This weekend marks the third year year in a row that I have worked our plant's annual electrical maintenance program. The narrative which follows describes our system in general and the maintenance tasks we perform. I was wondering what others are doing.

We have a 13.8 KV switchgear with two load breakers protecting our 13.8 KV ring which passes thru 9 transfer switches. Attached to each transfer switch is a 13,8 kv to 480 V substation. At each substation we have at least 4 and sometimes 9 load breakers {600 - 800 A} feeding our motor control centers.

Annual Maintenance starts with load shed of all 480 loads and then opening of the main switchgear. Our Electricity Vendor then disconnects the 13.8 from our Main Switchhouse and we ground the 13.8 loop at the main. This is followed by grounding the 13.8 at each and grounding of the 480 at each sub. Once the system is "cold" maintenance begins.

Maintenance at each substation includes un-racking each breaker, changing the battery in it's trip unit, cleaning with Windex to remove all dirt, dust and corrosion, applying a fresh coat of electrical grease and re-racking the breaker. This is followed by a cleaning of all buss bars and checking of a joints for torque. We also vacuum our and dust each transformer enclosure and check cooling fan performance.

Maintenance at each transfer switch includes, cleaning, checking for toque all connections, checking clearances and condition of all arc arrest insulators and re-greasing all switch surfaces.

Main Switchhouse 13.8 gear and breakers are treated in a manner very similiar to substations.

After cleaning is complete we reverse the de-energize procedure and restore function to the facility.

This has been standard procedure for the past 25 years or more and we have what I am told are some of the best electrical gear for their age. Our electrical service loss downtime is also extremely low.

Thoughts from others?

There-in lies the issue. Electrical equipment is funny stuff. It does NOT follow the standard assumptions for maintenance. Most electrical failures will happen within the first few weeks/months of operation. After that point, as long as you don't disturb it, it will typically run and run and run right up to the end of it's useful service life and then fail, sometimes with early warnings. This is contrary to mechanical equipment that is usually the most reliable and holds specs only when it is brand new and goes on a linear downhill curve from there.

The cycling of the equipment is far more damaging than not. A classic example is that there are two plants in central Georgia almost side-by-side. One does a similar outage procedure every year. The other does...nothing. The difference in maintenance costs? Significantly higher for the plant that does annual maintenance. Difference in reliability? Zilch. Roughly the same number of failures, etc.

If in doubt, doing nothing is far better with electrical equipment than doing anything at all, contrary to popular opinion in "proactive" groups. For instance, all the old text books (American Electrician's Handbook for instance) still recommend taking your motors apart annually and regreasing the bearings! If you follow the greasing schedules you typically get for mechanical equipment, you'll push grease up inside the motor and cause premature failures in nothing flat. The best grease schedule is absolutely minimal, and nothing is second best to overgreasing. One of the formulas I've seen is 1 pump (one stroke) per 25 HP. This is very little grease indeed.



I definitely stopped doing this. Depending on the trip unit, the only thing the battery does for you is store settings in the event of a trip and the breaker becomes unpowered. It's nice to do but rarely has any value. This is a nice to have but makes sense when you are already doing maintenance work.



Good idea, since you disturbed a connection that was gas tight.



Bad idea. You are actually gradually INCREASING the torque and not following torque specs. This is the same as taking your torque wrench and hitting it multiple times. There is plenty of mechanical maintenance literature talking about this fallacy.



This is a good idea. I highly recommend doing this as long as you're not disturbing any of the connections.

What I didn't see is any mention of doing thermal (IR) scans. Though it's not a 100% gaurantee and there are some basic rules about doing it right (such as not scanning your transformers at high noon when solar radiation accounts for more hot spots than actual problems), it does tend to catch joints and some component failures in the earliest stages of failure. I recommend every 6 months, and that seems to be the general rule.

Second, annual transformer oil analysis is generally a very good idea. It doesn't cost much. It will let you know what's going on in your transformer IF you keep records. Educate yourself on this stuff because there is a lot of "noise" in addition to the data. You can easily be fooled by simple garbage data coming from AA spectroscopy for instance (but very few companies run anything else). Karl Fischer is notoriously unreliable and the moisture data is difficult to make sense since it's an ongoing process.

Third, IEEE and NFPA recommend draw out circuit breaker testing every 3 years. Might as well do a Dobles test on your transformers (turns ratio test) while you're at it. These are the just about the only electrical tests worth messing with on distribution equipment because they directly measure performance against specs. If you've upgraded your circuit breakers to electronic trips, they frequently have built in tests. But in addition you are supposed to exercise the mechanisms periodically (manufacturers specify frequency; I believe S&C is the most aggressive @ twice a year) and sometimes grease them on a recommended schedule.


I know that plants that do no maintenance are always concerned about doing the right thing. The few that do far too much are just as bad but it's much harder to change the culture there.

I don't know what you would consider some of the "best gear for their age".

But my former employer's gear and buss duct was installed and 1968, was never maintained, was grossly overloaded, and hardly ever shut off. The gear looked like it had cottage cheese growing off the back off it. The buss duct was packed full of steel dust. Most of the knife switches on it hadn't been move in so long you had to put a bar on them to switch them off.

The system was thermally tested once, and it was easier to note what didn't show signs of heat. We shut it down a few years ago to temporarily move the main power. When we reengaged the gear, you could hear it sparking for about a minute while it "burnt" back in.

It was all taken out of service at the end of last year with a record of zero downtime in 40 years

Not to brag but I've got one Post WW II - Pre-Korea era Substation that you can see yourself in the brass on and has not given us any trouble with switch operation. Recent Thermal testing showed no hot spots and trip tests were still within spec.

I'll try and get some photo's next year.

In a word Nothing, we recently had an old Square D Panel which caused us problems. It all started when a 400A breaker went bad. We got a new 600A and turned off all 5 breakers then the main 2000A. We changed out the bad one, then switched on the main and turned on 4 of the 5. The last one a 600A Breaker would not switch on, just kept tripping. So we ordered another 600A. Again when it came in we turned off all 4 breakers then the Main and changed out the second 600A breaker. Came to turn on the Main, it would not set. Now we needed to overnight a 2000A breaker.
Bottom line non of these breakers have been turned off in 25 years, and that is as long as my boss has been here, how long before that, probably not since the initial installation and nobody knows that date.

What I'm finding with breakers that old is that the old dash pots in the trip units lose their oil and then they trip prematurely. Either this is brought on by cycling the breaker or by tripping it (something shorts out). Either way, unless you test them once in a while, you'll never find this stuff until it becomes a catastrophic problem. Weak trip units going out of calibration give you plenty of advanced warning when they drift out of spec but long before the relay mechanism is completely shot. Worst case scenario is when the trip unit doesn't trip for several seconds/minutes and the relay takes some damage on opening/closing. Then you've got a real problem unless you keep spares on the shelf.

Spares? What's them?
Unfortunately our company policy is spares sitting on shelves are a waste of money.
So what do we do, well, when we need something urgent we pay overnight shipping to get what we need and lose 24hrs+ of production
I really like my Job, but some of the management decisions seem to come from the twilight zone.

I'm not sure where you expect to find spare draw-out ITE circuit breakers in the 600-1600 A range overnighted from. Same with old GE AK-1 or AK-2 series (they are now on AK-9 or something like that).

Those management decisions are very simple. Let's take for instance a plant located some place in the continental United States within 40-50 minutes of a major metropolitan area. Let's assume we're talking about a very stock motor, say a 3 phase standard squirrel cage induction motor with a NEMA standard frame size (C-faced or standard), 2, 4, or 6 pole, and with a size of between 1/2 and 200 HP, you can probably get it within 2 hours. If it takes you 10 minutes to diagnose that you need to replace the motor and 10 minutes to order it, and it takes you more than an hour or two to actually physically change one out...

The loss in production is less than the cost to warehouse the motor and let someone else warehouse it. Everything sitting on the shelf in the warehouse costs roughly 10% of the cost to acquire it in general.

If you are talking about for instance a wire wound, double shafted, multi-speed AC motor or many of the DC motors, or even a very large AC motor (500+ HP), chances are that there are none anywhere in the entire world available. Rewind time/costs will take about 3-8 days depending on the size, etc. SO...again, it comes down to some pretty simple dollars here.

If you do not express your inventory requirements in this way (in terms of lost production dollars), you will never, ever win the stock/nonstock argument.

There is also a third way out. It's called VMI (vendor managed inventory), or consignment. Have a larger motor shop keep the motors on your shelves and manage the inventory FOR you. You will pay extra for those motors but the value on the books (in inventory) is $0.00. You pay for the motor only when you "purchase" it, which is when you pull it off the shelf and use it. For nuts, bolts, fittings, etc., this is the only practical way to go. With motors, you have to look more closely at it but the smaller sizes generally tend to lean towards VMI.

During the time that I spent at my last employer, we had an electrical preventative maintenance program like BobLFoot describes. The only difference is that we could never buy a complete outage from production scheduling. We had to settle for 1/3 of the gear 3 times a year. All of the loads are shed at the power distribution panels associated with the "to be PMed" transformer. The secondary mains on the transformers are opened, followed by the primary main. Then the outdoor 25kV switch which feeds the primary main for one or two 25kV to 480V transformers is opened. Grounding is done on the load side of the primary and secondary mains, as well as in the outdoor switch bay so that the workers are in-between the grounds.

We then remove all of the panels to the sub (indoor dry type). Buss ways are inspected for any discoloration. We do not torque the joints, as the flats on the nuts and bolts are marked with a hash mark. We'd just verify that there wasn't any movement. Terminations are checked on the breakers, and any hotspots detected in a prior IR scan are addressed. We then vacuum the bays, and wipe the insulators with damp rags, also checking cooling fans. The fuses are exercised in and out a few times, as well as dead transfer switches. The dead switchgear is exercised a few times before being closed permanently for another year.

We also use this time to make sure our entire emergency Life and Safety generators, auxiliary power generators, and UPSs are doing their job.

Maybe my frustration of lack of spares misled my post. I agree with what you are saying Paul, and I agree there is a practical reasoning behind keeping specialty (High dollar) items that might fail in a year or ten years is bad buisness. But I have had production down for longer than need be because of a 110V relay Double Pole, or a $20 Proximity Switch.
OK sorry to side track the thread.

I know this is a old post, but Paul brought up a good point about thermal imaging. I want to know if you guys do it yourself and hire an outside contractor to do it. We do it every year and we find new issues every year and some old ones that we think we fixed. We have someone come in and do it. It takes them alomost a week to go through everything. Mainly we like to the big stuff such as sub station, overhead lines and compressor cabinets.

We used to pay a company to come in and do a thermal imaging report on our power equipment, but the cost of a thermal imaging system has come down in price so we finally bought our own and now we do it in house. We set up a schedule so that everything is spread out but everything gets done within a years time. It actually works out pretty good and has helped a lot.

It's also very useful in a breakdown situation. Maintenance uses the camera a lot to check for bad motor bearings and such and we use it to diagnose heater head problems or hot spots in electrical contol panels (bad contactors, overloads, etc...). I've used it several times to check a MOD-5 box that we rebuilt to make sure that there are no bad connections. I think the camera was around $7,000 but I'm sure it's more than paid for itself in the few years we've had it.