I've got this circuit board on the boat that I work on that's giving me some trouble. This board is sensing switches that I've got in a tank, when the tank's full it should be triggering a relay to power a pump to empty the tank.

I've got two float switches in the tank, one at the bottom and on at the top. The pump is supposed to be turned on when the upper switch is activated, and the pump turns off when the tank is empty and the lower switch opens.
These switches float in that when the height of water rises so does the "arm" of the switch and mercury then completes the circuit.

I'll walk you around the board.

1. 24vac incoming on the 2 yellow leads.
2. Rectified to 32vdc. upper right hand corner.
3. Green/Black wires go to lower float switch.
4. White/Black wires go to upper float switch.
5. Power out to a 24vdc/120/vac relay to trigger pump. Red/Black next to the Yellows.

I've replaced the float switches, rectifier, the 2200uF 50V Cap and the diodes with no change: Nothing happens when the upper switch is closed.
But I've jumped from the red on the relay to bypass the MOV and resistor that goes to Green wire (I have no power to the green) Doing this I do get the pump to turn on at the upper float switch trigger, but it only runs until it falls back down, maybe 20-30 seconds. I need it to run for approximately 3 minutes. (This set-up worked beautifully before)
So is the MOV causing a problem?
I was told by the local electronic components store that the blue jobbies were the MOV's. There is absolutely nothing written on them.
The one inline from the switch to the green wire seems to be blocking the current, I can steal the other from the alarm side.

Incidentally the small Red/Black wires and Hasco relay are part of a high-level alarm not used now.
This is my first try at troubleshooting a board and it'll save the owner of the boat about $1100 if I can fix this.
Also this isn't my work of art.
Thanks in advance!
Oz

 

Evidence of the earlier failure can be witnessed in the apparent charring that occurred at the site of the filter capacitor.

The devices that were identified by the technician as MOVs are what I believe to be PTCs (Positive Temperature Coefficient) fuses. These are made by Bourns and Raychem (Tyco Electronics). The blue ones are Bourns and the yellow one that appears at the top right of the board(near the green LED) is more likely a Raychem PTC. MOVs are placed in parallel with the load they protect while the PTCs are circuit protectors in the same since as a fuse and so are placed in series with the load they are intended to protect. PTCs are also referred to self-resetting fuses as they are triggered to go open when the overload heats them up. Once the overload is removed these PTCs cool off and return to their low resistance state.

Since your system works temporarily, it is possible that one of these PTCs was damaged during the earlier failure and so it works briefly until it gets hot and then it opens up. Once it cools down the system will work once more for a brief period.

Check to see what part number is on the yellow component in the upper right of your picture adjacent to the illuminated LED.

hgmjr

 

Thanks for the reply, so the system is operating, but only while the upper float switch is closed... there should be some sort of "timing" relay in effect here because the system should operate after the upper float switch is open and until the lower float switch is open too.
So maybe it's the Zettler relay? It looks like that un-used Hasco relay would be a direct fit... could I interchange those to see if that's my problem?

 

Thanks for the reply, so the system is operating, but only while the upper float switch is closed... there should be some sort of "timing" relay in effect here because the system should operate after the upper float switch is open and until the lower float switch is open too.
So maybe it's the Zettler relay? It looks like that un-used Hasco relay would be a direct fit... could I interchange those to see if that's my problem?

I was a bit disappointed with the construction of the circuit, I must say.

You can hookup a relay in parallel with the other and see if the problem persists.

The way I see it, one relay should act as a memory, but I can see which one is feeding its own coil, if any. Perhaps none of them. I can't see any timer circuits, that is why I bet on the relay memory. What I am thinking is on two level switches, the lower level switch would activate the relay when the tank is empty. The circuit would open a valve to fill the tank. The upper level switch would deactivate the relay once the water reacher its level. The tank would refill again if the lower switch actuated again. Hence we have hysteresis and a memory.

Here is what I am thinking (simplified circuit):

I'm using contact symbology, since it expresses better my idea. So contacts are represented as capacitors with the plates abroad and coils are represented by circles with letters.

Note:
NC - Normally closed contact when not actuated;
NO - Normally open contact when not actuated.