# 15% drop in motor RPM when using PWM controller

I'm new to the forum and a total noob to electrical engineering.

I have a 120W, 24V DC gear motor. Runs at ~200 RPM no load connected straight to a 24V power supply.

However, when I connect it to a PWM controller, the RPMs max out at ~170. The PWM controller has an input range from 6V-90V DC, rated for 15A, PWM frequency of 15 kHz, duty cycle is supposed to be adjustable from 0-100%, uses a 10k potentiometer.

After doing a little research, I thought that maybe I needed IR compensation. But, it seems that IR compensation is more for keeping RPM stable with and without load. My drop is happening without load in both cases. So, is there a loss of power associated with using a PWM controller?

From what I've read, PWM should yield pretty close to full power. And, since my controller is adjustable from 0-100% duty cycle, 100% duty would be full power or the whole 24V right? Any insight as to why the motor is not performing at full potential would be greatly appreciated.

Update: Using a Fluke meter I took the following readings, as requested in this answer below:

With meter set to AC, measuring across the motor and varying the pot. The reading bounced around a lot when I moved the pot. With that said, once things settled down I got 0 at min, 0 at maximum and 2 at middle.

With meter set to DC. DC input = 24.5. Motor voltage = 21.2. +DC input to Motor+ = 0.6, -DC input to motor- = 2.7.

• Have you looked at the output of the PWM controller to see what the waveform looks like at 100%? – JYelton Feb 13 '19 at 22:52
• This is of course completely unanswerable without implementation details of the controller. As such it is an off-topic usage question rather than an engineering design question. – Chris Stratton Feb 14 '19 at 0:28
• The switching elements in the controller will, of course,drop some voltage. Maybe that is all that is going on. Hard to say. You can add an RC filter to the output of the controller and use that to measure effective DC voltage with you Voltmeter. Of course an oscilloscope would be better, but since you don't have one, we will make do. – mkeith Feb 14 '19 at 6:45
• In theory, you are right. 100% duty cycle with 24V should produce the same result as a direct 24V connection (with no controller). But in reality, the controller will absorb some power, causing a voltage drop, and therefore a slightly lower speed than direct connection. BUT, you are seeing a drop of 30rpm out of 200rpm, which works out to 15%, which seems excessive. – mkeith Feb 14 '19 at 6:49
• "-DC input to motor- = 2.7." - that's not good. What voltage do you get when measuring between input and output terminals on the controller itself? – Bruce Abbott Feb 14 '19 at 8:04

You mention that you don't have an oscilloscope. That's okay - you can do some simple measurements with your DMM.

Start by setting the DMM to the AC voltage range. First, connect the leads to the 24 Vdc supply that feeds the PWM controller. You should see very low voltage reading. This confirms that your DMM has AC coupling on its input for AC voltage measurements.

Don't laugh - some really cheap (worse than inexpensive!) meters do NOT have AC coupling for AC measurements!

Next, connect the DMM across the motor, still set for AC voltage. Vary the pot. You should see the AC voltage change. AC voltage will at a minimum at the extreme ends of the pot and maximum near the center of the pot. This test confirms that your DMM can measure voltage at the PWM frequency that your controller is using.

Now set the pot for maximum motor speed and note the AC voltage, if any. Do this again for the minimum pot setting and middle pot setting. You should have three AC voltage readings.

Now set the DMM to measure DC voltage. Set the pot for maximum speed and measure the four possible voltages: DC input, motor voltage, voltage from (+DC input) to (Motor +), voltage from (-DC input) to (Motor -). Modify your question with the results of those seven voltage measurements (three AC, four DC) and we will help you interpret those readings.

[Edit in response to information added to original question]

I'm going to suggest that you need a better PWM speed controller. I base this suggestion on the following:

1) The 0.6 Vdc drop from (DC+) to (Motor+) suggests either a lot of PCB trace resistance or a series Schottky diode with a lot of PCB trace resistance.

Although this voltage is much higher than I like to see, it's only an indication of problems.

2) The 2.7 Vdc drop from (DC-) to (Motor-) is much more concerning. We don't have any idea of the internals of your controller but I see (probably) a lot of PCB trace resistance plus a MOSFET that isn't anywhere near beefy enough: the RDSon resistance is far too high.

My suggestion is to cut your losses now and either build or purchase a much beefier DC PWM controller.

[Edit 2]

This site is not the place to ask for product recommendations. The reason for this is that information often goes out of date quickly whereas this site strives to accumulate information that is timeless.

That said: I will suggest that you can go to eBay or some other supply site and enter the following string into their search engine: "dc motor speed control". eBay has many offerings but the one that I would go for is good for 10-50V at 40 Amps.

The search string that I offered should meet this site's requirement for information that is timeless and doesn't expire.

• I took the readings you asked for. Hopefully I did it correctly. I modified my question to include the results. – 1056 Feb 14 '19 at 6:15

If it is an H-bridge (bidirectional), you lose two times the voltage drop on the semiconductor switches which could easily be around 2 V, very much dependent on load current and temperature. That is already close to 10% reduction in Uout and, therefore, in RPM.

• I'm pretty sure it's not bidirectional. Unless, of course, I manually swap the poles on the motor. – 1056 Feb 13 '19 at 23:21

You might have an old or poorly made potentiometer that isn't properly varying across the range. You can rule that out if you check that the middle terminal of the potentiometer goes all the way from the voltage of one side to the voltage of the other as you turn the knob.

• Hadn't thought of that, I will check it out. Thank You. – 1056 Feb 13 '19 at 23:30

If the input to the controller is 6V-90V DC and the maximum output is 24V DC, there must be a boost-buck conversion function in the controller. That may impose a constraint on the maximum output voltage. What does 100% duty cycle really mean? You need to determine what is advertised as the maximum effective output voltage with a 24V DC input voltage. The product either provides what is specified or it doesn't. If it doesn't, it is not a good product.

If the product has no voltage boost function, the maximum output voltage would be limited by voltage drops in the PWM circuit and also by the maximum pulse width. The output could transition to "always on," but that would mean that just less than the maximum voltage would be a step change from always on to the voltage at the minimum "off time" between pulses. As a result, the maximum output voltage is limited to the input voltage minus internal voltage drops minus the voltage lost due to the minimum "off" time. The product specification should make clear what effect these factors have on the maximum output voltage.

• I see what you're saying. The way I understood it is that the output voltage = the input voltage. It is a DC only controller. I never said that the maximum output was 24v DC. I only stated what the acceptable input voltage was. However, it is very possible that it a poorly made product. – 1056 Feb 13 '19 at 23:52
• The problem may be partly trying to make the product cover an input voltage range that is too wide, partly not providing a very good performance specification and possibly not the best design. See revision to my answer. – Charles Cowie Feb 14 '19 at 0:32

Is this a bidirectional PWM controller? If so, the high-side gate drivers might use bootstrap capacitors (more common than not) that require periodic recharge, thereby limiting 100% duty cycle. You can google it more if you want to understand it further.

15% seems pretty large though. It's hard to tell what's really going on if you can't put an oscilloscope to the outputs of the PWM controller.

• I don't' think it's bidirectional. The controller has 2 inputs for voltage in(+/-), 2 outputs for voltage to motor (+/-), and the external potentiometer. No switch for forward and reverse. And, unfortunately, I have no oscilloscope. – 1056 Feb 13 '19 at 23:20