Like you stated, that's kind of my experience too. Fortunately, industrial equipment is pretty robust and failures are usually rare. Repairs are expensive and difficult to arrange. Some devices allow for component replacement if you can find them but otherwise it's off to the trash bin. Depending on the equipment, the stuff usually becomes functionally obsolete before it fails outright.

AVOID BUYING JUNK

and you know what I mean; things that are substantially cheaper are often junk, since one real way to cut costs is to eliminate steps and components along the way. The corollary is to buy from reputable suppliers.

The world's standard of living has drastically improved from the time when my first computer, Z-80 based, cost $2000, while I can now get computer that is literally millions of times faster, has millions of times more memory, and runs way more stuff than that original. We've gotten there through a combination of wage increases, size reduction, and automation. This ultimately makes it uneconomical to repair things, even if you did, or could do, repairs yourself.

And this does not only apply to electronics. Consider when was the last time you saw a shoe repair store? If the cheapest shoes you can buy are the $20 shoes on sale at Costco, but it costs probably 2 hrs of equivalent labor to repair such a shoe, wouldn't it be more economical to buy a new pair?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

The problem is often SMT or surface mounted technology. If a part goes wrong it is often cheaper to throw away all the electronic that to trace signals on a board where the traces are only 8 microns wide and the leads are small. Replacing the parts is difficult if you don't know how. You will likely do more harm than good.

To do SMT board repair requires a significant investment in tools and training.

Another problem has been ROHS. Lead free solder or tin solder will grow whiskers that may short out leads or traces. Thank the politicians and regulators for that. Do a search for tin whiskers.

Peter Nachtwey
Delta Computer Systems
http://www.deltamotion.com
http://forum.deltamotion.com/

Quote:

Fortunately, industrial equipment is pretty robust and failures are usually rare.

I wish... Have a machine from abroad with one main panel and two remote panels. Each panel features a Siemens TP700. 2 of 3 are now dead. One never turned on, the other died after running for a week or so. Quality component? I though so.. at least before this. Appears genuine. Expensive too, $1000 each new. Of course by the time that all this revealed itself, any warranty is long gone.

I could even do some basic diagnostics like checking test points to identify the faulty PCB and order a replacement, but no. They don't do it that way. (Does anyone? I would surely buy from a manufacturer that does...)

So, I don't know... I guess it is off to ebay for donor units.

I have no experience with the Siemens unit so I can't offer any more than an uninformed opinion, but that sounds strange to me. I would first verify the supplied power is within limits. If so, then I guess I'd say they're not worth spit. I have lots of Allen-Bradley units, some are more than 20 years old, and I can't remember ever replacing one. We have some A-D Cmore units, and we've had to replace a couple after a few years, but still not an unacceptably high failure rate.

But, you say that one never turned on; why wasn't that flagged as a warranty return? And note that for a company like Siemens, which is relatively reputable, they might still honor their warranties and possibly send someone to check it out.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

The manufacturers pretty much refuse to provide information, so you just have to fix it to spite them.
I don't attempt to fix small surface mount microprocessor , the tracks are too tiny and components too complex.

Unless it's something obvious like a blown electrolytic, you'd be hard pressed to have the resources to adequately test a complex board to begin with. Moreover, repair comes with a certain level of risk of damaging the board or other components when removing and replacing a complex part.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

Quote:

I would first verify the supplied power is within limits

That would be my plan as well. Start from the power generation and follow it to the CPUs. Then ???, then profit. But I just won't have the time for it in this lifetime. I would love to pay someone, but I have not come across a bona fide specialist yet. Not for the lack of trying. Anybody want to recommend a friend ;) ?

Quote:

The manufacturers pretty much refuse to provide information

The only thing that has ever worked for me, is to lure it out of the salesman *before* you pay.

Your 2nd step, if you are still insistent on trying to repair it; is look for blown capacitors; that seems to be the favorite failure mode of many of my stuff:
> A/C compressor starter cap
> Circulation fan starter cap on heater (in two different houses)
> controller board for $$$ refrigerator
> graphics card for PC

If the boards are relatively recent vintage, the electrolytic caps all have nice score marks on the top, so it's pretty obvious if they've died a horrible death.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm

Variant on the same theme that I'm seeing a bit of at the moment:

The reservoir capacitors in the power supplies in a number of our 30 year old control systems have all reached the point where they can't quite sustain the dc link voltage all the way through a half-cycle. All of a sudden, the microprocessor gets a Reset signal fifty or sixty times a second - with entirely predictable results.

Once upon a time, I would never have dreamt of telling people to put a scope across their 5V power supply.

A.