Sparking seen on commutator of DC motor...






...Read further to know more about the reasons


Sparks are often seen on the commutator, below the carbon brushes. There are many reasons for this. Minute observations can reveal the casue.


One needs to check if.....


1.     Sparking occurs only when the load and speed are increased.

2.     Sparks are yellowish in color and do not have blue tinge.

3.     Sparking is intermittent and occurence is unpredictable.


Out of many causes, the following are a few.



a) Inter turn short in the armature winding. Current waveform in such cases is seen as under.

b) Defective interpole winding can also cause sparking. If the position of the brush holder is disturbed the following steps can be used to adjust it to normal position, assuming that the interpoles are ok.

-         Disconnect the motor totally.

-         Remove armature connections.

-         Apply 230 volts AC voltage to the field winding.

-         Measure the induced voltage in the armature on the carbon brushses. ( across the terminals where DC leads from the drive were connected )

-         Adjust ( rock the rocker slowly forward and backward ) the rocker till the voltage induced is lowest.

-         Tighten the rocker position at this point.


c) Sparks yellow in color indicate that these are  caused due to uneven surface of commutator. Particles of carbon from the brushes are flying off due to mechanical impact on the brush tips. Making the commutator smooth on a lathe machine may solve this problem. For motors with higher ratings, one will find that the carbon brushes are split in two slabs and are sandwitched together. Each slab has independent copper pig tails which are made parallel on the brush holder. This construction has following advantages.

    1. The air gap between the two slabs gives additional cooling area for the heat to dissipate, keeping brush cooler.

    2. Uneven commutator segments do not make shorting of segments impossible. With two slabs moving up and down

        independently, allow shorting of segments possible. For smooth commutation of current it is necessary that the neighbouring

        segments are shorted to each other during commutation.

    3. Split brushes become light weight. When brushes are jumping, lower weigtht helps in easy up and down movement.      

d)      Worn out brushes…

Worn out brushes also cause sparking on the commutator. There is an engraved line on each carbon brush which indicates the limit upto which the brush could be used.

e)     Low pressure on the carbon brushes….

Inadequate spring tension can also cause brushes to jump easily which reflects in sparking. Pulling the lead of a carbon brush by hand up to one or two centimeters and releasing it could give an ideas if spring has lost its tension. Comparative checking with other brushes can point out to a defective spring. Dust and accumulation of other oily or sticky material around the brush can restrict easy movement of the brushes inside the barss holders. Cleaning with a piece of cloth can help solving this problem.

Some brushes have additional leads of wire embeded in the brush. These are thinner wires as compared to the main current carrying braided wires. These wires are connected to a relay of PLC in the control panel. Wearing out of brushes beyond usable limit makes these lead touch each other. This is detected by a relay of PLC input to give an Alarm.















Here is a photograph of current waveform with all 6 thyristor firing. However, all the current peaks are not of same amplitude. This means that the motor is taking high current for some time and a lower current for some time. As explained above, uneven mechanical friction can NOT cause this. Because, as you can see, each current peak lasts for hardly 3 milliseconds. Mechanical non uniformity in friction or load can not change that fast. The reason for this uneven current peaks is always from electrical side.



There are two possibilities.


a)     First possibility is that the firing of all 6 thyristors itself is uneven or unbalanced. This can be verified by bypassing Ramp, speed amplifier, current amplifier and connecting the drive in OPEN loop mode. i.e. by giving a dc signal of firing thyristors directly from a potentiometer to the firing circuit. ( See the explaination in the link  "Fuses blow"  ). The motor must be operated very carefully in this mode as there are no protections of current limit etc. A small mistake or rough handling of the potentiometer can cause a big jerk or shock to the mechanical parts. If the current waveform does not change its shape ( uneven peaks ) then it can be concluded that the Phase-Shift circuit components in the Firing Circuit are not matched properly. In analog Drive, one can check values of all 6 resistors and 6 capacitors which give required phase shift for firing pulses. The resistors are usually of 1 % tolerence levels. All must be precisely same. If the current waveform becomes smooth with all 6 peaks of current of same amplitude, then it can be concluded that the cause for uneven peaks lies in the firing signal coming from the output of current amplifier


b)    Output of the current amplifier, when seen on oscilloscope, will show wavy dc signal. Incorrect adjsutment of Gain of the amplifiers( both speed and current ) causes this signal to be wavy. Reduce the gain to make the DC level more like a straight line. If the gain adjustment does not solve the problem, then another source which is likely is bad tachogenerator. Missing dc voltage on some commutator segments of the tachogenerator can cause the output of the current amplifier to becomve wavy.



For more on DC motors, and associated testing etc.Click here...





Download PDF version of the pages here...





Block Diagram



Ramp Circuit


Speed Amplifier


Current Amplifier


Firing Circuit


Power Circuit


Types of DC drives




Checking Thyristor


Voltage Feebback Vs Tacho Feedback


Current Limiting


Constant HP & Constant Torque


Field Weakening


Using oscilloscope




Motor runs at Full speed


Fuses blow


Hunting in speed


Motor gives jerks


Belts vibrate


Speed control not satisfactory




Speed drops on load


Motor overheats


Sparking on commutator


DC drives and Power Factor


Your comments and suggestions please



Drive stops all off a sudden..



 Motor not able to drive the load