Author Topic: Fake vs original Components  (Read 2063 times)

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 5,329
Fake vs original Components
« on: July 29, 2009, 12:47:46 am »
Yes.

Shonver - you and everybody else

The news regarding fake semiconductors is naturally alarming, especially if one also builds to sell. And 'complete' experience is hard to gain, short of buying a handful of all types of power devices and testing them. Then the very next one may not make it ....

Somewhat off-topic (can one hi-jack one's own thread?). I would really like to know from all who are able to contribute (I do not buy that many semiconductors at a time, but my older supply is running out): The shortcoming of these devices; is it simply unable to withstand rated voltage/current, or anything else? Is there any way of noticing fake stuff - annotations, construction, whatever? I could try getting 'original' markings, but who's to say that has not been imitated?
"Take care of the SENSE, and the SOUNDS will take care of themselves."  (Lewis Carrol: Alice in Wonderland)

Offline DRNB

  • Forum Moderator
  • Trade Count: (0)
  • *****
  • Posts: 2,669
Fake vs original Components
« Reply #1 on: July 29, 2009, 01:04:13 am »
The shortcoming of these devices; is it simply unable to withstand rated voltage/current, or anything else?

Its seems that votage/current ratings might be the main issue. But someone with first hand experiance might know if there might be other differences as well.

Is there any way of noticing fake stuff - annotations, construction, whatever? I could try getting 'original' markings, but who's to say that has not been imitated?

Here is some reading

http://sound.westhost.com/counterfeit.htm

"There is no shame in not knowing; the shame lies in not finding out." (Russian Proverb)

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 5,329
Fake vs original Components
« Reply #2 on: July 29, 2009, 01:27:03 am »
Thanks Norval.

Bowing out now for bed - hopefully warmer than in front of the PC.
"Take care of the SENSE, and the SOUNDS will take care of themselves."  (Lewis Carrol: Alice in Wonderland)

Offline ludo

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 1,598
Fake vs original Components
« Reply #3 on: July 29, 2009, 04:27:41 am »
Another long one. Sigh.

I sometimes wonder if the counterfeiters print the part numbers on their devices only after the big order is placed.

Some examples of what you may find locally.

2SC5200 and 2SA1943 from Screenvision. Total crap. Pops immediately. Package looks more like Fairchild/On-Semi parts, with a "shiny window" for the lettering. Toshiba doesn't do it like that.

MJ15003/4 MJ15024/5 from Communica. As above. Pops. See the link Norval gave. Same thing.

(I am aware that MJ15024/5 is a tricky one, because Motorola changed the die several times but not the number. Mix the types and kablooey. But if it's the real thing and you replace in batches, all is well. This is important for the guys with Phase Linear/Flame Linear gear who try to economise when the time comes. ;D)

2SB817 and 2SD1047 from Screenvision seem to work fine in their application (car amps and such), but I have no doubt that Sanyo never had a hand in these. The old ones were very different internally, and the packaging was never so crooked. The old ones had something like silicon gel (white) over the die on the heatsink tab, with the bakelite-like plastic over that. The new ones don't have the silicon gel over the die and the heatsink tab has sharp edges around the top that pierce mica washers. If you file those edges down, they work OK in small amps. YMMV!!!

Of course I'm in no mood to try and characterise everything I have to buy. Not that would I be very successful if I tried, as I know far too little about the technology. It's not just a case of voltage breakdown and current handling, one has to check if the gain seems reasonable over a wide BW, check switching times, monitor temperature while doing all this, and who knows what else.

What I do, is I break one open. With a genuine ON-Semi/Toshiba part, the silicon die cracks when you break off the black plastic. One part comes off with the plastic, the other stays on the heatsink. One can hardly chisel the die off the heatsink tab. With a dodgy part, the die comes off the heatsink tab cleanly. It may have been attached with Bostik. I think this pretty much covers everything from Inchange via Communica and the 2SB817/2SD1047s Screenvision has now. I presume that the silicon die was never attached properly. But I'm no silicon foundry expert.

