genuine OPA627 info

[jitter]

Since the OPA627 is prone to being a counterfeit when bought through alternative channels, I was wondering if it would be helpful to you guys if I were to maintain a list of datecodes of genuine OPA627s I come across at work from now on.

After having been confronted in the past with counterfeit OPA627s, the company I work for nowadays only orders them from a reliable (authorized) supplier. That fact together with successful tests on the instruments they're used in make it fairly certain that they are the real deal.

We use the following versions:
OPA627AP (plastic DIP);
OPA627AM (TO-99 metal);
OPA627BP (plastic DIP);
OPA627AU (SOIC).
Most of them will be the BP followed by the AU and occasionally the AM and AP versions.

We're talking of 1000-2000 OPA627s a year, bought in smaller batches, so we get to use quite some different datecodes. It's little effort for me to post them here, if you're interested.

[BrianL]

This is an admirable task, but difficult. I am pretty sure that there are too many made to be able to track all valid date codes (more correctly 'lot trace codes'). Also, having dealt with counterfeits, I have seen many cases where they copy known good LTCs onto the bogus units.

[jitter]

Those are things I'm aware of and are the reasons why I didn't just start posting the LTCs without asking first.

I agree that the LTC in itself doesn't give a definitive answer. It must be combined with a careful examination of features and, if possible, some technical specs.

I know that one of the counterfeit batches of the OPA627BP we received all had the same datecode. Comparing the bottom side of some ICs immediately made us suspicious: in the circle on the bottom it should say THAILAND, instead some just had two digits in them, others said TAIWAN or MALAY.

Another incredibly simple test revealed they were fake: current draw from the PSU wasn't anywhere near the 7 mA it should be!

[BrianL]

Country of origin is not a sure-fire indicator. Before becoming part of TI, Burr-Brown did assembly/test of plastic parts in subcons in Singapore and Malaysia. Their 'date code' scheme was different than TI's. Now that TI owns them, the parts can still be built at one or more subcons (various countries) and/or TI factories (Malaysia, Taiwan, most likely). If you are dealing with only recent vintages, consistency of visual appearance with known-good material is a good start. Checking very basic specs, such as quiescent current is also very good - you shouldn't expect anything outside the datasheet specs.

[jitter]

Originally Posted by BrianL

Country of origin is not a sure-fire indicator. Before becoming part of TI, Burr-Brown did assembly/test of plastic parts in subcons in Singapore and Malaysia. Their 'date code' scheme was different than TI's. Now that TI owns them, the parts can still be built at one or more subcons (various countries) and/or TI factories (Malaysia, Taiwan, most likely). If you are dealing with only recent vintages, consistency of visual appearance with known-good material is a good start. Checking very basic specs, such as quiescent current is also very good - you shouldn't expect anything outside the datasheet specs.

Yes, they can come from different factories in different countries, but in the case of the counterfeit batch all ICs had the same datecode/LTC on top yet several different countries of origin in the circle on the bottom. That tipped it off in this case.

Since we keep as little stock as possible, we're mostly dealing with recent vintages. Recently produced pcbs have OPA627BPs from January and March of this year. I will post some pictures of old (ca. 2000), young (ca. 2009) and recently produced OPA627AP/BPs just to show how the appearance has changed over the years.

[BrianL]

One more comment, Jitter. I sure hope that when your company gets counterfeit parts when purchased through a reputable distributor that they notifiy both the disti. and TI.

[jitter]

At the time, the counterfeit batches came from what we, and our customer, thought was a reputable distributor.
Informing them and TI/BB was a waste of time. TI/BB is aware of the counterfeiting problem, and the most helpful comment we got from them was "to buy from reputable distributors only". The only ramafication for the distributor was that we don't buy parts from them anymore.

Nowadays we buy the OPA627 from a big and well known globally operating distributor only. A good choice as they probably deal with TI/BB directly and not through a third party.

[jitter]

And now for some photographs of genuine OPA627APs from several vintages.

The first picture is a OPA627AP mounted on a board we use as a reference ("golden board"). This board was assembled in 2000 and the IC itself is presumably from 1999 week 31.
The second picture is from another golden board assembled in 2006. The LTC-scheme has changed to what is looks like nowadays and is probably the result of TI taking over BB. IC from 2006 month 4.
The last two pictures are from current production in Thailand. The one on the left is what the OPA627 on the left (2008 month 3) looked like until 2009. Since then, they look like the one on the right (2011 month 1). This also goes for the OPA627BP.

Buying online from eBay or other unconfirmed sources is risky with these expensive and therefore counterfeit prone components. Even if you have an opamp that looks similar to one of the pictures, that is no guarantee it is.
I posted these not only as a guideline what to look for but also as a warning not to judge genuineness by looks alone because, as you can see, looks change over time.

[benb]

Originally Posted by jitter

At the time, the counterfeit batches came from what we, and our customer, thought was a reputable distributor.
Informing them and TI/BB was a waste of time. TI/BB is aware of the counterfeiting problem, and the most helpful comment we got from them was "to buy from reputable distributors only". The only ramafication for the distributor was that we don't buy parts from them anymore.

Nowadays we buy the OPA627 from a big and well known globally operating distributor only. A good choice as they probably deal with TI/BB directly and not through a third party.

I think the buzzword you're looking for is "authorized distributor." The TI website should have a list of authorized distributors, and if the distributor in question (from which you got counterfeit parts) IS on that list, I'd think TI would want to know. If the distributor is NOT listed by TI as an authorized distributor (which is what it sounds like TI was saying), you might as well be buying off ebay.

[jitter]

You're right, "authorized" is the keyword here.
The fake batch was not from an authorized distributor. Nowadays we buy them exclusively from Arrow and/or Avnet, both are listed in TI's HiRel Authorized Distributors list.

Here is a picture of a OPA627AU and soon I'll complete this thread with some OPA627AMs, a version we've also had a counterfeit batch from.