Central plant LPS vacuum system

SDSteam

Recently began working for a small local university with a central steam plant vacuum system. We have two 1970's Kewanee (750 & 500 hp's) and a 10 year old 250 hp Hurst.

I'm told we haven't run the pumps for years because it just pulls air through the remote condensate pump tank vents...

Condensate pumps have been added here and there to "push" the condensate back to our Skidmore 36cfm cond/vac pump at the plant.

We're running 9 psi right now.

Pretty much constant water hammer in a condensate return coming from one of our buildings. We're also having a lot of problems getting enough heat from a couple of steam coils inside of air handlers in that building.

It pretty much sounds exactly like Dan's "25 steps..." article on repairing vacuum systems.


We have a fair number of leaks and it looks like the "vacuum" portion of our system has been ignored as buildings have been added and remodeled (hence the vented condensate pumps). One condensate pump tank was ran correctly for a vacuum return system (vent going back into the return line instead of venting to atmosphere), we have about 8 others that are vented. Some steam converters have vacuum breakers, some do not. We don't seem to have any steam main vents nor vacuum breakers on condensate return lines. Our blueprints from our most recent "steam upgrade" even shows a pumped condensate line magically turning into a vacuum condensate return. The only thing near this location what appears to be a bad attempt at a vacuum lift. There's also what appears to be a condensate receiver tank that was never used. I've attached a photo of that spot in the tunnel. Note that you're looking towards the plant. There's connections to another building in the photo as well. That building is pumped as well. It's actually above the condensate return line though, so shouldn't this be a simple gravity return?

Going to start getting a handle on tightening up our system, starting with leaks and then making the piping correct (correct lifts, remove vacuum breakers and vents from condensate tanks). Which leads me to my question. Is there a good resource that focuses on central plant vacuum systems? I can only seem to find a few things here and there from equipment manufacturers, but most things are about residential systems. I know it's pretty much the same thing, but I'd really like to have something with a little more detail on central plant systems so I can be sure I'm doing it correctly and prove that I'm not just making this stuff up to the higher ups.

Steamhead

Sounds interesting. Where is this job located?

SDSteam

Steamhead said:

Sounds interesting. Where is this job located?

Almost smack dab in between Baltimore and Seattle, Washington in sunny South Dakota.

Steamhead

I've been to SD- what part?

SDSteam

Steamhead said:

I've been to SD- what part?

Madison. We're about 45 miles North & West of Sioux Falls.

SDSteam

Or about 5 hours East of Sturgis!

ttekushan_3

Attached is an analog sketch showing how I've had to do this. It works very well. @Pumpguy at oak services is the vacuum return guru.

http://nash-jennings-htg.com/

Pumpguy

Typical campus arrangement I see is a central power plant with steam lines to outlying buildings. Each building has its own PRV for steam, reducing to 1-2 Psi.

Each building would have it's own vacuum condensate return pump, with condensate pumps discharging @ 40 PSI or so, back to central plant. Typical vacuum for each building's return lines is 3-8" Hg. Pump capacity for each building would be based on that building's EDR.

The mechanical lift piping shown above is correct, except for the check valve on the vacuum vent. This is typical of the "mechanical lift" for vacuum return drawing I have in my files.

This allows the vacuum to be applied to the returns that drain into the condensate pump.

Which boilers are you usually firing at any one time?

It would appear that the 36 CFM vacuum pump(s) you are using are undersize for this load, but I would need more information to confirm that.

Is this Skidmore a duplex vacuum unit with 2 vacuum pumps? If yes, have you tried running both vacuum pumps together?

Have you any way to determine if you ARE getting 36 CFM from each vacuum pump? This can be easily measured by valving off the vacuum pump's receiver tank, and fitting one or more orifices to the tank that are open to atmosphere. By knowing the size of the orifice(s), and the resulting vacuum, I can tell you your CFM. For this test to be accurate, all air entering receiver tank must come from atmosphere through these orifices.

Also, keep in mind that there is a direct relationship between the temperature of the condensate and the performance of the vacuum pump. Vacuum pumps like cold condensate. 90 degrees F. would be great, but 140 degrees F is realistic and OK. Now, 175,180, well it gets more difficult to make vacuum as the temperature goes up.

SDSteam

ttekushan said:

Attached is an analog sketch showing how I've had to do this. It works very well. @Pumpguy at oak services is the vacuum return guru.



http://nash-jennings-htg.com/

Thank you! I've stumbled on to their website quite a bit in my search for knowledge!

