Hi,

Appreciate if someone could share the knowledge regarding this:

The 110 VDC supply to the solenoid valves are common and coming from the control room. As I understand, the voltage polarity (+ve/-ve) of the coil doesn't matter when connecting to the 110 VDC supply. However, some mentioned that there is some effect of connecting the coil with different polarity in which it may shorten the life of the coil. Is this true? If yes, any explaination or reading material is very much appreciated. Looking at the coil wires, both are RED and no polarity sign is marked as is it wasn't really important.
We experienced the coil shorted after one day in operation. We are in process of determining the root cause.

TQ

 

Are you actually supplying dc to these coils? These sound like a standard ac solenoid which is non-polarized.

 

Some DC solenoid valves come with suppression diodes built in parallel to the coil. If you hook the solenoid up in reverse you will briefly have a short circuit through the diode until it fails open circuit. Since a suppression diode is typically quite small, this will not normally overload the device driving it and you may not notice it occuring.

Once the suppression diode has failed, there is know nothing to suppress the transient high voltages which result from the magnetic field collapsing in the solenoid each time it turns off. This can put stress on the coil, eventually causing insulation break down.

Note that not all DC solenoids come with built-in suppression diodes, so the above may not apply to your situation. However, all should *have* diodes, whether they come built-in, or you have to install them yourself. They don't need to be right at the coil, but they should be reasonably close to it.

 

I don't see how it would shorten the life of the solenoid. Either way, it's just trying to pull the plunger to the centre of the coil. If the dc current is the same value on either polarity, that means your heat dissapation/losses is roughly the same. so unless the manufacturer specifies other wise, i don't see a problem with it.

 

I see no issue with the polarity - but rather with the potential to ground of the 110 VDC supply - is one of the 110 VDC supply poles grounded? What is the tolerance of the coil versus the actual voaltage range of the supply? Remember, if the source is lead-acid batteries, then the voltage a "24 VDC" battery can go up to 27.5 VDC, and even higher, with proportional results on a "110 VDC" arrangement.

Meir

 

Responding to Ed's May 22, 10:25 pm query... Yes, at one time polarity had a deleterious effect on DC-operated solenoid coils. (I'll bet that the persons mentioning it were oldER!)

Because of material improvements, as well as conditioned-environments, the problem is rare today.

The problem is covered in detail in a 60-year old book (ca 1945) I have. If you want a copy of the article, send me your e-mail address?

Responding to Michael Griffin's May 23, 10:33 pm comment... the suppression diode will not affect coil-failure!

p.s. to all: Read the caveat covering transient-suppression techniques in my paper:

"Probabilistic Risk Analysis of SIS (formerly ESD) Systems"

Regards,
Phil Corso, PE {Boca Raton, FL, USA}
[[email protected]] ([email protected])

 

In reply to Phil Corso - Could you please elaborate on your point? The original statement was based on the assumption that without the diode turn-off would produce a (relatively) high voltage across the coil. This voltage could eventually cause insulation break down in the windings. The diode (if present and functioning) would prevent the transient voltage from reaching high levels. Most coils could operate without a diode for years, but one with weak insulation may fail prematurely.

 

Thanks to everyone for the response.
Some info to add:

1. Our 110VDC is floating from ground. The supply quality is good as measured at the field.

2. Yes, we have a suppression diode installed at he terminal inside the control room (just before the 2 wires going to the field - connected in parallel). As such polarity is not an issue at the solenoid coil.

The question now is the posibility of manufacturing defect. I don't have much data on this but I assume it is very unlikely. This solenoid though, are manufactured in batch and I believe the test was done randomly.

TQ

 

Certainly:
I understand and accept your premise that an open-circuited diode can't negate the transient voltage the coil's collapsing field produces. But I doubt this happened for two reasons:

First, Ed said it happened after just one day of operation. While the "last straw on the camel's back" scenario is possible, it is highly
improbable.

Second, diode failure-mode is more likely to be short-circuited than open-circuited. But guesstimating is not necessary. Ed can test my
thesis!

Comment To Ed,
I know the usual reason it is done but Control room placement of the diode is not the best location! Once again, I refer you to my paper for
transient suppression reasons and techniques.

Regards,
Phil Corso, PE {Boca Raton, FL, USA}
[[email protected]] ([email protected])

 

Wound coils are subject to two general classes of manufacturing defect. One is due to thin or bare spots in the insulation as delivered from the wire manufacturer, and the second is due to damage in the coil winding process or in subsequent handling.

The wound coils would typically be tested by the solenoid manufacturer. However the testing methods are not fool proof so even with 100% testing the manufacturing process requires carefull set-up and monitoring by properly trained personnel.

If you are very concerned about this solenoid, you should contact the valve manufacturer through the distributor (or rep) and ask if they could examine the failed valve and determine the root cause of failure. They should also be able to answer application questions with more certainty than our speculations could.

 

Try installing another diode at the coil. The inductance of the long wires may be enough that the coil still sees the spike from field collapse. It doesn't take much inductance before the diode is ineffective. If you were to connect a fast oscilloscope across the coil you would see that 6 inches of wire may be enough inductance to allow a surprisingly high voltage before the current through the diode begins to rise. That's why the diode is often connected right at the terminals. And if the driver is sensitive, it should be a Schottky or other hot carrier diode. As I've said before, it's amazing this automation
stuff works as lax as they are about electronics. The signal integrity is most often horrible.

Regards

cww

 

In reply to Phil Corso - While the normal failure mode for a diode may be a short circuit, that analysis applies to a diode which is installed correctly (reverse biased). If the diode is backwards (forward biased), it is already effectively a short circuit in parallel with the coil and will carry full available load current. In either case it will normally proceed to an open circuit fairly quickly due to over heating as the diode will normally be quite small and unable to carry full load current.

Having said the above however, I agree that this sort of explanation is really grasping at straws as the coil is unlikely to fail quickly (or even at all) unless it was already somehow defective. The real cause can only be determined through a tear-down analysis of the solenoid.