4v per second rise on chassis to earth - I cut it at 100v AC!

[Ace1]

3 1.5KW single phase motors 35uF CBB60 capacitors with stud to chassis.
The chassis is a 150L air receiver.
Chassis isolated by plastic wheels.

With all three motors running, the potential difference was metered between the chassis and a supply socket earth pin.

I watched the voltage rise until 100v AC, when I cut the power.
I was concerned about possible damage.

My problem is that I just don't have enough theoretical knowledge to pop out a definitive theory while sat on the toilet.

Clearly, the chassis appears to be behaving like a capacitor, gaining charge.
The motors are rated 35uF

I'm thinking/hoping that there is probably only one potential reason for this to be occurring.
Does that reason spring to anybody's mind?

 

[telectrix]

earth the chassis to the main earth of the property might be a good idea.

 

[Ace1]

earth the chassis to the main earth of the property might be a good idea.

Yes.... certainly recommended

 

[Lucien Nunes]

You give an AC voltage reading but an isolated piece of metal building up charge will give you a DC reading. 100V AC would mean the stray capacitance of the non-earthed metalwork was being charged, discharged, then charged in the opposite direction 50 times per second, i.e. not slowly building up charge.

An isolated air receiver building up 100V DC charge on its stray capacitance I can easily believe. Air and insulating liquids like fuels rushing through plastic pipes create static charge by friction, hence the need for grounding straps to pipe fittings etc. on fuel tankers and static dissipative air hose etc. Earthing the tank will solve.

If you did have a rising AC voltage then the cause is not charge building up but leakage from the supply increasing as motors warm up etc. It might actually be a very small and harmless amount of current, if the receiver is truly isolated, but it suggests that something that should be earthed for electrical safety, e.g. the motor frames, actually isn't. Again the solution is to fix the earthing.

 

[Ace1]

Here's a photo that I took, showing the multi-meter, prior to cutting the power.

The heating motors is an interesting theory.
The air valves were left open, to allow the motors to keep running - simulating say sandblasting.

The motors were not cold to start with.... they had run, while looking at this, a few times.

This was not a scientifically run test... so I can't say how long it ran before I cut the power at 100v.
I had the feeling that the rise had speeded up slightly.

May I ask.... are you confident with the heating motors theory?
Is this a known condition?
Is it possible to explain why this should be happening?

The thing is, that we have two compressors of the same type, and the other does not do this.
Does it mean that there is a fault?
It runs fine when earthed.

 

[Lucien Nunes]

We don't have enough information:

Is the unit class I or II?
If class I, what is its earth continuity like?
What is its insulation resistance at 500V DC?
Is the tank supposed to be electrically isolated from the motors?

A voltage reading from a point thought to be isolated from the rest of the world (e.g. the tank) is not usually a very meaningful indication, all it tells you is that something is able to source enough current to develop a voltage across your meter's input resistance. Since the input resistance of a digital meter is quite high, it does not take much current to create an indication, so it doesn't tell you whether the source of that current is a fault or just normal leakage.

 

[Ace1]

Thank's Lucien,
I understand the format of your questioning.

I'm a mechanical engineer, that has through life, needed to comprehend basic electrical scenarios.
One of my areas of interest is engine tuning... and I regularly post a response similar to yours, in order to receive an audit of an engine, and it's fuel delivery etc.

The difficulty here, is that, unlike engines, where the OP is shamed into carrying out the audit.... getting carb data, needle and jet numbers etc...
... I am struggling to answer your questions.

Background:
What I know is that the motors are 1500W 35uF 7A single phase.
They are fed via a staggered start control unit, monitored via a standard diaphragm pressure switch.

At pre-determined pressure, the contacts separate, triggering the supply to be cut to all motors.
IE. unlike many compressors.... the feed is not via the pressure switch.
The controller 'sees' contact open & closure, and cuts or feeds the power to all motors.

On cutting power, an N/O relay opens and air in the main supply feed hose is evacuated.
While pumping, the 'blow-off' valve remains closed.

RE: Earthing

No earthing existed.
I have instated the chassis earth.
Everything seems to be running fine.... the motors are even..... I believe it is entirely safe.
Advice to the contrary will be acted upon.

Other than that:

The voltage leak is a mystery, in terms of it's rising potential difference.
The fact that the other compressor isn't like this, confirmed that this might be worthy of a new thread.

 

[Lucien Nunes]

Ok, alternative questions:
Is it an unmodified commercial product?
How is it connected to the supply? (e.g. 3-pin 32A plug & 3-core cable, by permanent wiring in conduit etc)
By 'no earthing existed' do you mean specifically there was no connection to the tank or no connection to earth at all via the supply cable / connections?
Can you post pics of the electrical data plate and the complete machine?

We cannot attach much importance to the 'rising' aspect of your reading as it may be an insignificant parameter that is rising. Or not, depending on whether there's something wrong with the earthing configuration that might otherwise be conducting the associated leakage current harmlessly to earth.

If I may make an analogy to pneumatics, leakage current is not too dissimilar to a leak in a pipe joint from a system that runs a constant pressure. Suppose you find a leaky joint and hold a hose connected to a pressure gauge over the outside of it. The gauge gives you a reading in PSI somewhere between zero and the system pressure, depending on how well you're able to seal the open end of the gauge hose over the leak. The reading doesn't tell you what is wrong with the joint or how many cfm of air you're losing, it only makes a comparison between the system leakage and the leakage from where you're holding the hose over it. This is a direct equivalent of a multimeter on a voltage range connected to a high resistance source, as the meter itself presents a high but finite resistance.

What you actually want to do is measure the current not the voltage, and locate its source, which a proper ISITEE procedure would identify.

 

[Ace1]

Really nice explanation.
The analogy hit its mark.

I don't have the kit to measure the AC current.... the multi-meter provides DC Amp readings.
Sorry... it's what I've got.

The supply is apparently a three pin plug, but the earth lead was cunningly clipped, necessitating removal of sheathing to expose more earth cable.... that was then fixed to an abraided chassis mounting, via a grip washer.

Will report back in a few minutes with explanatory photos.

 

[Ace1]

Here you can see the front panel.
the far left pair senses pressure switch contact.
the four central (3) supply power to the motors.
the two right are neutral.


Here we see the circuit board layout.
you will appreciate the soldering.... it does make contact to motor two, as I heated the land and the solder flowed to it.
It's a full sized image so can be zoomed.



Here is the motor label:



Here is the motor fixation

 

[Lucien Nunes]

Hmm, no sign of a class II symbol on the motor so one would expect it to need earthing. It's not clear whether the compressor is electrically continuous with the tank, the resilient mounts are probably insulative but there might be casual paths through metallic pipework. How was the earthing arranged when the unit was supplied? Was there a 3-core cable fitted, if so where does its G/Y core go? Or was it just the L&N terminals in that DP MCB? Was this sold as compliant with relevant UK/EU directives and/or CE-marked?

 

[mache]

I don't want to be unkind but is this being used in a commercial / production process. Type of construction, quality of labeling, poor translation and lack of serial number suggests to me that even if you get the current situation cleared it probably won't be long before you get other problems depending on the duty cycle it will be under.

Someone earlier mentioned possible static charge buildup from the movement of air which sounds like it could be pausible. If it was me I think I would be tempted to use the current function on your multimeter. You will probably find nothing to measure. But I am no professional so go only with recomendations you get from the qualified members.

Hope you get your concerns sorted.

 

[Ace1]

Thanks for the comments.

The earth cable had been cunningly clipped to the sheath.

I am old enough, and experienced enough to know better, but I didn't see it on first appraisal.
The fact is, that there were other problems absorbing the mind, so the issue of earth was seen later, as the primary problems were resolved.

I believe that one of the design failures was the lack of an earth bar.
Not having a 'panel mounting point' meant, out of sight, out of mind.
.... and no place for the assembly factory to connect the earth cable.... so clip it.

To be honest... in the overall scheme of how the world seems to work now, I actually don't blame the factory.
I'm certain that they will not have an engineer on the premise.
Their task is to buy in the components, and assemble.

It's clearly all a bit crap.
I think how it is, is that "it works, so it's fine".

We look at the labels... invariably skew-wiff (but it doesn't matter).
No engineer, means no jigs... so drills walk... so slot a bit... the component then fits (so it doesn't matter).

I'm not going to get on my high horse here... I had a priveleged education in the 70's by top engineers at Armstrong Patents.
Anything that left my hands had to have my name on it, and reflect the highest level of skill.

China simply jumped that stage of 'engineer creation' - while we dumped ours too.
So we reap what we sow.

That means: no comprehension of skill nor detail.... and if it works, then it is fine.
Right... I get that!

But perhaps it filters through, further up the chain.
The soft staggered start controller, clearly works, but by failing to see the need for an earth, meant that no connector was provided... meaning the assembler doesn't have anywhere for the earth cable.
They know that in many cases, an earth is not provided... so this must be one of those cases.
Hence the clipped earth wire

We are now at the control panel proper.
My guess is that the design is probably fine... it all works as it is meant to do.
The soldering in places was abysmal... but it works.

We move further up the chain to the motors, and the parallel capacitor.

The capacitor, I see like all these individual components.
They will fail quickly, or run continually.

However, that may be mistaken thinking.
Perhaps a capacitor could partially fail, yet keep working, and cause problems.

Then the motor.
My guess is that this will be the best component.
To produce it in the quantities they do, and the work it must do... it must have the tightest engineering controls.
But there is still that concern that 'out of tolerance is still probably fine'... so let it go.

Note: I don't know that... but it is a concern.

RE Luciens remark over motor-chassis connectivity.
I can say that it is present, because the rubber isolation blocks are merely for vibration.
To fully insulate would require a plastic stud tube, and an insulating washer.
Neither are present.
The motor is directly connected to the chassis.
-------

This leaves us with the mounting chassis voltage.

Having two kits, and one not having this problem, I am personally not leaning towards the static build up option.
If one, why not two?

The problem is apparently mitigated (not solved) by earthing the chassis, only I still don't know.
It may remain a mystery.

What i was hoping, was that there would be an immediate... "oh, I've seen this before... it is this"
Life eh?
It constantly tries us.

 

[Lucien Nunes]

Take a continuity reading with the multimeter from the incoming earth lead to each motor and to the tank. If they're not all off the bottom of the scale on the meter, add earth continuity straps until they are, you won't do any harm. The resilient mounts are often two metal discs moulded into the ends of a rubber bush - not designed as an insulator but most of the ones I use are good insulators by default.