Electric motor troubles.

[andjygen]

Hi,

I have an old sheet metal shear from around 1960. It has an 240V AC motor with brushes. The brushes are connected to an odd looking cylindrical three wire capacitor marked 0.2 + 2*0.001 uF, where two of the leads are connected to each of the brushes and the third lead is connected to earth. I think the cap is made from oil saturated paper, and used for spark suppression (I'm guessing here). A few days ago the cap started to smolder (when the tool was turned off ) - I guess it finally dried out and shortcircuited. The tool is very good and in working condition, so I am trying to fix it. Needless to say, a google search on three wire capacitor came out blank. It should be possible to replace the cap with three regular capacitors if I could only work out how to connect them. Any clues on how I can repair the tool are most appreciated.

Thanks,
//A

[TonyDuell]

It's almost certainly a 'delta capacitor', the 3 capacitors form a triangle when you draw the circuit diagram.

The '0.2uF' should be a class X2 connected between the brushes. That is, one wire of the capacitor to each brush. The value is not that critical, a 0.22uF will be fine.

The '0.001uF's (1nF) should be class Y2 and are connected between each brush and ground

[andjygen]

Ah, that makes sense.

Thank you so much, I will post the result.

BR,
//A

[Lucien Nunes]

I agree this is a delta capacitor, as @TonyDuell describes, for RF interference suppression. Search terms to find them might also include 'suppressor' and 'filter'. They are available in a range of capacitance values and case styles, including PCB, wire-ended and power input modules integrated with a connector. You can, as you say, substitute three separate capacitors.

Also as above, the capacitors must be Safety Capacitors, because of the connections to the mains supply and to metal parts of the machine via the centre lead. Delta-caps and filter modules will have the appropriate type of capacitor in each position. The class-X and class-Y designations refer to the safety performance and there are subcategories, see:

Safety Capacitor classes

The capacitance values are not usually critical, however where they are used on Class II equipment without a physical connection to earth / ground, the value of the two smaller caps L-E and N-E must not be so high as to cause excessive touch-leakage current.

Wire-ended type:
Delta cap at Amazon

[Craig Sawyers]

Or pass enough current to trip an earth leakage circuit breaker on the consumer unit.

Craig

[Lucien Nunes]

True, although a much less stringent limit. Touch leakage is typically limited to 0.25mA whereas most 30mA RCDs trip between 20-25mA.

With increasing domestic use of small SMPSUs, many RCDs and RCBOs serving general-purpose circuits are being pushed closer to their threshold by aggregate functional capacitive leakage. One or two class-I appliances each adding a few mA leakage can be the straw that breaks the camel's back.

[Roger Ramjet]

Hi Andy,

The piccy's show capacitor that I rescued from an kaput Servis washer in the 80's that contained a brush motor.

I presume the earth connection was achieved via the stud & nut mounting. Not sure if modern washers have a similar set up but clearly scope for rescuing one from a scrap machine......

Rog

[Lucien Nunes]

The centre connection is indeed made via the metal case.

Whilst the values of the capacitors are often un-critical as far as suppression performance is concerned, beware the difference between those suitable for Class-I and Class-II equipment (respectively with or without a connection to earth.) @Roger Ramjet's suppressor is from a Class-I 'white goods' application, so its Y-caps can be of high value because the functional leakage they cause from line to earth at 50Hz is returned via the supply cable without affecting the touch voltage on accessible metal parts. Depending on which standards you are working to, either 3.5 or 5.0mA limit is typical. If @andjygen's machine is of Class-II construction, the absence of an earth connection means that the entire leakage from the suppressor is 'touch leakage' and limited to 0.25mA.

At 230V, 50Hz, Roger's 25nF Y-caps leak 1.8mA, so comfortably within normal leakage range for a Class-I device but excessive for Class-II (noting that touch leakage is actually measured in a slightly different way.) Andjygen's 1nF Y-caps leak only 0.072 mA, therefore suitable for Class-II applications.

[Roger Ramjet]

Originally Posted by Craig Sawyers

Or pass enough current to trip an earth leakage circuit breaker on the consumer unit.