Your diagram actually looks quite a bit like what we have. The building that we're having the most trouble returning the condensate from is even on the west "wing" of the tunnel...

SDSteam

Pumpguy said:

Typical campus arrangement I see is a central power plant with steam lines to outlying buildings. Each building has its own PRV for steam, reducing to 1-2 Psi.

Each building would have it's own vacuum condensate return pump, with condensate pumps discharging @ 40 PSI or so, back to central plant. Typical vacuum for each building's return lines is 3-8" Hg. Pump capacity for each building would be based on that building's EDR.

The mechanical lift piping shown above is correct, except for the check valve on the vacuum vent. This is typical of the "mechanical lift" for vacuum return drawing I have in my files.

This allows the vacuum to be applied to the returns that drain into the condensate pump.

Which boilers are you usually firing at any one time?

It would appear that the 36 CFM vacuum pump(s) you are using are undersize for this load, but I would need more information to confirm that.

Is this Skidmore a duplex vacuum unit with 2 vacuum pumps? If yes, have you tried running both vacuum pumps together?

Have you any way to determine if you ARE getting 36 CFM from each vacuum pump? This can be easily measured by valving off the vacuum pump's receiver tank, and fitting one or more orifices to the tank that are open to atmosphere. By knowing the size of the orifice(s), and the resulting vacuum, I can tell you your CFM. For this test to be accurate, all air entering receiver tank must come from atmosphere through these orifices.

Also, keep in mind that there is a direct relationship between the temperature of the condensate and the performance of the vacuum pump. Vacuum pumps like cold condensate. 90 degrees F. would be great, but 140 degrees F is realistic and OK. Now, 175,180, well it gets more difficult to make vacuum as the temperature goes up.

Thank you, thank you!



The 500hp is our workhorse. When the temps drop to under about 5 F, the 750 takes over. The 250 only ran a couple of times in the fall. I'm told that we don't use it because it isn't big enough and "slings water" when a big load comes online (steam coils in air handlers and make-up air for the kitchen).

Only one boiler runs at a time and the guy that's been here for the better part of 3 decades decides which boiler should be running according to the forecast. The lead boiler is manually set at 9 psi while the lag boiler is manually set at 6.

We have no PRV's anywhere in the system. The strategy here so far seems to have been "more pressure, more better".

Most of our buildings have a tube and shell HX in the mechanical rooms where the steam piping enters. The only actual steam equipment that we have are 5 air handlers (3 of them pretty large, ~16,000 cfm each) with steam coils, steam coils in our kitchen's make-up air units and two dorms with baseboard steam radiators where each room has their own Danfoss RA-2000 thermostats.



The Skidmore is a duplex unit with 2 vacuum pumps and 2 condensate pumps.

Invoice for the unit states:
(1) PRV-36 duplex vacuum return producer 3 HP, 3450 RPM, TEFC, 208-3-60, rated for 36 CFM at 5-1/2" Hg
(1) 110 gallon CI receiver

Have not done any testing of the vacuum pumps. They've only been on once since I've been here ( just to show me how they can't pull more than 1" vacuum...). I don't figure there's much sense in worrying about them until I can get the rest of the system sealed up and properly piped. I'll do a little more digging and test the pumps out as described in the next couple of weeks.


Interesting that you mention that the condensate should be cooler. 4 years ago there was a "steam system upgrade" where all of the condensate return lines were insulated for the first time ever. Also, when I asked why we're running 9 psi I was told it was to get our condensate temperatures up so that our water treatment works better...

I'll measure the temp of the condensate main heading into the plant's condensate pump tomorrow. Judging by feel, it can't be over 140. But I've been know to be wildly off in my estimations in the past!


Thank you again. I really appreciate you even taking the time to read all of my blabber.

Pumpguy

Being a pumpguy, my concern is about pump performance. I believe I have read that high temperature condensate is a good thing for condensate chemistry, but would have to search again for that information. Actually, I believe I read it here on heatinghelp.

Our sizing standard for vacuum pumps for 2 pipe steam heating systems is 1 CFM per thousand square feet EDR. Many other pump manufacturers use 1/2 a CFM per thousand square feet EDR. Ultimately it depends on the air tightness of the system and the temperature of the returning condensate.

As systems get older, the load on the vacuum pumps increases due to minute air leaks and also rising condensate temperature as traps begin to leak steam.

Assuming all the steam is going to direct radiation for building heat, and not for process or heat exchangers for hot water heating,, and based on the 750 HP boiler, I would recommend 100 CFM vacuum pumps.

