Two 500 KVA units down within a week Dirty load

[micksaudi001]

Hi everyone i just lost two 500 kva units within a week one has 5000 hours the other has 3125 hours
units are dirty loads on flashbutt welding machines
anyone had issues with stator hittig rotor on OLYMPIAN units DE and NDE its still looking good i wonder aboutmounting on these older units if its effecting the internals

 

[Besoeker3]

Hi everyone i just lost two 500 kva units within a week one has 5000 hours the other has 3125 hours
units are dirty loads on flashbutt welding machines
anyone had issues with stator hittig rotor on OLYMPIAN units DE and NDE its still looking good i wonder aboutmounting on these older units if its effecting the internals

Harmonics maybe?

 

[micksaudi001]

Hi I understand about harmonics but how to test to find problem could an electricsl engineer view the load schedule

 

[micksaudi001]

Hi I understand about harmonics but how to test to find problem could an electricsl engineer view the load schedule

 

[Besoeker3]

Hi I understand about harmonics but how to test to find problem could an electricsl engineer view the load schedule

It's what I usually I did but you need specific details.

 

[zbang]

Units? Are these transformers or generators? Need more specifics.

 

[micksaudi001]

So my understanding generator s is two phases loaded to weld via transformer and the other phase small ancillary load running welding head to weld rails

 

[petersonra]

Seems unlikely a generator would be affected by harmonics on the line since it's not attached to the grid.

It's hard to know what they might have done to break it. Could be just coincidence that they failed at the same time. Or could have been the same operator running them both wrong that caused them to fail.

I think you will have to do some more investigation if you want to find out what really happened but chances are low that you will find out the real scoop.

 

[micksaudi001]

I understand the units were running quite sometime with moderate loads then the load factors where changed to give better. Welding result this meant shorter loaded time of around 140 seconds compared to 220 but the amps was increased almost twice the amount. I am told this is Common welding technology
It was noticed the generator trying to jump out of the container and then sparks comming from fan end
The nde and de bear ii ng looks good i have 5 500 kva units all the same function to loose two is almost unheard of i really at a loss to explain

 

[zbang]

Large single phase loads? Ya, the current unbalance could potentially pull the rotor into the stator (flexing the shaft). It might have worked before because the load wasn't enough to mechanically pull things out of whack. Add to that- most engine-driven generators aren't good at going from low to high loads quickly; the prime mover usually doesn't have enough torque to keep the rotor moving at speed when a large load is thrown on. They can also overspeed when the load is quickly removed.

My understanding, from a colleague a while back, is that constant-speed engine-driven generators are terrible for large welding loads and one is better to use purpose-built welding machines. Or to keep the welder current to a fraction of the capacity of the engine/generator to avoid the problems above.

Give is specifics! What are the load currents and voltages? What are the actual generator models? More information is better, don't make us guess.

 

[micksaudi001]

Ok so in 120 seconds the voltages vary from 380 to 420 volts the amps goes up to ,900 to zero from what i see but very very short times these are olympian 500 and 550 kva units perkins 2500 series engines from all you have said i dont see too much droop but im a mechanical person i can shoot you a chart if i can work out how to add it here i know the damage is done now on these two units just seems to intrigue me that technology cannot design a better unit or does bigger generator kva solve this problem

 

[micksaudi001]

Ok so in 120 seconds the voltages vary from 380 to 420 volts the amps goes up to ,900 to zero from what i see but very very short times these are olympian 500 and 550 kva units perkins 2500 series engines from all you have said i dont see too much droop but im a mechanical person i can shoot you a chart if i can work out how to add it here i know the damage is done now on these two units just seems to intrigue me that technology cannot design a better unit or does bigger generator kva solve this problem

 

[micksaudi001]

Ok so in 120 seconds the voltages vary from 380 to 420 volts the amps goes up to ,900 to zero from what i see but very very short times these are olympian 500 and 550 kva units perkins 2500 series engines from all you have said i dont see too much droop but im a mechanical person i can shoot you a chart if i can work out how to add it here i know the damage is done now on these two units just seems to intrigue me that technology cannot design a better unit or does bigger generator kva solve this problem
Sorry i cannot load the weld run chart im still yrying

 

[micksaudi001]

Ok so in 120 seconds the voltages vary from 380 to 420 volts the amps goes up to ,900 to zero from what i see but very very short times these are olympian 500 and 550 kva units perkins 2500 series engines from all you have said i dont see too much droop but im a mechanical person i can shoot you a chart if i can work out how to add it here i know the damage is done now on these two units just seems to intrigue me that technology cannot design a better unit or does bigger generator kva solve this problem
Sorry i cannot load the weld run chart im still yrying
My email is micksaudi001@gmail.com if your interested i can PM you

 

[gar]

22071-2105 EDT

micksaudi001:

Given a fixed resistance, and you double the current thru that resistance, then you quadruple the power dissipated. Given an equal repletion rate for a pulse, then average power dissipation remains constant for a constant period if the average power dissipations of the pulses remains constant. You can do the math to tell you how current pulse amplitude has to change to make this constant.

So in your application assuming source voltage remains constant, and repletion rate is the same, then 140/220 = 0.64 is your duration ratio, but power dissipation ratio in the generator goes up by 2^2 = 4 because of doubling the current. Thus, temperature rise in the generator goes up by about 0.64 * 4 = 2.55 times.

If your original loading was somewhat near maximum for the generator, then clearly with the indicated load change you will burnout the generator.

.

 

[zbang]

seems to intrigue me that technology cannot design a better unit or does bigger generator kva solve this problem

That's the answer, but a purpose-built welding generator will work much better than a constant-speed/frequency generator. (Welding generators do not try to keep a steady frequency or voltage, but control the current to the weld specs. Red-D-Arc has a diesel-powered 1000amp welder , there's video of it on youtube.)

(someone please check me on this)
Assuming a 500kva generator and 400v phase/phase or 230 phase/neutral, that's ~725amp available from each phase. 900 amps is definitely over that, but maybe not enough to trip the breaker fast enough (if the generators have electronic controls, that should trip the unit offline). This squares with that the weld current was recently doubled; before while being unbalanced it would have been only ~63% of rated current, not 125%.

Add to that the single-phase load and you have severely unbalanced current flows. Note that while the draw (~360kva) isn't over the rating, but that's deceiving because it's all on two out of three phases.

 

[Jraef]

...
anyone had issues with stator hittig rotor on OLYMPIAN units DE and NDE its still looking good i wonder aboutmounting on these older units if its effecting the internals

So DID your stator hit the rotor? If so, and the bearings are good*, then the only way that can happen is shaft sag, which I have seen on some really dirt cheap motors, so it's possible for it to happen on dirt cheap generators I suppose.

*But how was it determined that the bearings were good? "Looking good" is not the same as BEING good...

 

[paulengr]

Hi everyone i just lost two 500 kva units within a week one has 5000 hours the other has 3125 hours
units are dirty loads on flashbutt welding machines
anyone had issues with stator hittig rotor on OLYMPIAN units DE and NDE its still looking good i wonder aboutmounting on these older units if its effecting the internals

Hitting? Mechanical issues. Olympian line was discontinued over a decade ago. The generators are I think Katolite. The engines are Perkins. Those are typically very solid engines. I’m guessing coupling or wear. When were they ever greased? Can you even get it to purge? Sounds like bearing failure, not a power problem.

 

[micksaudi001]

Thanks everyone for your comments
i think yes Olympian maybe run away
anyway some interesting comments to absorb
one simple question for the audience
if one assumes a generator can handle at 300 percent overload for a short time as manufactures suggest then other than the frequency time of the overload what could go wrong

 

[micksaudi001]

Hitting? Mechanical issues. Olympian line was discontinued over a decade ago. The generators are I think Katolite. The engines are Perkins. Those are typically very solid engines. I’m guessing coupling or wear. When were they ever greased? Can you even get it to purge? Sounds like bearing failure, not a power problem.

well i would maybe consider the 5000 hour unit as a bearing issue the axil measurement on the fly wheel was around 2mm so in theory it seems excessive
The other unit 3000 hours is a 2005 unit which the bearing unit looks and feels perfect with no movement and .5mm axil movement on the Engine flywheel
ATM its a bit of a fight with manufactures to get any answers about there product or any support