General > General Technical Chat
Right to repair, my problem with it
<< < (8/39) > >>
bdunham7:

--- Quote from: pqass on July 18, 2021, 06:27:52 pm ---"...ask Intersil to create..."  or mark chips with my proprietary part numbers all sounds like collusion to me as they seem to rely on IP that is already mostly developed by the OEM and is being incorporated in other chips they sell to distributors/general public. So I, a third party, should not be barred from buying same (subject to one-time, license, and minimum quantity charges).

--- End quote ---

I disagree with robint91, but you are badly misstating his argument here.  He wasn't referring to re-marking, but to a contractually supplied custom chip where the customer holds the IP.  Applying FRAND to that is ridiculous.


--- Quote ---Right to repair is mostly about stopping stupid shit like re-labeling and deliberate obfuscation. 

--- End quote ---

Not at all.  There are two avenues available to protect your IP--the copyright/patent route and the proprietary/trade secret route.  Both are legitimate.  You can design a product with magic module that has all the IC labels sanded off and then the whole thing potted in cement if you like--there's no R2R law I've ever seen that would prohibit that.  What R2R might require is that if you sell a product containing the magic potted module to an end user that is a consumer,  you or your dealer supply that module to the consumer or an independent repair facility on non-discriminatory terms.

T3sl4co1l:

--- Quote from: robint91 on July 18, 2021, 12:53:59 pm ---I think that people are overlooking a few aspects of the whole right to repair legislation. I see two big asterisks in the whole discussion.

The first one is "product conformity" and how much repair can deviate from the original design. I know that it takes very very little modifications to a design to invalidate any EMC/Intended radiator/Safety/... test for FCC/UL and CE. The producer has to be sure that every product he makes is exactly the same as the one which is tested. He is liable if it doesn't adhere to the same standards. So they have meticulously create internal assembly guidelines on how to create the exact same product, how to open/close the enclosure, how to apply shielding tape,... Just to ensure "product conformity".

How can somebody without proper training do this correctly?
--- End quote ---

Exactly.  This is another excellent point!

How can repair people figure these things out, correctly, without instruction from the manufacturers?  It's a crap-shoot at worst, and best-practices at best.

If manufacturers responded to certain detailed questions about repair, the repairs could be done in a manufacturer-approved way.

This isn't a denial-of-service thing.  They're big, they have plenty of time to respond to questions.  More to the point, repair people only need one answer, and they can share it among themselves (e.g. via Rossmann's wiki).

That communication could take many routes: it could be ad hoc, everyone for themselves; it could be structured, through designated channels e.g. a few people in the community volunteer as liaison; it could be formal via an industry group/consortium that collects community questions, sorted by priority, relevance, etc.; lots of possibilities.

The point is to have anything at all, and not be just guessing all the time.

Nor is this an IP thing.  Specific, pointed questions about component availability or replacement, reveal absolutely nothing about the IP.  There's very little protectable about PCBs anyway, and in the recent past, no one had any problem with distributing schematics and even board and layout photos.  (The schematic might be protected by patent if novel, and layout is protected by copyright either way, IIRC.  Doesn't matter how you publish it, or in what forms.)

So you make an excellent point.  I'm not sure how you came to the conclusion that it works against RtR.  Such a conclusion almost seems it has to be a case of, looking at the current state of industry and saying "well this is terrible, how could it ever improve, I'm not going to do anything to fix it!"



--- Quote ---And this training is not about if they can replace a SMD resistor or do rework on super small BGA component, that doesn't matter. The only thing that matters if the repairs happens accordingly to the same way a new product is assembled. A repair guy can be very skilled at dissembling and reassembling a laptop, but the main idea here is, did he used the same methodology as they did in the production line? Must that repair guy guarantee "product conformity"? Or is the original producer still liable for the product conformity when a third-party repair has happened?

--- End quote ---

How can an auto repair guarantee that emissions control systems work as intended?

Well, there are means of testing this, but before these were available in the 90s, what did they do?  Well, they didn't do anything, they just replaced parts more or less to spec and that was that.  Best practices.  The same works here.

