Testing a Three-Phase Gen Set with Acid Water and Clamp Meter

This is called water load.generally the water loads are considered to be resistive loads and pf will be unity.Since the DG is rated for 35 kVA and 0.8 pf.The DG can deliver 49 amps at 0.8 pf,@415 Volts Since the load is resistive and upf, you can load only up to 39.2 amps.further for the load to be balanced the probes (3 nos) to be 120 deg apart and it should form equilateral triangle.also tips should be in the same plane.

further for the load to be balanced the probes (3 nos) to be 120 deg apart and it should form equilateral triangle.also tips should be in the same plane.

I am attaching an image of my setup. I would ignore Power Factor, so that I can better understand how KVA calculations are done on a 3 Phase System. As the image below suggest, I would be having three circuits, I1, I2 and I3. If each circuit is at 440 Volts, than I would be having a KVA of (440 * 54) at each of the three circuits. Hence I would infact be doubling from 35KVA to 70KVA. Is this possible or is my calculations screwed? Image is as below:

PS: I will explain why I need to test in next post. Thanks for the inputs so far.

A 35 kVA unit is 28 kWe (Kilowatt electrical) at 0.8 PF, the typical PF shown on normal generator tags. To run a 28 kWe alternator ("generator" term used from now on-wards) you need a engine capable of 30 kWb power (kilowatt brake; means the brake power available at the flywheel after all losses of friction, water pump, radiator fan and lube oil pump inside, etc) at your altitude, and your ambient temperature plus humidity assuming a generator efficiency of 94% (just a guess). At unity power factor (1.0 resistive load of a brine tank) your balanced amps will be 36.78 Amps on each leg which will give you a full 30 kWb engine load at 440 volt three phase. It will not fully test the generator winding temperature or generator performance as your generator is capable of 45.98 amps if you were able to create a power factor load of 0.8 PF ; keep in mind I assume your generator nameplate is 35 kVA, 440 volt, 3 phase, 0.8 PF . The power factor load is recirculating magnetism load by using reactors, NOT a brine tank, or some inductive load like an induction motor with that motor under a heavy load). I repeat that the engine will NOT see any reactive load, only resistive (watts) load. If this machine is rated as a standby (emergency black start) then the temperature rise in the windings might be rated 130* C rise over 50* C ambient so I wouldn't run it for more than a hour at full load. If it is rated at continuous 24/7 duty, it will likely be a 80* C or 105*C over 50*C ambient rating and some overload for 30 minutes or so is possible. Check with your dealer.

I am trying to test a gen set by actually shorting the three phase nodes into a bucket of acidified water...

Don't! This is very, very dangerous. Why do you need to test the gen-set rating anyway, do you not trust what is printed on the name plate? All other information should be available in the gen-set's data sheets and from the gen-set dealer or sales person.

I cannot over emphasise just how dangerous your water load approach is (especially the way you want to use it). There are far better and safer ways to do this.

More information is required!

Can I watch?

Yep. I live in Burma.

Hello Guest, I will not suggest a water load and would prefer running a load bank test. A load bank typically consist of heaters ( resistive load) and hence pf is not an issue, and load can be increased in increments till you reach 54 amps. Maintain this load for 10 minutes, and measure your output voltage and frequency, and all other critical engine parameters, like oil/coolant temperature, oil pressure, exhaust temperature noise/vibration level etc. Now, remove the load and check the output voltage and frequency. This is to check the regulation of AVR and engine actuator response. Otherwise, you can connect 4 X 7.5 hp motors and start them in sequence. Good luck !! Chaterpilar

A load bank typically consist of heaters ( resistive load) and hence pf is not an issue,...

Would it be possible to increase up to 54 amps with water load as well, since the first comment mentions that water loads are resistive loads? I have also attached an Image regarding wheather or not it will be 54 amps for 35KVA generator. My three phase system is a 440 Volts system. To get up to 35 KVA from the generator, what should the ampere be? Since I am getting three 440 Volts circuits by dipping all probes into the same bucket, multiple by three to get total KVA, right? Hence 440V * 26.5A * 3 = 35 KVA. Should I be running the test up to 26 Amps only?

Regarding why I am running a "water load" test instead of any other test? First, yes I do not trust the name plate (long story). Second, I have very limited options. Currently my gen set is with the dealer (it's been repaired under warranty). I wish to make sure that it is correctly running before moving it to my site. Currently the total capacity at my site is just 18 KVA, hence I wonn't be able to make an onsite test. I do not have access to other solutions like rigging up heaters (it would get too hot?), or hiring motors as suggested. Water is my only friend, by slowing increasing the acidicity I can control the current. But I am clueless as to which current reading should I ramp up the acid to? Basically it falls down to 26 Amps or 54 Amps? (I hope).

Water used as an electrical load will eventually heat up and eventually boil . Even more quickly if acid is added. Extreme care should be exercised.

Who's going to hold the acid jug while pouring it in to increase load? (same story : who holds the voltmeter leads while testing voltage one 15,000 volt line? ) Answer: a dead man! laugh.

If after all the "don't do it" warnings, (to which I fully agree) you still insist on doing it your way, make sure your system includes a mechanism for quick retraction of the electrodes. Do you know exactly how much of the sulphuric acid you should add without increasing conductivity to dangerous levels?, The rate in which the acid dissolves in the water is a function of the water temperature.This test (water load) is being performed only when all else fails. Why not using a resistive load load bank? Not only MUCH safer but will let you load the gen set to it's maximum of 35KVA with a fully controlled and safe environment.

Wangito.

Thanks for all the warnings. That's my only choice. Most of the generators in my country are tested this way (Burma). It's the Ampere value that I am not clear about. I am pasting an updated graphics below.

I believe that my test can be considered reactive load and balanced. Hence if I measure at at any node it should be 54 Amps. Would I need three clamp meters? Is expecting 54 Amps acceptable (the risk being that if the Gen wasn't rated for 35KVA, it would burn out).

Regarding Dissolving the acid, what I would be doing is pouring acid a little by little. At 20 amps, I will stop and run full test for 30 minutes. Check oil temperature, noise etc. And again continue adding acid until 30 Amps, 40 Amps, 45 Amps, 50 Amps. (I would not got up to 54, but stop at 50).

In Summary:

• Is a water load test consider a resistive load (Hence full KVA)
• I am weak in physics/electrical, is 54 Amps per node correct for my setup (shown in the image)
• Thanks for all the warnings, all I have is a clamp meter.

This forum has been very supportive and informative for me. Especially in a place where I live, information like this is much needed. You guys are a tremendous help. :)

First:

Given / data are:

• Line voltage E =440 V, probably 50 Hz so phase voltage is:254 V
• 3-phase nom power S = 35 kVA
• FLA / Nom current = S / (1.73 * E) =35 000/ (1.73*440) = 46 A

You want to draw 54 A so voltage must be lower, right? To keep mover in its limit.

FLA (full load amperage) = 54 A will require (at power factor PF =1)

the apparent power of: 3 phases*254 v *54 A = 41148 VA = 41 kVA. It is obvious overload so you have to keep current lower (control your load!) or keep excitation current of the generator for lower output 3-ph voltage.

I did 72 -hour 100 kVA generator test using the swimming pool as a water load. The water evaporated pretty fast! But it is possible to do the test in the way you want.

Second:

I believe above calculation before you design your adjustable load is necessary.

Third:

Start build your test station and do first tests with v. low loads, said 3 kW. Resistors or acid in water (be sure safety is observed)

If you use a 200l drum as your load it'll heat up fairly quickly and start to boil in about 30 minutes. So use the biggest (insulated) tank you can get.

Can you find a large motor (or lights in a sports ground) to run instead?

The long term rating of the generator is probably limited by internal heat build up, so you'll need to leave it running and then check the winding temperatures.

Please be careful. Jeff

I would run at 10-20% below full capacity for long term continuous use. Things last longer when not strained to the limit. Example : Never run a space heater to max of 1500 watts for more than 10 minutes.. the plug end heats up terribly. (thats what starts fires). In fact that high of a heat on that plug and wire is too much.. but they sell em all the time. Conversely you can run it on low , forever.

Something no one has mentioned: The three phase is essentially three single phase circuits. You don't HAVE to test all three at the same time. This is simpler and very likely safer. Test L1-L2 to 54 amps. Test L2-L3 to 54 amps. Test L3-L1 to 54 amps. You'll only need one current tester instead of 3.

IF you use clamp on current gauges.. you will have to get close enough to read it. This makes for danger (you'l wake up dead). Maybe a better method would be to use binoculars from a safe distance.. I had to do this once or twice myself. cheers !

This can be done safely, but only by experienced people. None us want to go to your funeral and tell your wife "He gave his life for a gen set test".

Use salt, not 'acid'. In a 55 gallon drum (250 l), I used about a spoon the size for a cup of coffee. I was amazed at how little salt it takes. One spoon was like a 100 kW or something strange like that (30 years ago so I can't remember the ratio). There were commercial "Brine Tank Load Bank" set ups sold which were wooden, clamped on the top of the drum, and lowered 3 metal paddles down in the water by a hand cranking mechanism and the deeper you went, the load increased plus with more salt it increased also. I never had the courage to hold the wooden crank handle while the gen set was running, so I shut it off, lower it 5 cm, and start it again.

