Semson's CF to PC Card adapter - PCMCIA Hard Drive Problem with 2215

[kagayaki1]

I've been doing some interesting experiments with the Semsons CF to PC Card adapter in my new HP 2215, and I've found the results to be disappointing. Here's why:

I've long used my 5GB PCMCIA card to store my PPC DivX movies. I would put the card in my PCMCIA sleeve, which was attached to my older 3870. Then, I would transfer the movie to my SD card. Unfortunately, with the infamous 1-bit bus, it took FOREVER!

Naturally, I wanted to move to a smaller device. So, I bought a 2215 and the CF to PC card adapter with this very thing in mind. When I first put the CF end in the device, and then attach my PC card, File Explorer initially sees the drive as "CF Card." In fact, for about 10 seconds, it's browsable. But inexplicitely, it powers down after about another 10 seconds, much like it's not being accessed, so it goes into standby mode. Then, it makes the OS slow way down until it's disconnected.

I wondered if this was the adapter, so I pulled out my old 3870 with a CF sleeve (unpowered, original version). There it worked just fine, no problems. I could copy, paste, stream, everything. So, that kind of rules out the adapter.

That either means there's a power output difference between the PDAs, or there's a bug in the OS. What do you all think?

[ctmagnus]

Not owning any of the components you mentioned I can't say first hand, but I would guess that the 2215 CF slot doesn't support the higher voltage requirements that PC card devices require.

[kagayaki1]

Perhaps, but if that were true, would it power up at all (it currently spins and is accessable for a very short time)?

[ctmagnus]

Perhaps, but if that were true, would it power up at all (it currently spins and is accessable for a very short time)?

Good question. Ya got me. :confused totally:

[Mark_Venture]

Perhaps, but if that were true, would it power up at all (it currently spins and is accessable for a very short time)?Hey, anything is possible when it comes to bus powered devices...

While not the same bus, some bus powered oddity examples...

I have 3 different portable bus powered USB 2.0 hard drives. a Buslink, Storix, and I-O Magic. Now, on one system (Abit BX6r2), the Storix and I-O Magic work just fine, but the Buslink requires me to use the PS/2 keyboard power tap to give it more power. The drive also requires more power even if it is plugged into a USB 2.0 hub that is powered by an A/C adapter. The same drives, all can work at the same time on my Asus P4PE board's USB ports without additional power. And I can use the Buslink and one of the other drives at the same time in the two USB ports built into my Thinkpad T23 without addition power.

I just bought a SanDisk 6-in-1 memory reader/writer. Its USB 2.0 and bus powered too. When plugged into the ports on my system board (Asus P4PE/L/R/SATA), It is recognized by Windows. Stick an SD memory card in the slot, and for about 30 seconds it works, then drops off like I unplugged. Put a CF card in it, or connect it to one of the ports on my USB 2.0 hub (powered via a/c adapter) it works fine. Neat considering I can hook up 3 of the bus powered hard drives at the same time to the system board ports and they should require more power than the 6-in-1 with SD card.

So bus powered devices can give you STRANGE symptoms or just work.

[kagayaki1]

Has anyone tested this with a 55xx series device? I would be really interested to see if it was a global product line decision.

[kagayaki1]

I just heard back from Semsons and Co. They were unaware of the problem, but plan to work on it now that they know. Here's what they say:

"Our CF to PC Card adapter is designed as a pass-through adapter. There is no internal battery nor capacitor to supply the power. The power is provided entirely by the PDA. However, according to Compact Flash Association. Any CF compliant slot should be able to provide 3.3V of power."

So, is it possible HP did not make a CF Association compliant slot? It works just fine with my CF Wi-Fi cards, which indicates the 3.3V is working. But the Kingston 5GB PCMCIA Card (rebranded Toshiba OEM) runs off both 3.3V and 5V. This should be working...still no definitive comment from HP. The only thing I heard back from them was "the Semsons adapter is not yet tested and approved for the 22xx series" ya-da ya-da. :roll:

This is a fairly minor topic, but so far I've not seen anyone else comment on it. I hope this makes it to the front page. Heads need to roll if the slot is not compliant.

[kagayaki1]

Just in from HP, they provided me with the following information:

"We would like to inform you that this behavior of the iPAQ H2200 is due to power output of the iPAQ. The iPAQ H3800 comes with a battery power of 1400mAh while the Power output available from H2200 series battery is only 900mAh."

:? I don't buy it. Someone correct me if I'm wrong, but I belive mAh deals with how long the energy can go on, not how much energy is delivered at any given time.

Still fighting to get the right answer.

[droppedd]

:? I don't buy it. Someone correct me if I'm wrong, but I belive mAh deals with how long the energy can go on, not how much energy is delivered at any given time.

That's right - i'm no electrical engineer, but from what i recall from physics, mAH is a measure of essentially how much juice the battery can supply while running at the battery's rated voltage. So mAH just affects runtime, not the "bandwidth" of power (more voltage is the equivalent of "wider pipes" in plumbing, as i understand it, where mAH would measure how big your water tank is relative to those pipes).

Thing is... the battery on the 2210 is rated 900mAH at 3.7V... so if anyone who knows electrical engineering can tell me... if the unit itself uses even .5 volts of that at a time, can't the battery not supply the required voltage to the CF slot for any reasonable period of time? Do other iPaqs have higher voltage batteries?

(btw for comparison's sake the average AA rechargable is 1300-2000 mAH at 1.2 volts, and my laptop's battery is 1800 mAH at 14.8 volts).

