UPS for Exhaust Fans

Status
Not open for further replies.

kaichosan

Member
Our architect has requested that the exhaust fan (on emergency) be on UPS. Apparently, they have done it in past projects. I've only used emergency diesel generators in my past to provide emergency backup for the fire pump, exit signs, emergency lights, smoke control, limited elevator power, etc. This design-build housing facility is considered a low rise building with 320 residential units plus six retail areas.

Have any of you used UPS in lieu of emergency DG? I am configuring the UPS similar to an emergency diesel generator requirement.

1. If so, do you size the UPS like the generators? That is, 100% load, 90 minutes run time, and 25% for derating due to battery life (Add another 33% for future load?)

2. I think I will also need an emergency shut-off.

3. Aso add a general UPS trouble alarm tied to the Fire Alarm. Will I also require control for the exhaust fans for the fire fighters?

Thanks in advance...
 

tkb

Senior Member
Location
MA
It's a bad idea to put motor loads on a UPS. They are not designed for motors.
Also the batteries would have to be serviced yearly and replaced about every 5 years.

The UPS would have to be sized 2-3 times the load to be able to start the motors.

Why would it need to be on a UPS and not the generator anyways?
 

Besoeker

Senior Member
Location
UK
It's a bad idea to put motor loads on a UPS. They are not designed for motors.
Also the batteries would have to be serviced yearly and replaced about every 5 years.

The UPS would have to be sized 2-3 times the load to be able to start the motors.
I somewhat agree.
If it is true on line UPS it would be supplying the load all the time rather than being like a back up generator.
There is the issue of starting the motors. Most commercial/industrial UPS systems have a "walk in" feature that mitigates inrush. For sure that won't avoid the direct on line start currents for a motor. The inverter section of the UPS needs to be short term rated for that but the storage batteries don't.

Horses for courses.
If you want say half a MW for some hours a static UPS probably isn't the ideal solution. A public building requiring maybe 20kVA for three hours for emergency lighting is UPS territory.
 

tkb

Senior Member
Location
MA
There is also the duration of operation to be considered.

A UPS is limited to the battery size as to how long the load will be energized during an outage. Maybe an hour.

A generator on the other hand is only limited by the fuel supply. Typically a 24 hour tank.

A generator would be outside the building and a UPS will need to be inside and taking up valuable floorspace.
 
I somewhat agree.
If it is true on line UPS it would be supplying the load all the time rather than being like a back up generator.
There is the issue of starting the motors. Most commercial/industrial UPS systems have a "walk in" feature that mitigates inrush. For sure that won't avoid the direct on line start currents for a motor. The inverter section of the UPS needs to be short term rated for that but the storage batteries don't.

Horses for courses.
If you want say half a MW for some hours a static UPS probably isn't the ideal solution. A public building requiring maybe 20kVA for three hours for emergency lighting is UPS territory.

Agreed that it is wasteful to power motors from UPS'. Soft start can help with the motor inrush.Without is the FVNR start may force the UPS onto the bypass mode IF that is available. UPS battery capacity, as a general rule, should be sized for 10-30 min. max. continuous use anything beyond that should be supported by a generator. You also have to think about that if you oversize the batteries, prolonged use will cause overheat of the location if you lost power and have no AC available for the removal of heat. So prolonged UPS operation should be supported by a standby generator that is sized to support the UPS and the associated HVAC. In this case, if you only have fans to support and no other critical electronic loads that require truly uninterrupted - less than 1/4 cycle - support, a genset would be sufficient.

Lastly the walk-in feature has nothing to do with the load, it is the front end design of the UPS so it won't overload the generator that may be required to supply the UPS. Centralized emergency supplies designed for lighting and other non-electronic equipment support are not true UPS's, their output quality and load transfer capability matches the requirements of those and inferior to the UPS's quality of the same parameters. Even HID lighting can tolerate a full cycle interruption. Again the same rule aplpies for battery sizing, a standby generator supporting the Centralized Emergency supply may be better economic alternative than oodles of batteries that require replacement.
 

kingpb

Senior Member
Location
SE USA as far as you can go
Occupation
Engineer, Registered
Are we getting the full story?

The reason I ask is the first line of the OP says the Architect has done this before on past projects, with projects being plural, therefore more than once.

Why are they doing this? What type of application is the exhaust fan(s), i.e. a fume hood in a lab?

Before getting all excited about what to do, why not sit down with your Architect and go over the application and find out why he feels this is necessary. Maybe somebody sold him on this before and now they think it's the best way, or just maybe the terminology they are using is not correct, or they just don't know what to ask for.

You might be surprised how far a little communication goes.

