Episode #22: How To Mitigate Risks In Mechanics

Contact Valin today for more information at (855) 737-4716, or fill out our online form.
 

The Motion Control Show

 

I would like to thank my good friend and colleague, Ben Furnish, and all the great mechanical engineers that he works with over there in Irwin, Pennsylvania, for teaching me about the ways to mitigate risk when sizing, selecting and designing mechanics.

I have a question for you: what is more important to you, your time or your money? Or how about the question: what is more important to your company, your time or your money? I ask customers this question all the time because it really depends upon the business model. Are they designing just one? Are they designing a hundred or thousands? Are they wanting to design in the inexpensive stuff that they then must spend a lot of engineering time on to design in? Or do they want the more expensive stuff that is going to minimize the amount of the engineering time? It really just depends upon their priorities. So, I asked that question. I am curious as to how you answer this poll up here. If you like what we are talking about today, follow my #MotionControlShow and check out my website.

We are talking about mitigating risks when selecting mechanics. There are a few different areas: ownership, technology, specification, connectivity and support that we can consider. The ownership: Who has the liability? Who supports failures? Who responds to the customer's needs? The technology implications: When to select which technology? Technology strengths? How to balance the cost versus performance? Specifications: How do I understand the various marketing speak of manufacturers? How do the specifications affect application performance? The connectivity: Will the parts mount with minimal effort? Will communication between electronics be seamless? The ease of setup? Support: Is there a local contact wherever your machine ships to support it? Will the on-time delivery hold your project up? When you call, do you get answers? Those are basically the topics we are going to be talking about.

Make or buy pros and cons: Let us talk about making your own mechanics. The pros: you can design it exactly the way that you want to; you can design it with a particular cost-effective scenario and with high-volume. The cons: the functional and life testing of the design; managing the development efforts otherwise it is going to consume your engineering resources; testing; the inventory. Many times, the total cost of ownership is higher than buying. How about buying it? The pros are your reducing your functionality risks, you have a proven warranty, it is more modular, and there is a known and fixed product cost which is typically less than the in-house design. The cons though are your typically buying features not needed, the size may be slightly outside what you are looking for, and then you have to worry about somebody else having control of the product obsolescence.

There are some ways to mitigate risks if you are making it yourself. You can use the Failure Mode and Effect Analysis (FMEA) tool which I will show on the next slide. You need to perform the Design For Manufacturability (DFM) reviews by bringing your teams together and be able to go through all that. You can perform lifecycle testing. You can verify the manufacturing of the tolerances and the fits and then you need to run alpha units to verify the performance. Here's a sample FMEA form you could check out. There are ways to mitigate risk if you are buying a product: you can request the products lead time and it's on-time delivery percentage; you can ask whether the supplier can provide the validation of specifications from laser interferometers and torque graphs; you can ask what the products PPM (their parts per million of products that are returned) for problems which may not be available for a few years if it's a new product; and what is the typical life cycle of the products that you are buying?

Here are some technology implications: selecting the right components, which is important if you are making or buying; learn the technology basics, the pros and cons of each of the products that you are using in it; use the right technology; learn the detailed specifications of the technology to be able to determine their performance. You must learn a lot if you are making it from a component level.

For instance, here are several different linear drive trains for changing rotary motion to linear. Then there are different components if you are using a direct drive linear drive train like linear motors and piezo as a couple of examples. Then there are the linear bearing technologies. And there are linear feedback technologies. You must pick from the different pros and cons of all these different technologies.

Understanding component or product specifications is the key to determining your application performance. However, the standards are lacking on many of the technologies and products. We talked about specsmanship in a previous episode. It really comes down to the life of the product, the mechanical life of the product with its bearing life curves if you are talking about bearings for instance. The in-depth detailed understanding of the components of products is needed to make an informed decision. Remember that the system level or multi-axis performance is not the same as the component specifications when you bring a bunch of components together. You can't necessarily rely on just the component specifications.

