boiler water pump relay burned out - again

Zephyrprime

Last spring, my boiler's pump stopped pumping. This pump is about ~5 years old. I replaced the relay controlling it and it worked again. Now the heat went on for the first time this year and the pump was inactive again. I replaced the relay again and it has started working again.

However, I wonder if something more significant is going on here since the relays keep dying.

The relay is very hot to the touch. Definitely more than merely warm. The pump is 0.75 according to its label. There are two wires leading to it and I took an ac magnetic ammeter and measured to see if it was drawing too much power. The readings were strange. One of the lines read about 0.75 amps. The other line read about 1.5 amps. ****? Should they be equal (with opposite polarity)? There is large water pipe leading out of the pump but it is too large for me to clamp my ammeter over. I tried taking a reading from one of the smaller water lines and it read about 70 milliamps. What is going on here? Do I have a short of the electrical line to the water lines or is something else happening here?

JUGHNE

Try clamping over both lines together, should be very near to zero amps.

Teemok

Is the relay coil voltage correct? Most pump relay contacts can handle 5 amps. You are way under that. Is this a 120v pump? Power off, check for continuity to ground on both of the disconnected pump wires. There should be none.

neilc

the inbalanced amperage,
something is grounding out,
replace motor?

Jamie Hall

You may have two problems, not one. As @neilc said, the unbalanced current on the pump is a problem -- though if you are using a clamp type ammeter be sure that have zeroed it for each measurement, not just one. I have seen very large errors with clamp type ammeters by just moving the clamps a few inches.

However, if the relay coil is burning out, that's not from the pump current. That suggests an overvoltage problem on the coil. Check the voltage between the coil terminals when the coil is switched on. It should be within 10% of the rated voltage for that relay.

HomerJSmith

You have it wired wrong, betcha. It depends if the coil is toast or the points are toast. It is usually the coil goes out right away when you make the connection. The coil is usually 24Vac although there are line voltage coils. If the points are gone, use a two pole relay and connect both points in parallel. The pump maybe drawing too much current, a locked rotor for example.
If you measure the voltage between the hot lead and ground the voltage should be about 110V to 120V. The neutral lead to ground should be 85V to 95V. The hot lead to the neutral lead should be 110V to 120V.
It could be the pump not circulating at design speed which may increase the current draw. A Taco pump?
If the line voltage is over 125V call the power company and have them fix it.
One last question does anything else burn out in the house, light bulbs etc.
You say the relay goes, does the transformer go?

mattmia2

i would insert a shunt type ammeter in series before i'd conclude there is actually leakage current, clamp meters can read very differently depending on how the clamp is positioned.

motors have some amount of inrush when starting, relays usually have a horsepower rating that more or less accounts for this, the 5a or whatever rating is for purely resistive loads.

Zephyrprime

JUGHNE said:

Try clamping over both lines together, should be very near to zero amps.

That's what it should be but it's not actually that way. When combined, it's about .75a.

Teemok

A hot relay from 1.5 amps at the contacts? Maybe but I don't think so. Is it an AC or DC coil relay? Assuming this is switching 120V, neutral wire is connected to ground at the panel. You shouldn't see voltage over 2v from N to ground with a load. Is it possible the neutral is shared with a hot wire on the other phase. Strange things are afoot.

mattmia2

the coil dissipates a few watts which makes it hot.

EdTheHeaterMan

Ask @109A_5 about this. I'm sure he will find a SHORT somewhere.

Zephyrprime

Thanks for all the input guys. I have investigated and this is what I have found.

The relay is ac and the pump is ac as well. 120v. It is a Taco brand.

The old relay and the first replacement were the same model and were rated for 5.5a inductive load. The new relay is rated for 12a inductive load.

The coil is 24 ac. I tested the coil terminals and it is getting 26v ac.

I disconnected the power lines from the pump and tested the two wires. The hot wire has about a 100ohm short to the water pipes and a 10ohm short to electrical ground. The neutral wire has about a 100ohm short to both the water pipes and electrical ground.

So what do you guys think? Sounds like a mess to me.

mattmia2

Is that with both the wires of the circulator disconnected from everything else?

EdTheHeaterMan

Can you post a picture of the pump with the wiring cover off. Some of those pumps have a capacitor under that cover. If you take the AMP reading from the 2 motor leads (A & C)you may get different readings. if you take the readings from the two wires that get connected to L1 and Neutral (A & B ) they should be the same.



If you are taking a reading from A and C you will get different readings
If you are taking a reading from A and B You will get different readings

Just want to be clear on where you are reading your numbers.

109A_5

Hello @Zephyrprime,

Zephyrprime said:

I disconnected the power lines from the pump and tested the two wires. The hot wire has about a 100ohm short to the water pipes and a 10ohm short to electrical ground. The neutral wire has about a 100ohm short to both the water pipes and electrical ground.

So what do you guys think? Sounds like a mess to me.

Yes, it sounds like the circulator would miserably fail a Hi-Pot test. I would expect either power wire measured to the circulator housing or ground wire to be near infinite Ohms on most multi-meters. That may explain your original odd current readings and relay failure.

Jamie Hall

Readings like that are really bad. The hot wire -- when disconnected at the switchboard (you can just open the breaker or take out the fuse) -- should read at least 1 megohm and preferably better than 10 meg to either ground or neutral. To check the neutral wire it will have to be disconnected in the power switchboard (CAREFUL IN THERE! but when it is, it should have the same resistance. On the other hand, when it IS connected at the switchboard it should read at most a few ohms to ground. The water pipes could be several hundred ohms to ground.

I might add that if that 10 ohm hot to ground reading is real, if that allows a current path through the relay contacts that might explain a short and miserable life for the contacts,,, but it WON:T trip a 15 amp breaker or fuse.

Zephyrprime

EdTheHeaterMan said:

Can you post a picture of the pump with the wiring cover off. Some of those pumps have a capacitor under that cover.

Yeah there is a humongous capacitor in that housing. Much larger than the one in your picture. I am testing at points A and B.

Zephyrprime

Jamie Hall said:

I might add that if that 10 ohm hot to ground reading is real, if that allows a current path through the relay contacts that might explain a short and miserable life for the contacts,,, but it WON:T trip a 15 amp breaker or fuse.

Yeah it doesn't really make sense. With only 10 ohms of resistance, it should be flowing 12amps through hot and not just 1.5 amps.

mattmia2

10 ohms is the resistance of the coil, it doesn't account for the reactance, the interaction of the magnetic field and the motor will resist ac current. It is likely a few ohms dc to neutral so you must disconnect all of the motor leads when testing if the leads are shorted to the case since it will be connected to ground through the neutral at the service if you do not disconnect the neutral.

109A_5

Hello @Zephyrprime,

Zephyrprime said:

I disconnected the power lines from the pump and tested the two wires. The hot wire has about a 100ohm short to the water pipes and a 10ohm short to electrical ground. The neutral wire has about a 100ohm short to both the water pipes and electrical ground.

With reading like this I would suspect the coil of the motor is compromised to the housing. The point of compromise could change resistance with power applied, (Ohmmeter probe voltage and current versus the AC power) like an incandescent lamp changes resistance between a cold and hot filament. Corrosion and/or failing winding insulation can do odd things and give odd meter readings.

I also would suspect that the boiler (its pipes) may not have a good bond to the buildings electrical system ground. And/or any parasitic AC voltage in the buildings grounding system may be messing with the Ohm meter reading.