In the end if that's what they truly want, then it can be designed. I have done many 480V 3ph UPS systems where motors are on essential service buses, and are started an run on batteries due to loss of normal AC; some had to run for up to 3 hours off of batteries. The system just has to be designed properly.
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
I once did a project for the DOD on mobile emergency air traffic radar systems. The critical nature of the units and the predominance of power electronics inside meant they required continuous air conditioning, so we had to provide a UPS to operate the AC units for another 30 minutes just in case the generators failed (typical redundant military stuff). To power a 15HP AC unit (compressor and blower combined), we had to use a 75kVA on-line UPS. We tested it with smaller units first and the motor starting current would bury the UPS and engage their crowbar circuit to protect the transistors in the inverter. In other words they shut themselves off when the motor stating current was too much, as in both the compressor and blower starting together. I'm only telling you this story to give you an idea of how much of an issue the starting current is for UPS systems running AC motors. The only solution is to size the UPS to provide locked rotor current for all of the motors that may start at one time, and with the worst case on-line load. In your case you could use staggered start timers on each fan to prevent simultaneous starting, but you still would have to use the total combined FLC of all the fans + the LRC of the largest motor. No real way around it.

By the way we did try putting soft starters on the motors; no good. The soft starters did not like the PWM output of the UPS. Soft starters have to sense and know where the "zero cross" is on the incoming sine wave as a reference point for determining their firing sequence. On high harmonic supplies, like PWM from a UPS, they have trouble finding that true zero-cross point. We were going to try VFDs and that would have worked for the blowers but not the compressor, the A/C unit mfr said absolutely not.

Bottom line it can be done, but it is NOT cheap because you have to SERIOUSLY over size the UPS compared to lighting / PC power supply loads. In my case it was the military, so regular maintenance would be mandated and very likely to take place. But I would question that in a commercial environment; I rarely see anyone even change filters on a cooling fan after the first year and replace batteries 5 years later? Probably will not happen. What is more likely is that they will not need it for 6 or 7 years, then when they do, the batteires will be long dead and the system will not be there anyway. Just my opinion though.
 

tkb

Senior Member
Location
MA
I once did a project for the DOD on mobile emergency air traffic radar systems. The critical nature of the units and the predominance of power electronics inside meant they required continuous air conditioning, so we had to provide a UPS to operate the AC units for another 30 minutes just in case the generators failed (typical redundant military stuff). To power a 15HP AC unit (compressor and blower combined), we had to use a 75kVA on-line UPS. We tested it with smaller units first and the motor starting current would bury the UPS and engage their crowbar circuit to protect the transistors in the inverter. In other words they shut themselves off when the motor stating current was too much, as in both the compressor and blower starting together. I'm only telling you this story to give you an idea of how much of an issue the starting current is for UPS systems running AC motors. The only solution is to size the UPS to provide locked rotor current for all of the motors that may start at one time, and with the worst case on-line load. In your case you could use staggered start timers on each fan to prevent simultaneous starting, but you still would have to use the total combined FLC of all the fans + the LRC of the largest motor. No real way around it.

By the way we did try putting soft starters on the motors; no good. The soft starters did not like the PWM output of the UPS. Soft starters have to sense and know where the "zero cross" is on the incoming sine wave as a reference point for determining their firing sequence. On high harmonic supplies, like PWM from a UPS, they have trouble finding that true zero-cross point. We were going to try VFDs and that would have worked for the blowers but not the compressor, the A/C unit mfr said absolutely not.

Bottom line it can be done, but it is NOT cheap because you have to SERIOUSLY over size the UPS compared to lighting / PC power supply loads. In my case it was the military, so regular maintenance would be mandated and very likely to take place. But I would question that in a commercial environment; I rarely see anyone even change filters on a cooling fan after the first year and replace batteries 5 years later? Probably will not happen. What is more likely is that they will not need it for 6 or 7 years, then when they do, the batteires will be long dead and the system will not be there anyway. Just my opinion though.

Did you ever consider a rotary UPS diesel generator combination as Pillar offers?
 
It's a bad idea to put motor loads on a UPS. They are not designed for motors.
Also the batteries would have to be serviced yearly and replaced about every 5 years. The UPS would have to be sized 2-3 times the load to be able to start the motors.

We put motor loads on UPS all of the time. They work fine, when sized properly. Our application requires it to keep chilled water pumps and fan units running during generator start-up. (Chillers on generator only.)

We size the UPS to ensure that it has enough spare capacity, to support 5 times the largest motor load. This way, the inrush of the largest motor can be supported. (Motor start-up is staggered so that inrushes are not combined.) And we base the size of the UPS by dividing the resultant capacity needed by 90%; in other words, there is additional capacity on top of this. In addition, larger 3-phase UPS can typically handle overloads of 110% continuously, 150% for 90 seconds, and 300% for one cycle (the last one for clearing faults). (And we spec & test this so there are no surprises.) Bear in mind, UPS'es power transformers without problem; Transformers can have larger inrush requirements than motors.

Unfortunately, the typical batteries need to be replaced in 3 years (5-year rated). However, batteries can be 10-year, 20-year or longer (at an additional cost, of course).

Why would it need to be on a UPS and not the generator anyways?

This is my question, as well. Generally, there is no reason for exhaust fans. We have hydrogen exhaust fans in the UPS battery rooms, that don't need to be on UPS. Could be a special application. If the exhaust fan is the UPS' only load; its going to have to be sized a lot larger than 2-3 times. More on the order of 6 times.
 
Last edited:
Status
Not open for further replies.
Top