Chuck Yung 
EASA Technical Support Specialist 

We have a DC motor that arcs when loaded. We checked all the usual suspects: brush neutral, interpole polarity relative to the armature, brush spacing around the commutator, etc. How can we determine the correct interpole circuits? 

I’m excited to be able to share a brand new DC repair tip. A conversa­tion with two EASA members led to a method for determining whether the interpoles are connected with the correct number of circuits. 

Not only is this new method easy, it’s a refinement of the interpole polarity test we routinely perform on every repaired DC machine. To explain why this method works, let’s review some design basics and then use that information to determine the correct interpole circuits.

Back to basics 
The interpoles provide a magnetic flux equal to, but opposing, the armature flux. (See Figure 1.) This minimizes distortion of the field flux, so the brush neutral position remains fixed. Before interpoles were developed, the brushholder position had to be manually shifted each time the load changed. A DC motor with interpoles should not arc within the normal range of load and speed.

Because the interpole and armature circuits are connected in series, the current—which varies with the load— is the same in the interpoles and the armature. The magnetic strength of a DC coil can be calculated in ampere-turns per pole. An interpole with 25 turns, carrying 100 amps, has a field strength of 2,500 ampere-turns. Using 50 turns at 50 amps would result in the same 2,500 ampere-turn field strength. (25 x 100 = 2500 = 50 x 50) 

How did the designer determine the number of turns to use for each interpole? There is a “square of the inverse” relationship between distance and magnetic strength. Because the interpole iron is farther than the armature from the field iron, the interpoles normally require more ampere-turns/ pole than the armature. That gives us a rule-of-thumb that the ampere-turns of the interpole should be approximately 1.2 times the ampere-turns per pole of the armature. And that leads to our first good test to determine the interpole connection: By applying AC voltage to the armature-interpole circuit, we can determine the turn ratio between the armature and interpoles. 

Verify the relative polarity 
One important step after assem­bling every DC machine is to verify the relative polarity of the interpoles and armature. The simplest method for doing so is to apply 20-30v AC to adjacent brushholders, and measure the output voltage between leads A1 and A2 (Figure 2). The measured voltage should be lower than the input voltage. If the voltage across A1-A2 is higher than the input voltage, we should exchange the leads at the brushholders to obtain the correct relative polarity. 

The simple test just described uses the armature-interpole circuit as an autotransformer. Because the polarity of the interpoles must oppose that of the armature, our autotransformer should “step down” the voltage. If the output voltage “steps up” instead, the armature and interpole polarities are the same, so we swap the position of the brushholder leads to reverse the relative polarity. 

Here is where knowledge of that 1.2 ampere-turn ratio is useful: When the polarity relationship between the armature and interpoles is correct, and 20v AC is applied to adjacent brush posts, we should measure approximately 12 – 16 volts AC across A1 and A2. 

If the interpole circuits are incorrect, there will be a corresponding change in the ratio of input-to-output voltage. For example, if the interpoles should be connected series-parallel, but instead are connected in series, the output voltage will be approximately one-quarter to one-half of the expected value, or approximately 3-6 volts. That is a strong indication that the interpole circuits are incorrect. 

The exception 
I mentioned earlier the relationship between distance and magnetic pull. That explains an important exception to the input-output ratio. When a machine has compensating windings (“pole-face bars”), the output voltage will be unusually low. That is because the compensating winding, an extension of the interpoles, is imbedded directly in the face of the field poles. For the same reason, the European design which looks similar to an AC core (stacked laminations, with slots instead of individual poles) also results in very low output voltage. 

What happens if the circuits change? 
The parallel DC circuit is a current divider. Figure 3A illustrates the series circuit of 4 interpoles, and Figure 3B shows the series-parallel connection . For a machine rated 100 amps, and each interpole having 15 turns, a series connection results in 1500 ampere-turns per interpole. The series-parallel circuit results in 750 (50 x 15) ampere-turns, while a 4-parallel circuit yields only 375 (25 x 15) ampere-turns per interpole. With the wrong interpole circuit connection, the interpoles will either be much too strong or much too weak, and the brushes will arc as the load changes. 

