Frequency of test runs on Emergency Generator 3

[BamBam.a]

Hi Guys,

I've got a Waukesha VGF F18 GLD that is in a standby application at a Chiller plant. I wanted to see what your
ideas would be in regards to test running the engine? I like to do a weekly load test (%80 load (max sight load)) for about 30 minutes.

I've heard the stories; its not the running of machines that wears, it's the starting. etc etc. I dont want to hurt the
engine unnecessarily, but I still need to make sure it starts when it needs to start.

Any thoughts?

 

[catserveng]

Since its a gas engine you may want to consider running it less often for a longer period. Ideally you'll want to run it long enough to get the lube oil up to normal operating temp.

If you're worried about excessive engine wear that could be caused by engine starting, consider adding a prelubrication pump, and if you're in an area with low ambient temperatures, possibly adding a lube oil heater as well.

Since you stated this was a standby, I'm assuming you already have a jacket water heater installed.

Do you have exhaust aftertreatment installed, such as a 3 way catalyst or SCR? If so these systems would also benefit from a longer test run period.

Hope that helps, MikeL

 

[BamBam.a]

Hi Mike,

Yes, I usually run the engine till it reaches operating temps and then for an additional 10 minutes or so.

I already have a prelube pump that runs for 2 minutes prior to start, jacket water heater as well. No catalytic converter though.

Thanks

Dan

 

[TugboatEng]

30 days is the standard test interval for the marine industry. Is your engine air, hydraulic, or electric start? Does it have redundant staters? Are there redundant engines? Battery starting systems are the only starting systems that wear while sitting. Other starting systems wear only when starting so it would be preferable to have fewer starting cycles.

Catserveng, I had a pair of 3512B standard engines that the keys didn't get installed in the pre-lube pumps. I believe they ran the entire 15k overhaul period wbich I'd guess does 3800 starts in that time period (2 starts per day and 2800 hours per year). The rod bearings weren't as good as they normally are but they were definitely fine.


I like the systems that run the pre-lube continuously and with heated oil. I'd bet you could offset those energy costs moving to a 30 day test interval.

 

[catserveng]

For a lot of turbocharged engines an intermittent prelube system seems to be a good choice. Continuous prelube is nice for several applications as long as you don't flood the top end or the turbo bearings. Some larger engine continuous prelube systems utilize a dump valve to drain off prelube flow to just maintain oil in the filter housings and lower galleries..

Plant I'm working in this week has four 2MW gas engines and three 2MW diesel standby's, all have intermittent prelube running 5 mins/hr, if a gas engine is called for "normal start" it prelubes for 3 minutes before cranking, if it is called for a "fast start", usually means one of the online units has a problem, it bypasses prelube and starts ASAP. During a normal stop cycle we also do a post lube for 5 mins after stop, have found it seems to help with turbo bearing life. In a total of 386,000 operating hours over 16 years we have only had one turbo failure on a gas engine unit.

The diesels all start without prelube prior to crank. Each diesel unit can be manually prelubed for a maintenance run if selected by the operator.

In all the work I have done over the years I find that engines that sit a lot and have a "fast start" requirement such as life safety standby units really see the impact of not having a prelube pump. Their life saving is that fact they typically run less than 100 hrs/year. The standby market is usually very cost sensitive, so trying to add a feature like a prelube pump is often considered non-essential.

I recently got called in to do a failure analysis on a 2001 vintage 3516B genset that was a standby for a cold storage facility in central California, the unit ran an average of 28 hrs a year for maintenance and emergency runtime, last year in the fire season it was required to run several days continuously a couple times, for a total of about 300 hrs. During a normal service that included an oil and filter change in December, the service technician noted the oil filter housing was full of "glitter". As they dug into it the rear main bearing was heavily damaged and starting to fail the two rear con rod bearings. Crank had to be pulled, fortunately the block was ok, but the owner was having a hard time replacing a crank and lower end bearings on an engine with less than 1000 hours on it. It is very likely a properly applied prelube system would have prevented the failure, and was even suggested in 2008.

MikeL.

 

[TugboatEng]

I found a really interesting article about EMD engines in standby service from the early 1990's. The don't get oil to the wrist pin/piston carrier during pre-lube. At that time chlorinated engine oils were banned and the standby EMD engines stated experiencing piston carrier failures.

With that said, I feel engines shouldn't be exposed to rapid starts (direct on line) for routine testing. Short run times with minimal load are really bad for these engines. A 2MW diesel engine is going to have a 50-350 gallon sump and it takes quite a bit of time for that to get hot enough to remove water. If your test procedure to improve reliability is causing failures perhaps you need to review your test procedure.

 

[BamBam.a]

Tugboat: I currently have electric starter, 2 x 8D batteries with a IOTA DLS series battery charger. Back when the same engine was used for peak shaving, as you mention too, I would replace
the batteries every 14-15 months. And that is with the 2 starts every day between November-March, and one start per day in the summer period. Excluding of course weekends and national holidays.

As that particular engine does not have a catalytic converter I can no longer use it for peak shaving and it was decided that the price to install one was currently too high, so it was "converted" to a standby application. Aside from actual emergencies, I'm only allowed to run the engine for 50 hours annually.

