240V VFD substitute? 3

[TurbineGen]

I'm having a problem with a control system which requires speed control of 2 fans for coolers on a 25 MVA transformer. The fans are 5 hp and will be run at two frequencies: 45 hz and 60 HZ. The manufacturer has suggested using a VFD to achieve this. However, I really don't have the room in my control box. I'm already trying to fit 10 gallons into a 5 gallon bucket. Unfortunately I am unable to install and additional or larger control cabinet due to space constraints.

I was thinking of an saturable core reactor control or something to that effect. Does anyone forsee any problems with that? Are there any other possibilities?

I know it's seems ridiculous to run a cooler fan at reduced speed, but the reason for the dual speed is for noise limitations.

Any help or advice would be greatly appreciated.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[Skogsgurra]

Saturable core reactor aka transductor or magamp. If that would solve your problem, I suggest a simple triac or thyristor device instead.

OTOH, working with reduced voltage isn't a solution if your fan motors do not have a suitable rotor (like NEMA D) so they can be voltage controlled.

What kind of motor do you have?

Gunnar Englund

http://www.gke.org

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

 

[rhatcher]

What rpm is the motor that is to run 45hz? If it is 1800rpm, you could substitute a 1200rpm motor running 60hz line power and get a fan speed equivalent to 40hz at 1800rpm.

Otherwise, you may consider getting a VFD with a proper enclosure that could be mounted outside of your existing control cabinet.

 

[LionelHutz]

2-speed motors?

 

[jraef]

They would have to be custom wound 2 speed motors to get that specific low speed.

If, as in most of the transformer cooling fans I have applied VFDs to, they both run all the time, you can use one 10HP VFD running both fans. Keep in mind that means you only need one CB to feed it and you get rid of the contactors. Chances are you can shoehorn that single 10HP VFD in place of 2 x 5HP combination motor starters. There are some very nice "bookshelf" style VFD packages out there, meaning narrow deep boxes. I have used that trick several times because I can often go with a deeper box when footprint is restricted.

You can also look at mounting the VFDs outside of your box, using an outdoor rated VFD package. There are also some VFD/Motor combinations that you can use on the fans themselves, where the drive is integral to the motor.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[rhatcher]

I was having a 'duh' moment. For some reason I though that the OP wanted to run one fan at 45hz and one at 60 hz. My proposal was with that thought in mind. I wasn't suggesting two speed motors. Although that is one possible solution, it is not cost effective.

As jraef pointed out, he can run both motors on a single drive mounted outside if the existing controller.

 

[ScottyUK]

If you can use simple voltage control as mentioned by Gunnar, and dependant somewhat on on the motor and the fan, what would using a Y-D give you in terms of speed? You'd have full line voltage in delta and [√]3 line voltage in star configuration.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[ozmosis]

Couldn't you get VFD's of a higher NEMA(NEMA 12) enclosure rating and mount them outside your control box, nearer your control fans?

 

[ScottyUK]

If you do, don't use A-B's Powerflex 70. The cooling fans mounted on the heatsink are basically just electronics-grade fans like you would find on a CPU and the bearings last a few weeks. The site engineer than gets a call in the middle of the night when the drive trips on over-temperature. Guess how I know?

Are the cooling fans fitted to any of the other manufacturer's IP-rated drives any better than those used by A-B? And by 'better', I mean a properly protected fan which will survive in the type of environment where you need an IP55 or better enclosure, like the steelworks so beloved of A-B's promo graphics people and which pops up on the cover of a lot of their literature. Rockwell did show a little interest initially but never followed it up other than to tell us that it had 'passed tests'. We ended up using venturi airmovers to cool the external heatsinks, and a product which could have been a real winner was blacklisted.

I'd have a really long hard thing before using a VFD on a critical transformer if it had to be mounted in the local control cubicle. Use a multi-stage switch to bring in cooler fans as required instead.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[davidbeach]

Perhaps we could take a step back and question the whole need for multiple speeds? What I like best in a transformer fan control is simplicity and ... The two things I like best in a transfomer fan control is simplicity and robustness. Turn on the fan(s) and let 'em rip. If one fan is enough then just run one fan. Trying to exactly match the cooling needs strikes me as a solution in search of a problem. What happens if you run two fans full on when running at reduced speed would provide adequate cooling? Well, you use a bit more power on the fans and you get a bit more life out of the transformer insulation. Transformers are spendy and the added life from over-cooling may well more than pay for the added energy consumed. Failure of the cooling system due to lack of simplicity or robustness will have a far greater cost.