Maybe they got a bum part sometimes, or didn't install it correctly, and you had some stinky cars on the road.  Maybe they were later discovered, and fined, or repaired.  Responsibility goes to the owner first, then repairer, then manufacturer, as noted by others.

Have you had a product returned by FCC complaint?  Are you familiar with the procedure?

In principle, a sale of 1 should be tested as much as anything else.  But there's The Law, and there's "The Law".  In practice, you can have... thousands, hundreds of thousands even, of products out in the wild, without testing.  Maybe they even generate complaints, maybe they pass after all whether by good luck or intent.  I've seen it before.  The first step is typically a licensed user making a complaint -- the FCC has little time to wardrive anymore (they're more concerned with [not] regulating telecom these days).  A C&D is sent to the user.  On repeated complaint, the user may be investigated or fined.  Only when a pattern of behavior is found, does the manufacturer get notified, and then a recall, or fines or other remediation, might be used.


As for actual repairs, stuff like pictured, doesn't bother me in the slightest.  Antennas?  Perhaps.  The wires are low to the ground plane, hardly different from the traces they're extending.  They're short, only relevant in the GHz.  (I would go a bit further than pqass, and say no, it won't have any effect at AM BCB at all.)  Perhaps this would raise the radiated floor a few dB in the 1-10GHz band.  It's not even obvious if they'd do anything at all -- there might be an absorber or shield over the location in question.


You're also missing something about the nature of this kind of repair -- many of them are "heroic", patching traces from water damage, shit like that.  No one's going to guarantee that kind of work.  Well, maybe they do, I don't know.  It would be foolish to give a guarantee on something that's cursed with contamination and chronic corrosion.  If it can't be cleaned up, and it's going to keep spreading, just get out of there -- back up your files and get a new machine.

The purpose of such efforts is to recover anything.  Remember, Apple thinks YOUR FILES DON'T EXIST ANYMORE if you so much as break a key off your keyboard, let alone any kind of damage like this.  If your alternative, as a customer, is to literally throw away everything on that machine; or pay much less than the cost of a new machine, to a maybe un-reputable (certainly not manufacturer-approved) service -- what are you going to do?  It's an insane, retarded situation to be put in.  And it's an entirely, intentionally constructed situation.  The manufacturers MEAN to do that to you.

Probably, a lot of people do continue to use machines after such treatment and repairs, anyway.  That's fine, they do so at their own risk.  Maybe they take the lesson and use NAS backups (or backups of any kind at all).  Maybe they follow through on the action and actually get a new machine.  Whatever -- it's on them, the repairer has done their due diligence.


One more point about EMC.  It can fail spontaneously over time.  The manufacturer tested one article and put it into production.  What happens to the components in operation, who knows.  In particular, X1 type filter caps are made with very thin materials, and are exposed to mains surges; over time, they self-heal, meaning some of the material breaks down and vaporizes, "healing" the wound.  After enough time, whole sections of electrode can be "healed" away, significantly reducing the capacitance.  They also corrode: the metal layer is so thin, it can be fully oxidized by exposure to oxygen or moisture -- which diffuse in through the plastic slowly but surely.  So, they even have a shelf life.  All in all, they can end up with very little value after a decade or so of service.  You'd think this should generate huge amounts of reports, but, eh?  Just seems to be part of modern life, having awful reception in the MW to SW bands.

Also, besides capacitors, I suppose stuff like EMI tape might fatigue and break, if subjected to flexing; or EMI springs may not wipe cleanly, or get clogged with gunk.  (Both of which can be repaired, if a tech is instructed to look for such things!)



--- Quote ---Or do we want year checkups of all our electronic device, the same way we do for our Car's in the annual vehicle safety inspection? A yearly EMC inspection?

--- End quote ---

It's funny because that would actually be pretty feasible... get a LISN for mains checks, a CDN or cable clamp for peripherals/telecom lines, and a TEM cell for radiated.  Hook it to a spec -- pretty cheap these days for the couple GHz you need -- and you can do these checks in under an hour.  Nothing like the certified lab experience of course, but if you have a baseline, and you have a device that exceeds it considerably, well that's a good hint that something might be wrong!

