Motor selection and VFD design questions

[stabmaster]

Hi. I'm trying to design a controller to control a VFD that is intended to drive a circulation pump. The pump circulates water in an "open loop" 1/2" pipe through solar panels. The setpoint is a differential temperature accross the panels. The control method will be P-I-D.

First, is there any reference guide for determining what RPM range and load range is required of the electric motor based on the initial head that is required to overcome, the dimentions of the pipe, and the max/min flowrates?

Also, I'm having a difficult time figuring out what kind of electric motor to select. Correct me if I'm wrong, please. I need to drive this pump from domestic 120VAC line power. From what I have gathered, I have two options: Permanent Split Capacitor (PSC) motor, or a phase splitter to drive a 3 phase motor. I've also noticed that a "universal motor" (DC motor) would also work but for some reason doesn't seem too appealing. The pump will be somewhere in the range of 1/8HP.

I haven't analyzed any initial or lifetime costs, but it appears to me that the output waveform for a PSC motor must take into account many nuaunces that I have yet to understand. One assumption here is that the 3-phase motor is not as sensitive to these nuances. Another assumption is that the 3-phase motor is more energy efficient, even when coupled with the phase converter. I would like a software developer to program a control algorithm to drive the electric motor, and from what I gather the software required to drive a regular 3 phase motor is probably readiliy available on the internet, whereas the PSC motor control requires a lot more experience and knowledge.

So please set me straight. Is there a lot more than meets the eye to controlling a 3 phase motor, and if not than is there some open source code laying around?

Finally, are there any guidelines regarding the VFD for different motors? I gather that the output waveform may be far from ideal, and the switching frequency is another variable that I have yet to fully comprehend. Is there anything else I should be aware of?

Any help is greatly appreciated!

 

[Skogsgurra]

If you are going to use a VFD, you can forget about capacitor motors. Just use an ordinary asynchronous induction motor. It is the cheapest and most robust motor available. And VFDs output three phases. They all do.

Some VFDs come with built-in controllers. Especially the HVAC models. So you probably do not have to write any code at all. Just input the parameters needed to make it run the way you want it to run.

Simple as that!

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[dcasto]

The only problem with motor selection in VFD use is that the motor may not get enough cooling at reduced speeds. That is one of the things looked at in a VFD "certified" motor. At low horsepower motors, I'd consider getting a 15 HP motor if only a 10 HP is required or get a motor with a 1.15 service factor or higher.

 

[CJCPE]

You can probably get information from pump manufacturers regarding pump selection for flow control by VFD.

In the USA and Canada, you can get a 1/4 Hp, 120V 1-phase input / 230V 3-phase output VFD with built-in PID for US$150 from AutomationDirect. You can get a 1/4 Hp motor for US$99. You can probably find a local supplier that is competitive.

 

[stabmaster]

Thanks for the help! I think that, from a product development standpoint, it is still beneficial to write the PID control into my controller, and save $50 on the VFD. Also perhaps i can find a more "cost effective" drive. I'm trying to shave every dollar i can off of this at the price of a higher initial development cost.

 

[CJCPE]

There are a few VFDs made for use with single-phase motors. I am pretty sure they are not as versatile as those made for three-phase motors, but they might be worth investigating. Check: Anacon Systems OPTIDRIVE 1

 

[jraef]

The lowest cost drive/motor solution will be a small 3 phase VFD that will accept a 115V 1 phase input and provide 230V 3 phase output, then use a 230V 3 phase motor. They are less expensive than the 1 phase VFDs noted above and the motors are less expensive and more reliable as well. But you are correct in that the least expensive of these VFDs will not have built-in PI or PID control. Here are some examples;

Teco FM50 link
Automation.com link (Note; direct link to 115V drives page conflicts with TGML code)
Hitachi link
KB Electronics link
Delta Electronics link

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[DickDV]

Also, check out ABB's ACS50 or ACS150 product line. 120V single phase in, 230V three phase out in the smaller sizes.

 

[itsmoked]

For an 1/8hp a single phase VFD would probably be fine. I also expect that you are looking at a steep learning curve to develop one.

You say differential temp control. How does the system ever start? 6AM input and output are both the same temperature.

Don't even consider really low speeds.

That OPTIDRIVE is obscenely expensive!! Wow!

What about running an 1/8hp motor on/off and controlling the flow with a controlled valve. Remember as you throttle the flow the motor load will drop so it isn't as big a waste as might be expected.

The least expensive design that covers all your desires would be using a PM brushless DC motor and micro to control it. You could run from 1RPM up to xx,000.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Yes, PM BLDC motors and controllers have come way down in price in the past few years. I often forget about that since I don't deal in small motors much any more. It used to be that DC was undesireable for something like this because of brush maintenance, but PM motors were way too expensive. I chased some OEM grocery checkstand conveyor business last year with AC and had my a** handed to me by someone proposing a PM BLDC solution. I should have remembered.



JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

Hahaha

Thanks for the real life storey jraef.

I'd totally agree with you on 1HP and up or maybe even 1/2HP a but 1/8Hp... PM BLDC all the way. They are THE standard in front loader washers and are probably in the 'almost' 1HP realm.. They variable speed them from stationary to (on my washer) 22,000RPM. And the motor is silent. It's the clothes that make all the noise.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Why am I not surprised that you have apparently torn down your washing machine Keith? Given that these new VSD driven machines are relatively new in the US, I'd be willing to bet there was nothing wrong with it when you did it too!

 

[itsmoked]

Jraef => thread1088-172491

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[HCBFlash]

#1: -"First, is there any reference guide for determining what RPM range and load range is required of the electric motor based on the initial head that is required to overcome, the dimentions of the pipe, and the max/min flowrates?"

A1 : - yes, and it's something that you can only get from the pump manufacturer, because the pump characteristics are unique

#2: -"I've also noticed that a "universal motor" (DC motor) would also work but for some reason doesn't seem too appealing."

A2 : - Can anyone else really know what may or may not "appeal" to you?

#3: - "The pump will be somewhere in the range of 1/8HP."

A3: Have you already dealt with #1 above, or is your determination of mechanical power used as your design starting point? Have you determined the maximum, normal, and minimum amount and rate of fluid to be moved and the resistance of your channel (piping etc.) and process elements? 1/8 HP is rather small to be contemplating a polyphase SCIM, especially in the absence of other associated polyphase loads and power source, unless cost and size are no consideration.

#3: - "Finally, are there any guidelines regarding the VFD for different motors? I gather that the output waveform may be far from ideal, and the switching frequency is another variable that I have yet to fully comprehend. Is there anything else I should be aware of?"

A3: certainly. motor manufacturers designs are developed around the characteristics of the power source and the driven load. NEMA standard designs are nominal. Additionally it seems that you've jumped to "VFD" (and thereby SCIM) without considering other speed controlling methods. A PMDC motor/power supply/controller is a far more commonly used solution for what your application seems to be.

#4: -"I think that, from a product development standpoint, it is still beneficial to write the PID control into my controller, and save $50 on the VFD. Also perhaps i can find a more "cost effective" drive. I'm trying to shave every dollar i can off of this at the price of a higher initial development cost."

A4: At this point you are building or refining a prototype. When the actual flow (and thereby power), requirements have been obtained and measured for multiple variables, you'll then be able to evaluate the load/speed/power characteristics of the pumpmotor required. At that point the power supply characteristics can be evaluated against those requirements. Since the cost of the finished product in large quantities is evidently your concern you shouldn't forget that motor/control packages purchased in large quantities will also have a lower (negotiated) cost. I certainly think that there are more "cost effective" system drives than A/C induction motors and VF type electronic motor drives.

What you refer to here as "my controller" must be your system operators controls, where your delta-t setpoint is set, alarms can be reset, and system turned on and off. If this is the case, it is most likely the appropriate place for your "PID control" to be located. It may or may not be where your want the power to be supplied to the motor. Signal wiring run along with motor power "leads" is a poor engineering practice, and for good reasons.

Please let us all know how this works out for you! and don't forget faq731-376

 

[jraef]

Regarding the issue of PID control in the drive or your controller:

My take is to look at the failure mode and replacement issue. If the PID control is in the drive, and the drive fails, a new drive MUST be sent out with the PID loop tuning preloaded by you. The average end user will have no clue. If the PID is in your system, the replacement drive (whatever the type you settle on) can be "off the shelf" if necessary and many users are capable of the simpler setup requirements for that basic function. If your controller fails, you need to send that out anyway, but no re-tuning of the drive will be necessary.

In all fairness though, some drives have removable control modules that can be swapped out, so all the user need do is replace the base (power) section if it dies. Still, those drives are usually more expensive.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[stabmaster]

Thanks everyone for your help!

HCBFlash
first, i remember a piping and instrumentation class in chem engineering - my book gave guidelines for motor sizing based on viscosity, pressure head, etcetcetc.

anyways i did not expound too much because sometimes i can be wordy, but in short the dc motor doesn't "appeal" to me because it won't last as long. The pump is the only mechanical piece in the entire design, so it is the first and only thing that requires service.

as per pump sizing- we use 1/8hp pumps right now. They're oversized for some projects and undersized for others; however, the thing about these pumps is that they need to overcome an initial head and then they really don't need to do squat.

as for 3 and 4.. thanks for the tips. that's why eng-tips lives up to its namesake! never ran accross PMDC control. Also, you pretty much guessed what "my controller" is without me saying anything really. It pretty much turns the pump on and off based on conditions that (1) add heat to a storage tank and (2) avoid overheating, flashing, or freezing of components. Also monitors liquid level, and triggers alarms.

Thanks again i'll be looking around!

 

[itsmoked]

Whoa stabmaster; a Brushless Permanent Magnet DC motor will last as long or longer then most others. An O L D 'brushed' DC motor would last the least amount of time of any mentioned. Don't confuse the two!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[stabmaster]

itsmoked- yes i realized that after a little searching. i really know little about motors- seriously i've pretty much been reading wikipedia and howstuffworks lately. I've never actually taken a look at brushless DC motors. Thanks for the advise.