DURApulse GS20(X) VFD: Overvoltage Protection from AutomationDirect

So, you are trying to decelerate your motor rapidly and you get the dreaded overvoltage fault. Now what do you do? Before we get into that, let’s step back and make sure we understand what’s actually happening inside the drive. In the simplest terms, a drive just takes an A/C input, rectifies it to a DC voltage, and then uses some large transistors called IGBTs to pulse width modulate – or chop - the DC voltage and send it to the motor on three wires each with different phasing of the pulse width modulation. But when the chopped waveforms are used to slow a motor down, the inertial energy built up in the motor has to go someplace. So, the motor becomes a generator and it pushes current created by the rotating magnetic fields back into the drive. That additional current adds voltage to the internal DC bus which makes the bus voltage increase. The problem is, the semiconductors inside the drive can only handle so much voltage and if that voltage gets too high, well they burn up and you end up with an expensive paper weight. The good news is, the GS20 drives have several ways to deal with that. In the extreme case where the deceleration rate is raising the DC bus too fast for drive to manage, the drive says, “I can’t handle that!” and it shuts down it’s outputs, displays the overvoltage fault and leaves the motor to spin down by itself. Which in my little test rack here takes a couple minutes. We can actually see that using the free GSOFT2 scope utility. You can download GSOFT2 for free from the AutomationDirect.com website. I’m assuming you know how to use this, if not then check out the GSOFT2 quick start video. This is the hardware I am using. I have the free GSOFT2 utility running on a PC and a USB cable connecting that to the drive and the drive is connected to this motor. I love that the GS20s have a built-in USB port for programming and diagnostics and I don’t have to pay extra to get that. The key thing in all of this is this large 10-pound cast iron pulley I’m using to create some inertia in the motor for the drive to overcome. Once you get that massive guy spinning it takes some effort to get it to slow down. Exactly what we want. I reset the drive to factory default, so you know exactly where I am starting from and then I set up the drive with my motor parameters and these ramp rates. I also enabled the frequency potentiometer on the keypad and I’m sending the drive's internal overvoltage signal to output 2. Notice that the deceleration time is set at 3 seconds. That’s not really reasonable for this large pulley’s mass but will make for a good first test for us. I’m monitoring the DC bus with a red trace. The desired, or commanded frequency, with green, the output frequency to the motor with blue and the overvoltage signal with yellow. I’ll reset the overvoltage fault and hit run to do the exact same test, so we can see it on the scope. I'll wait for it to ramp up to speed and for the DC bus to settle out. You can see the DC bus voltage drooped while the motor was accelerating, but once it got to speed the DC bus leveled off because the motor doesn’t need as much energy to just maintain the motors speed. I’ll hit the drive's stop button and let it run for a second. Let’s stop the scope and zoom in on the results. When we hit stop, you can see the DC bus which was at around 335 volts rose to over 400 volts. The drive saw that too and gave up – it said I can’t handle that and dropped the output to zero and says, “good luck!” you are on your own. You can also see when the drive hit the over voltage condition on digital output 2. To get rid of that fault you can simply back off on the deceleration time to give the drive a chance to manage the extra energy. Let’s double our deceleration time to 6 seconds. And I’m going to change output 2 to monitor something else – I’ll tell you what it is in a minute. I’ll start the scope and hit run. Let it ramp up, wait for the DC bus to settle out and hit stop. Hmm, this is taking a lot longer than the 6 second deceleration time we asked for isn’t it? We’re not getting the overvoltage fault – that’s good - and it IS giving us a deceleration ramp, but it looks kind of bumpy. What’s going on? Let me stop the scope and zoom in. Ok, here’s the deal. What you are seeing is the drive automatically slowing the ramp down to keep the DC bus from rising too far. We asked it to ramp down over 6 seconds. And it started to do that. But when it saw the DC bus get too high – let’s zoom in on that – looks like around 370 volts – the drive automatically held off on the deceleration until the DC bus got back down to a more tolerable level. And then it re-engaged deceleration until the DC bus got too high again, and just kept repeating that over and over and over again until things settled out. The good news is we didn’t get a fault. The bad news is it took almost 20 seconds to ramp down. You can change the voltage level this trips at in Parameter 7.00. And look at that! The default is 370 volts – exactly what we saw. I wouldn’t mess with that unless you really know what you are doing. Just know that it’s there is you want it. And look at this. Output 2 – which I changed a minute ago - is now showing us every time the protection kicked in. Cool. OK, that’s the drive’s default protection method, which it calls the traditional over voltage stall prevention and you set it in Parameter 6.02. If we change that to a 1, then we get what the manual calls “smart” over-voltage protection. Let’s see what that looks like. I’ll start the scope, hit run, wait for it to ramp up and the DC bus to settle out and hit stop. Interesting. It looks like it tries to manage the deceleration – and it does hold the fault off a little longer, but the DC bus still got too high and the drive eventually faulted out. I did some experiments and I had to change the deceleration time all the way out to 21 seconds before the fault went away. That looked like this. You can see the fault protection is engaged in here and that the drive is manipulating the deceleration to accommodate the changes in DC Bus voltage. So the traditional method decelerated in around 18 seconds and this method decelerated in about 21 seconds but that was for this system. Your system might be different, so try both. Could I have turned this automatic stall prevention off and just set the deceleration myself? Sure. You do that by setting Parameter 6.01 to a zero. I did that and I found I had to set the deceleration to 22 seconds to prevent an overvoltage fault. But, if my load changes I’ll have to go back and change that manually myself. The automatic stall prevention methods adapt the deceleration to whatever load you have so you don’t have to mess with it and waste time trying to figure out what the right deceleration is. So it’s usually a good idea to leave automatic stall prevention enabled. Of course, if you really need short acceleration times, then check out the video on dynamic braking. It’s simple to use and very effective at getting you fast deceleration times without over voltage faults. If you do use dynamic braking be sure to turn off these stall prevention methods by setting parameter 6.01 to a zero. That way they won’t interfere with your dynamic braking. Click here to see all of the videos in this series. Click here to subscribe to our YouTube channel so you’ll be notified when we publish new videos and click here to learn about AutomationDirect’s free award-winning support options.