Sunday 16 December 2018

The concept of TPM (Total Productive Maintenance)


Purpose of Total Productive Maintenance

Total Productive Maintenance (TPM) is the concept of involve maintenance all employees with the purpose of effectiveness on overall production systems through participation and maintenance activities that are productive.

The two main words on one of the science management of maintenance it was "productive" and "total". It means that maintenance of these types want to appear different from other maintenance concept in a second term that.

The Total Productive Maintenance also gives consequences to give priority to the prevention of disruption, so that its motto became "maintenance is free". How can a device operator know doing activities which is preventive and as early as possible know of any potential distractions, before the disorder actually occurs?

How is it possible we could do prevention of disorder on our device?

This is very possible, although within the life-time of a 15-year-old device, could have been if it continues to do maintenance in chic, then age it would be exceed. Besides that, in fact, the cause of the disorder is actually partly not coming from within the device, but from the outside, dust covers, loosen bolt (so interfere with PCB holder), scratches, shock, temperature, humidity, water, and others.

Therefore, use all the means that can make the device become durable. It was said even if an electrical device is not turn on for too long, it will be quickly damage. If you have more than one signal generator, for example, use interchangeably.

So what should be done with these Total Productive Maintenance in doing technician

Learn to do periodic maintenance on a regular basis.

Check the condition of the room every day, how the State of temperature, humidity, and hygiene conditions in Lab calibration conditions. in the on-site, it sometimes becomes quite serious issue, where the standard devices cannot be condition on the appropriate temperature as conditions in the Lab.

Receive and perform device usage guidelines are reasonable.

How to use the device properly, either standard or device customer (unit under test), usually already given in detail in the manual (user guide or the calibration guide). The technicians have to accustom ourselves to always use this manual consistently, in any circumstances, too. Not to because it felt "memorize on the outside of the head", the measuring instrument is not use in a reasonable, for example, should warm up needs to be 30 minutes, new 5 minutes alone is directly use. Or rather, do not be lazy view usage instructions gauge through the manual, for example in doing the self-calibration, which sometimes requires a lot of steps before the device is ready to use measure.

Get use to identify as early as possible signs of device experience decrease performance.

In the ISO 17025 quality system, one of the means to do this is with the identify drift monitoring, that is by checking the performance of the devices, are still within the limits of reasonableness or already are out of specification.

Drift monitoring is generally carry out on a periodic basis, could, be able to annually per six months, depending on the type of device. Even for devices that use high frequency, this activity should be done more often again. Drift monitoring can also be done accidental, for example, on the condition that requires further examination on performance of the device.

However, associate with point 3 above, I am more likely to be at once using the calibration activities every day to do drift monitoring, because it often happens that there is a measuring instrument customer calibrate using one standard device as the meter and one standard device as the source. If the parameters are equal, why not all just check (crisscross) between devices such standards?

There are even times when measuring instrument customer can be use to check the accuracy of the standard devices belonging to this all of course Lab. There are rules and procedures that must be obey by the technicians.

Meanwhile, technical manager or senior technician at least should also support the Total Productive Maintenance on this in the following way:

  1. Help engineers learn maintenance activities they can do themselves, or in other words improve their capabilities in terms of maintenance.

There needs to be an increase knowledge for engineers, even to the already accustom to work daily in a Lab. This increase through training, seminars, training, even a built in informal discussions in everyday Lab.

The topic of discussion is not just how to do the calibration, but also how to perform maintenance correctly.

  1. Fix the device capabilities of a decline in standards and other accessories, for example, through the process of repair, adjustment, even changing methods of measurements in calibration.

Not the functioning of the fan (fan) the devices do not underestimate, because if left on and the device is use in a long time, then it could be the temperature inside device will exceed the limit. This potentially shorten the lifespan of the device.

  1. in a more advance level again, find the flaws of the standard device architecture and other accessories and depart from here to do planning for potential disorder can anticipate on the device later.

From your experience, it may appear a foregone conclusion that the device type A with brand X has a less good performance, for example, excessive heat and the fit is not stable. Perhaps you would not recommend procurement tool similar to another purpose in other departments in your company, especially if it is to be use in the calibration operations in your Lab.

However, if you already have it, the only thing I can do is to maximize the ability of the device or restrict its use on certain conditions. Examples such as my Lab ever did, in cases where we have four signal generator of the same type, and the same brand. The four turn out to be a new discovery is the device that the accuracy of the frequency is sometimes unstable.

Specifications - Total Productive Maintenance

However, there's one fourth of which is really already outside its specifications. Then we decide to use the device on this one specifically as one of the signal generator to the main Lab, while another third can be use in calibration on-site. The accuracy of the frequency we keep by way of continue put it together of reference out of Rubidium Frequency Standard over a distribution amplifier. Then these devices now have accuracy equal Rubidium, a third higher than other devices.


The Total Productive Maintenance is actually has a close relationship with preventive maintenance. His passion is so that technicians are not preoccupy with things that are corrective, an improvement because there is damage. Unwittingly, often the effort for this too big so as to make all members of the Lab so busy but the results a bit.

All technicians need to be given the awareness that there are actually some of the main causes of interference of an electrical device, even a passive though.

And the cause of it in part is due to human factor, not the factor for the physical device. If the device is fast because it is often left in a State of dirty, it's not the wrong device, but because of the let it dirty. Understanding this simple should be impress although a basic understanding for all Calibration Lab personnel, not just the technicians only.

