When Engineers’ Worlds Collide

I'll start this story off with some questions and I want each of you electrical engineering guys and gals to take a moment to think about these questions and how they relate to your past experiences.

How well do you work with mechanical engineers? Do you ever battle it out regarding design changes? Do other departments and managers seem to favor mechanical engineers' opinions over yours? Does anyone at your company understand your views or why you need to do things a certain way? Here's an example to explain what I mean. Has this happened to you?

The project is halfway through the schedule and it's a whirlwind in every department because tooling release is coming up. Mechanical engineers and drafters are putting final touches on their designs and tweaking the 3D CAD models and drawings. Tooling is being quoted. As for the electrical team, first PC boards have already arrived for initial testing. You had to get layouts and boards done to stay a step ahead of the project because there are some long lead times on several critical components and you needed to give your layouts, schematics, and board samples to UL.

It looks like everything is running fairly smoothly, until the lead mechanical project engineer walks up to your desk. After checking 3D CAD assemblies and SLS prototypes, it seems that one of the plastic ribs in the enclosure is interfering with one of your tall electrolytic capacitors by about 1mm. The request is for you to move the capacitor over. In turn, your request is to move the rib or shorten it. The duel begins, back and forth, managers get involved, panic mode sets in After all, tooling fabrication has started. Who will give in and change their design?

Now let me take you through a chain of events from both sides and you decide who should make the change. The mechanical team runs to their management and to the VP and convinces them why they can't change their enclosure. Besides, if they do, it will delay the tooling, which may delay the project. They just want you to move a cap. As things work out, they convinced the entire management team.

As an aside, here is what would have happened if the mechanical team had not convinced management. A drafter would take an extra hour to shorten the rib and change the drawing. The drafter would have created a work order and sent it to the tooling department. The tooling department would have quoted the change (which is quite simple) and reported a cost of about $500 and one day lead time. Not surprisingly, no one wants to pay and possibly delay the schedule a few days.

Now, here's the other side — what happens if the electrical engineer moves that capacitor over enough to clear the rib. Move it 5mm — if you are going to make a change, it's best to add a little margin so you don't have to do it again. How hard could this process be? You move the capacitor, which changes the board trace patterns. That causes interference with several other components. You now need to move those around, which causes more traces to change.

Before you know it, you are relaying out the entire PC board. Because of that, you need to modify the board tooling, which will cost several hundred dollars. Because the traces were changed substantially, you will need to have all new board samples assembled and sent to you from your assembly house. That will take two to three weeks. On top of that, you need to tell UL to stop their preliminary review because the board layouts and silk screens they have don't match the new board layout.

You will have to redo some of your testing because of the trace changes. You will need to recheck signal integrity. You are hoping your assembly house has enough components on hand from the first build and won't have to reorder anything. Yikes — what a nightmare just to move that capacitor a few millimeters.

Now, who should make the change? I know there are a lot of variables in this scenario. Perhaps you have gone through changes comparable to this. Have you had a similar situation where you were forced to be the one to make a change? Did you know that at the end, it would probably cost more and be riskier to the production schedule? Have you had to do battle with the mechanical group?

Related posts:

15 comments on “When Engineers’ Worlds Collide

  1. Dirceu
    July 31, 2013

    A rather silly question: A smaller capacitor would not solve the problem? It would be better than to redesign the board. In any case, I think these problems have a minor impact on small companies.

  2. Jason Bowden
    July 31, 2013

    Exactly, the point however, is how lack of understanding and company panic can cause unnecessary changes or changes that at the moment seem easier turn out to be harder and more costly.  I think many times the larger companies can be more difficult as too many hands in the fire and too many decision makers and not enoughdecisions based on the ones directl involved with the issue.

  3. David Maciel Silva
    July 31, 2013

    I dealt with it for over 10 years until it changed the way I develop products, tightening the timeline of electronic design, to have a sample at hand before developing the mechanical design. This may not always resolve at the last minute we have several surpressas this is common, most often made the modification on the PCB so you can somehow save the mechanical designer.