If the sizing is based on the 500 HP boiler, I would recommend 70 CFM vacuum pumps.

Based on the above numbers, and your 35 CFM vacuum pumps, it would appear they are sized for 1/2 a CFM per thousand square feet EDR. and operating the 500 HP boiler.

This undersized vacuum pump scenario is something I see fairly often. When it comes to sizing vacuum pumps for steam heating systems, the basic advice we offer is IT IS DIFFICULT TO ERR IN PROVIDING TOO MUCH AIR CAPACITY.

As a general rule, as the vacuum on the returns goes up, the supply steam pressure can be lowered to maintain a given overall system pressure differential.

So, if you are heating with 9 PSI and have no vacuum in the returns, you have a 9 PSI differential. Now, if you carried 10" Hg. vacuum on the returns, you could drop the supply pressure to 4 PSI and still have the same the same 9 PSI system pressure differential. (10" Hg. vacuum = a negative 5 PSI). Of course, this does require a good air tight system, condensate temperature low enough for the vacuum pumps to achieve 10" Hg. vacuum, and enough air capacity from the vacuum pumps to deal with the system air load.

Something else to keep in mind is the required steam temperature of the various heat exchangers, radiators, fan coils, etc. to put out 100% of their rated heat output capacity. Typically, this is in the 215 degree F range, which just happens to be the steam temperature @ 1 PSI. So, if these numbers are correct for the heat emitters you have, all you need upstream of last heat emitter is enough additional steam pressure to allow for any pressure drops through the system.

Hope this information is useful to you.

SDSteam

Pumpguy said:

...

This undersized vacuum pump scenario is something I see fairly often. When it comes to sizing vacuum pumps for steam heating systems, the basic advice we offer is IT IS DIFFICULT TO ERR IN PROVIDING TOO MUCH AIR CAPACITY.

As a general rule, as the vacuum on the returns goes up, the supply steam pressure can be lowered to maintain a given overall system pressure differential...

...Hope this information is useful to you...

Incredibly helpful!

Most of the steam goes through heat exchangers in the individual buildings. Hot water heat is used in most of our buildings coming off of shell & tube HX's. We have three dorms that also utilize HX's for domestic hot water.



Assuming our pumps are undersized even when we get the system tightened up, would it be possible to replace our standard remote condensate pumps with vacuum/return units to make up for the plant pumps? Or would that mess up any lifts in the system? Is there a stand alone vacuum pump that we could use to augment our existing units? I'm assuming there is no need to completely replace the unit that we have since we're only trying to suck more air out and not moving fluids.


Thanks again for your help and also for maintaining a great website. I had previously found it as pretty much the only source of decent information on vacuum systems.

One last thought. Your website really stresses the lower condensate temps aspect that you mentioned. We have a feedwater tank in the plant where the condensate is pumped to from the vacuum/condensate unit. Would it be OK to maintain the tank at a high temperature and the actual returning condensate at a low temp or would elevating the tank water temp be detrimental to the vacuum system?

Pumpguy

Thank you for the complement.

To address your comments in order,

There is really no benefit to have vacuum pumps where the condensate is draining from hot water heat exchangers. In these applications, the traps are few and large, the return line is usually very short, and the condensate is too hot. Only caveat is the condensate must drain from the traps in a downhill run to the condensate pump's receiver tank.

Yes, it is recommended that the vacuum pump be in the individual building, with the condensate pump sized with enough pressure to send the condensate back to the main power plant. Once the condensate leaves the vacuum pump's receiver tank, so long as it has enough pressure, uphill or downhill doesn't matter.

Yes, we do offer stand-alone vacuum pump kits for use with existing condensate only pump and receiver units. Only thing that might matter is the air tightness of the receiver tank, and the existing pump be capable of discharging from a vacuum. Generally cast iron tanks are better suited for this conversion than thin wall carbon steel tanks.

Keep in mind that with a vacuum in the receiver tank, the pressure differential across the condensate pump will increase, so it really depends on how closely the condensate pump was sized to the actual conditions.

In cases where we have done this before, the vacuum switch is set to off @ 8" Hg. and back on @ 3" Hg., maintaining an average of 5.5" Hg. vacuum. This is an additional 3 +- PSI pressure differential, which had very little consequence.

Once the condensate leaves the vacuum pump, then yes, the temperature can be raised in the boiler feed receiver. Many boiler feed units will have steam heater tubes for preheating the condensate. The most sophisticated types are deaerator units that can get down to .005 cc of dissolved oxygen per liter of condensate.