z600 / z620 Front Panel USB 3.0 Upgrade

[D-Fi]

Does anyone have the pinout for the Front Panel USB 2.0 connector on a HP Z620 Mainboard (the one that the Yellow front panel connector plugs into)?  I am trying to use a z620 front panel assembly on a z600, and I cannot figure out what the two orange  / black connectors are on this cable header.  They are not present on the z600 cable header.  I will be using an add-in card for USB 3.0, and the 20 pin header for the USB 3.0 on the front panel will connect to this card, but there is a USB 2.0 port on the front panel as well, with the "extra" orange / black wires on them.  If I can see a pinout of the front panel USB 2.0 connector for the Z620, I can possibly translate these to a pin on the z600 and then have front panel USB 3.0 on the Z600 using a "stock" front panel.

Here are the pinouts that I am looking for:

Z620 - P27

Z600 - P27, P26, P24

Thanks!

[SDH]

Did you try using the "Search the community" search box just above?  It defaults to searching only this forum, which is handy.

For example, I just searched for Z620 Pinout, and this may help:

LINK

http://h30434.www3.hp.com/t5/Business-PCs-Workstations-and-Point-of-Sale-Systems/Front-panel-connect...

The pinouts have stayed similar over the different generations of workstations so you may find added info from prior posts too.  Dan is a very helpful HP engineer who has been able to provide lots of pinout solutions.  If you need me to get into one of our Z620s to take a look let me know.

Also, be sure to get the "2 x 2" version of the HP USB 3.0 card, which is the later one based on the TI chipset.  It has 2 ports on the rear, and the header on the card for two up front.  It is best to put that PCIe card into one of your PCIe generation 2 slots because that will have the appropriate bandwidth.  You want a Gen 2 slot to match USB3 speeds, and in a xw6600 only the two video PCIe slots are Gen 2.  I can't remember which of the PCIe slots are Gen 2 for the Z600, but that info will be in the Z600 Maintenance and Service Guide which you can find via Google.  Sounds like a great project.

Many 4-pin to SATA power adapters don't fit well on that card, but the Molex brand ones from Mouser.com do fit perfectly.  If you get the full kit from HP it will have a power adapter that will fit fine, but if you buy the card bare you'll need to get one.  Best to tap into a 4-pin spare power connector up by the optical disk area if you have one, so you may also need a 4-pin male to 4-pin female extension.  For me in the xw6600 a 12" extension plus the Molex brand adapter gave enough room to get the power feed down easily.

Finally, power the card before you attempt to load drivers...... parts of the card won't be seen by the driver installer until you get it power from the 5v feed from above.  The PCIe slots don't provide that.

[D-Fi]

I too saw that post (after searching the forums), but the connector that I am interested in "decoding" is the yellow one that can be seen to the very far right of the front panel connector.  The Z620 USB front panel cable has the standard 20-pin USB 3.0 blue connector (which goes to the 20-pin connector on the USB 3.0 card), but the USB 2.0 cable has one row for the USB 2.0 port (red, white, green, black), and another row for "something else" (orange, orange, black, black).  The "something else" row leads to some type of chip which is encased in plastic, which also aggregates the USB 3.0 connectors from the 20-pin USB 3.0 header cable.

Yellow Header (p27) that the USB 2.0 cable connects to:

Normal USB 2.0 Pins on Z620 USB 2.0 Cable

What are these Pins on the USB 2.0 cable / header?

Here is the "chip" that the orange, orange, black, black pins on the USB 2.0 cable / header lead to.  The large cables on the right are the input leads from the USB 3.0 20-pin header on the board.  The two large cables on the left are the leads to the USB 3.0 ports on the front panel.  I cannot see what is inside the casing w/o destroying it, so hopefully a pinout of the P27 yellow USB 2.0 front panel header on the Z620 board will lead to clues as to what these four cables do.

[SDH]

I got into a Z620 here, and it is unclear to me too.  There are the two blue USB3 ports available at the front face of the worksation, and the one black top USB2 just above them, so you have accounted for those.  The 4 wires you mention go out from that header's bottom row of pins via a small diameter insulated wire bundle, but then enters that black box you show.  You'd need to get inside of that box to figure out where they come out, but that will probably break it.

A few ideas.... in the service manual about header #27 HP talks about the front power button, LED, and speaker.  But not Mic.  But those 3 wires usually would come back to the motherboard via the front cable to header #27 (from past workstation engineering) and maybe they just left that off the "Mic" notation.  There also are rarely used options such as the side cover alarm device that alerts if the side door had been opened, etc.   Look into those options.

Honestly, your best bet would be to get ahold of Dan, and if you had a bare Z620 motherboard there might be some hints in tiny printing on the PCB at that far lower edge just beneath the header.

Let us know..... very creative solution to getting front USB3 ports into a Z600.  Another way would be a 5 1/4" to 3.5" adapter for one of your optical drive bays, and put in the Akasa 3.5" adapter that comes with a nice long cable going back to feed off the standard motherboard/card USB3 header.  That cable is more than long enough... they PhotoShopped it for this pic. The quality is A+, the device is designed to mount a 2.5" drive if you want (I did not) and that is exaclty what I used in my USB3-in-xw6600 project:

[Dan_WGBU]

The Z600 has USB 2.0 ports only, no USB 3.0. 

     Blue P27 is a single USB internal port.  It follows the standard USB pinout:  Pin1 = 5V USB power, Pin 2 = Signal D-, Pin 3 = Signal D+,  Pin 4 = Ground, Pin 5 = empty = key. 

      - Blue P26 is a dual port USB 2.0 internal header.  It has the standard USB dual pinout, similar P27 above (with one port on a row of pins on one side, and the port other on the other row of pins), with the exeption of pin 10 being a cable detect pin.  Pin 9 is a key.  Most third party adapters will fit on this header.

       - Yellow P24 is a 14 pin header, with 3 USB 2.0 ports, all routed to the front panel.  Pins 1 to 10 follow the P26 pinout, with the extra pins were added to add the third port to the front panel.  Simple. 

The Z620 is very different.  It has both USB 2.0 and USB 3.0 ports routed to the front of the system.  The Z620 has a separate USB 3.0 controller IC, which is not on the Z600.

      - P27 pins 1, 3, 5, 7, and 9 (one side of the header) are a single USB 2.0 port that routes to the USB 2.0 port onthe front panel (with cable detect on pin 9)

      - P27 pins 2 and 4 (orange) are +3.3V power, and pins 6 and 8 are Ground.  Do not connect a USB connector or adapter to these pins!

      - P29 is the blue 20 pin USB 3.0 connector.  It follows the industry-standard pinout, so, in theory, one could connect an adapter to this and use it for another USB 3.0 device or different panel. 

HOWEVER, KEEP THIS IN MIND:  The Z620 front I/O USB cable is an active cable, with USB 3.0 redriver ICs in them.  That is the big square in the middle of the cable.  Redrivers are needed to provide USB 3.0 signal integrity.  The orange power wires on P27 are used to power these redrivers. 

     - If one does not use the Z620 cable, and runs ordinary cables from the USB 3.0 connector P29, USB 3.0 will (very) probably be intermittent, flakey, and unreliable.  That is why HP put active redrivers inside the cable, to make them reliable. 

USB 3.0 is 5Gbps, which is really high speed and requires careful system and cable design. 

I am not sure what you are trying to do with the Z600.  Are you trying to put in a USB 3.0 front panel, and connect this to an add-in expansion card in the Z600?  If so, long cables (long is greater than 4 inches) will probably (certainly?) have an unreliable connection.  I strongly discourage doing this. 

