Comments by "" (@JakeRoeder) on "The Engineering Mindset" channel.

Many of our drives (I'm from Danfoss) include some kind of rough power meter so you can simply read it off of the display. No calculations necessary :)
2
Generally, yes. The motors and inverters are a bit different in design due to the torque requirements (same in modern electric vehicles) but the same principle applies.
2
Filtering before the motors isn't always necessary. It very much depends on the application and a number of other factors. As for EMC filtering, it's a great topic that probably deserves its own explainer. EMC isn't needed in many cases but we (Danfoss) put a lot of time and effort into offering integrated EMC solutions for long motor lengths and Residential/Commercial environments.
1
Yep. Very quick pulses, but that's all it takes to "fake" an AC signal from a motor's perspective.
1
So what we do is utilize chokes, most typically on both sides of the DC bus to help mitigate harmonics. Your average drive will be somewhere around 45% THDi. You can further reduce harmonics through several methods, generally externally, with active or passive solutions or by sort of "stacking" up diodes on the front end (you may have heard of the terms 12-pulse or 18-pulse). We just actually published a video on Harmonics here: https://www.youtube.com/watch?v=1eY_lidO0K4
1
@SamiKhan_15 Depends on which direction you want to run. If you were developing your own PWM controller, you don't really need to spend time to think about it. Many controller manufacturers have some kind of PWM controller reference design with source code.
1
Depends on your definition of efficiency. Your typical drive loses about 1% to do the AC/DC conversion then 1.5-2% on the DC/AC conversion. Motors, on the other hand, lose anywhere from 15-30% (rough window, may not be fully accurate) when operating. Adding a drive allows for a lot of improvements.
1
Reactors (used on the input side or DC bus) are typically used to reduce harmonics. It might not be such an issue with smaller drives or in places with a lot of other switching loads, but for larger drives or more isolated applications, it's very helpful. Reactors on the output side are typically used to reduce the amount of dU/dt (dV/dt) caused by the switching of the IGBTs. Excessive dU/dt, especially over long distances between the drive and motor, will help protect your motor's bearings.
1
You can find VFD dev kits/reference designs from companies like Texas Instruments, Infineon and Xilinx. You don't need to do a lot to make a basic one these days.
1
Great to hear! Pumps and fans are some of the best applications for VFDs where money can be saved. A simple reduction in speed of 10% is a bit more than 25% reduction in power consumption. Another thing you can think about with those is that not only are you not blowing out pipes, but you're increasing the lifetime of your pump impeller by likely limiting cavitation.
1
This is a simplified view, actually...haha.
1
That part's a secret ;)
1
There are variable speed air conditioning units. The drives are most commonly used on the compressor of the A/C unit, but in large buildings, they're also used on the fans to circulate the air. I'd say it's not so cut and dry to modify an A/C unit that you already own, but it could be done.
1
I tend to see color codes changing a bit depending on where you are in the world. Regardless, checking motor direction before attaching a load is fantastic advice. In many of our drives (I'm from Danfoss), you can simply change the direction of the motor switching pattern in a parameter so you don't have to rewire everything as a backup :)
1
True for induction motors. A bit different with permanent magnet motors.
1
The PWM signal ends up being smooth in current, but a bit rougher in voltage. It's smooth enough because the switching frequency is fast enough for motors to spin, but it does come at a cost which is voltage spikes on the output. If this becomes an issue, adding an output choke, dU/dt (dV/dt) filter or Sine filter on the output will take care of that to increasing degrees. We (Danfoss) offer those filters along with our drives if this is an issue.
1
They are. Sort of. Electric vehicles primarily use only the inverter side (the part that does the switching) as their power source is already DC. Because of this, they don't use off the shelf ones like what is shown at the top of the video. The concept is the same, though.
1
Full disclosure: I work for Danfoss. That out of the way, we have single phase input drives primarily for HVAC and Water applications. Most single-phase drives are single-phase in and three-phase out. This is mainly because three-phase motors are more prevalent and it takes less overall current to get the job done. In many cases, you can also run a three-phase drive with a single-phase input as well, but part of the three-phase waveform reduces the amount of capacitance needed on the DC bus, so you really have to derate the current quite significantly (50-80%) which we don't really recommend. Hope this helped :)
1
Power factor correction is an often overlooked source of savings as well. In residential/commercial applications, average power consumption is billed and, with newer meters, you'll likely get charged for reactive power consumption. In industrial applications where we apply drives (I'm from Danfoss), these companies often get charged based on peak demand, meaning their bills reflect the most energy they use at a given time. There, the drives can significantly reduce the peak demand. There are many other places where savings comes from, but I'll leave it here for now.
1
@kevinn2759 Actually, this might not be necessary. It depends on whether you want to reverse direction while under operation, in which case you just tell the drive to run backwards, or if you want to change the direction that the drive considers to be "forward," which is something you can do in some drives by simply changing a parameter.
1
Technically, yes. It's quite expensive and the engineering is a bit different, but Yaskawa (I'm from Danfoss) has been selling their Matrix converter for some time now. It's a lot easier to do with the transition and the losses are only around 2-3% on average, so I'm not quite sure what the benefit would be.
1
For solar sources (or any DC source in general), we just basically ignore the diode bridge. It's not doing much anyway. But the IGBTs work the same way without issue.
1
Well done, Paul. Basically the same story I used to train people on for years :) I'm with the Drives group of Danfoss so I've taken the liberty of going through the comments and answering the questions I could since I've been doing this for about 10 years now. If you don't mind, I'd also like to plug the Danfoss Drives blog: www.focusondrives.com and point you guys to our "Facts worth knowing about Drives" booklet: https://danfoss.ipapercms.dk/Drives/DD/Global/SalesPromotion/FWK/fwk-2019/#/ Feel free to add these to your already massive list of links if you'd like :) Thanks again for the wonderful video.
1
@australiangamer7956 You can technically do it without the transition, but it's a bit trickier and more expensive.
1
There's on average 1% lost in changing AC to DC and, typically 1.5-2% lost in switching back to AC. When you look at the other losses in the system, the motor efficiency is the biggest factor. Then the mechanical losses of connecting the motor to the load. Where a lot of the savings comes in is by running the system more effectively. Reducing speeds, especially in pumps and fans gives a huge savings. There's also savings by reducing the current required to start up the motor (induction motors can draw as much as 10x their name plate current when starting up across the line). There's a ton of other benefits, but a bit too much to list. Danfoss actually released the first commercially available VFD back in 1968 and, as I understand it (I work for Danfoss), it was done because there was a meeting on electric motors and people said it couldn't be done. We proved them wrong in around a years time.
1