SCSI HBA Recycling?

Several weeks ago I bought this Adaptec 39160 64-bit PCI SCSI HBA in order to experiment with different HBAs:

Adaptec 39160 SCSI HBA, perhaps 2007

The motivation was that although I’ve been a happy user of LSI HBAs (SCSI and SAS, PCI and PCIe) based on the MPT Fusion architecture, the Linux driver for the LSI SCSI HBAs is dumb, broken, or buggy, and ignores certain firmware settings. I have several older SCSI drives (early 1990s) which badly fail if the host tries to negotiate wide and/or fast transfers; the drive in some cases resets but usually hangs and must be power cycled.

All that can be avoided if the LSI HBA is set up to not use wide transfers and not negotiate fast transfers for the given SCSI ID. That works well with the LSI HBAs at POST time, but the Linux driver then just ignores the firmware settings and hangs the drive anyway—unlike the LSI Windows driver. So far I have not found any way to convince the Linux kernel drivers not to do that, and after perusing the source code I’m pretty sure there isn’t.

At any rate, I wanted to try an Adaptec SCSI HBA to check if it has the same problem in Linux (spoiler: it doesn’t) and ordered the 39160 pictured above off eBay. This is a nice dual channel HBA and (for an U160 SCSI controller) somewhat unusual for having both 68-pin and 50-pin connectors on the first channel; that feature looked useful.

The HBA I ordered was sold as new and at 10.50 Euro, it was a bargain. It arrived very promptly, in an original-looking Adaptec box, sealed in an anti-stat bag, and included a printed installation brochure as well as a driver CD.

The adapter works fine, and although it exhibits this extremely annoying error in my Intel Stormville board, I have no reason to think that this particular specimen is defective, especially given that the error does not show up on a Supermicro X7DWN+ board.

Whenever I get a new piece of hardware, I like to have a close look at it simply out of curiosity. When it was made? What chips were used? That’s when I noticed that the HBA just does not look like new.

The connectors and some of the chips have clear signs of wear that would not show on a new, unused HBA. There is some sort of whitish powder on the HBA, and the solder joints have odd cracks and don’t have that shiny new look.

A lot more scratches than a new connector should have
The solder joints look a little funny and one of the caps isn’t straight

Most bizarrely, the date stamps on the components just don’t make any sense. The PCB has a 1607 date stamp, which I assume must mean week 16 of 2007; the HBA is far too old for 2016, but it’s really kind of old even for 2007 given that the Adaptec 39160 was introduced circa 1999. The chips appear to have been manufactured between 2000 and 2002, and the product sticker (which covers the main chip) says 0105, indicating week 5 of 2001 (not week 1 of 2005; looking at photos of other 39160 HBAs makes it clear that it’s year first, week next).

A PCB presumably made in week 16 of 2007

It is common that some chips are older than the PCB they’re mounted on, but 5+ years does not add up. Especially when the chips really don’t look like new.

Scuffed and scratched, but also 6 years older than the PCB it’s soldered on

The terminator chips are another mystery. There are two sets of three chips (one set per channel). Typically, the chips would be all identical because of course they all came from the same batch, or maybe from two batches manufactured not far apart. But on this particular HBA, the date codes on the terminator chips, if I’m reading them right, range from week 33 of 2000 to week 34 of 2002. They’re all the same model (DS2119M) but only two of them have the same date code, and those two use a slightly different font and don’t look like they were made on the same production line.

Three (of six totall) not terribly identical terminator chips

So what happened with this board? I can’t quite figure it out. It works fine, does not look like a fake, and 2007 is really kind of late for an Ultra 160 HBA. Yet the CD in the package shows a 2004 copyright, and the files on the CD are from 2006, which actually would make sense for a HBA manufactured in 2007.

Does anyone know if this sort of recycling is or was a common industry practice, and what the point was? I just can’t figure it out. It’s hard to imagine how it could have made economic sense, but if that’s not what happened, how did this Frankenstein HBA came to be?

This entry was posted in Adaptec, Fakes, PC hardware, SCSI. Bookmark the permalink.

10 Responses to SCSI HBA Recycling?

  1. Chris M. says:

    What you got is a counterfeit card. Its plainly obvious by the sticker over the main chip that it was rather poorly copied (that “barcode” shouldn’t be a blotch of ink!). That sticker should be on the PCB, not on the controller chip. Peel it off and see what the actual silkscreen on the chip is.

  2. Chris M. says:

    Here is the real card for comparison (late model RoHS): https://www.newegg.com/adaptec-2253700-r-scsi/p/N82E16816103075

    This auction shows an older pre-RoHS card that yours attempted to copy: https://www.carousell.sg/p/adaptec-39160-dual-channel-ultra-160-scsi-card-pci-32bit-or-64bit-slot-16187924/

    There is a silkscreen error on your card too. “JP1” on the upper left corner should be “TP1”.

  3. Julien Oster says:

    That soldering on the mismatched terminator chip looks terrible.

    PS: I also ran into the problem now where my browser was on the http URL of your page and ticking the “Really, I am not a spammer” box failed. I came through the RSS feed, and had to switch manually to https to be able to post. (Writing this I don’t actually know that yet, but since the post now exists it must have worked!)

  4. Michal Necasek says:

    Counterfeit card made from genuine chips? That would actually make sense. I have to say that the driver CD, if it is a pirate copy, looks amazingly well done.

    There are other errors on the silkscreen, for example the top connectors are labeled “CH 27B” and “CH 17B” when it should be “CH 2/B” and “CH 1/B”. And I think the latter should have been “CH 1/A” anyway.

    So who made this card? The same people who make those AMD-only DDR2 4GB DIMMs? In this case it is a copy of an existing design, not something that otherwise does not exist. The adapter works just fine, too, I didn’t find any functional problems with it. It’s a lot like those fake Yamaha OPL3 chips, if you didn’t look closely you’d never know.

  5. Michal Necasek says:

    I’m not sure how to fix the stupid HTTP/HTTPS mess. I still don’t want to force the site to be HTTPS only but at the same time people should visit over HTTPS whenever possible. The thing is that if I default the URLs to HTTPS, it won’t work over HTTP at all, while the other way it at least somewhat works. If anyone has brilliant ideas, I’d love to hear them.

    I suppose I could change the checkbox to read “I am visiting over HTTPS” to remind people?

  6. Zir Blazer says:

    I don’t know about SCSI HBAs, but NICs are well known to be a favourite counterfeit target. At times they may be made out of genuine main chips but the rest of the chip cast sucks.

    https://www.servethehome.com/investigating-fake-intel-i350-network-adapters/
    https://www.servethehome.com/identifying-risky-counterfeit-intel-gigabit-ct-network-adapters/
    https://www.servethehome.com/3rd-party-intel-x710-da4-quad-10gbe-nic-review/

  7. Michal Necasek says:

    That’s very interesting. Selling a counterfeit card for 200ドル makes much more sense to me than the circa 10ドル I paid for the fake Adaptec. What I could not find in those articles is whether the fakes are built using new Intel chips, recycled Intel chips, or if the chips themselves are counterfeit as well.

    I’m also still mystified by the installation booklet and CD that came with my Adaptec card, because those really did not look fake at all.

  8. Chris M. says:

    That fake card likely sold for more when it was a current product. Even at 10ドル they are making money on the thing, whatever nefarious group of people that is.

  9. Nils S. says:

    ~10 years ago I worked at a company building those really big LED displays used in a car wash.
    The mainboards of their controllers were gigabyte Pentium MMX systems they’ve ordered in far excess when they weren’t out of date. IIRC these boards were AT form factor but used PS/2 connectors for keyboard and mouse, had 4 PCI and 3 ISA slots.

    I got a few of them on my desk, “not doing anything” was the short error description.
    The ISA connectors were very badly soldered, stopped the whole thing from starting POST etc.
    Our salesman tried to get replacement from gigabyte for a product that was over it’s age for years, but that never worked. We’ve sent them a few of the broken boards for investigation.
    The result was that we got counterfeit boards. The salesman at gigabyte told us, that the only explanation(s) they could think of, is that someone got either the original drawings, schematics and other manufacturing files or someone has stolen broken boards from manufacturing.

    The latter is more likely, as it was obvious at a second look, that the parts were of different age and some then already looked used (especially the probritary chips).
    Another sign for the second explanation was, as far as we’ve been told, there were no other counterfeits of these boards around.

    The seller of those counterfeit mainboards was a big german retailer that now is closed – don’t know about any connection to this incident. But to keep bad talking away from the dead, I’d rather omit their name here.
    Also, this weren’t the only counterfeit parts. There also were problems with SODIMM-systems (the whole thing on a SODIMM, german product) and pirated software.

    For smaller electronics, like a PCI card – more or less the reference design, I actually think it might be a job of a few days for a good designer to copy the PCB.

    My conclusion is, even if it looks like an awful big pile of work and not profitable, every product may be counterfeit.

    An interesting thing, I learned from this, was, that replacing the board with a different type, but still PC compatible, violates a lot of industry’s engineering “rules”. They’d rather send out a trainee with a lot of flux, tin and a hot air gun to repair boards in field – not replacing with spare parts and repair at home…

  10. Michal Necasek says:

    Fascinating. Sometimes you wonder how much the companies selling these things to end users are part of the scam vs. victims themselves.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.