I wonder if I could make a sales case for that... heh...

The funny part is, it's literally just making more work for the end user.  So, I think not.  Unless the FCC requires it, which I've seen nothing to hint at that.  So, possible, but highly unlikely. :)



--- Quote ---Apple invested in that part, why should Apple sell you that part?
--- End quote ---

Well, "invested" is maybe an overstatement.  But sure, they customized it, and to the extent of NRE and minimum order quantities, you could say that.

Why should they sell it?  They shouldn't, and Rossmann has noted this before.  You can hold onto all the chips you want.  You're an asshole, but you have the right to be an asshole in this country.

Securing parts supply is probably a pipe dream, but it's still one of extreme importance to board-level repair, so he will always be bringing it up.



--- Quote ---That this happens is just NORMAL way of doing business. I design and manufacture boards for a customer, and a customer of my costumer comes directly to me for a replacement board, should I sell it to him? Is it my place to sell a board where I don't have the intellectual property anymore? If I did this and my customer knows about it, I would a serious court case against me, with a very slim chance in winning. Same applies to Intersil. If the original customer, which hold the IP, doesn't want to sell the board, then it is too bad I can't help you.

If I will be legally forced to sell those boards to customers the value of doing design work and contract manufacturing will drop. I think this will render most "Do not compete" void.
--- End quote ---

You're extrapolating pretty far down one track here...

First off, if such a law were enacted -- it would either have to grandfather in prior terms like you're working off of above, or provide some graceful means, or remediation, to break the clauses relevant to the above.

So, going forward, such agreements would simply be made with different terms, allowing for whatever that law would cover.

More likely, such a law would target the product OEM, not their suppliers.  They ordered custom parts, so they should retain extra stock of them, or make them available as long as they support the product themselves, or retain some amount of stock thereafter; whatever.  It needn't be the 10 year sunset period that auto parts have.  It needn't be much of anything at all, just so long as the parts are available at all, through some means other than desoldering junked boards.

They can even turn a profit on them.  Hefty one at that, it seems.  Or they can set the price sky high, and even write press releases saying We DiD tHe ThInG gUyS, reLaX!!11  They can continue to, you know, be an asshole, as is their right.

Anything like that, would be a step in the right direction.  Right now, it's outright impossible, they refuse to release any stock.  At least then, it would be a matter of negotiating price.

Tim
bsfeechannel:

--- Quote from: robint91 on July 18, 2021, 06:16:44 pm ---But that makes you, the repair guy, liable for the conformaty. Don't know if you want that. The manufacturer doesn't grant you conformaty. Without that right to repair is only a hollow vessel.

--- End quote ---

Those who are demanding the right to repair don't seem to have a problem with that liability. Quite the opposite. They're confident that the product will be back to the same (or even better) performance it had before it failed.
coppercone2:
you know you are supposed to repair those things in the proper manner, if there was serious repair shops they would have the best in PACE tools and things like that do do board level repair properly. I can imagine a TON of stuff that could be improved with the idea that volume brings capital.

These repair shops don't have volume because consumer behavior/culture sucks balls , so they can't raise capital for the best tools, so you get half assed jobs.

Also don't think we are in some kind of apex of repair tool technology, its just that it is not developed because there is no demand (related to volume), so no one will ever get their R&D back.


I.e. you can repair a trace with a floating bodge wire or you can repair a trace with a heat glue pace cutout with epoxy overfill, new vias, etc. The 2nd one is expensive and rare because its like military tech. But so were half the things we use often right now. PCB repair tools are an unexplored market IMO

Those repair shops right now are practically poorly funded counter culture run hold outs. try to imagine what would happen if they got big.

Think about the 50's. When you had demand to repair televisions and radios, the pharmacy turned into an analog testing center with high voltage component testing machines and vacuum tubes. Right now the best tool you can get in a pharmacy is a primitive cheap screw driver in a blister pack, and its 70 years later. Someone might think they walked into dystopia had they walked into a modern pharmacy. They would have expected fusion reactor parts.

