Tear down of failed motorized ball valve

[calzap]

On my ranch, we do a lot of drip irrigation with water flow controlled by electric valves.  In the past, we’ve used pilot-operated diaphragm valves … the type commonly used for lawn sprinklers.  The problem has been that the irrigation water is high in particulates, mostly bits of plant material.  Diaphragm valves have a small hole that is opened or closed by a solenoid for operation.  This small hole is easily clogged by particulate matter interfering with operation.  And yes, the water is filtered.  Problem is the particulates can agglomerate once through the filter.  In addition, some of the particulates are relatively long and stringy.  So, by chance, some will go through the filter by having the narrow dimension enter a filter pore, but they lodge crosswise in the hole in the diaphragm valve.

The solution has been to use motorized ball valves as replacements, which we have been doing gradually.  They’re more expensive but worth it.  In the past few years, motorized ball valves have become available that operate electrically in the same way as pilot-operated diaphragm valves.   Apply 24 VAC, they open; remove power and they close.  Voltage spec is 9-24V AC/DC.  Manufacturer did not supply current specs, so I measured them. Peak current flow at start of motor (in-rush lasts 1 second):  200 mA.  Steady state current flow while motor is operating: 30 mA.  Steady state current flow with power applied but motor not operating (i.e. open cycle completed): 9 mA (takes about a minute to decrease from 30 mA).

The ball valves have a drawback in that they cannot tolerate water intrusion into the electronics.  The diaphragm valves are immune even to immersion … their electronics consist of one small solenoid that is completely sealed.  This means that the ball valves cannot be installed in underground valve boxes that might flood … and flooding is always a risk.  All our ball valves are above grade.

First ball valve was installed in 2019 and failed this year due to water intrusion.  The valve was installed upright, above ground, but not in a cabinet.   The valve supposedly has an IP65 rating, but water got in anyway.  Pictures of a complete valve and the guts of the failed valve are below.  The water rusted the motor.  There are some faults in the case design.  The motor sits in a pit at the lowest point in the case.  Any entering water will go immediately to the motor.  The case seal obviously failed.  Further, the tongue and groove interface between the upper and lower halves of the case is the opposite of what it should have been.  The tongue is on the rim of the upper half and the groove is on the lower half.  With any failure of the seal, this arrangement promotes water ingress. 

I tore-down the failed valve.  The rusty motor is obvious.  The valve has one circuit board with components on both sides. See pics.  There are leads soldered to the board, and there are connectors that attach to the same CB traces as the leads.  Presumably, the connectors are for testing.  There are two leads for power input, two that go to the motor, and two pairs that go to limit switches. The side of the CB with the connectors will be called the top side. 

On the top side, in addition to the connectors. there are 4 electrolytic capacitors.  The 2 largest, are mounted horizontally, each rated 2.7 V 2F, and are wired in series. They are JH brand supercaps.  Wiring in series doubles the voltage rating of the pair to 5.4 V but halves the capacitance to 1 F.  Presumably these store power for closing the valve when power is removed. When 24 VAC power was applied to the input lines, the pair of supercaps charged to 5.6 V.     The motor is 5 VDC.  The other 2 electrolytics are 35 V 220 uF and 35 V 100 uF   There is a tan disk component with label Q 72 XF050 on the disk and F1 on the CB.  It’s a polyswitch resettable fuse rated 500 mA at 72V.  There is a blue disk component with no label by a RU1 mark on the CB.  It’s probably a MOV.  There is an inductor (L1) with marking of 220; it gets hot when power is applied.

On the bottom side, there are two surface-mount ICs, each with 8 leads.  One (U1) has markings of 34063 PNJZA, and is a 1.5-A, boost-buck-inverting switching regulator, 3-40 V input, 1.25-40 V output.  These are made by many companies.  The other IC (U2) has markings of RZ  RZ7888  1846.  It’s a Ruizhi 2-A DC, bidirectional motor driver.  There is one small surface-mount transistor (N1).  There are 3 large diodes (D1-3) and a full-wave rectifier (BG1).  And there are a bunch of small, surface-mount resistors and capacitors.

When I applied 24 VAC to the power input of the failed valve, the voltage across the motor terminals was zero.  If the motor was disconnected, the voltage on the motor leads was still zero.  Presumably, there is a failure in the motor drive circuitry in addition to the motor itself.  Resistance across the motor terminals was 10 ohms with the motor disconnected.  Interestingly, when power was removed, the voltage on the motor leads jumped immediately from zero to 5.5 VDC and then began slowly decreasing.  Presumably, the voltage is coming from the pair of super caps which were discharging internally or through a bleed resistor.  So, it appears that the reversing circuitry was working OK.

When I applied 9 VDC to the motor terminals, there was a barely audible click with one polarity; silence with the other.  The motor shaft could be turned easily by hand.

To prevent water damage in the future, all ball valves that are not in cabinets are covered with a small, transparent acrylic box.

