Author Topic: Pump motor postmortem  (Read 2399 times)

0 Members and 1 Guest are viewing this topic.

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Pump motor postmortem
« on: January 04, 2020, 02:14:16 am »
The pump is a two-stage centrifugal pump and was powered by a 3 HP, 240 VAC, 60 Hz, 1 Ph motor running at 3450 PRM.  It had been in use for 10 years, used to keep water pressurized in tanks, and controlled by a pressure switch.  It had over 6000 hours on it, but I still consider it a premature failure.  Just before failure, it made a bit of noise, which increased suddenly with a small amount smoke seen before it was shut down.

The pictures below show a roasted motor with giblets.  Main course is burnt stator windings, followed by a melted fan blade, and finish with a thoroughly baked run capacitor.

Run capacitor is same brand as motor ... WEG.  Start capacitor is a Lorenzetti and tests OK.  So, what failed first?  Probably not the stator coils.  If the insulation on them had failed first, I would have expected to find a small area that was severely burned, maybe with some melted metal, with progressively less scorching further out.  Instead, it appears that there was relatively uniform heating with little or no arcing between conductors.  My vote is for the fan blade.  If the fan blade fails, everything overheats including the coils and capacitors, which toast and die.  The fan blade was nylon and force-fitted to the motor shaft.  My guess is that it started losing its grip on the shaft which started things heating, including the shaft, which caused more slippage, and so on.

For a motor/pump combo that costs US$1000-2000, would it be asking too much for a decent run capacitor and a metal fan blade that is solidly attached to the shaft?  Maybe so.  And a thermal fuse  ... how much would that have cost?  The motor bearings were in good shape, and the pump is fine.

One other gripe is the motor shaft is extended out so that it is also the pump shaft.  This saved some money because the manufacturer didn't have to buy a coupler or a frame capable of holding a separate motor and pump.  However, it makes the motor unique to the pump.  Unless I want to saw the shaft from the motor, buy a coupler and build a frame, I'm stuck buying the same motor to run the pump.

Mike in California

 

Offline Gregg

  • Super Contributor
  • ***
  • Posts: 1156
  • Country: us
Re: Pump motor postmortem
« Reply #1 on: January 04, 2020, 02:35:24 am »
Sometimes you get what you pay for, i.e. cheaper product = cut corners (often times because bean counters can’t find the funds until things fall apart and then funding suddenly appears out of nowhere).   |O
The motor doesn’t look like it has been in ideal operating conditions or it wasn’t designed for the conditions where it was placed.  Note the rust on the laminations and dirt inside.  A totally enclosed fan cooled motor would be much better.  How are the bearings?  I’d bet they are shot like the rest of the motor and may have been the main cause of death.  Bad bearings may have helped the cheap plastic fan to spin on the shaft and no longer provided proper cooling.  But it doesn’t look like the rotor crashed into the stator.
Was the pump run continuously or were there a lot of start / stop cycles?
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #2 on: January 04, 2020, 02:52:07 am »
Pump was use for irrigation and operated to keep pressure tanks charged, so operated by a pressure switch.  Lots of on/off cycles 7-8 mo/yr; not used during rainy season.  Actually had excellent environmental conditions.  Was completely indoors in a pump house and run only in the dry season.  There were no pipes or valves running above the motor.  But 10 years is a long time, and I would expect any non-stainless, non-protected steel to rust in that time.  Huge amount of air would have been blown through it in 6000 hours of operation accounting for dust and dirt inside.

Mike in California

 

Offline andy3055

  • Super Contributor
  • ***
  • Posts: 1159
  • Country: us
Re: Pump motor postmortem
« Reply #3 on: January 04, 2020, 03:04:18 am »
You forgot an important part. The centrifugal switch on the motor shaft that disconnects the start capacitor from the circuit after speed is up. Most often those fail due to continuous on/off operation. If the switch was stuck in the off position or pitted so bad that it does not make contact, the next time around when the motor has to start, the start winding is off circuit and the run winding will be getting hot as the motor does not spin.

If you say the run cap has failed, it could also be that the run cap just failed due to age. In the 10 years of operation, its life span would have been already finished.

The pump may also need replacement eventually. A better bet would be to buy a motor and pump that are externally coupled so that only one part needs replacing anytime. Ultimately, it is about the $$.
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #4 on: January 04, 2020, 03:40:32 am »
Good point on the centrifugal switch.  I'll inspect it before tossing the motor.  The pump metal parts are in great shape.  Case is cast iron and vanes are stainless.  I'm impressed with the pump, but think they could have chosen a better motor to attach to it.