Linear positioning products define the accuracy in one of two ways: one is based on the accuracy of the drive components; the other one is based on the accuracy of the position or assembly with a fixed point of measurement. The Point of Measurement (PoM) is usually centered left to right and 35 millimeters above the mounting surface. But neither method tells the whole story. The component specification is the least representative, but neither method considers the Point of Measurement of your application. So, to determine the accuracy of the position, or at your Point of Measurement, you need to know your PoM and as well as the angular errors of the positioner.

There is the pitch, there is the roll, and there is the yaw. When translated to the Point of Measurement, they affect the positional accuracy. Let us look at an example. With the load mounted centered above the bearings, the primary error contributor is pitch which will affect the accuracy in the X-direction. As it pitches, it is going to cause some angular error there. With the load mounted off to the side of the bearings, the primary error contributing factors are yaw and roll. Yaw will cause the error in the X-direction and roll will cause some in the Z-direction. Remember as it shifts you are going to get some angular error in different directions.

The Point of Measurement accuracy is complicated to calculate because the angular errors vary from positioner to positioner. Angular errors are not linear errors and are normally estimated. They are very difficult to measure over the course of travel. Determining multi-axis accuracy is much more complex. I have taken multiple axes, given the specifications of repeatability and accuracy of the different actuators, and put them together in XYZ configurations. You must consider varying bearing deflection and the deflection of the actuator itself. It gets to be very complex.

How you should mitigate risk is to find ways in your process to use the repeatability of the system instead of the accuracy. That goes back to a previous episode. Mapping the process works well. Work with a professional motion supplier when you can. Understanding specifications or product performance is a key to a successful application. Knowing everything about all the technologies is not possible. Just recognize you need help. Work closely with the manufacturers. Get training and work with companies that have expertise in system-based solutions if you really want to know what is going on.

Let us talk about the interconnectivity between the mechanical components. There are the tolerance levels and the repeatable mounting for which you could use dowel pins if you must know about the XY orthogonality. You might have to worry about the misalignment which causes premature failure whether from the motor to screw, the screw to the bearings, or the brake to the screw. Then you might wonder whether it is field repairable or whether it is easily assembled and disassembled in the field, whether it is the brakes, the parallel motor mounts, or the motor replacement.

You might want to know about the electrical components. Remember in a previous episode I talked about looking holistically at the system. Do not just look at the components and make decisions. You want to take the holistic approach to the system. You must worry about the electrical components even though you are just looking at the mechanics. You might want to worry about the motor feedback to drive, the signal frequency and its quality, and the noise and bandwidth. You might want to worry about the drive to controller communications, whether it is the motion bus or its communication speed. This goes back to the electrical noise presentations I made back in Episodes 9, 10 and 11.

How to mitigate some of these problems: approval drawings, minimize supplier variety, single source where possible, partner with a local motion expert who are people who know motion and know about these problems.

Some questions to ask: If a machine goes down, can a supplier support your needs locally, nationally and globally? How about what type of technical support is available in different time zones and overseas? Does your motion supplier understand a complete motion system? I get calls all the time for motors that are making noises where they think something is wrong with the motor. It turns out usually that there is something wrong in the programming. It is the software because it is the tuning. Or someone might think that it is a problem with the mechanics where it is actually a problem with the controller or vice versa. You have to look at the whole system when you are troubleshooting. Do you have that expertise? Or are you working with somebody who does?

Some tips on mitigating the risk: Partner with a name-brand company. You pay for what you get; most people don't believe this, but I have seen that time and time again where people try to cheap out and then they end up paying more in the support and the problems that happen down the road. Ask for references; ask people that you know that have worked with companies and see if they've been pleased or not. Develop relationships with specific people; then you learn how to trust them and trust their knowledge and what they can and can't do.

In summary, risks to consider are ownership, technology implications, specifications, connectivity and support.

I am Corey Foster of Valin Corporation. Follow my #MotionControlShow. Check out this website. There is a lot of good material in the episodes prior to this that I referred to. There are more coming down the road. I hope you learned something.

If you have any questions or are just looking for some help, we're happy to discuss your application with you.  Reach out to us at (855) 737-4716 or fill out our online form.