The Secrets to Successful Startups | Automation.com

The Secrets to Successful Startups

October 182015
The Secrets to Successful Startups

By Patrick Supanich, Contributing Author

This article was originally published in the 2015 Summer Edition of PULSE.

Introduction

Equipment startups are like the "weeder classes” in college. Some people make it, some people don't. During the startup is where you establish your reputation as an engineer. The engineer who succeeds will come away with credibility, as well as stronger client and employer relationships. The ones that do not succeed will be asked to leave and could potentially find themselves looking for alternative employment.

The primary perspective of this article involves shutting down an existing manufacturing line; one that is producing a product and profit for a client. In my opinion, this is the most high risk shutdown scenario. It often involves control systems upgrade, new equipment, retrofits to existing processes, as well as catching up on deferred maintenance. There a lot of people competing for scarce downtime resources. As a controls engineer you are usually the last one on the line. Whether you like it or not, you are affected when others are not prepared. Therefore, there is little room for error. If things go poorly, leadership will be looking over YOUR shoulder.

In a startup there are a lot of variables that you cannot control and there will be surprises. This is why you need to control the variables that you can control. Startups are no laughing matter, but for the prepared engineer they can be a rewarding experience.

Believe it or not, depending on your team, covering all bases and being as prepared as possible can be a little unconventional. You may even get pushback from people who want to "leave it to the startup." My advice is if it can be practically addressed prior to the startup, you need to address it before the startup. If something is not practical to address then be sure there is a plan in place and time scheduled for it during the startup.

Before the Shutdown

Study the Existing System

 A lot of existing systems have been patched together over the years, and they likely have undocumented equipment, wiring and software. They may have patched in connections to business or enterprise information systems. You need to develop a mastery of the existing machine before shutdown. Dedicate time in the field prior to shutdown to identify panels, dig inside them, and talk to maintenance people and the operators. This is time well spent and will improve your comfort level and knowledge. You will, in all likelihood, identify a number of surprises that you can address now rather than later.

Hardware Factory Acceptance Test (FAT)

There is no excuse for not completing a thorough test of the panel and associated hardware prior to arriving onsite. Any major hardware such as PLCs and drives should be fully configured and communicating. Networks between panels should be temporarily connected and communication established. Your time at startup should be about installing the equipment, debugging the software in real-world conditions and addressing the unexpected issues. It should not involve tasks that could have been done before the line was shut down. So make a detailed list of tasks and complete it in advance. This may seem like obvious advice, but failure to perform a thorough FAT is all too common.

Software Factory Acceptance Test (FAT)

Software is another critical component of any system. It is also the most common point of failure. Too often engineers go onsite with incomplete and untested software. During a startup you do not have time to debug your whole program, let alone rewrite it. In the worst cases, this procrastination is because the engineer does not understand how the system needs to work and they decide they will figure it out when they get there. The most egregious instance is when the system requires a level of theoretical expertise that the engineer does not possess, and they do not admit it until it becomes obvious in the field.

You eventually will need to master the system you are programming as well as its theoretical underpinnings. Based on this requirement, you must decide where you would rather learn the system; in the office with the support of your peers, or in the pressure cooker during the startup. If you lack the skill or knowledge, do everyone a favor and let them know before the machine is shut down. Nobody can be an expert at everything. It’s better to admit your shortcomings – others will respect your honesty and can offer supportive mentoring. The alternative will cost you your reputation and maybe your job. There is really no choice here.

Be sure to involve the client in the review process, even if they do not ask for it. Eventually, they will want to review both the software and interface. You can choose the informal route where they wander up and take a look during the shutdown, or you can choose a formal FAT where they sign off on the software prior to system shutdown. Based on my experience, I can guarantee that the client will have changes and often they will be major changes. In all situations, ask yourself the question, “What approach will lead to the best outcome?”

If you have followed the above advice, you will likely be in good shape. That does not mean you should be complacent. Have a backup plan for the worst case. This would include having spares of long lead time items. The plant will likely stock these items anyway so order them up front rather than waiting until after the system is back up and running.

General Preparation

You will hit the ground running when you arrive on site. Here are a few things to do before you arrive:

Even if it’s not in your job description, get a set of tools. At a minimum the trades people will respect you more for it (scuff them up a bit to make them look used or you may get some grief). This should include a quality meter, as well as any calibration or loop check hardware. You should have a set of long leads with clips, a set of electrically safe screwdrivers, standard and metric Allen wrenches, wire strippers and cable cutters. Finally, do not assume that the electricians will have every specialty connection tool or diagnostic device that is needed, especially when it comes to networks. Make a list of these tools and confirm that they have them or get them yourself.

