IR Testing of Denergized Equipment

[Natfuelbilll]

Hello -
I am involved in a new build project that specifies IR testing of the gear. The manufacturer did not install IR windows and they have work policy that forbid them from opening energized gear.

My question is if the industry has used an approach to perform IR testing as follows:

1. The load should be operated at full load for three hours, or until the breaker reaches normal load temperature;
2. Deenergize the system;
3. Open equipment doors to expose previously energized gear quickly enough to limit temperature degradation;
4. scan the breaker with an IR type non-contact thermometer and record the readings.
5. Reenergize the system if no significant problems are present

.

 

[charlie b]

I don't believe that process would give you an accurate indication of any potential overheating issues. It is just not possible to be fast enough in getting the doors open.

Is the constraint that they don't want the equipment to be energized at the moment the covers get removed? Then how about getting the IR equipment ready, turning off the gear, removing the covers, reenergizing the gear, allowing the internal temperatures to stabilize, then and only then entering the room to take the IR measurements?

 

[Chamuit]

If they want to do it that way, it is pointless. Normally, you want panels, equipment, whatever your are taking readings on under normal load and warmed up. Just need to suit up appropriately. If they aren't allowed to open them up, can you get a different contractor to do so? That's what we have to do.

 

[mgookin]

Do an RFI to the specifier and inform him of the conflict.

 

[David Goodman]

Kill the power to the panel, verify, install IR inspection windows on panel, close panels, return power, run load for reasonable period of time, inspect with IR instruments.

Another option would be to install remote sensors that would be left inside the panel. Likely would need to use an individual sensor for each device to be monitored inside the panel.

 

[David Goodman]

Hello -
I am involved in a new build project that specifies IR testing of the gear. The manufacturer did not install IR windows and they have work policy that forbid them from opening energized gear.

So, the work policy states that opening energized gear is forbidden. Does the policy prohibit energizing gear once opened if the flash protection boundary is secured and no entry is made into that boundary for the IR inspection? A basic IR thermometer would not likely be sufficient for this purpose. However, the higher end IR cameras should perform well.

 

[iwire]

Kill the power to the panel, verify, install IR inspection windows on panel, close panels, return power, run load for reasonable period of time, inspect with IR instruments.

Having done a fair bit if IR surveying I would say the likelihood of getting the viewing windows in locations that are of some use is slim at best.

 

[scotton]

I do a lot of electrical infrared scans and on occasion I run into an equipment panel that can't be opened without de-energizing the equipment. If the equipment has been under load for a while, there is plenty of time to see heat differentials without needing to re-energize the equipment. This is using a good infrared imager, not an IR thermometer. For me, an imager with a resolution of at least 160x120 is the minimum.

However, I don't think it would be feasible to do an entire plant in this manner. The logistics make my head hurt. You would have to work up stream from the equipment to the MCUs, and every time you de-energize something, the load goes away. So then you would have to wait for the load to warm back up before opening the next panel, and on and on. And the entire place would have to be down while everything is being shut off over and over again.

So on a case by case basis where dictated by the equipment, you can get results by de-energizing equipment and scanning. But as a system wide policy, I think you will be very disappointed and waste a lot of person and equipment time.

I would recommend having a qualified infrared thermographer (accompanied by a good electrician to open panels) scan the facility while everything is running and in use. That is the best way to find real issues. Give yourself the best chance for success.

 

[Natfuelbilll]

"If the equipment has been under load for a while, there is plenty of time to see heat differentials without needing to re-energize the equipment. "

It is this idea that got me to write the OP.

It would be interesting to know the rate of cooling following denergization. Perhaps some smart physics savvy responses are in order.

 

[Natfuelbilll]

I don't believe that process would give you an accurate indication of any potential overheating issues. It is just not possible to be fast enough in getting the doors open.

Is the constraint that they don't want the equipment to be energized at the moment the covers get removed? Then how about getting the IR equipment ready, turning off the gear, removing the covers, reenergizing the gear, allowing the internal temperatures to stabilize, then and only then entering the room to take the IR measurements?

Hi Charlie - you started it's not possible to get doors open fast enough. I am wondering a couple of things:

1. How fast must the doors be opened in order to get any meaningful temperature reading?
2. How much accuracy is needed to identify hot spots?
3. Would you say there would not be any value to performing this testing method?

 

[mgookin]

Hi Charlie - you started it's not possible to get doors open fast enough. I am wondering a couple of things:

1. How fast must the doors be opened in order to get any meaningful temperature reading?
2. How much accuracy is needed to identify hot spots?
3. Would you say there would not be any value to performing this testing method?

If you had data logging equipment and that equipment could time stamp the shut down and record the thermal data, you could plot your data and run regression to see what that data says it would have been (might have been) at T=0.

But as stated previously, the conflict is in the specs. Someone has specified a procedure which is against company policy if I remember this thread correctly. It should be on the specifier to develop a solution to a conflict which he created.

 

[Natfuelbilll]

If you had data logging equipment and that equipment could time stamp the shut down and record the thermal data, you could plot your data and run regression to see what that data says it would have been (might have been) at T=0.

But as stated previously, the conflict is in the specs. Someone has specified a procedure which is against company policy if I remember this thread correctly. It should be on the specifier to develop a solution to a conflict which he created.

Ooooohhh I like this logical conclusion that the specifier needs to develop the solution.

 

[big john]

You will not be able to give any meaningful quantitative values from the shut-it-down-and-look-quickly method.

While I agree that overall the thermal mass of an object will show that it is hotter than it should be, you can very rapidly dissipate many dozens of degrees from a hot spot on a good thermal conductor like copper.

So you might see that A pole in a disconnect was hot, but you would have no way of ascertaining whether that was the result of a rapidly cooling much more severe hotspot, or just overall heating.

This would be a slow, cumbersome, and inaccurate survey. Far better to open the panels during the shut down, reenergize and then do the survey from outside the flash boundary.