  4. David Maciel Silva
    July 31, 2013

    Always falls on the question “Hey did u here that caused this problem as the solution will cost me,” I believe that when this information is verified, either large or small businesses which will be more expensive?

    Any feedstock much labor.

  5. Vishal Prajapati
    August 1, 2013

    I think the mechnical change is better and easy to tweak than electronic change specially when design is in review stage of certain certification like UL. Changing board layout with 4 layers or more at this stage can create havoc. Because some RF circuits are very susceptible to layout changes and your time frame might get heavily disturbed to complete the next iteration.

  6. eafpres
    August 1, 2013

    I have worked in very small startups where there was one VP of everything and that was me so this would be a non-problem.  I worked at a bigger startup where everything was controlled by one head of engineering who would make the call; no recourse there either.  I worked at another medium sized startup that was populated mainly by veterans from much bigger companies, and they brought a lot of their behaviors with them, and it was fight tooth and nail all the time.  The EEs would constantly need changes, the MEs would go crazy, the Project Managers lost years off their lives, I'm sure.

    Later in my career I found the solution to this–true cross functional teams on every project, with a project manager that had both responsibility and authority.  We had mechanical, electrical, quality, purchasing, manufacturing, validation, sales, all on the team.  They had to sign off on things; no passive participants.  This worked very well in fact, but also stopped a lot of bickering later in projects becuase somebody in every department had been part of the decision process.

  7. Davidled
    August 5, 2013

    Once issue is occurred, all engineers are arguing each other to avoid their responsibility like defense attorney.  I think that the bottom line is cost issue. Regardless of which direction they go to, in the most time, they are going to direction which is less cost and less hard to modification. I realized many time that owner thought is slightly different from that of employee every aspects.

  8. Brad_Albing
    August 6, 2013

    @Maciel – To your point about saving the mechanical engineer, that was my experience back when I worked as an anlog design engineer. It seemed like our group always had to make changes to work around the mistakes that the mechanical guys had made. Similarly, we had to make changes to deal with problems that the software guys created.

  9. Brad_Albing
    August 6, 2013

    @DaeJ – You're right that the owner's point of view will likely be different than the employee's point of view. So what should happen, when all the engineers are arguing about who should make the change, is that the owner – or the project manager – should calmly look at the problem, assess the different possible solutions, and pick the best one. “Best one” should be the appropriate blend of lowest cost, most likely to succeed, fastest to implement, and most reliable.

  10. Brad_Albing
    August 6, 2013

    @VP – yes – for boards containing sensitive analog circuitry, RF circuitry, multiple layers, or that have already been submitted to a certifying agency (like UL), any changes can easily create big problems.

  11. Brad_Albing
    August 6, 2013

    @eafpres – I like it – that's the way things should be run.

  12. WKetel
    August 7, 2013

    The obvious solution is to get a shorter capacitor. That kind of change would solve the problem easily, if it were allowed. Some companies just use one supplier and those are the parts that we get stuck with, probably the reason it is that way is related to the purchasing agents nice Rollex watch. That stuff does happen. But often we find that a slightly more expensive part has a slightly different size, usually a bit smaller. 

    On the other side, better communications at an earlier stage could have prevented the whole thing, and not added cost or time.

  13. Josh Forgione
    August 24, 2013

    @eafpres – was it a nightmare to get drawings signed and released if every last person wasn't present? 

  14. Josh Forgione
    August 24, 2013

    It sounds like the only leader on the team is the project manager.  Is there a project engineer or systems engineer present who is responsible for managing interfaces and crossing disciplines? 


    It also sounds like there was nobody present to concisely lobby on the EE's behalf when the problem first arose.  Managers who don't understand technical do understand money, schedule, and risk – was there nobody present to speak that language from the EE perspective?

  15. eafpres
    August 25, 2013

    Hi Josh–we did a lot electronically, which helped.  In addition, the entire process was a stage gate process, and this was well embedded into the culture, so for the 4 key stage gate reviews people would show up.  All in in it was very effective. In addition the Project Manager, typically the mechanical engineer on the project, had the role of tracking down approvals etc. to keep schedule.  Worked pretty well.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.