There is a rebuild kit for the seals and gaskets on the pump.  I may go ahead and cut the shaft from the motor, buy a better motor and coupler, build a frame, and get the pump going again.  There is already a replacement pump/motor in operation, but we like to keep a spare ready to go.

Mike in California

 

Offline unitedatoms

  • Frequent Contributor
  • **
  • !
  • Posts: 324
  • Country: us
Re: Pump motor postmortem
« Reply #5 on: January 04, 2020, 03:54:25 am »
It is amazing that AC motors even work, because nobody can explain how exactly they work. As amazing as few remaining domestic telephones working over wires. That motors mystery was noticed in the past by someone, who found that AC motors in refrigerators never fail. Never. That's is another mystery.
Interested in all design related projects no matter how simple, or complicated, slow going or fast, failures or successes
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8526
Re: Pump motor postmortem
« Reply #6 on: January 04, 2020, 04:48:14 am »
If the start winding is either stuck on or off, the motor will be drawing a lot more current than usual and the overload protector should trip... did it have an overload protector?

That price you mentioned sounds like it could be worth rewinding the motor --- and fitting a more permanent fan (one made out of sheet metal).
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Pump motor postmortem
« Reply #7 on: January 04, 2020, 07:12:29 am »
It is amazing that AC motors even work, because nobody can explain how exactly they work. As amazing as few remaining domestic telephones working over wires. That motors mystery was noticed in the past by someone, who found that AC motors in refrigerators never fail. Never. That's is another mystery.

I don't think there's any big mystery how AC motors work, they are quite well understood as far as I know. Certainly the basic idea behind an AC induction motor is not difficult to grasp and I don't find it all that amazing, I mean physics do tend to just work.

Refrigerator motors are very reliable but they do fail. Just recently I tore down the compressor from a water cooler, it's the same kind of compressor you'd find in a small dorm type refrigerator, I was surprised to find that it had only a single bearing between the compressor and motor section, the bottom end of the motor shaft was free hanging. The bearing seized and caused the compressor to lock up. I've seen compressors in domestic refrigerators fail on a few other occasions too but most refrigerators are replaced due to broken shelves, clips, other hardware, or simply because they're cosmetically out of style.
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #8 on: January 04, 2020, 08:41:50 am »
Yes, the motor had overload protection, but it never tripped.   It had a fairly sophisticated controller that not only protected against overload, but also against low and high voltage, too frequent cycling, and no water at the pump inlet.  It didn't have over-temperature protection or fan draft sensing.

Motor value is less than one half the motor/pump combo, so not worth rewinding.  Would prefer to buy a better brand of motor and couple it to the pump.

Mike in California


 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 7375
  • Country: au
  • Toys so very many Toys.
Re: Pump motor postmortem
« Reply #9 on: January 04, 2020, 09:55:15 am »
......

One other gripe is the motor shaft is extended out so that it is also the pump shaft.  This saved some money because the manufacturer didn't have to buy a coupler or a frame capable of holding a separate motor and pump.  However, it makes the motor unique to the pump.  Unless I want to saw the shaft from the motor, buy a coupler and build a frame, I'm stuck buying the same motor to run the pump.

Mike in California

Dealing with this issue. This is standard and normal on large numbers of Industrial single and multistage pumps up to about 3kW. They work reliably in 24/7 duty and providing the seal doesn't fail it is a non issue. On higher stage (impeller) count pumps bearing thrust issues can be an issue but generally 6000 series bearings cope well.

Some of the controllers 'claims' of protection I have always taken with a grain of salt. There are mechanical failures that can give false positives depending on your plumbing. Generally part of the sensing of these is pressure and temperature a simple outlet check valve failure could trick the pump into thinking everything is ok while the pump continues running dry and getting hot.

Do you have a decent size pressure tank in the system and was it charged with air?
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #10 on: January 04, 2020, 06:10:44 pm »
With the exception of overload, the various protection functions of the controller have been put to real-world tests including low and high voltage from the power company.  There is a test button on the overload module, and it seems to work, but I don't know how realistic the test is.

The outflow of the pump goes to a pair of 86 gallon (327 liter) tanks that have air above diaphragms.  We always use two tanks in case a diaphragm breaks, and yes, it's happened.  Air pressure is checked and adjusted occasionally and has never been an issue short of a broken diaphragm.

Outflow goes through a flapper-type check valve to the tanks.  Pressure switch is in the plumbing Tee assembly that conducts water to/from tanks.  Distance from pump to Tee is about 2 m.  Pressure switch is dual-acting, low pressure safety type.  If tank pressure falls below a certain  level, the pressure switch flips to an off-state that requires a manual reset.  Most common causes of this have been loss of prime, controller detecting something amiss, failure of a controller component, or attempt to draw too much water from the system.

If the outflow check valve stuck in the open position, it would probably not be a big deal because there is a check valve in the inflow plumbing.  If it failed in the closed position, yeah, might destroy the pump  ... any significant obstruction between the pump and pressure switch has that possibility ... might even be treated to a steam explosion.  Possible saving grace in our system is that when we're irrigating, it's usually 24/7.  If the check valve stuck closed, the pressure on the other side would soon drop to the safety trigger level requiring a manual reset.  I've never seen a flapper check valve get stuck in the closed position.  I've seen them get stuck in the open position.   Check valve in this system has worked fine in the past and is working fine today.

The system is pretty well tuned to avoid short cycling too.   During high demand, like sprinklers, the pump runs continuously.  During low demand, like drippers, it cycles on and off based on pressure switch cut-in and cut-out pressures.

Mike in California

 

Offline andy3055

  • Super Contributor
  • ***
  • Posts: 1159
  • Country: us
Re: Pump motor postmortem
« Reply #11 on: January 04, 2020, 06:26:01 pm »
If you are planning to buy a separate motor and couple it to the pump, try to buy a motor that has only the run capacitor. That will eliminate the centrifugal switch issue. You could always buy one that has no capacitors (TEFC). The advantage is that the interior of the motor will be safe from the elements. I think it depends on the power/size. The motors with run caps are smaller in size, if I remember.
« Last Edit: January 04, 2020, 08:24:00 pm by andy3055 »
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Pump motor postmortem
« Reply #12 on: January 04, 2020, 08:49:21 pm »
Lots of motors without a start capacitor still have a start winding and centrifugal switch, as far as I know the only motors that avoid this are fractional horsepower shaded pole types. To avoid all this on larger motors you can use a 3 phase motor, if you only have single phase power available  you can use a VFD to power it. Those seem to be quite reliable, there are lots of them running things at my friend's machine shop and we have not had one fail yet.
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #13 on: January 05, 2020, 12:15:26 am »
Just noticed something on the motor label.  See the picture below.  Well, it didn't burst into flames!   I think "Thermally Protected" refers to the Klixon CSJ31SV overload protector at the end of the motor opposite the fan.  This OP device works by passing the motor current through a wire that, if current is too high, will heat a bimetal disc which will change shape and  open  contacts.  It's really a self-resetting, thermally operated circuit breaker.  Theoretically, it can also protect against high heat in general in the area where it's located.  Unfortunately it was at the end of motor away from the fan and not that close to the stator.  With fan blade failure, heat generated in the coils would not reach the device sufficiently for it to offer thermal protection.

I also examined the centrifugal switch.  The centrifugal mechanism on the shaft was in good shape.  When at speed, a plastic disc (opposite end of the shaft from the fan as seen in pics in my first post) moves along the shaft and presses on a pair of non-rotating plastic posts that move and open contacts.  The contacts were OK, but the plastic posts had some wear.  I don't know if the wear was sufficient to prevent the contacts from opening.

Mike in California

« Last Edit: January 05, 2020, 12:18:49 am by calzap »
 

Offline andy3055

  • Super Contributor
  • ***
  • Posts: 1159
  • Country: us
Re: Pump motor postmortem
« Reply #14 on: January 05, 2020, 02:37:03 am »
 

Offline calzapTopic starter

  • Frequent Contributor
  • **
  • Posts: 539
  • Country: us
Re: Pump motor postmortem
« Reply #15 on: January 05, 2020, 08:08:34 pm »
Thanks for all the replies and the link to a motor supplier.   I've asked for advice on a shaft coupler as a separate topic.  Here's the link:  https://www.eevblog.com/forum/repair/(b)advice-wanted-on-motor-shaft-coupler(b)/msg2858712/#msg2858712

Mike in California
 

Offline DBecker

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Pump motor postmortem
« Reply #16 on: January 06, 2020, 12:28:51 am »
It's not likely that you'll be able to diagnose the failure with a melted fan, but pressed-on nylon parts have a failure mode where the hub splits from long-term stress.  Humidity changes increase the stress.  A spring steel collar around the hub mitigates the problem.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Pump motor postmortem
« Reply #17 on: January 06, 2020, 01:27:28 am »
I've seen similar fans come loose before. I used to have a dishwasher with a metal fan pressed on the motor shaft that came loose, the motor was spot welded together and the fan was inside the shell so all I could do is dribble some glue on it now and then and it would last a while.

As far as big motors go, I've been impressed by Baldor but there are doubtless many other good quality motors to choose from.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf