the correct distance somewhere farther 1 or 2 mm away, otherwise the reed will no be activated either if magnet is too strong, you need to try and error on this. if magnet is not strong enough, there will be no click. I'm surprised there are 2 reed switches one at top and another at bottom. i guess in normal situation, reed1 will be activated, when over pressure and the rod is moving away, it will activate reed2 and deactivate reed1.

if the magnetic force is diminishing, just like my wife's water heater did sometime ago (water flow detector), what i did is i replaced the corroded/broken magnet with the new one, luckily i have the same size neodymium magnet (eBay link below) so i just put it in without further modification, problem fixed. in your case its like the rod is one solid matter, not sure where the magnet is i'm suspecting it will be at the rod's tip closest to the circuit, you may clean the area with alcohol or something to reveal whats buried underneath the mud accumulating in the whole years. if the prognosis indicates the magnet is indeed diminishing, you may want to replenish the magnet power at that location, either by re-magnetizing it if its applicable, or find the magnet in question, it must be tucked in somewhere and replace it. if there is no separate magnet, then perhaps you can  file sand the rod and glue the small neomed magnet there, but this is destructive non reversible method i will do as a last resort if the only solution remains is "buy new", this is more of an art work rather than technical work. hope it helps ymmv...

http://www.ebay.com.my/itm/100pcs-2mm-X-1mm-Tiny-Disc-Rare-Earth-Neodymium-N35-Permanent-Strong-Magnets-/301624205443?hash=item463a342083:g:1ugAAOSwstxVS5E-

edit: wait a minute. i believe the 2nd reed at the back is the water flow sensor, since another protection for this device is turn off relay when there is no water supply in, not just over pressure. so i suspect there will be a 2nd magnet at the water inlet, besides the 1st magnet at the pressure rod tip... you can tell that the 2nd magnet to the 2nd reed (RD1) is damaged if fault red led is lit when you turn on this device. if the red led is turned off, the pressure magnet will be suspect.. ymmv...

[tautech]

wait a minute. i believe the 2nd reed at the back is the water flow sensor, since another protection for this device is turn off relay when there is no water supply in, not just over pressure. so i suspect there will be a 2nd magnet at the water inlet, besides the 1st magnet at the pressure rod tip... you can tell that the 2nd magnet to the 2nd reed (RD1) is damaged if fault red led is lit when you turn on this device. if the red led is turned off, the pressure magnet will be suspect.. ymmv...

Correct, the OP's "pressure switch" is more than just a pressure switch so that it does not cycle the pump on and off while there is draw off. Instead it senses that there is still flow and keeps the pump on until the flow ceases.
To operate in this way it needs to sense two parameters; pressure and flow rate.

[tfnw]

@Mechatrommer:
Funnily enough I had just finished testing the reed switches and noticed that I could hear them working as you described

I also noticed that the rod was not activating the second reed switch unless i moved it closer than it would be when installed. Your comment made me realize that the piece of metal at the outlet might be magnetic too, so I tried testing the reed switch with it and it seems that your and tautech's explanations of it being a flow rate sensor are correct.

The magnet on the rod seems to be embedded inside the rod, along it's length, maybe it is a cylindrical magnet. It also seems to have a non-magnetic end cap.

The failure case was that the relay was not closing, and the red failure led was lit, but resetting did not help. So it does seem that the most likely cause would be demagnetization of the flow rate magnet.

@tautech:
I didn't even think about how when pumping the pressure would rise above the set pressure of the diaphragm but you would want it to keep pumping until flow stopped.

[tautech]

@tautech:
I didn't even think about how when pumping the pressure would rise above the set pressure of the diaphragm but you would want it to keep pumping until flow stopped.

I think you will find it even offers dry run protection by way of sensing low/no pressure and no flow.
Then it will stop the pump so the pump shaft seals are not damaged.
When powered again the dry run protection will be reset if there is no manual reset button.

[tfnw]

Yeah, I've run it dry a few times, so I'm familiar with resetting it

It's impressive that the designer got so much logic without any complex chips.

[Mechatrommer]

It's impressive that the designer got so much logic without any complex chips.

its only AND circuit... output = reed1 AND reed2, if both ON, relay is ON. you can get by even without a chip, only few transistors for the logic and latch. most probably with other hysterisis/timing circuit. other than power supply to relay's coil... the other yellow caps and the neighboorhood seems to be power factor correction or somesort...

The magnet on the rod seems to be embedded inside the rod, along it's length, maybe it is a cylindrical magnet. It also seems to have a non-magnetic end cap.

the non magnetic end cap could be a screw end cap, try unscrew it bit by bit, there is no way a magnet can be magically got inside. it must came in and out from somewhere. if its long cylindrical, there are few cylindirical neomed magnet in ebay too, for example i did order this earlier years ago size 5x10mm (for my overunity project, another funny story but nevermind they are still there waiting to be implemented)...
http://www.ebay.com.my/itm/1-10pcs-N52-Super-Strong-Magnet-Rare-Earth-NdFeB-Neodymium-Permanent-Magnet-Hook-/311761475797?var=&hash=item48966e98d5:m:mDnnSXw0hnSPBWCn2o-MQUQ