You can't look at what you see right now if you want to change culture by making repair popular, economical, etc. IMO we get 'partisan' (in the sense of soldiers hiding in the woods in enemy territory) technology in these places right now because the law is unfriendly, and business even more so.


Without repair, in a modern house, you can easily be screwed for a years worth of savings with just random appliance and technology failures if they sync up right, double that with automotive problems. Its ridiculous.

I would say the repair quality is very promising , it basically says you can fix things even if the odds are totally stacked against you if you are creative enough. Things would only get better with better and cheaper repair tech. These repairs now seem to require a dentist (precision parts casting and x-rays (this one is hard without communal ownership in a pharmacy or something, scan you parts for 3$) lol

And how much easier would repair get if you had some kind good streamlined part duplicator? I.e. glue broken part back together with shitty glue and put it into a machine that will actually scan it good without needing a associates degree in CAD analysis and make you a replacement part? Its clearly possible, you can kinda do it, but its super thorny. If you had good cheap wax - zinc castings made for duplicate parts you could repair a TON of mechanical BS with higher quality replacements. I.e. stupid printer parts remade out of zinc. It has to be easy though, like an appliance, that actually works.

At best the best repair technology right now moved from skilled engineering team to somewhat specialized technician, IMO with UV-resin printers being somewhat the exception (that thing actually worked brilliant out of the box with no work, but its damn weak, but its cheap enough and you don't need to mess with alignments and beds, face it, thats technician setup work). It can go ALOT further.

Also with EMC, you can do ALOT better then chase repairs. Try not designing the product so its passing the test by a bee penis laid against the compliance trace. Then if you replace it with something a little different, guess what, your fine. Not to mention it will protect your company from parts obsolesce and design changes in the future. That's where consumer law and good engineering decisions can satisfy safety requirements. Minor changes causing huge problems = symptom of hyper focused cost engineering.. that stuff is done by rabid people.
pqass:

--- Quote from: bdunham7 on July 18, 2021, 06:51:53 pm ---
--- Quote from: pqass on July 18, 2021, 06:27:52 pm ---"...ask Intersil to create..."  or mark chips with my proprietary part numbers all sounds like collusion to me as they seem to rely on IP that is already mostly developed by the OEM and is being incorporated in other chips they sell to distributors/general public. So I, a third party, should not be barred from buying same (subject to one-time, license, and minimum quantity charges).

--- End quote ---

I disagree with robint91, but you are badly misstating his argument here.  He wasn't referring to re-marking, but to a contractually supplied custom chip where the customer holds the IP.  Applying FRAND to that is ridiculous.


--- End quote ---

I fail to understand what IP the contractor to Intersil might have in a battery charging chip that doesn't involve a new mask.  Binning? Blown fuses in an existing Intersil design?  Smells like collusion to me.  If Apple wants to keep it proprietary they should pay Intersil for the IP and use Apple design people to add their 2cents and contract TSMC to produce a few million.  Otherwise, it should be purchasable by anyone.

In the case of a board with MCU, sell the board with unprogrammed or missing MCU, and I'll take care of the rest.


--- Quote ---
--- Quote ---Right to repair is mostly about stopping stupid shit like re-labeling and deliberate obfuscation. 

--- End quote ---

Not at all.  There are two avenues available to protect your IP--the copyright/patent route and the proprietary/trade secret route.  Both are legitimate.  You can design a product with magic module that has all the IC labels sanded off and then the whole thing potted in cement if you like--there's no R2R law I've ever seen that would prohibit that.  What R2R might require is that if you sell a product containing the magic potted module to an end user that is a consumer,  you or your dealer supply that module to the consumer or an independent repair facility on non-discriminatory terms.

--- End quote ---

Yes, consumer protection law should kick-in to compel the 1st party to sell components to affect a repair for a set period of time after the original sale.  And none of this replacement board which costs 85% of the original price shit either.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod