How would you know if you bought mislabled IC's?

[skiraly017]

There's been more than one thread about Ebay sellers selling mislabled IC's (mislabled MN3007's come to mind). My question is how can you tell if what you bought is what was advertised? Thanks in advance.

[Mark Hammer]

In the case of BBDs, the answer is pretty simple.  You pop the chip into a pin-for-pin compatible socket, feed it a known clock signal and you simply HEAR the difference.  How on earth you might be able to tell that the flatbed truck worth of TL072s you bought were REALLY mislabelled TL082s is well beyond me.

[Connoisseur of Distortion]

"ey bob, fella in 'ere sez these ain't ev'n in th' 0x2 series"

"reckon we oughta design a simple test circuit to measure various parameters of the units, chuck?"

[Paul Perry (Frostwave)]

Which is why anybody doing quantity production has a 'test' unit with sockets for everything....
Some data sheets have test circuits shown, that allow you to confirm crucial parameters.
I might point out that it often isn't the fault of the person you buy from, most retailers buy bulk from larger wholesale companies... their only fault is that they havn't tested the chips.
I personally bought over $500 worth of NOS octal sample and hold chips stamped PMI SM2300 from a wholesaler. No knowing where they came from originally, but I doubt it was Precision Monolithic Industries, because they sure wern't s&h chips.....
I guess it is another reason for buying from someone like Steve at Smallbear, who guarantees to supply the right chip & actually knows how to test one, if he has to!

[Mark Hammer]

You don't want to know how many Yankees, Knicks, Islanders, and/or Giants games Steve Daniels has watched while sticking  BBD after BBD into a Memory Man or Electric Mistress!!   (and neither does his wife Judy!)

[smallbearelec]

I personally bought over $500 worth of NOS octal sample and hold chips stamped PMI SM2300 from a wholesaler. No knowing where they came from originally, but I doubt it was Precision Monolithic Industries, because they sure wern't s&h chips.....

The pain...the pain. My own heartaches with such problems have been detailed here. I am very reluctant to buy the MN3205, for example; there are just too many out there that are really re-badged MN3208 or BL3208. Anything that's obsolete and relatively scarce seems to be fair game for such thievery. I have learned to insist on getting photos and samples. Then I check date codes and condition of the pins before I test: If the pins are tinned, it may indicate that the parts have been removed from gear and are not new. If they work, they might be worth buying, but not at the price originally quoted. It's a bazaar...I have been patient, and the many good deals have far outweighed the occasional agita over a ripoff.

[Elektrojänis]

How on earth you might be able to tell that the flatbed truck worth of TL072s you bought were REALLY mislabelled TL082s is well beyond me.

And the fact that the manufacturer says that the TL072 is a low noise version of TL082 but still puts the same noise specs on the datasheets of both devices does not make it any easier...

If anyone wants to compare check out their equivalent input noise voltage/current:
http://www-s.ti.com/sc/ds/tl072.pdf (page 9)
http://www-s.ti.com/sc/ds/tl082.pdf (page 7)

I just don't get it. 

[Paul Perry (Frostwave)]

And the fact that the manufacturer says that the TL072 is a low noise version of TL082 but still puts the same noise specs on the datasheets of both devices does not make it any easier...
I just don't get it. 

I don't know whether it is the case here, but as chip manufacturing has improved (cleaner assembly, higher purity materials, more exact litho & doping) 'standard' chip specs have improved, to the extent that they have caught up with 'improved' versions.  And another thing is that economies of scale mean that it's sometime worth just making the 'improved' chip & labelling some with the earlier chip number.
For almost all applicatins this kind of jiggery-pokery is OK (provided the chips match the data sheet specs). It's only a problem if we use them in 'unexpected' and 'unintended' ways... which coulld happen in an overdriven fuzz, for example.

[Paul Marossy]

I don't know whether it is the case here, but as chip manufacturing has improved (cleaner assembly, higher purity materials, more exact litho & doping) 'standard' chip specs have improved, to the extent that they have caught up with 'improved' versions. 

Yeah, and that is why is still think that a newly manufactured 741 sounds pretty quiet compared to an old one. 

[Elektrojänis]

I don't know whether it is the case here, but as chip manufacturing has improved (cleaner assembly, higher purity materials, more exact litho & doping) 'standard' chip specs have improved, to the extent that they have caught up with 'improved' versions.  And another thing is that economies of scale mean that it's sometime worth just making the 'improved' chip & labelling some with the earlier chip number.

You're probably right. Probably a combination of both.

They would probably make them in the same process anyway and then just test them and label the better ones 072 and the others 082. Testing costs too and the improved manufacturing process could be able to get them all to 072 spec anyway.

Usually 072 Still costs more though.