Make sure the electricians have the latest prints and you do a walkthrough of the job with them. The plant will know who their best guys are. Push for that team and be sure your job is staffed well for the undertaking. Your input and oversight, while it may be informal, can save you a lot of headaches later. Remember if the Project Manager fails to resource the job well, the timeline will suffer. Any experienced Controls Engineer knows that it will probably come out of their budget. Do yourself a favor and make sure you are contributing in areas where you hold the expertise. This involvement will improve your standing with the whole team.

Small, but Critical Factors

Prior to your arrival, be sure you will have Internet access in the plant. Understand the plant’s Personal Protective Equipment requirements and come to the site with the necessary gear. If possible, arrange for a place to work away from the plant floor. When you are on the plant floor you should have a place to sit and work. Don't stand with your laptop on a garbage can or the machine. Worse yet, don’t hold your laptop in your arm all day. Again, these suggestions may seem obvious, but I’ve seen and experienced these situations all too often. Where ever possible, wireless connectivity to the intelligent devices is extremely valuable, as is a dual monitor.

Anyone who has been involved on startups knows that long hours are common. You need to judge when you are causing more harm than good. Problems that seem intractable can suddenly have a solution after a good night’s sleep. A double shift may be tolerable with proper breaks but longer shifts than that are impractical and counterproductive. If the startup expects 24-hour days, the person on the overnight shift is best given low level tasks unless they are seasoned in night shift work. Oh, and not to sound like your Mom, but eat well and take regular breaks.

Keep moving forward

A lot of time can be squandered during the startup by waiting. It’s likely the Project Manager on the job will not be running around making sure everyone is productive at all times. They will likely just be checking in periodically or you may be on a job with no real Project Manager at all.  Don’t be complacent when things have stalled. Solve the issues you can, especially when they involve people under your direction. Alert the Project Manager immediately to areas that are holding you up and push that they be solved. In other words, take control. While you are idle, you should look for small things that can be done. Maybe it’s documenting software, adding a requested feature, panel cleanup, etc. There are always checklist items. You might as well address them during the otherwise unproductive times.

If you find you are in over your head, ask for help

If you were truly prepared you should not get to this point. Admittedly, however, it can happen to the best of us. I will not get into the excusable causes here. The most critical time to ask for help is if you are in over your head technically. This is because now you are in a "random, try things mode.” Here you can squander a lot of time, lose your clients confidence, and also destroy product or equipment. In the end, even if you do get it working, you probably will not know why. So call in the big guns when you need them, otherwise the client will.

If you have extra time, deliver extra value

What??!! A startup with time to spare?? If you are prepared, you may find yourself in this situation more often than not. Rather than rest on your laurels, this is the time to impress. Add features such as better diagnostics, more documentation to the program, or automate the startup sequence. Provide extra training to maintenance people and operators. This attention to the end client, whether internal or external, will be very much appreciated. This is not to say that if you are a paid consultant, you give away all your value. You need to use your discretion, but it’s a great opportunity to over-deliver to your valued clients.

In Summary

Startups are a rite of passage for any controls engineer. There will be ups and downs and a lot of surprises. However they can also be very rewarding experiences if you are prepared and remain diligent throughout. Your reputation is often built on these experiences so take them seriously. Always remember that some of these machines make more money in an hour than you do in a year. Your client trusts you with their means of profit, so don’t take it lightly and don’t be the reason for failure.

About the Author

Patrick Supanich is a BSEE graduate of Michigan Technological University. He chose an automation career after graduation, when he went to work for 3M Company as a Systems Engineer. After 6 years at 3M, Patrick took a year-long sabbatical to travel the world. Upon his return, he founded and built a successful systems integration office for Tegron. After 8 years with Tegron, Patrick joined Automation.com and Automationtechies. At Automation.com he served as the Vice President of Sales and Operations. After the acquisition of Automation.com by ISA in late 2014, Pat remained with Automationtechies, a full-service recruiting and contract staffing company focused on automation professionals. In this role he balances his time between automation consulting, recruiting and business operations.

MORE ARTICLES

VIEW ALL

RELATED