It works the other way, too 
If we know the armature and interpole winding data, we can calculate the ampere-turn ratio as follows: 
(Armature total turns x armature current)/(number of poles x plex) = ampere-turns per pole Interpole ampere-turns per pole = turns of 1 interpole x armature current/parallel circuits.

The armature plex is easily overlooked in this comparison. If the armature data is unknown, use a digital low-resistance ohmmeter (DLRO) to determine the commutator pitch. Measure the resistance between bars 1-2, 1-3, 1-4 for a lap winding. The pair with the lowest resistance is the commutator pitch, or lead throw. The most common armature connection (commutator pitch of 1-2) is a lap simplex (1 circuit per pole), while a lap duplex (1-3 commutator pitch) connection has 2 circuits/pole. 

For a wave winding, use the following formula to determine the probable commutator pitch: (Bars +/- plex )/poles Example: An armature has 41 slots, 123 bars, wave simplex 4-pole: (123+/-1) /4 = 31; the commutator pitch must be either 1-62 (retrogressive) or 1-63 (progressive) 122 / 2 pole pairs = 61; commutator pitch of 1-62 or 124 / 2 pole pairs = 62; commutator pitch of 1 63 Measure the resistance between bars 1-60, 1-61, 1­ Apply 20v AC to positive & negative brush posts (INPUT). Measure voltage across A1 and 62, 1-63, 1-64. The lowest A2 (OUTPUT) at the terminal box. resistance pair confirms thecommutator pitch. The plex is the number of circuits, which we can use for the purpose of determining the ampere-turns of the armature. For a simplex (whether the armature winding is lap or wave wound) connection, the ampere-turns per pole equals the total number of armature turns divided by the number of poles. For a duplex connec­tion, divide that number by two, by 3 for a triplex, etc. 

A less reliable method for determining the interpole connection varies with the age of the machine. Designs built prior to approximately 1980 typically had 600­1000 circular mils/amp (CMA) for the interpoles. Later designs are sometimes as low as 300 CMA; hence, the uncertainty in the results when using this method. 

Turn-ratio most reliable 
If both methods indicate the interpole circuits are incorrect, that makes a very strong case. Of the two methods described in this article, the turn-ratio is the most reliable. In two recent cases, I have helped members where a typographical error on the OEM paperwork resulted in incorrect identification of the interpole circuits. Within the past four months, this new test has helped nine members! 

A significant advantage of the turn-ratio method is that it can be incorporated in the routine interpole polarity verification that should be done as part of every DC machine repair. Just standardize the voltage at which the test is done, and have the technician document the measured voltage. 

This test is a simple expansion ofa test you should already perform on every DC machine repaired. The results will not only confirm when the interpole circuit is correct; when incorrect, the results will help you determine the correct number of circuits. 

A word of caution: There is no correlation between the plex of an armature and the number of interpole circuits. For example, a duplex armature (2 circuits) might use interpoles connected in series (1 circuit), series-parallel (2 circuits), or 4-parallel (4 circuits.) In other words, the interpole circuits and armature circuits are arrived at independent of one another, during the design process. The armature plex is influenced by the armature voltage and speed rating, while the interpole circuit decision is influenced by wire size, ease of manufacture, and the armature current rating. So this test can reveal incorrect number of circuits in the interpoles or the armature. If the armature is where the error is, the rpm will be off by a large factor. 

Members helping members 
One of the great things about EASA members is their willingness to help one another. A discussion about a particular DC motor resulted in this new method for determining the number of circuits for a DC machine. I especially want to acknowledge Trevor Meyer of Rexel United (Motor Repair) in Alton, Illinois for his keen powers of observation, which culminated in this article. 

Categories: DC motors, Interpoles

Tags: Interpoles

Rate this article:

3.5

Print