For years, Stewart and Steven Houston, one of the largest gen set suppliers in the world use brine tanks on their roof for several megawatts of load (cheap, low maintenance, easy to control). But as pointed out, the water will heat up. It will be a resistive load at unity PF (1.0). If you are testing the engine, keep in mind 1.0 PF or 0.6 or 0.8 all produce the same watts (engine load) so by finding a way to increase power factor type load (recirculating magnetism) your need reactors and reactive load will ONLY test the generator, not the engine. The engine will never 'see' reactive load. Reactive load heats up the generator windings and distorts the wave form, so a high reactive load is ONLY to test the dynamics of the generator, not the engine response.

Be safe. IT IS NOT WORTH IT if you don't totally understand this old way of loading a unit ! There is a high chance of injury and death and virtually NO WAY this set up meets any safety codes these days.

As others have said, although this approach is commonly done (with plates lowered into a tank of liquid) numerous safety features are employed to prevent death (like being no where near the damn thing when the plates are remotely lowered). If you insist on doing the test remember the following.....

1) A 55 gallon drum makes a nice container. Most of these are metal. Metal conducts, and if you touch the metal drum when the generator is running you will probably be killed.

2) A 55 gallon plastic drum makes a nice container also. They are harder to find. Plastic melts. Assume the outside of the plastic barrel is also covered in spilled liquid. Liquid conducts and if you touch the wet plastic drum when the generator is running you will probably be killed.

3) Secure the electrodes in place so they don't wobble around and come into contact with the side of the container.

4) Assume any water leaks out of the container are at the same potential as the electrodes (ie- you touch the water when the generator is running, you die).

5) Don't perform this load test alone, and keep the other person well back but near a point where they can shut down the generator in the event of an emergency!

6) Get advice and help locally from an electrician or engineer. Two heads are better than one.

7) Don't go anywhere NEAR the load when it is on. If you must do any modifications, turn the generator off. I don't care if it wastes a lot of time or is more difficult to fine tune the load, ignoring this basic rule will likely result in death (accidents happen after all, especially with the inexperienced).

8) Rubber kitchen dish washing gloves will NOT protect you from electrocution.

9) Have a way to immediately shut down the generator in the event of an emergency. Ensure you can do this AND evacuate the area in seconds in the event of a major problem.

Most of these are obvious, but generally it is the obvious things that get overlooked and get people killed.

Also, if you want to test the generator you will need to measure the voltage also. As the current increases the voltage will start to drop. You may get the current you want but if the generator output voltage drops to 380V then the generator is not much use to you.

Additionally I would advise possibly coming up with some way to raise and lower the electrodes remotely (perhaps a pulley with a rope). Just make sure the three electrodes are fixed and can only move up and down (not side to side or wobble).

Final words.................

As I have said when training electricians, that exposed terminal there right next to your hand is live. If you touch it you WILL die. There are no second chances with electricity. ALWAYS remember this and treat a live circuit as one simple mistake away from death. If you have not personally isolated the circuit, always assume it is live. If you have personally isolated the circuit, ALWAYS check to see if the circuit is dead, even if you are sure it is.

Play safe

________________________________________________________________

Jack - Power and distribution engineer (and former electrical safety testing laboratory technician)

BE CAREFUL NOT TO TOUCH ANYTHING>> Back feeding has killed many a lineman who thought that a ground wire was safe to touch!! Wear rubber boots at all time is possible and rubber gloves ( or lineman gloves).

I saw a guy one time who accidentally touched something with a tape measure (unreeled 6-8 ft) in a 440 system, it burned the tape measure thru in about 5 places and knocked him down backwards. He was lucky that tape measure was thin metal since it acted as a fuse saving his life.

We used to use a 6000 gallon tank of water with open top to atmosphere. 2MW would boil the water in 4 hours. By recirculating the cold water from the bottom of the tank to the top and adding a couple of shovels of salt (about 10kgs) we could extend the time before boiling to 9 hours, this is enough for the standard test of 8 hours at full load plus one hour at 10% overload.

You may need to have your electrodes independantly adjustable in their immersion length to balance the phases. We used iron plates suspended on pulleys to enable independant adjustment of the phases.

Please ensure that no other CR4 reader is within range of this equipment during the tests. If no further posts are forthcoming then one might assume the worst.