How bad could this get (VFD Overheating)

[JasonSelf]

So, my boss quoted a panel to a client with a ControlLogix rack and (4) 30HP Powerflex 70 VFD's. The enclosure is completely enclosed and the VFD's are surface mounted, not flange mounted. I asked him if he might want to put some cooling or ventilation in the panel, maybe a few large fans and he told me "No, it isn't going to get that hot". This thing is in an equipment room for a water treatment plant. I did some calcs to see how much heat gain I am going to have and after seeing the numbers I am wondering how bad this is going to get.

I am not sure if that is a question or a statement...maybe just a vent (pardon the pun).

 

[boneless]

First of all, I would recommend to separate your drives and controls into different cabinets.

Second, 30 HP, would be about 22 kW if I am correct? I would definitely put some ventilation in there.

If I look at this document, page 10, I see that the maximum temperature is 40/50 C. I have no experience with the Powerflex VFD's, but I imagine if you run all 4 off them at full capacity, in an fully enclosed panel, you will run up the temperature pretty darn fast.

 

[just the cowboy]

I asked him if he might want to put some cooling or ventilation in the panel, maybe a few large fans and he told me "No, it isn't going to get that hot".

Sorry he's wrong, it will get hot. I just put 4- 5hp drives in a enclosure and had to increase the fan size "because I thought it would not get that hot"

 

[JasonSelf]

I can't argue with him over it, he won't budge and becomes....unpleasant.

Did I mention I am starting my own business....lol.

 

[Bullzi]

Just like my tag line says: There is never enough money to do it right but always enough money to do it over. He will regret this decision!!

 

[jraef]

The best drives out there are at best, 97% efficient. So the worst case scenario that should be used as a design point would be full speed, fully loaded. So the enclosure cooling must be based on those drives rejecting a total of 2640W inside, not counting other devices in the box. So think of it this way, that's roughly the the equivalent of 2 hand held hair driers running at high heat in that closed environment, with only the steel (presumed) walls radiating heat as a dissipation method.

As an extremely gross rule based on experience, I have had success in sealed enclosure sizing by looking at the manufacturer's offering of a NEMA 1 vented package, calculate the volume of that box, and quadruple it for a sealed box (type 4/12). In your case, your overall enclosure volume would need to be 16x the standard N1 volume. This method excludes the influences of external heat loading from sun exposure or proximity to hot equipment (ie ovens), and a minimum 20 degree F delta from max drive operating temperature and max ambient. So the PF70 drives are rated for 40C when enclosed which is 104F, therefor this will work ok for ambients of up to 84F, with all sides except the back exposed to dissipate.

So given those qualifications, the box volume would need to be a minimum of 13,600ci, assuming a Frame D drive (30HP, Normal Duty, 480V) and given a height limit of say 48" and let's say 12" depth, it could be as small as 24" wide and it might work. YMMV. 84F is not very warm in the real world however. If you are in Main or Alaska, maybe...

 

[dlweber]

About the third time the fire department gets called there for these things tripping heat sensors and he is threatened with fines for false alarms, he might consider ventilating then.

Or if they do actually start a fire the fire department will be happy to put big ventilation holes in the walls and ceiling for him

 

[RET]

Salesmen are going to sell.

 

[Tom Jenkins]

There is really no need to guess. Hoffman and Hammond and Rittal all have sections in their catalogs showing how to calculate panel temperature and how to size cooling fans or A/C units. Here is a start:
http://www.hoffmanonline.com/produc...39055&cat_3=162530&catID=162530&itemID=162584

Use 5% of the VFD rating to be safe, and use the pdf you can download to get a much better estimate than "I think it will be OK."

 

[the_msp]

In my last company we had 3x22kw drives inside a large enclosure. It had two fans, they were about a sq. ft each. And it still got warm!

 

[JasonSelf]

I did exactly what you said Tom with the Hoffman calculator before I went to him. I got numbers that suggested I needed like 3 of Hoffmans largest 24v fans with exhausts.

Oh well, you can only lead the horse to water. I know he is concerned about chlorine and to have this system sealed, and only running 2 VFD's at a time (sure they will), you would be looking at like a $3.5k AC unit if I am right.

 

[Tom Jenkins]

First of all, I would recommend to separate your drives and controls into different cabinets.

And I forgot to mention that I totally agree with this advice. You might even be advised to examine why the VFD needs to be in an enclosure at all.

 

[Tom Jenkins]

I did exactly what you said Tom with the Hoffman calculator before I went to him. I got numbers that suggested I needed like 3 of Hoffmans largest 24v fans with exhausts.

Oh well, you can only lead the horse to water. I know he is concerned about chlorine and to have this system sealed, and only running 2 VFD's at a time (sure they will), you would be looking at like a $3.5k AC unit if I am right.

As you say, you can only do so much. Actually, if you've got chlorine in the air in any significant concentration then corroded VFDs are the least of your problems! I always put the control panels in a separate room from the chlorine storage or injection areas, an if I recall correctly that is required by codes.

If you can't escape in time and need to retrofit, you may consider something like this: http://www.automationdirect.com/static/specs/enccoolers.pdf The compressed air can be pulled from a safe area. You need a vent, but the pressure will keep the chlorine out.

 

[DickDV]

One comment each for jraef's and Tom Jenkins' last replies: For jraef, it is worth emphasizing that the 97% efficiency (3% losses) is calculated from LOAD KW, not drive output or input rating. So, if the motors don't work very hard, then there is less waste heat to deal with. For Tom, compressed air injected in a box works very good but compressed air is never cheap. And, it needs to be clean, very clean. No oil, especially!

 

[jraef]

One comment each for jraef's and Tom Jenkins' last replies: For jraef, it is worth emphasizing that the 97% efficiency (3% losses) is calculated from LOAD KW, not drive output or input rating. So, if the motors don't work very hard, then there is less waste heat to deal with. For Tom, compressed air injected in a box works very good but compressed air is never cheap. And, it needs to be clean, very clean. No oil, especially!

Dick,
Yes, I know, but I always size for the worst case, because Murphy rules...

I once had to find a way to cool some VFDs on cranes over an aluminum smelter pot line, where right above the pots it was upward of 180F ambient. We looked into using vortex coolers since they already had instrument quality cleaned dried air lines near by. To do the job I needed 1500BTU of cooling for each enclosure. A standard Hoffman A/C unit capable of 1500BTU was about the equivalent of 1HP, whereas each vortex cooler capable of that needed 25SCFM of air. That equated to about 6HP worth of air compressor...

Side note: they went with the vortex coolers anyway because the pot line was a different cost center from the part of the plant where the compressors were, so the cost of running the cooling system didn't come out of their operating budget, whereas the cost to run the Hoffman A/C units would...