[SDH]

Dan,

Thank you once again for your help and insights.  It is such a benefit to have your input.  The OP wanted to add in a HP USB3.0 card, which was released by HP for the Z600 era workstations as an option, and also use a front panel from a Z620 fed USB3 from that card because it would be a clean install and a great look.  Those Z600 option cards from HP had 2 forms.... the rare early card was made by NEC/Renasus, and the later common version of the card was made with a TI chip.  It was the winner because it upped the USB3 ports from 2 total (on the card's backplane) to 4 total (2 on the backplane, and two more via a standard 2x10 USB3 "motherboard type header" for running a USB3 dual cable set to out front.  HP called that their "2x2" card, and later used the Texas Instrument chip on the motherboard of the Z620 era workstations.  That is why there are TI USB drivers included for the Z620, to run its on-board TI chip.

Having said that, from your post the powered redrivers in the Z620's small black box receive one +3.3vDC supply and one ground for each of its two USB3 ports built in up front.... hence two total orange power feeds and two total black grounds.  Those 4 wires come into the little black box via the thinnest black cable going forward to that.

What I have done with my xw6600 work is to put in that same TI based 2x2 HP card running it in my lower PCIe x16 video slot (which is PCIe Generation II, as is the top video port).  That way I can get the bandwidth needed to/from the PCIe bus to match USB3 capabilities.  That has worked very reliably, and I have used high quality USB3 thumbdrives from Sandisk to do speed testing to prove I'm getting USB3 speeds.  Key point.... I believe HP engineered the redrivers you described into that 2x2 USB3 option card, because it works just fine over those longer cables you see in my post above.  Quite a few USB3 cards on the market now include that type of 2x10 header (1 row actually is 9, with the 10th key hole epoxy-filled).  HP did not specify length of the two cables that extend from that card's header forward, but there presumably is a published maximum in the USB3 standards.

So, for the OP the use of the front panel from a Z620 in a Z600, even if it fits, adds in the complexity that his redrivers in that small black box won't be powered correctly because the Z600 does not have the power feeds/grounds off of the motherboard like the Z620 does.  He might be able to hack those in by down-converting to 3.3vDC from power supply feeds from above, but who knows what risks that entails?

Hence, maybe my off hand suggestion that he use the HP 2x2 card and the Akasa front USB3 interface shown in the pic above might suffice better.... it sure has been working very well in my xw6600 builds.

For completeness.... the rest of the PCIe slots in the xw6600 are PCIe Gen I, and all the PCIe slots in a xw6400 are PCIe Gen I.  That is why I've only done this mod on the xw6600s here.

Always good to get your input......... and thanks again.

[SDH]

Just found the 2x10 header and cable info in the Intel USB 3.0 specifications document, as a PDF here.

http://www.intel.com/content/dam/doc/technical-specification/usb3-internal-connector-cable-specifica...

The 2x10 header is described in section 3, and the max recommended cable length extending from that header is specified:

"The maximum cable assembly length shall not exceed 457.2 mm (18”)."

[D-Fi]

Yes, using a Z620 cable and an add-in card to provide USB 3.0 to the front panel of a Z600 is exactly what I am trying to do.  I would use an add-in card and a 5.25" usb 3.0 add-in "bay" but both of my 5.25 slots are currently filled with other peripherials.  I was going for a "clean" look and trying to see if I could swap out the cable assembly and use the 20-pin USB 3.0 connector on a add-in card to drive the USB 3.0 ports on the front of the PC.  Now that I see that, due to the cable length, the assembly has a redriver that needs 3.3V power, this has become even more of a challenge.  I am well versed in electronics, so it would not be difficult to use a 5V to 3.3V converter to step the voltage down off of the 5V rail from either the power supply (preferred) or a 5V connector on the mainboard.  Sounds like a interesting project to tackle.  The easy route would be to free up a drive bay for a USB 3.0 interface (such as the Akasa that SDH reccommended), but I think I like the idea of tackling an actual electronics project.  I can break off the two 3.3v and ground pins from the USB 2.0 cable, and then use the remaining pins to connect to the USB 2.0 motherboard header to provide the USB 2.0 service to the front panel.  After that, it is just a matter of getting 3.3v and ground to the USB 3.-0 redriver from a stable source.

I appreciate all the info and advice given, and I will post back when I have figured out how to do this.

[Dan_WGBU]

Thanks for the replies.  They help me understand what is being asked.  Some points:

Using the HP 4-port USB 3.0 card and an Adapter, like the akasa, is an alternative.  The USB3 card's internal connector matches the Intel specification, and is compatible with the adapter.  It is good data that SDH has success with it.  However, you discuss header #27 with power button, LED, etc.  On a Z600, P27 is a 1x5 (with missing pin key) header for internal USB 2.0.  P5 is the 2x9 front panel header with these signals.  The front panel audio signals are on header P28.  Some earlier workstations name headers differently.  FYI. 

Some cautions about signal integrity and cables.  Poor signal integrity can cause all sorts of problems, like dropping devices, incomplete transfers, etc.  There are many items to consider, including the motherboard/card USB3 printed circuit board signal trace lengths, cable length, and extremely important, the quality of the USB 3.0 cable from the motherboard or expansion card to the USB 3.0 connectors.  The USB3 card has short PCB trace lengths, and the akasa cable seems to be shorter than 18 in.  Shortening PCB trace lengths and cable lengths, even by an inch or less, can make a significant difference in performance.  Cable construction is critical, especially how the wires are separated from the bundle and then soldered to the connectors.  I have seen some poorly made cables that have an error rates up to a billion times higher than those that are well made.  I can't go into details, but trust me on this.  Intel's 18 in. spec assumes an (unrealistic?) best case for cards, motherboards, and cable construction.  My $0.02.  

BTW, the USB3 card does not have any USB signal redrivers.  The PCB trace lengths are very short, so that is why I recommend using well-made cables that are as short as possible.  FYI.  

As SDH wrote, make sure that the USB3 card is plugged into a PCIe Gen2 slot.  It will work in a Gen1 slot, but at half the bandwidth.  

The Z620 motherboard has longer USB3 PCB signal trace lengths.  With the longer cable to the front IO connectors, HP determined redrivers, located in the cable, were necessary to give reliable operation.  HP testing requires exceeding the USB standard bit error rate by a significant amount, and to not have any USB errors, caused by signal integrity, when running constant data traffic for days at a time.  HP spent a lot of time making this work.  Trust me on this.   

Back to the Z620 front panel.  The panel's blue 20 pin cable has the same pinout as the Intel standard.  However, since the cable has the redrivers "between" the motherboard and the USB connector, if these redrivers are not powered up by the 3.3V wires, they will be "off" and the USB 3.0 signals will not work.  

The Z600 power supply does not provide 3.3V, only 5V, 12V, and -12V.  The 3.3V power is created by circuitry on the motherboard.  There is no easy way to tap into the 3.3V power, and I cannot recommend soldering wires to the motherboard (I am an HP employee, and must warn that mods can damage the system)   

On a Z620, the redriver 3.3V power is on during normal operation, and also during sleep.  Connecting 3.3V power that is active only when the system is "on", and not in sleep, will cause any USB devices to drop off during sleep.  Wake, such as by a keyboard, wil not work.  If 3.3V power is supplied all the time, it might work (again, cannot recommend), but system power consumption will be higher.  

The Z620 front panel has 2 USB 3.0 and 1 USB 2.0 port.  The USB 2.0 signals are routed from the "odd" pins on the connector going to P27.  I absolutely cannot recommend removing the 2 orange 3.3V wires and ground wires, and plugging the connector with the remaining wires into blue 2x5 internal USB header like P25 and P26, to make the USB 2.0 port on the front panel work.  (cough)  Making USB 3.0 ports work is not as simple.