General > General Technical Chat

Engineering code of conduct

<< < (8/10) > >>

bdunham7:

--- Quote from: NANDBlog on June 03, 2020, 08:29:47 am ---Thats a lot of reply. I will give some background.
The company I work for typically deals with ATEX/IECEX equipment. We design intrinsically safe electronics devices, that go into areas where explosive atmospheres are likely. When we design stuff here, we follow the law, we follow the IEC rules, we follow the national standards, and everything is checked and double checked by a notified body. Otherwise I would've left a long time ago. I personally oversee a lot of these projects, and designed electronics here.

There are a few projects, that dont require this. So this project doesn't go into an ATEX zone. But it is an electrical box that is bolted on an IBC tank, controlling 230V/380V heating equipment on that tank. Few KW goes here. The tanks can carry anything from milk, to paints, fuels, chemicals. Anything liquid that would need heating. We made a GPS tracker, and we connect a 3rd party heating controller to the cloud. It needs new wiring, 230V and low voltage. Setting the tank on fire is very very unlikely, but even so, I wouldnt take any chances. Even simple things, like missing ferrules, or having insufficient isolation thickness on a cable.

We are not going to do these installations/modifications. So there is the problem, that we dont even have the expertise to make electrical drawings. The engineers here, we make PCBs, not electrical boxes. We dont even know, which standard to bring up. And management's mentality is: If the client havent asked for a standard, it is not required.

--- End quote ---

Thanks for the detail.  It's an interesting dilemma and I can't advise you as to how your laws work on this, but just generally it appears that this electrical box ends up being integrated into an end product that is at least nominally portable, right?  I'm assuming IBC tanks aren't used in permanent installations, someone correct me if I'm wrong. So the building electrical codes wouldn't really apply, but there would be a bunch of other ones that now would.  You mentioned the obvious, CE, but the whole notion of using it for such a wide variety of uses would really have me worried.  Even jellybean ICs often have restrictions on using them in aviation, medical, military, mining, critical process control, etc.  You need an additional certification (and $$$ more) to get essentially the same part but for that application.  I don't know how you go about designing and certifying even a simple product for use in that many areas. 

nctnico:

--- Quote from: NANDBlog on June 03, 2020, 08:29:47 am ---We are not going to do these installations/modifications. So there is the problem, that we dont even have the expertise to make electrical drawings. The engineers here, we make PCBs, not electrical boxes. We dont even know, which standard to bring up. And management's mentality is: If the client havent asked for a standard, it is not required.

--- End quote ---
Perhaps it is a test to see how resourceful the engineering department actually is. Engineers are supposed to come up with solutions. For starters: figure out which standards apply. Look on internet for pointers, call a few companies. My simple solution for these kind of situations is: acquire the knowledge to do the job. An engineer is never done learning.

Edit: Gregg's solutionm (outsourcing) is also a good one but you'll still need to know some basic regulations / requirements to verify the solutions offered is the right one.

SiliconWizard:

--- Quote from: NANDBlog on June 03, 2020, 08:40:25 am ---
--- Quote from: NANDBlog on June 03, 2020, 08:29:47 am --- And management's mentality is: If the client havent asked for a standard, it is not required.

--- End quote ---
Actually, I just had an idea. We need CE on the product, that means, that it needs to confirm with the LVD.
"About the low voltage directive (LVD)
The LVD covers health and safety risks on electrical equipment operating with an input or output voltage of between

50 and 1000 V for alternating current
75 and 1500 V for direct current"

We just need to have someone state that this is a requirement, and start from there. It will be revealed, that we don't have the expertise for that.

--- End quote ---

Well, obviously if it has to comply with the LVD (and all the more if you don't have much in-house expertise on that/and thus internal knowledge and appropriate means of testing), you are STRONGLY adivsed to have your device tested in an external lab that will analyze the design and test it and eventually give its compliance approval or not. If your management half knows how to deal with CE marking, they will likely tell you that. And if you manage to get something to pass the conformance tests, although not a 100% guarantee that it is well designed, it will be at least a guarantee that it is compliant with the required standards. Now if you never manage (at least in reasonable time) to get something to pass, then OTOH you'll have proven you're not qualified for that.

Of course never ACCEPT to release a design if your management doesn't want to pay for external conformance tests.

But as long as some engineering task doesn't formally require a specific qualification that you don't have, I don't think you can really refuse the job. Just make it clear, as others have said, that you have limited knowledge and experience for this, and that using an external lab for compliance testing won't be optional. That will put all the responsibility in the hands of your manager(s).

0culus:

--- Quote from: engrguy42 on June 02, 2020, 11:49:53 pm ---Geez, guys, don't you think you're going a bit overboard?   :-DD

Signing petitions??? Contacting lawyers?? Documenting everything??

Don't you think you should see if there's a problem first?

--- End quote ---

From a non-life-critical systems perspective, often the most effective course I've seen used in situations where you're asked to do something you know isn't going to work or will cause more problems is to explain why it won't work (in writing) and (in writing) ask the person requesting the work to provide (in writing) that they are taking responsibility for it. Usually you don't hear another word from them about it. Office politics...  :palm:

tggzzz:

--- Quote from: 0culus on June 03, 2020, 06:29:36 pm ---
--- Quote from: engrguy42 on June 02, 2020, 11:49:53 pm ---Geez, guys, don't you think you're going a bit overboard?   :-DD

Signing petitions??? Contacting lawyers?? Documenting everything??

Don't you think you should see if there's a problem first?

--- End quote ---

From a non-life-critical systems perspective, often the most effective course I've seen used in situations where you're asked to do something you know isn't going to work or will cause more problems is to explain why it won't work (in writing) and (in writing) ask the person requesting the work to provide (in writing) that they are taking responsibility for it. Usually you don't hear another word from them about it. Office politics...  :palm:

--- End quote ---

Yes.

That's a variant of the "requesting written guidance" technique I outlined earlier - but more passive-aggressive and therefore better suited to the UK civil service :)

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod