Announcement

Collapse
No announcement yet.

Sweet spot for inexpensive VFD?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #61
    Generally speaking, when capacitors are put in series to increase the Voltage rating, it is a good idea to add a pair of resistors, one across each capacitor, to keep the Voltage division equal. I do not know what the recommended values for these resistors, but it probably is a fraction of the equivalent series resistance of the capacitors, which may not be well controlled in the manufacturing process. By using external resistors that are lower in value, those external resistors will control the Voltage division.



    Originally posted by PStechPaul View Post
    Good application note, but it doesn't go into the capacitance requirement, and I'm not 100% sure of my math. Yes, two 450 volt capacitors in series would probably be less expensive, and probably would not need to be low ESR types if the inverter is already equipped with its own. I found some 450V 10,000 uF capacitors for $100 each. Two of them would probably be fine.

    My Toshiba H3 480V 7.5 HP VFD has two 1800 uF 400 VDC capacitors. It may help to use a PFC circuit ahead of the inverter to minimize the capacitance needed. The energy needed to smooth the ripple would then be stored in an inductor operating as a high frequency step-up converter which will draw more current as the input voltage drops, eliminating the very high current surges that otherwise would occur as the input voltage exceeds the DC on the link.
    Paul A.
    SE Texas

    Make it fit.
    You can't win and there is a penalty for trying!

    Comment


    • #62
      Originally posted by Paul Alciatore View Post
      Generally speaking, when capacitors are put in series to increase the Voltage rating, it is a good idea to add a pair of resistors, one across each capacitor, to keep the Voltage division equal. I do not know what the recommended values for these resistors, but it probably is a fraction of the equivalent series resistance of the capacitors, which may not be well controlled in the manufacturing process. By using external resistors that are lower in value, those external resistors will control the Voltage division.
      It is basically required, not just a good idea.

      The resistor value used is related to the leakage currents. You want the resistors to be as large as possible, but small enough that the worst case leakage (a high value equivalent "resistance" internal and across the capacitor terminals) will not pull the voltage at the center point high enough to exceed the voltage rating under any conditions.

      The two leakage resistances form a voltage divider, and the two external resistors form another. The external resistors should hold the center voltage to a safe level.
      1601

      Keep eye on ball.
      Hashim Khan

      Comment


      • #63
        Originally posted by PStechPaul View Post
        Something else to consider would be using a joystick to control a motor's speed and direction. It might be ideal for tapping.

        Pretty cool, but that looks like a lot of work! I've found a simple SPDT, Mom ON-OFF-Mom-ON toggle switch with the mushroom OFF works fine for tapping on my drill press with the correct speed set via VFD, belts or both. Best part of the mushroom OFF is that I have it right on the front of the tool where I can hit it with my forehead if things are going sideways, literally. The best hack would be a rocker foot pedal so you have both hands free.
        Location: Jersey City NJ USA

        Comment


        • #64
          IIRC, the reason for the external circuitry was to supply +/-5 or +/-10 volts to the control input, using the potentiometer of the joystick, which returns to center when released. This was using my Fuji AF-300 E11 VFD.It looks like it is possible to set a "bias frequency" so that a 0-10 VDC signal will give an output from reverse to forward, with midpoint 5V giving 0 RPM. Releasing the joystick quickly stops the motor, but perhaps not as quickly as a stop command. I bought this joystick many years ago for just a few dollars from a Radio Shack clearance, but similar units now seem to be rather expensive. But you can get tiny thumb operated joysticks, as used for R/C planes and cars, for just a couple dollars.

          I tried to copy and paste an image from Banggood, as I have done many times, but it produces something like this (looks like base64):

          Code:
          [IMG]OI3XFTwW5cPVp9+Uj5Yfz3yw erTT8oH669HPlh9+kn5YP31yAerTz79RuSD1SeffiPyweqTT78 R+WD1yaffiHyw+uTTb0Q+WH3y6TciH6w++fQbkQ9Wn3z6f+3TA QkAAACAoP+v2xHoBydkhQlZYUJWmJAVJmSFCVlhQlaYCBREbC/svMrlAAAAAElFTkSuQmCC[/IMG]


          Click image for larger version

Name:	Screenshot_2020-01-18_joystick_Banggood.png
Views:	56
Size:	55.9 KB
ID:	1849717
          It looks like fancier joysticks only have switches, and there are some wireless versions that may use accelerometers or movement sensors with pulse trrain outputs. Ones similar to mine, with 15 pin D-sub connectors, are now "vintage":
          Click image for larger version

Name:	Screenshot_2020-01-18_Vintage_Gravis_Analog_Joystick_Controller_PC_Flightstick_15_pins_eBay.png
Views:	48
Size:	359.0 KB
ID:	1849718
          http://pauleschoen.com/pix/PM08_P76_P54.png
          Paul , P S Technology, Inc. and MrTibbs
          USA Maryland 21030

          Comment


          • #65
            Originally posted by martik View Post
            I've used this one for years on a 1HP mill and lathe, absolutely no issues:

            https://www.ebay.com/itm/220V-2HP-1-...bg5rV&LH_BIN=1

            The 3HP version is $6 more
            I bought the same thing about 2 months ago and have had no problems with it, was 57$ including shipping.

            Comment


            • #66
              Originally posted by johansen View Post

              I bought the same thing about 2 months ago and have had no problems with it, was 57$ including shipping.
              As much as I love a bargain I wasn't that brave. I ordered the $130 Lapond from Amazon. If I couldn't actually read the manual, I couldn't be sure it would work in the config I wanted.
              Location: Jersey City NJ USA

              Comment


              • #67
                Saturday afternoon fun, Teco L510 my goal was to get everything mounted ,wired and test ran. However soon after starting I found out I was out of #8 ring terminals, 12ga control wire, oh and two sizes of shrink tube
                I just need one more tool,just one!

                Comment


                • #68
                  Originally posted by lakeside53 View Post
                  That is true, but Hitachi guidelines (their devices) imply single phase is only 2x the capacitance of 3 phase. i.e. there array above from a 15hp vfd is good for a 7.5 hp vfd on single phase. Someone with EE 101 can calc the ripple voltage. 1/2CV^2 etc. Assume 17.5 amps each phase out, 60hz. (or Paul can put it into his fancy spice program).

                  .....
                  They may use considerably more than the minimum C for the 3 phase case.

                  I designed a 3 HP "semi-MIL" VFD (under the "COTS" program) which used only a 50 uF high current film cap for the filter. We were eliminating all electrolytics throughout the unit for reliability and temperature performance to the -50C requirement.

                  It worked great for the 3 phase input, but as you can guess, had nil for output with single phase input. Single phase was an "after the fact add-on requirement" for one foreign military customer of our customer. We had to add an external capacitor bank for that specific usage, and design it into their product. I do not recall the total capacitance needed, but it was many times the original film cap. That customer did not require -50C operation.
                  1601

                  Keep eye on ball.
                  Hashim Khan

                  Comment


                  • #69
                    In addition to making their (Hitachi) devices "single phase ready" (derated), they may want to keep the ripple current through the main caps down to some acceptable limit.

                    The ABB's I looked at were quite different, likely putting in minimum of capacitance for three phase. Their 7.5 hp on three phase becomes 1.5 on single.
                    Last edited by lakeside53; 01-19-2020, 01:57 PM.

                    Comment


                    • #70
                      You definitely need a low ESR high frequency (10 kHz) rated bus link capacitor to handle the PWM current surges, but there can be other capacitors in parallel to handle the 100-120 Hz ripple from a single phase mains supply,and they can have higher ESR and not be rated for high frequency. It might be a good idea to add an inductor between the low ESR link capacitors and the power line electrolytics to avoid overheating them.
                      http://pauleschoen.com/pix/PM08_P76_P54.png
                      Paul , P S Technology, Inc. and MrTibbs
                      USA Maryland 21030

                      Comment


                      • #71
                        Don't buy some offbrand Chinese ****. Buy a teco, a hitachi, an AC tech, a lenze. literally anything that has a real CE marking (not the China Export one) and a UL listing.

                        The thing the Chinese one will skimp on is capacitor make and capacity (they use fake ones or shoddy substitutions relative to the design they knocked off), heatsinking, and FETs. All of which will cause premature release of the magic smoke, which is also shoddy...
                        Last edited by psomero; 01-19-2020, 08:42 AM.
                        -paul

                        Comment


                        • #72
                          Originally posted by lakeside53 View Post
                          In addition to making their devices "single phase ready" (derated), they may want to keep the ripple current through the main caps down to some acceptable limit.
                          Naturally. Use of several in parallel adds up the ripple currents (if ESR is consistent, which it generally is not). They may be ripple current limited, as opposed to diode current limited. (Diodes are cheap in comparison).

                          Originally posted by psomero View Post
                          Don't buy some offbrand Chinese ****. Buy a teco, a hitachi, an AC tech, a lenze. literally anything that has a real CE marking (not the China Export one) and a UL listing.

                          The thing the Chinese one will skimp on is capacitor make and capacity (they use fake ones or shoddy substitutions relative to the design they knocked off), heatsinking, and FETs. All of which will cause premature release of the magic smoke, which is also shoddy...
                          There are some chinese companies making VERY good parts indeed. You may never see them, because they are sold only on the home market, and the spec sheets may not even be published in english etc. A company I know of uses some that have an excellent ripple current rating (quite real), and very good performance in general. When I dug around I did find specs in english for their parts.
                          Last edited by J Tiers; 01-19-2020, 12:45 PM.
                          1601

                          Keep eye on ball.
                          Hashim Khan

                          Comment


                          • #73
                            Yes, don't knock "Chinese" without looking at the product in question. We tend to see them as making the low end garbage that any county "can" make. Check your phone, laptop or your Hitachi VFD (and may others others) for county of origin Same with INDIA - they make garbage and some really good products.

                            Interestingly.... In Seattle are not allowed to use a device that does not have an NRTL listing, and CE is NOT acceptable. I run into this daily.

                            We all have to remember, if they (or any company/county makes crap and we keep buying it, they will keep making it and we will keep-on complaining.
                            Last edited by lakeside53; 01-19-2020, 05:01 PM.

                            Comment


                            • #74
                              Yep, feedback is important, if a product is crap or needs improving the vendor needs to know so they can relay that information to the mfg. That is how it has always worked in the past.

                              In China it also makes a difference who owns the factory. If it's a private concern then odds are the product will be the best they can produce because they are subject to market forces. If it's a government or PLA controlled factory then not so much.
                              I just need one more tool,just one!

                              Comment

                              Working...
                              X