HP Thin Client - Kirkwood core

Sounds like you need a good solder station and a solder sucker! I’d be scared to drill the old one out, personally!

A solder sucker does the job (with a crocodile clip heatsink) when removing a RiscPC CMOS battery. But that’s probably not a fair comparison, considering the t5325’s small size and number/close proximity of connections.

A solder sucker does the job

Yeah, I was thinking something more like this. Cheaper alternatives exist as well.

Cheaper alternatives exist as well.

For a one-off, a friendly TV repair shop might do it for a small fee.

Maybe I should get a good solder sucker, I’ve been using desoldering braid, but that just wasn’t cutting it.

And, yes, my iron can do RoHS work, I had it up at 400 C, and it could melt the solder without using the braid no problem, but no go with the braid.

no go with the braid

I rarely have a lot of luck with braid. It always seems like a nice idea, but doesn’t work well for me in practice!

I used to have a solder sucker which worked with a springloaded piston. that one helped tremendously with things that were hard to desolder.. Now where did i leave that thing.. it would come in handy again soon enough.

Also, no response to my request for source code.

Wondering if the U-Boot and Debian teams should be notified of the potential GPL violation. ;)

(Attempting to contact HP again.)

Maybe I should get a good solder sucker, I’ve been using desoldering braid, but that just wasn’t cutting it.

And, yes, my iron can do RoHS work, I had it up at 400 C, and it could melt the solder without using the braid no problem, but no go with the braid.

Your iron isn’t powerful enough, and/or the bit is too small. The copper planes in the board, plus the braid, have got such good thermal conductivity that too much heat is being sucked away.

A spring-loaded solder sucker is probably your best bet. It’s what I usually use. Braid can be useful, but its thermal conductivity has drawbacks. It is available in numerous widths; if you don’t want to buy a solder sucker, a thinner braid may just be enough help.

One other distant possibility is to do all the desoldering with the board on top of a hotplate. It’s a real work-up, though, and there is the very real prospect of burning yourself quite badly.

Got a solder sucker, got everything unsoldered, and got the SATA connector soldered in.

It does NOT detect anything plugged into the SATA port, although a laptop HDD I plugged in did spin up. (And, the CF to SATA adapter that I got does work in my netbook, so I know it’s good.)

It could be my soldering job (I had trouble with the solder not sticking to the pins or board (yes, I used flux,) but instead sticking to the iron,) although it looks like everything’s connected to me.

Could a U-Boot setting be blocking that port? I don’t have a good way of checking the U-Boot messages at the moment.

Could a U-Boot setting be blocking that port?

Possibly, yes. If the Kirkwood is anything like the OMAP then the chip supports more I/O signals than the number of pins that physically exist on the package, so many of the pins are multiplexed. The versions of U-Boot used with OMAP boards configure the multiplexers so that the right pins are routed to the right place within the SoC.

A look through the Kirkwood TRM would probably help, along with a look at the Linux sources to see what kernel modules are needed for the SATA driver (even if you can’t get the exact sources from HP, the driver will hopefully be in the mainline kernel by now)

The Sheeva plug seems to have 400MHz DDR2 memory, doest the HP t5325 also have 400MHz DDR2 memory?

I looked around on the web and it appears that there are newer revisions of the Marvell chip that can handle DDR2 800Mhz and possibly even faster memory.

The U-Boot log I posted a while back matches the numbers I’ve seen online for PogoPlugs.

I did manage to get into the t5325’s U-Boot again over serial.

Not detecting anything on the second SATA channel. Wonder if U-Boot needs to be recompiled.

As for source code, they sent me a link to the binary image, but not the source.

Well. HP said that the source isn’t available.

So, e-mail sent to the debian-project mailing list (I figured that was the best place to aim at, rather than each individual package’s owner.)

The e-mail follows:

I recently purchased a Hewlett-Packard t5325 thin client. It uses HP’s ThinPro operating system, which is a derivative of Debian Lenny.

A written offer for the source code is included, stating that source is available at ftp://ftp.hp.com/pub/tcdebian, but there is nothing available for the t5325, only other models.

I have attempted to contact HP, and got a response that the code is unavailable, which is included below.

If additional information is required, please e-mail me at bhtooefr@gmail.com.

---

Forwarded message

---

From: HP Support <NA_ACTIVE_MAIL@hp.com>
Date: Sun, Apr 25, 2010 at 06:19
Subject: HP t5325 Thin Client Email Support <<#3945929#>>
To: bhtooefr@gmail.com


Thank you for contacting Hewlett-Packard.



This is in response to your e-mail regarding the HP t5325 Thin Client.




Hello Eric,

Thank you for contacting HP e-Solutions.

This e-mail is with reference to the case number: 3945929 for HP t5325 Thin Client.

According to your e-mail, I understand that you are looking for the source code for Linux Operating system.

I will be glad to assist you regarding this issue.

Eric, I would like to inform you that the Source Code for Linux Operating system is not available and also it is not available as download. In order to make sure about the above information you may refer the given below web link:

http://search.hp.com/query.html?cc=us&lang=en&charset=utf-8&qt=t5325&la=en

Please click on the above web link to get the web page.

Please feel free to contact us for any further assistance; we will be glad to assist you.

Once again, thank you for contacting HP e-Solutions.

Regards,
Ramachandra Aithal. B
Technical Support Engineer – HP e-Solutions

Do help share feedback with my Manager by mailing us at (Feedback_pt@hp.com)

NOTE: Our advice is strictly limited to the question(s) asked and is based on the information provided to us. Problems and solutions may depend on the nature of your system environment and various other parameters that are unknown to HP; therefore, HP cannot assume any responsibility or liability. Hewlett-Packard shall not be liable for any direct, indirect, special, incidental or consequential damages in connection with the use of this information.



Sincerely,

HP Email Support



Please include the following identifier in the subject line of all future correspondence relating to this ticket.



Incident ID: <3945929>>

Is this what you are looking for:

ftp://ftp.hp.com/pub/tcdebian/dists/t5325/

I just got an e-mail 10 minutes ago from Bdale Garbee with HP stating that it got put up. :)

So, yeah, this is new.

U-Boot sources are in there, too. :)

...I doubt the HP Labs folks [in Bristol] were involved...

I just got an e-mail 10 minutes ago from Bdale Garbee with HP stating that it got put up. :) So, yeah, this is new. U-Boot sources are in there, too. :)

• t5325 netboot on (reorganised) hp.com forums
• [Ltsp-discuss] HP t5325, ARM, gPXE (reported processor – linking to your blog, which I can’t access here – is ARMv5 not ARMv7!)
• [Edit: And this device manager update (which AFAICT doesn’t seem to add anything directly relevant)]