Ammonia Mass Flow Meter? 2

[Herman2]

I have an engine room with ten (10)ammonia screw-type compressors (400 - 1250 hp) pulling suction (5# and 25#) on an ammonia system in a large manufacturing facility.

I want to measure the cooling load. I can get compressor motor currents, ammonia temperatures and pressures. But don't I need to know the ammonia mass flow rate? I don't want to assume some compressor performance curve, I want to know actual results.

The ammonia is distributed to a series of fan coils in the plant, and we have an evaporative condenser outside. I know how I would do it on a chilled water system, but this situation has me puzzled.

Can I just put flow meters in the ammonia lines?

Thoughts from the experts please.

 

[quark]

hmmm.... not convinced. (though nothing can stop you from an ammonia mass flow meter, absolutely possible but more headaches)

Do you check return condition of ammonia from every fan coil unit? How will you average it?

Why can't you check cooling load at air side or water side?

 

[Herman2]

There are dozens of fan coil units in the plant, there are air-handling units, refrigerators, freezers, and also units that make chilled water, some cool a brine loop. Not much chance of getting my arms around all that equipment.

I'm interested in energy management, just want to know the cooling load and electric power used to meet it.

How far-out is the idea of an ammonia mass flow meter? I'm thinking some type of averaging pitot tube?

 

[quark]

Not a bad idea, but I didn't see it so far (ofcourse, I am just 9 years experienced). The flip side is that you may not come to know where exactly you are wasting energy, with a centralized data recording.

By averaging pitot tube, do you mean an annubar? Could be ok. What accuracy you anticipate? It is better if you install it in the liquid line. Just take care of the pressure drop. With more pressure drop, flashing may take place and this will reduce accuracy of the reading.

If you can feed Ammonia properties in the instrument computer, with liquid and gas temperatures and pressures, it is possible to read direct BTU readings. May be Fischer Rosemount can help you.

Vortex flow meter is another good option.

I don't say far-out, because you are pioneering something. You are welcome incase you need any specific help.

Good Luck

 

[chillin]

I've not seen anyone in the ammonia industry attempt to measure flow. Most energy analysis is done by monitoring compressor performance (i.e. temp in/out, press in/out and motor current).

The most effective and simplest way to cut energy costs is to float head pressure based on wet-bulb temperature (assuming you have evap condensers). VFD drives on your condenser fan motors will maximize the results (and there may be an incentive from your power company to install them if you haven't already). Huge energy savings can be seen by operating at the lowest possible head pressure, instead of the base case design pressure (probably 181#).

You can also experiment with floating the suction pressure based on plant demand (i.e. room temps). If your product is temperature sensitive, then you have to be careful how much you allow it to float up. Unfortunately, this method for coolers will work only when when rates are lowest (at night and on weekends), when transmission or infiltration loads are at a minimum.

Since you have freezers, you can try to lower the room temp off-peak and let it float during the on-peak to run fewer compressors. Again watch product temps as well, but there will be a lag in product and air temps. If you run a -10 freezer (likely at 5# suction), then try to get it as low as possible during off peak, say -20, and let it float back up during the day. This will be easist on the weekend.

Look for simple ways to maximize efficiency. Attempting to measure the flow through the system will be problematic at best. It's best to deduce flow based on equipment operating conditions.

chillin

 

[NH3Valves]

Along the lines of what quark has mentioned, one can obtain information about flowrate from the air or water side of the condenser rather than measuring cooling load. You would need inlet and outlet conditions of both the NH3 and the medium to which it is rejecting heat, as well as, the flowrate for this medium. This is easy (I have done it) for water cooled condensers with a cooling tower. This would be harder for an Evap. condenser, though. Another way would be to use the suction conditions and the swept volume, volumetric efficiency, and rpm of the compressor to get a mass flowrate. This information can likely be obtained compressor manufacturer. At a compression ratio of 5 or 6 to 1, you can expect a volumetric efficiency of 85-90%. Either of these methods are likely to give you ballpark numbers similar to those you would obtain from an averaging pitot tube manometer. I have used those before (never with NH3), and while they may not be the most accurate, they certainly are cheap and durable.

 

[Macbeth]

I have used FCI thermo mass flow meters on about 25 NH3 screw Comp. These work very accurately although they needed to be corrected for enthalpy. Although I must add these would only end up being toys in a warehouse. I generally do very much as NH3Valves described to prove compressor performance. And program systems very much as CHILLIN describes. So they both must be very intelligent people.

 

[chillin]

Most compressor manufacturers have computer simulation programs, or at the very least capacity tables with correction factors, for their products. The data provided by the programs/tables is backed up by the factory's own R&D lab tests. If you doubt the accuracy of the data, you can always evaluate system performance on the evaporator side (i.e. air/water in/out temps and cfm/gpm).

If you have the suction and discharge temps/pressures, along with the motor amps, you can use the program to identify the actual operating capacity, and therefore the ammonia cfm or lb/m. Most of the data you need can come from the on-board microprocessor (for big machines).

Your evaporator fans are a minor part of your overall electrical consumption. Knowing the load on the process/air side is important, but only in determining if you have enough compressor to balance the load! The best ways to save energy are as I described in my first post. You can also consider VFD drives on evaporators, but the cost may be prohibative if you have many fan coils.

Measuring the actual flow via flow meters is a great scientific exercise, and would be useful if there were unexpected performance problems. However, if the goal is to trim electrical consumption, I think that much less effort is needed.

With the options I've described (float head/suction pressures, VFD condenser fans, VFD fan coils), I've designed system(s) that were higher in first cost than without those measures. But my client(s) received upwards of $125k incentives and less than 5 year payback on the design improvments (the plants were on the order of 1000TR).

keep it cool
chillin

 

[jhospedales]

We also experienced dificulities in determining the condition of the the screw compressors and did install flow meters.