 

[jraef]

As he said, it's a noise issue. I have seen this before when power transformers are near residential areas. When the ambient noise is low, i.e. at night when people are asleep, they must lower the fan speed to reduce the db level and they usually don't need the full capacity from the transformer then anyway. During the day when the hubbub of nearby life makes a loud din anyway, they can run the fans at full speed, which is most likely when they need to. I have a PG&E switch yard about 1000 yards from my house across a relatively open field. On summer nights I can hear those fans whirring away.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[waross]

I would investigate noise abatement techniques as well as having an oversized stator wound with an 1800RPM winding and a 1200 RPM winding. Called a two winding two speed motor when you need other than a 2:1 ratio. This is not to rule ut a VFD. I would investigate all three approaches and quite possibly one will be the clear best choice on comparison.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

Note that Bill's idea results in a slower motor than what was recommended. 45Hz equates to 1350 Srpm, something you can't get by changing the number of poles (in other words, you can't wind 5.33 poles). 1200RPM equates to 40Hz.

Should have been said earlier; voltage control on an AC squirrel cage motor, be via it mag-amp, SCR or even connecting in Y, will produce lower TORQUE, but speed is based on frequency and the number of poles; voltage reduction does not change that. Speed reduction as a result of voltage reduction is consequential to the lower torque and the amount of load. In other words, you are not really changing the speed of the motor, you are weakening it so that it slips more. So a basic Design B every day motor will actually over load if you do this unless the torque requirements of the load drop off at the square of the voltage reduction. Even then, any change in load will cause the speed to vary. What Gunnar was alluding to is that there are AC motors, such as Design D, that can operate at high slip speed continuously without damage. That still is not really "speed control". How that would relate to your situation is that if you did use a Design D motor and dropped the voltage to a level that resulted in a 1350 Srpm speed when turning those fans, if the air density changes, the speed will change too. You could, with an SCR controller and a feedback loop, alter the voltage to try to maintain speed, but for all that trouble, a VFD would be a LOT simpler to implement.

By the way, your mag-amp idea would not likely be any smaller than a VFD, and certainly would pose as much if not more of a ventilation problem.

I say don't make this more complicated than it needs to be. Find a way to take advantage of established technology and make the VFD work. take a look at other things you have in the control panel. Many VFDs now come with a lot of 'smarts" built-in and may be able to take on some of that functionality themselves, freeing up some real estate in your box.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[rhatcher]

for davidbeach (*), your idea of staging the fans (hot=2 fans, medium=1 fan, cool=0 fans) is a good obvious choice that nobody else has mentioned. Assuming that the individual fans already have individual starters, the only addition would be two thermostats that, by definition, would be outside of the controller. This addresses all of the OP's original requirements. The only flaw in this approach would be if the two fans are distributed so that they each cool only 1/2 the transformer and if the transformer is large enough (wide enough) such that this uneven cooling could allow 1/2 of the transformer to overheat.

(*) = star

 

[TurbineGen]

As strange as it sounds, rhatcher just stumbled upon something I hadn't thought of before. The coolers are run in separate loops. Cooling on this unit is critical. If it weren;t for the noise, I'd just run the fans at 60 hz with a standard starter and be done with it.

The fans are set to run at 45 hz all the time until the transformer gets to a certain temperature, then they go to a full 60 hz.

The problem is that this is a mobile substation. Mobile units tend to attempt to fit 10 gallons into a 5 gallon bucket. Space is very limited. So is weight. I'll see if I can mount the VFDs outside the control box. That is looking like my only solution. These mobile units are turning into a nightmare.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[itsmoked]

You can get VFD's that are NEMA 4 that could be mounted externally.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[DickDV]

One caution here: Just because a VFD is rated Nema 4 does not mean there are no temperature limits.

Despite the absurd claims of some manufacturers, VFD's operating in an ambient temp below 0 degrees C (32 degrees F), need a space heater. Drive life falls off dramatically when subject to colder temps than that.

Also, on outdoor units, extra care must be taken to avoid condensation inside the drive. As a general rule, if the drive is running and making heat, condensation will not be a problem but, if idle for a while and then started, a flash can easily spell the premature end for a drive with condensation in it.

 

[sreid]

VFDs have goten pretty small. A 10 HP Yaskawa V1000 is 10 in H x 5.5 in D x 5.5 in W.

http://www.yaskawa.com/site/product...al AC Drives~V1000 Drive~V1000Dimensions.html

 

[TurbineGen]

Wow! That is really small. The ones I was looking into were 4 times that size. You might have just solved my problem. I can likely fit two of those in the box. I already have cooling and a heater in there so temperature should be ok.

Thank you all for the help.

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.

 

[ozmosis]

I think I would look at the J1000 instead of the V1000 Yaskawa unit if looking down that avenue. The "J" is more suited to the basic applications whereas the "V" is high performance. Dimensions are the same.
Sledgehammer to crack a nut sort of thing.