Mike in California

[NiHaoMike]

Have you considered modifying a diaphragm valve to be operated by compressed air? Then you can just route thin tubing indoors and have a small compressor and solenoid valves there.

[Infraviolet]

Wonder if the motor was made from a much worse quality more rust prone steel than everything else. Because it looks like while the motor is rusted, all the screws, and the black plastic gearbox's output spur gear, are in perfect condition.

Dead electronics might be due to whatever wa driving the motor having been exposed to a very high current draw when the motor failed? Some sort of low resistance path being created between the motor's terminals?

[calzap]

Have you considered modifying a diaphragm valve to be operated by compressed air? Then you can just route thin tubing indoors and have a small compressor and solenoid valves there.

Appreciate the thought but it will be quicker and cheaper to install motorized ball valves.  I’ve attached a diagram of a pilot-operated diaphragm sprinkler valve below.  When the solenoid is not energized, the tiny channel to the outflow will be closed.  Via the tiny hole in the diaphragm (also subject to plugging by particulates!), the water pressure will equalize on both sides of the diaphragm, and the dominant force will be the spring, which will keep the valve closed.  When the solenoid is energized, the tiny channel will open between the top chamber and the outflow which will drop the pressure in that chamber.  The pressure of the incoming water will be the dominant force and push diaphragm open against the force of the spring.  This design is popular because it’s cheap to manufacture.  For example, the same small solenoid can be used on the smallest to largest valves.  It’s also completely tolerant of immersion.

Not sure how air could be used to improve the operation.  One thing that does work to lessen the particulate problem is to use a larger channel between the upper chamber and the outflow.  This requires a correspondingly larger solenoid.  We have such a valve on the main, unfiltered supply to the system.   Has never had a problem, but it’s large, made of brass, and costs as much or more than a ball valve.

All valves are controlled from the pump house near a pond.   The longest plumbing and electrical runs are hundreds of meters, so routing air tubing would be a huge, expensive job.

Mike in California



A- Input side
B- Diaphragm
C- Pressure chamber
D- Pressure relief conduit
E- Solenoid
F- Output side

[calzap]

Dead electronics might be due to whatever was driving the motor having been exposed to a very high current draw when the motor failed? Some sort of low resistance path being created between the motor's terminals?

I think you're right. 

[Stray Electron]

  I would open up the plastic box on the motorized values and coat everything with LPS-3 spray. I've had great luck with that preventing rust.  I would test it on one valve first just to make sure that it doesn't affect the electronics but I don't think that it will. But I would be careful not to get it inside of the electric motor as it does dry to a thick wax coating.

[Stray Electron]

  Looking at your second picture, it seems that there is a plastic cup formed in the blue piece of plastic and that the end of the motor sits in that cup. I believe that the cup is trapping any moisture that gets inside of the plastic housing.  If I'm right and that cup is on the bottom side of the housing then the end of the motor is sitting in liquid water most of the time since they're no place for it to drain. I would drill a decent sized hole in the bottom so that any water that gets inside of the housing can drain.  In my experience, water will  eventually get inside of housings of that sort and you're better off to put a few holes in the bottom of them so that the water can drain rather than remain trapped inside.  Also when you close up the housings put a good thick layer of silicon grease around the joint to help seal it.

[calzap]

  Looking at your second picture, it seems that there is a plastic cup formed in the blue piece of plastic and that the end of the motor sits in that cup. I believe that the cup is trapping any moisture that gets inside of the plastic housing.  If I'm right and that cup is on the bottom side of the housing then the end of the motor is sitting in liquid water most of the time since they're no place for it to drain. I would drill a decent sized hole in the bottom so that any water that gets inside of the housing can drain.  In my experience, water will  eventually get inside of housings of that sort and you're better off to put a few holes in the bottom of them so that the water can drain rather than remain trapped inside.  Also when you close up the housings put a good thick layer of silicon grease around the joint to help seal it.

I considered drilling drainage holes, but was concerned about drilling too far and damaging the motor.  And most of our valves are mounted vertically (the one that failed is an exception).  That leaves the question of where to drill.  I think the acrylic boxes will work nicely to keep moisture away unless the valve and its connections leak (it's never happened).  See the pic.  Putting sealant on the seam of the valve case is a good idea.

Couple of other things to note in the pic.  We put all electrical connections in a piece of PVC pipe, mounted vertically with a cap.  In addition, every valve has an LED wired in parallel (it's just above the black number tag).  If a valve is not working, we can decide instantly whether it's getting power.  It's also useful when doing dripper checks and other tasks. The LED is encased in a polycarbonate tube with a polycarbonate cap cemented on.  We tried clear heat shrink.  But it doesn't tolerate UV.  In a year or so, it becomes opaque then disintegrates.   

Mike

[jbb]

AFAIK, there is a tendency for water vapour to get into enclosures (can slowly diffuse through solid plastic) during the day and then condense into water droplets when things cool down at night. Over time things fill up unless there’s a drain hole.

Of course, a drain hole may backfire in other ways.

[calzap]

AFAIK, there is a tendency for water vapour to get into enclosures (can slowly diffuse through solid plastic) during the day and then condense into water droplets when things cool down at night. Over time things fill up unless there’s a drain hole.

Of course, a drain hole may backfire in other ways.

You may be right.  Here in N. California, there are enough bugs (spiders, moth larvae, mud wasps, etc.) that clog small openings that we would have to clean small drain holes frequently.  I'll experiment using the acrylic cover boxes and see what happens. If there is another valve damaged by water, I'll try a large drain hole with a screen on it.

Mike

[mk_]

AFAIK, there is a tendency for water vapour to get into enclosures (can slowly diffuse through solid plastic) during the day and then condense into water droplets when things cool down at night. Over time things fill up unless there’s a drain hole.

Of course, a drain hole may backfire in other ways.

You may be right.  Here in N. California, there are enough bugs (spiders, moth larvae, mud wasps, etc.) that clog small openings that we would have to clean small drain holes frequently.  I'll experiment using the acrylic cover boxes and see what happens. If there is another valve damaged by water, I'll try a large drain hole with a screen on it.

Similar Problem with moisture insinde an IP67 (we only need sealing against dust and rain from above) housing here. We solved this with 1mm holes - tiny enough to block most of  anymals - and fill this whole with some anorganic fibre like glasfibre as used in FR4 or similar products (I don`t know the english word). So water can drain out without letting animals or dust in. Since then every unit hasn`t seen any problems moisture related.

[amaschas]

This is an interesting analysis, thanks for sharing. Out of curiosity, is there a reason you chose not to use a DC solenoid? I'm using them for liquid distribution currently, and IIRC they can provide a fairly large aperture depending on the model.

[Doctorandus_P]

I also guess that either an hermetically sealed metal enclosure (expensive, especially with moving seals) or a semi-open enclosure with drainage and ventilation to prevent moisture build up is the solution.

I once tried to put the electronics of a bicycle rear light  (batteries, LED's, switch, some electronics) in a sturdy plastic bag and put that back in the bigger enclosure, but it did not work at all. The plastic bag just filled up with moisture, probably condensation, and with no way for drying it just kept accumulating.

Putting some water repellent grease (such as silicone grease) on the motor itself may help. Silicone grease has a tendency to spread around, and thus prevent rusting. That wax used to coat insides of hard to reach places in cars may also work well to repel moisture from the motor and increase life expectancy.

Enclosures for outdoors electronics also often have heating elements to keep the internals dry. Keeping them just a few degrees above ambient may be enough. In hermetically sealed enclosures, sometimes bladders are used to compensate for air expansion without creating differential pressures.

[calzap]

This is an interesting analysis, thanks for sharing. Out of curiosity, is there a reason you chose not to use a DC solenoid? I'm using them for liquid distribution currently, and IIRC they can provide a fairly large aperture depending on the model.

In the US, a 24 VAC solenoid is what comes with most diaphragm valves.  The solenoids are pre-threaded with a water seal. And 24 VAC is what most irrigation controllers use to power valves.  So, changing to any other voltage would be an expensive, time-consuming exercise.  That said, the solenoids work fine on DC.  In fact, a common way to test solenoids in the field is with a pair of 9 V batteries and some leads.

I was happy when motorized ball valves became available that operate electrically like the diaphragm valves … apply 24 VAC, they open; remove power, they close.  That meant no electrical modifications were needed to substitute ball valves for diaphragm valves.  I had a few DC-powered ball valves in use earlier that required one DC polarity to open, the opposite to close.  So, I designed and built interface boxes such that when a box received 24 VAC, it would put out the correct polarity DC to open a ball valve and do the reverse when 24 VAC was removed.  Don’t have to do that anymore.

As mentioned earlier, 24 VAC diaphragm valves are available with much bigger channels.  But they have a much bigger solenoid, cost a lot more and consume more power.  And there are direct-acting solenoid valves (no diaphragm) available with just about any operating voltage you can reasonably choose.  But they are also more expensive and use more power than pilot-operated valves with the same pressure and flow specs.

A big advantage of pilot-operated valves is that power consumption of the solenoid is very low because it’s small.  The pressure of the inflowing fluid supplies most of the energy to operate the valve.  While this is nice in many applications, there are disadvantages.  Clogging of the narrow channels with debris is one.  Another is that they don’t operate at all or poorly at low pressures.  If the force exerted by the inflowing fluid starts to get close to the force exerted by the spring, the valve starts to close or actually closes regardless of state of the solenoid.

Mike

[calzap]

Enclosures for outdoors electronics also often have heating elements to keep the internals dry. Keeping them just a few degrees above ambient may be enough.

This is doable in my situation.  We have spare electrical lines available wherever there are valves and plenty of 12 VAC and 24 VAC power.  So, powering heating resistors could be done in the humid part of the year (summers are bone-dry here).  I'll consider it if we have damage to an additional valve.

Mike