With some fake TO-3 cans, if you cut the lid off, you can scratch the die off the heat spreader with a fingernail, if there is a heat spreader. Same with an unused one as with a dead one.

My solution was to get out of power amp repair as much as possible. Mostly because I did lots of PA stuff. Failure is expensive with many pairs, and it always happens on stage. Lots of bad vibes then. News of a good tech travels well, but news of a bad tech travels faster. Musicians like complaining at least as much as I do, but they add beer. ;)

There's a little stash of genuine parts from EBV and Avnet Kopp just in case. Sold a lot to a friend who runs a repair shop. It may have prevented murder that week. For TVs, he's on his own. What I use from Screenvision (2SB649/2SD669) seems to work, but I doubt that it is genuine Hitachi.

It was never feasible for me to buy everything I need from EBV. Their prices are not totally unreasonable, I like them, but there are just too many different parts. TO-3, TO-3P and friends, MT-200 (Sanken) which they don't do, TO-126 and TO-220, some Darlingtons, Mosfets of all sorts, MJL21193/4/5/6 which they cannot source (yes, really, official agent or not). And you have to take quantity of them all. No telling when the broken amps will come by. With all the new SMT parts with unknown 2-digit codes, and manuals only for agents, not for sale, it just became a sh!t job.

Digidesign apparently wanted a non-disclosure agreement for the schematic of their charred mic amp full of SMT. So I just traced it out. Turns out it's no different from what Behringer have done slightly better for years. No different from everyone in the early 80s. Just the gain pot is a lot worse. But it's sacred knowledge. It's Digi. By the grace of God, it was an analog problem.

So now I'll be Mister Mixer Fixer for a bit. Till it too gets intolerable. When the fake TL072s and NE5532s come.

People say if you can't stand the heat get out of the kitchen. I intend to. But it won't make the kitchen any more liveable for anybody.

Because there's a buffet that needs replenishment out there. You see?

If it 's going vrot, put it on the buffet ! - Fats

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 5,329
Fake vs original Components
« Reply #4 on: July 29, 2009, 08:45:44 pm »
Another long one. Sigh.

Havn't you heard they say bigger is better?

Quote
People say if you can't stand the heat get out of the kitchen.

But when the kitchen is invaded by bugs, You stay in - you squash THEM.

In the 'old days' I was on a Store's Supply Committee at the CSIR (there were internal stores those days for convenience). Wish that still existed. We quickly sorted out many a supplier there, with enough capital behind us to do decent tests. But what to do these days. Ludo, can you mention specific types that I could go to Communica for to make a stink? I gather the 2SB817 and 2SD1047 would do, if Communica stocks them. Would one see the mentioned symptoms in MJ5003/4?

Time I ask among my professional colleagues' pool about this. We also have other professionals on this forum. What sayeth thee?

I just wonder if Deon Schoeman is prepared to stick his head out and publish conclusive results (with photos etc.) Joel occasionally reads here. He might have something to contribute.
"Take care of the SENSE, and the SOUNDS will take care of themselves."  (Lewis Carrol: Alice in Wonderland)

Offline Ampdog

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 5,329
Re: Fake vs original Components
« Reply #5 on: July 29, 2009, 09:02:50 pm »
By now all would have noticed that this topic was split off 'Old and new'.

As said there, bunkum sold on the market for good money should be of serious interest to all of us - as to the general public. We on this forum have the wherewithall to distinguish; kindly let us do that.

Thanks
"Take care of the SENSE, and the SOUNDS will take care of themselves."  (Lewis Carrol: Alice in Wonderland)

Offline Schalk

  • Commercial Member
  • Trade Count: (+6)
  • *****
  • Posts: 1,223
  • valveaudio@absamail.co.za
    • Valve Audio
Re: Fake vs original Components
« Reply #6 on: July 29, 2009, 10:15:10 pm »
'Tis easy me merry gentlemen ;)

Just get out of repairs altogether, I did. Most of us tech type outjies get branded as ripoffs, "mechanics" and so on. When the customers walked in the door and looked at me like I had "service shark" tattooed on my forehead, I just weighed anchor and departed for friendly shores.