Catserveng; it has intermittent prelube system and a post lube. The prelube runs for 3 minutes every hour and postlube for 5. If my memory serves me right it takes about 45-60 seconds for the prelube to build up oil pressure. so 2 minutes of full on prelube.
Further, I was in Wisconsin late last year at the Waukesha (now Innio) Factory Training Facility, we started talking about prelube systems, naturally. I mentioned that my prelube pressure is about 25-40psi depending on oil temp. Some of the guys where very surprised that it was that high. I dont have any comparisons so I wouldn't know.

Once I arrived at a plant that was using an H24 GLD for peak shaving as well. Turns out the prelube pump was wired backwards since the beginning of time, so the prelube had never functioned properly and no-one ever
noticed. The engine had about 13.000 hours on it. I wonder...

 

[TugboatEng]

The default Cat 3500 engine starts when it sees 2 psi oil pressure. Pre-lube seems to be mostly there for filling the oil cooler housing which drains out while the engine sits. It takes just under 2 minutes to fill. An oil filter change takes closer to 3 but the engine is set to start after a 2 minute cycle if pressure doesn't build first. Of course this can all be programmed (except the cool down cycle can't override the idle speed setting on the new engines which is SUPER annoying).

 

[catserveng]

I'd agree that your prelube pressure seems high, most systems I deal with are set at the pump at 10-15 PSIG and supply 2-4 PSIG in the upper oil gallieries. As TugboatEng said, the function of a prelube system is mainly to keep galleries and cavities like the oil cooler and filter housing full of oil, so once cranking begins oil is present at the crank bearing to get it "floated" on an oil film quicker. I used to have some trend charts of oil pressure at main oil galleries on system with and without prelube, was interesting to see how long it took to get oil pressure to a crank on an engine that had been sitting awhile and then called to go right up to speed and take load for emergency service. You need to make sure the prelube system is not flooding areas like the rocker mechanisms or the turbocharger. I have seen where a prelube system actually caused a cylinder hydrolock and bent a rod, not a common occurrence but can happen if all the right things line up. So after you test run and engine, after its first intermittant prelube cycle, pull a rocker cover and see how much oil is being delivered to the tops of the valves, if it isn't very much, then you're likely ok at that point, if you get a lot of oil flooding the valves, consider lowering your prelube pump outlet pressure.

Last night's job brings to mind another comment by Tug,

He noted you want to run long enough to "it takes quite a bit of time for that to get hot enough to remove water", this is an issue I'm seeing more problems associated with as more gas engine driven standby units are in place, especially in places like California that limit maintenance run time for an emergency unit.

A 1.2MW standby gas engine driven unit, weekly test runs for 30 minutes last 4 years, no full load test since original commissioning. Unit was called to start for a utility outage, ran about 15 minutes then shutdown on a detonation alarm. As we went thru logs found the actual load on unit pretty low, so couldn't understand why a detonation alarm occured. Found that other than a single oil change about 2 years ago, no other maint performed. We pulled a valve cover, found sludge built up around rocker mechanism, pulled all the covers. Found our culprit, one of the rockers had spit its push rod, the hammering from that is what triggered our detonation shutdown. The amount of sludge built up around the rocker mechanism was extreme. This cylinder's valve cover had one of the two crankcase vent connections to the vapor recovery system, the other cylinder that had the vent cover also had a huge amount of sludge. The crankcase fumes extractor was packed with sludge and the motor burned out. All this on an engine with less than 150 run hours.

Gas engines make a LOT of water as a byproduct of combustion, especially when cold. I'm thinking as the fault analysis goes further on this site they will find other contributing factors, like I suspect the wrong lube oil was used and I also suspect the "oil change" wasn't actually done due to some descrepincy in the records, but the point of this is for the OP.

Gas engine powered gensets in standby service need to be properly maintained and operated, newer gas engines are especially less tolerant of no/low load operation than older gas engines and way less tolerant than diesel engines. Plugs foul, the engine internals get varnish and sludge buildup, and newer fuel control valves get sticky and start to bind if the sit too much and because the service manual says the recommended service intervals in in the hundreds or thousands of hours, they tend to get ignored until something goes wrong. Last nights fiasco will end up in a lawsuit, they lawyers are already sharpening their knives.

For the OP, please review your operation and testing procedures, since it is in standby operation not generating revenue and running under higher loads for longer periods of time, your needs for maintenance will actually increase. Oil and coolant sampling, and visual inspections should shift to calendar based intervals instead of using operating hour guidelines in the manual, and your actual operation, installation and environment will dictate the needed maintenance and testing that actually needs to be done.

Hope that helps,

 

[Lou Scannon]

catserveng Caterpillar has published at least one white paper geared towards application and operation of standby gensets. Perhaps you have links you'd like to share, otherwise I will dig them up while at work tomorrow and post them.

"Schiefgehen wird, was schiefgehen kann" - das Murphygesetz

 

[catserveng]

CAT EPower White Papers Link,

https://www.cat.com/en_US/by-indust...n/electric-power-industries/white-papers.html

Mostly marketing info, but a lot of good information.

Hope it helps, MikeL.

 

[catserveng]

Some similar info from Kohler if anyone is interested,

https://kohlerpower.com/en/generators/industrial/brochures-literature

and from Cummins,

https://www.cummins.com/generators/white-papers

MikeL.

 

[TugboatEng]

With modern technology I think the better solution would be to motorize the generator for a short period at the weekly intervals while monitoring all important pressures including maxing out common rail pressure (without injection). Maybe, a couple of injector pops to verify operation much the way a diesel engine finds TDC without a cam sensor. The actual start could be at much less frequent intervals. *Patent Pending*