Shoves five main causes of interference on the machine, which I adaptation for device calibration, as follows (Total Productive Maintenance):

  1. failure of fulfilling the basic maintenance need device

  2. Fault condition the device correctly in accordance with the manufacturer's instructions

  3. lack of skills, for example, the technician shall pronounce setting config files on the measuring instrument OS-base computers

  4. Condition of the devices that are old or obsolete

  5. Departures from the intent of the architecture of the device, e.g. power supply use for grant as a voltage source

Maintenance will certainly increase its requirements if a device design as part of systems, e.g. devices must be connect with the computer data to be accessible the results of measure it.

For these cases then the maintenance should also be done on the computer, including the maintenance of computer hardware (RAM, hard drive, and the motherboard) and software (operating system and software application-specific device).


Prevention of nuisance or damage the device

By looking at the picture above, does not mean that the program always starts with maintenance preventive perform periodic inspection. You can start step-the step from anywhere, depending on conditions and priorities in your lab.

Let's try detail step-step prevention of disturbance of the device.

  1. Keep the device reasonable condition (Total Productive Maintenance)

Do the money on a regular basis, either in the form of calibration, drift activity monitoring, even daily calibration activities. I prefer to do calibration of two to three devices customer (unit under test). A kind all at once rather than one by one. Predictably, because that way I can compare the results between measuring gauge. So that if any one of the device experiencing interference, more easily identify. Even if the customer device has a great specification, a standard device we can check its accuracy rate.

Tightening the connector that is connect to a measuring instrument must be done proportionally, generally not too tight but not too loose. Do connection cable connectors and gently, especially on devices that need soft handling such as optical devices.

Total Productive Maintenance - Sub operations

Operating procedure must also be follow properly. Follow the manufacturer's instructions properly so that the device can deliver its performance to the maximum. If the warm up is reveal about 15 minutes, then before this deadline should not be use for the measurement, since the devices could be unstable so that his appointment can be fickle.

Related operating procedures, I recommend that the nonprofit activities also include the maintenance in it, not just give priority to description of the stages of its calibration only. Often it is overlook because it considers all the techs have enough knowledge about handling measuring instrument, when in reality it is not.

Don't forget stabilize electricity PLN, try using a stabilizer that is able to change the voltage of nets-nets (PLN) into the voltage with a purer form. This can save electricity, and is technically would be able to extend the life of the device. In addition, and this is more important, is able to minimize the uncertainty in the accuracy of the device. Just look at the specs of the device. Usually there is the addition of the uncertainties which have to do with getting the voltage supply labile.

Calibration Activities & Total Productive Maintenance

For onsite calibration activities, need to be taken in the transport. Packing is wrong, use the crate design potluck will can lead to the device is broken. Better buy a crate rather than the specific devices that cost hundreds of millions to be corrupt thereby. Make sure that you as an engineer supervising the process of packing the device on the sidelines of a flurry of you take care of the administration of the departure.

  1. Find a morbid condition of the device as early as possible

Sometimes the human senses sharp enough to figure out the distractions of a measuring instrument. For instance odor PCB burning, sparks, the sound of the drone, or heat outside the naturalness. It would be very good if this is support by special diagnostic devices to discover. This condition unnatural, though still I suggest human skills is more mainstream. Do not solely rely on the diagnostic device that also has limitations, no matter how powerful.

  1. Develop and implement countermeasures for recovery device condition

Ask a question Why as much as five times it actually teaches us to find the actual root cause of a problem, as it could be being simply the result of any other cause. So does not have to amount to five, depending on the conditions of your problems.

I have a unique story about this. A time when we perform calibration on-site at a company music equipment. We are busy for hours with our standard device conditions are unnatural. Until finally the conclusion while stating that the device should be "repatriate" and replace with another device that has the ability equivalent. However inadvertently their fellow technicians move the action in power supply of the other. And it turns out that the device be recover its accuracy.

After a discussion, it turns out that the device power supply is not connect to a power supply. To other devices (coincidentally in the room there are several stop contact and we use more than one). And a grounding power supply conditions are not great, condition. That turn out not to be realize also by technicians of the company. Because that room had was room for our administrative magic while being a work room calibration.

Total Productive Maintenance technical

There is one more story that might I share here. A technician state that a measurement error of frequency parameters of a signal generator belonging to customer "out of spec", where the standard device use is a frequency counter with the reference input of the Rubidium Standard.

But technical manager was suspicious because the signal generator is still in relatively new condition. After the measurement is repeat, the results show the same conclusion. Then the result of comparison is done using a frequency counter is different, but still with the same input reference (Rubidium).

Engineering of Total Productive Maintenance

Then ask to engineer a change of setting what is done compare to the previous condition. (e.g. conditions yesterday). From these questions, can finally found the root of the real problem. It turns out that the cable that connects between the Rubidium with frequency counters are already experiencing. A decline in performance, attentions is too large so that the input level on the "reference in" frequency counter doesn't quite qualify. So the accuracy of frequency counter so decrease.

Conclusions - Total Productive Maintenance

Finally, in closing this paper, overall there needs to be an overall device performance analysis, tailor to the life time of the device. It's good on every device list year purchases and life time (if any) match the manual or catalog. Give a particular sign to devices that are already in the area of "yellow".

Total Productive Maintenance: Which is already approaching the limit of the end of his life-time. For this device they should be distinguish.

For example, by monitoring the frequency drift do more often. Information about the life-time of the manufacturer are certainly not the only benchmark should be so. Because of the many other factors that affect the performance of a device. However, the manufacturer must have had his own argument to provide such data.


No comments:

Post a Comment