Craig

On a technical note & I freely admit being slightly out of my depth....

My understanding is that a RCD or RCBO activates due to any inbalance between the phase & neutral current feeding the circuit and /or appliance. Therefore if the Delta filter connects both phase & neutral to earth, can there be any actual current inbalance to be detected by a RCD or RCBO ?

Rog

[Station X]

My understanding is that ELCB's are just that and require an earth connection to actually work. An RCD on the other hand detects current imbalance and doesn't require an earth to work. An earth fault will trip it non the less.

[Lucien Nunes]

if the Delta filter connects both phase & neutral to earth, can there be any actual current inbalance to be detected by a RCD or RCBO

The delta capacitor is symmetrical and has the same capacitance line-to-earth as neutral-to-earth. But on all normal electrical supplies, the voltages across these capacitors are different: 230V line-to-earth and 0V neutral-to-earth, because neutral and earth are held equal at source and should be nearly equal at the suppressor. Therefore the line-to-earth capacitor creates leakage 'into' the earth but the neutral-to-earth one does not leak it back 'out' again, which an RCD sees as an imbalance.

Obviously this asymmetry applies to the functional leakage resulting from the 50Hz supply. At RF, which is what the suppressor is there to deal with, the capacitors' low reactances short everything together resulting in a symmetrical behaviour.

On a floating (IT) supply where neither of the two circuit conductors is earthed, the two capacitors form a voltage divider and the middle wire takes up a potential midway between them. This effect is also seen during electrical testing where a protective conductor is faulty and disconnected from earth.

My understanding is that ELCB's are just that and require an earth connection to actually work. An RCD on the other hand detects current imbalance

Yes, although the original type of ELCB is now so rarely encountered that it can be ignored for practical purposes.

Historically, there was a device often just called an 'ELCB,' connected between the Main Earthing Terminal and an earth rod, which detected the current flowing into the rod and was designed to trip if the resulting voltage between the MET and the rod exceeded its threshold. A single ELCB was usually installed in the supply tails feeding the consumer unit, on installations with TT earthing where the external earth resistance of the rod was too high for ADS to work in the event of a line-to-earth fault. I.e. where the rod would not pass enough current to blow a fuse or trip an MCB. These ELCBs were not sensitive enough to offer additional protection against shock, had many problems with both failure to trip and nuisance tripping, hence are no longer used. Occasionally, one finds an old example still in service.

RCDs, operated by current imbalance (residual current) and offering much higher and more stable sensitivity, came later. From the increased sensitivity came the ability to protect system users against shock by detecting the unbalanced shock current itself, vastly expanding their range of applications. Over time, their use has been mandated for an increasing number of circuits and situations.

Re. terminology; as RCDs became widespread, they were seen as another kind of earth leakage circuit breaker despite having quite different construction and function. To avoid ambiguity the old type of ELCB connected to the earth rod became more specifically known as 'Voltage-Operated' hence VOELCB. The distinction highlighted by @Station X is therefore between VOELCBs and RCDs. In contrast, the terms RCCB (Residual Current Circuit Breaker) and RCD (Residual Current Device) are exactly synonymous, although RCD is the correct modern name.

[mark_in_manc]

May I ask a related question on this subject? Where a trip has something written on it which looks like an MCB (say C32, which I would interpret as C-curve, 32A) and also something written on it which looks like an RCD (say I_subscript-something-or-the-other = 30mA, plus a 'test' button) ... can it be both, in one package?

[G6Tanuki]

Yes, many of the things fitted in breaker panels/ consumer units these days combine both overcurrent and earth leakage detection/protection in the one unit. Search for RCBO if you want more info.

Arc fault detection and protection is also becoming more commonly applied in breaker panels/consumer-units too.

somewhat more advanced than the old rewireable ceramic fuse-carriers of times past

[Lucien Nunes]

Indeed, RCBO = combined RCD + MCB. These are increasingly popular for protection of individual final circuits rather than lumping them together on shared RCDs. But we should avoid hijacking this thread about motor suppression.