Pointless to try and explain about the pirate parts you have to contend with, or the complete non-availability of original spares from appointed importers. Having to scrounge around for likely equivalents AND having to give a guarantee... :'(  Even back in the 70's and 80's fake spares were pretty much a problem already.

And heaven forbid, how COULD you charge extra for time spent in chasing these odd spares  :o

Like I said, easy...just let go. You'll be missed for a bit and then the joys of all that extra spare time you now have will be like a new life altogether.
Home of the Black Widow . You owe it to yourself to compare.

location : Pretoria

Offline ludo

  • Trade Count: (+1)
  • AVForums Veteran
  • *****
  • Posts: 1,598
Re: Fake vs original Components
« Reply #7 on: July 30, 2009, 04:40:58 am »
I think Schalk has a very sensible take on this.

Ampdog, there's that optimism again.  :)

It's not that I don't trust you or your colleagues or AVSA on this, not at all. I just think it's going to be a lot of trouble, possibly expensive trouble, to ascertain what we already know. I just mentioned a couple I know of, there are probably stores full... If you google "counterfeit semiconductor" you will see that there are companies that specialise in the detection of these parts for manufacturers, for a fee of course. Any number of industry associations claim to be deeply concerned, deeply.

It's a lot like the total lack of security in the Microsoft operating system. It never gets fixed, but it goes from strength to strength in the marketplace. Extra industries are created to cater for the problem. Money is spent, and lost, because of the problem. It's the original Win-Win. And guess who keeps paying, to be abused. It's nothing but the buying publics perversion. When a manufacturer gets to hate his clients' ignorance and gullibility, he sees the light. He'll do well. That is what business is now. Thanks to the "it's new, it's surely better, it's a bit expensive but we'll trust it again" crowd.

My suggestion to anyone interested in repairing things audio: Only deal with the official agents for semiconductor manufacturers. The manufacturers don't distribute fake parts, at least not through their own channels. You'll get the agents addresses from the manufacturers website, not from sales staff at a retailer. If you're a small-timer like me, don't you also be a damn fool about it, just stop, like Schalk says.

If you prove conclusively that a certain retailer knowingly distributes fakes, what will that do? Will we get a press release from Communica and Screenvision stating that they are deeply concerned? Deeply? That they actually care a whole lot? That they really believe in dedication, attention to detail, break-throughs and above all, partnership? That they will surely look into this "new" evil?

Do you suppose that the SABS will be interested? Dream on!
 
I cannot continue with the rest of the story here. We don't want the law involved. It has not helped us much of late anyway. I don't think it wants to. Besides, can you imagine the cost?


Ek is nou al lankal 'n hard-gebakte suurknol Ampdog. Jy gaan my nie sommer op-cheer nie. But thanks for trying. I do appreciate it. ;)

If it 's going vrot, put it on the buffet ! - Fats

Offline frikkie

  • Trade Count: (0)
  • AVForums Veteran
  • *****
  • Posts: 2,637
Re: Fake vs original Components
« Reply #8 on: July 30, 2009, 11:40:47 am »
I have to agree with Schalk and Ampdog.

Repairing stuff for people unknowingly using fake parts results in the thing coming back to you in a couple of days/weeks time, now often blown even worse than before. It's simply not worth it.

I still do car-amp repairs etc as a hobby, but I don't do it for gain. If someone asks me to have a look at something, I tell him straight out I'll give it a bash, parts for his own account - go source them and bring them to me. I'll install & test - if it works, great, and the moment it's off my property it's no longer my problem. If it blows again two hours later, sorry mate - I tried.
Sometimes I drink a glass of water... Just to surprise my liver.