Applicability of VFD to a small (1/4 or 1/2 hp) pump

Discussion in 'General Electronics Chat' started by kingdano, Aug 2, 2010.

  1. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    I have been researching use of VFD (variable frequency drive) for a project at work.

    We need to regulate pressure in a mixed fluid/air system which will have essentially continually varying amounts of fluid and air.

    Due to this, we will need to quickly and precisely adjust the current to the pump producing the pressure (approx 4.7 bar) to keep the pressure constant with the changing volume (load) to the pump.

    I have done a lot of reading on wikipedia and the web in general after realizing the pump was just a motor with a load on it - i came across VFD and read the following:

    Variable-frequency drives are widely used. In ventilation systems for large buildings, variable-frequency motors on fans save energy by allowing the volume of air moved to match the system demand.

    This statement seems to match the principle which we are trying to design to.

    I have also read that typically the VFD control is for 3-phase induction motors. "Some types of single phase motors" can be used (Wiki) but I dont know which type, or where to figure this out.

    I also feel like VFD is probably generally used for large motors/pumps - so far the mechanical engineer has said the pump will likely be 1/4 or 1/2 hp.

    Can this type of control be applied to a single phase and small AC motor?

    If not, is there an alternative method for controlling single phase AC motors which offers the same type of performance (referring to the HVAC statement here)

    Is there any comparable control algorithm for PWM controlled DC motors? Or is the PWM responsive enough to have the same type of effect?

    Any thoughts?


    Dan
     
  2. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    There are many single phase input, 3 phase output small and cheap VFD's. Some even have PID controls built into them. There are also many 90V/180V DC motor drives out there too although they're a bit more expensive.
    Google works good for this.
     
  3. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    Well, i should have been more clear in what i was looking for.

    I have found many manufacturers and models of VFDs - i guess i was hoping for some experience in using them for small motors and pumps since they seem to be geared towards high voltage, high HP applications in the readings i do (HVAC seems to be the major one)


    Hitachi and Toshiba are in the mix, so clearly its a big industry.
     
  4. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    They work fine on small motors, AC or DC. Your problem will not be the VFD, it will be controlling it to maintain that constant pressure in your process. The VFD will do whatever you tell it within reason.
     
  5. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    what do you mean by your statement?

    the hard part will be controlling it...

    if i buy an off-the-shelf unit which is sized properly for output current torque etc wont the unit essentially do the controls on its own?

    all i need to supply is a pump, pressure sensor and power right?

    i am working through some articles from ec&m magazine about VFDs which have been very helpful in understanding the various topologies etc - they talk about sizing based on current and torque, not HP, which is very useful.

    were you thinking i was doing the control algorithm at this point?

    all i attempting to show is feasibility/proof of concept - is it possible that an off the shelf unit wont provide the necessary algorithm or "hooks" into the inner guts of the controls to function as i may need it to?

    please expand if possible...


    thanks,

    dan
     
  6. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    How are you going to connect the pressure transducer to the VFD? You can't just wire the two together, you'll need a PID (proportional, integral, derivative) controller in between them to maintain your pump pressure at your desired set point with no, or at least very little deviation from changes in load. Some VFD's have built in PID controllers which might make things a little easier.
    Look at TECO/Westinghouse among many others for small vfd's. I haven't been in the market for a PID controller in 25 years so you'll want to Google that. Foxboro used to make good stuff.
    There's a reason you don't want to use an automatic control valve?
     
  7. jpanhalt

    AAC Fanatic!

    Jan 18, 2008
    5,696
    904
    Hi Dan,

    The answer to that question is usually "no."

    AC/DC (universal ) motors can be controlled, usually with PWM. For VFD, consider a fractional HP 3-phase. They are made and are becoming quite common because they are brushless, easily controlled, and provide high torque at low rpm when using a VFD.

    John
     
  8. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234
    I have used VFD's in the past, and I also tried to find one that could control a single phase AC motor, but with no luck. I have to go with a 3 Phase motor, and I also used a VFD called Penta-Power from here >>http://www.kbelectronics.com/Variable_Speed_AC_Drives_Inverters.html

    The model number I used was the KBAC-24D model it can handle 1/8hp to 1hp AC Motor....

    B. Morse
     
  9. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    Firstly, thanks to all for replies.

    Let me address them as I need to:

    I was planning on having a PCB made up for this task to do some valve control timing and ink level monitoring - the idea was also to mount a pressure sensor on the PCB and have the output "massaged" (ie - signal processing) to be a useful range of voltage or current for the VFD controller's built in I/O and PID functions.

    I have read in some recent IEEE publications that PID is not the ideal approach for VFD control, but fuzzy logic is.

    There will be an mBed uC on the board, so it may be possible to code our own control loop if necessary - i will simply need to make the design flexible enough to accommodate the possibilities of either massaging the sensor output to the VFD or routing it to the micro so that i can spin off a daughter board with the control hardware if need be.

    In short, i was planning on buying a VFD with built in PID and using a pressure sensor, with a processed output to be usable to the controller.

    Ideally, from a development standpoint i would like to do the VFD hardware layout myself, but in the interest of project deadlines that isnt a great idea necessarily. Time will tell as i dig deeper into the problem as the mechanical engineer solidifies his design. I am waiting for him to finish before i start my end. He is the type of engineer who flip fops on his design numerous times - number of valves, type of valves, how many sensors etc. Shop air vs pump, 1 pump - 2 pumps...so from an electronics support perspective its impossible to get a jump start.


    thanks for the vendor tip.

    I will google/wiki this - i am not familiar with the device and if it will fit our need.

    If anyone is interested, or thinks they can offer more suitable advice i can describe the functionality required. If not, id rather save the carpal tunnel.
    :)


    The reason I am looking into VFD is that it seems to be best suited for constant pressure/variable load applications. We need to maintain a constant pressure (within some TBD error bar) at a point in the pneumatic/hydraulic (mixed) circuit. I have read VFD was well suited for this. Originally the older engineers steered me away from AC motors - especially polyphase - hence my inquiry.

    After more research i do not fear them at all, and they (3-phase AC) seem to be best suited to this, and most efficient from a power perspective as well. Seems like the correct fit for the application.

    Fractional hp 3-phase has been what i have been leaning towards, i need to size based on the torque requirements and current still though, again - in a holding pattern at the ME's mercy essentially.

    Again, thanks for the vendor tip - this is exactly what i was looking for, vendors people have used and had success with. there are so many options out there that its almost overwhelming.

    thanks again to all.

    Dan
     
  10. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    Look at page 9 of this pdf. The B95 pressure reducing vale could solve your problem with minimum effort. It can be made for 720psi inlet and 400 psi outlet pressure.
    There are dozens of similar valves out there.

    http://www.cashvalve.com/PDF/CAVMC-0509-US.pdf

    Variable-frequency drives are widely used. In ventilation systems for large buildings, variable-frequency motors on fans save energy by allowing the volume of air moved to match the system demand.

    This statement seems to match the principle which we are trying to design to.

    Ventilation systems in buildings are predominately a static load totally out of the ballpark from your pump system. Once the required air flow for an air supply fan is set by the VFD, it will barely require adjustment. Even then, that adjustment would be made manually. In the old days we did this by changing pulley ratios which would happen very rarely if the initial design was done right.
     
  11. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19

    I think in regards to the first device (which i did check into) we are trying to avoid wasting shop air or any other consumable - customers tend to not like this.

    Although for a proof of concept model it may suffice, so thanks for the heads up.

    Regarding your second point - i would disagree albeit ignorantly, that the airflow into and out of a building is static. especially in an industrial environment with ventilated ovens, ventilated workspaces, atmospheric control chambers, A/C, exhaust fans in kitchens etc.

    From a non-informed perspective it seems to be anything but static, but if you can show me otherwise, im open to hearing it.

    Everything i read tells me that the best application of a VFD is to a pump/fan application with a variable load. This essentially describes our problem to a 't'.

    thanks again to all.
     
  12. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    kingdano likes this.
  13. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    awesome find, definitely worth checking into.

    thanks!
     
  14. BillB3857

    Senior Member

    Feb 28, 2009
    2,400
    348
    Did you watch the video with him running the demo motor with two flashlights? Now THAT was impressive!
     
  15. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    not yet, i just checked out the website quickly.

    unfortunately it looks like they dont offer a pre-assembled pump version of that technology, they might, but i didnt see it in my quick glance.

    doubtful that this place would let us design our own pump around it, despite the clear technological advantages it offers.

    deadlines destroy innovation. its really a shame.
     
  16. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234
    I happen to be talking to my ABB rep today, and he happened to mention their VFD drives..... and this thread poped right in my head!

    Anyway, check out these VFD's specifically for controlling low voltage 1 phase pump motors >>> http://www.abb.com/product/seitp322/dced894f9370b1be8525720a004cdaf8.aspx may have to look under the data tab and select the appropriate model from your spec requirements....

    B. Morse
     
    kingdano likes this.
  17. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    I don't see what plant air has to do with this regulator. It plugs in at the discharge of your pump and provides the 400psi you need. Simple.

    99% of the air consuming or air using equipment in a factory are very small in comparison to the building air supply fan. Turning some of these smaller fans on or off will be practically unnoticeable.
    I worked for 25 years in big paint shops, really big shops. For Friday night shutdown, almost every fan in the shop would be turned off for energy savings. That's about 75 fans ranging from 5 to 150 hp. Not once did we need to make any adjustment to the building air supply.
     
  18. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    my only arguement to your point would be that maybe the building air supply was running in a variable drive format so that it adjusted itself as the load changed?

    maybe?
     
  19. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    I think VFD's on supply fans are great. Besides providing a soft start they are much easier to adjust than louvers and they are stable. Set'em and forget'em til the next time. Adding PID control would be like peeing in a dark blue suit- you'll get a warm feeling all over and nobody will notice. Your pump control project will be a real challenge because of the two phase flow and the tight loop you'll need to maintain 400psi. What kind of pump is it and what is the liquid being pumped?
     
  20. kingdano

    Thread Starter Member

    Apr 14, 2010
    377
    19
    there is no pump selected yet, my research has pointed me to using a 3-phase synchronous AC motor with a VFD control topology.

    i would like to use a motor control uC, but people at work are steering me towards off-the-shelf controllers and sensors for this project.

    i am pretty sure that this:

    http://www.freescale.com/files/microcontrollers/doc/data_sheet/MC3PHAC.pdf


    would be a pretty hand device to wire-up and try to use.

    its fairly low cost in small volumes ($10USD/ea) - but seems to be a non-stock item.

    at any rate for a proof of concept application i think you want to maintain flexibility - hence my adverse reaction to off the shelf controllers, i have no access to the "guts" if i need it.

    i was told that concerns for something like this include the high voltage and uC on a single PCB layout - i have read though, that you can feed a VFD with a constant DC bus and avoid those issues.

    i will probably have to appease the bosses and go off-the-shelf btu it does go against my better judgment.

    the fluid being pumped is waxy material liquefied at high temperatures.

    its being moved via air pressure though - so you could say that in a mechanical load sense, pressurized air is the load to the pump.

    i am not sure why you refer to 400 psi - i thought i mentioned 4.6-4.7 bar, which is ~ 70psi, is it not?
     
Loading...