[kagayaki1]

Amperage technically defines the flow of the "juice," while voltage defines the pressure (also trying to recall some physics here).

What still blows me away is let's assume their conclusion is correct about the device. It still doesn't add up as to why it won't run the PCMCIA hard drive. The specs from Kingston's site suggest there should be plenty of power left.

SpinUp Current 370 mA (3.3V) or 380 mA (5.0V)
Active Read Current 360 mA (3.3V) or 370 mA (5.0V)
Active Write Current 390 mA (3.3V) or 400 mA (5.0V)
Idle Current 150 mA (3.3V) or 160 mA (5.0V)
Standby Current 70 mA (3.3V) or 75 mA (5.0V)
Sleep Current 15 mA (3.3V) or 20 mA (5.0V)

Anyone have more experience with power statistics?

[kagayaki1]

Also, I should note it still doesn't work when plugged into the wall with the battery removed. And I don't think there's a lack of power amperage from the wall. Must be an internal throughput problem.

[Jason Dunn]

Hey guys - I've emailed a contact at HP, we'll see what they have to say about it.

[ctmagnus]

Hey guys - I've emailed a contact at HP

I'd love to get a peek at your contacts list. ;)

[gry]

I'm with you.

My 3955 used my Toshiba 5Gb without any problems and the 2215 with the CF-to-PCMCIA adapter SHOULD work!

All of the excuses cannot be completely correct. I'm seeing other users at other sites reporting the same issues and claiming the same voltage problem from HP.

But why does the hdd even start spinning? Why can the PCMCIA card be read as a CF storage device? Why does explorer read the files?

It's got to be that the drivers don't know how to properly access the disk past initial read.

I'm seeing 5555 running wm2k3 users complaining also and I'm sure that PPC pumps out the right voltage and amps to the PC Card.

This has to be a romfix and driver issue. They simply forgot to put the drivers on wm2k3.

[kagayaki1]

I'm seeing 5555 running wm2k3 users complaining also and I'm sure that PPC pumps out the right voltage and amps to the PC Card.
Do you have a link to another discussion forum I could check out about the 5555? This really changes everything. This BS excuse I got from HP yesterday about the power doesn't hold water if there are 5555 users having the same problem.

[gry]

SOrry to cross post:
http://discussion.brighthand.com/showthread.php?s=40f0de29c998f7f2c2e8364b6518f1f7&threadid=84427&highlight=PCmcia

It's wm2k3...HP bastages. ;)

[kagayaki1]

Does anyone have a 5555, a CF sleeve, this adapter, and a PCMCIA hard drive to test with? We need an answer!!!

[kagayaki1]

I got another reply from Semsons. Detailed below:

"Hi Jason,

Thank you for your update.

It is possible that HP have changed the power requirement on the new iPAQs.

In our opinion, the best way to isolate the source problem is to test the
adapter with a PDA running on PPC 2002 and make sure it works. Then upgrade
to WM 2003 and test again to see if it still works.

If the adapter works before the upgrade but not working after the upgrade,
then we know for sure it is problem with WM2003 not the hardware. Otherwise,
it is possible that iPAQ 2210 does not supply enough power to run PCMCIA HD.

We will try to obtain a HP 2210 for testing when we have a chance.

Regards,

Support Team,
www.semsons.com"

This is a really, really long shot. But has anyone upgraded their device yet who had this kind of setup working in the past?

[kagayaki1]

I saw this post in a nearby thread about the external hard drive driver (http://www.tjotala.com/software/wince/storagebrickdriver/default.htm) that people had been playing with in the past. If it truly is a driver issue (and not power), this driver might accomodate the card better. I have absolutely no basis for conjecture here, but it shouldn't hurt to try it out. Let me know what you guys get.

[kagayaki1]

Well, the external driver didn't help. You know what, it just occured to me. Is there any way to "overclock" the battery throughput? That is, aside from physically soldering it, can more power be drawn from the battery than it's rated for?

[Zathras]

I started this Brighthand thread on their 5000 series iPAQ board. Two guys said they didn't have problems with their 5gb Toshibas and their 5555's with a PCMCIA sleeve. I don't know about trying this with a 5555, a Semson's adapter, and a CF sleeve.

http://discussion.brighthand.com/showthread.php?s=&threadid=84427

I've used my 5gb Toshiba with a 3955 using the Semsons adapter and a CF sleeve and it worked fine.

I contributed to a similar thread on the 2200 iPAQ board at Brighthand. I've emailed Semsons, Toshiba, and HP and no one will claim responsibility. I don't think the Semsons adapter is at fault. I wish HP would do some research and support their customers on this.

[droppedd]

Well, the external driver didn't help. You know what, it just occured to me. Is there any way to "overclock" the battery throughput? That is, aside from physically soldering it, can more power be drawn from the battery than it's rated for?

Even if you could, i would absolutely not play around with that. Computer circuits are VERY sensitive to voltage (as anyone who's ever fried a PCI card with static electricity can attest to). A few volts difference in the power going to the processor could leave you with a nice silvery paperweight. :?

Most desktop PCs do allow you, for example, to adjust the power (in fractions of volts) going to the PC... but really miniscule adjustments, only for overclocking when you need a tiny bit more juice for that extra 20 mhz.

Unless HP can change it themselves as a firmware release, if it's a voltage issue it's probably not user fixable unless you have an engineering degree, a soldering iron, a few capacitors and a willingness to void your warranty :).

[kagayaki1]

Unless HP can change it themselves as a firmware release, if it's a voltage issue it's probably not user fixable unless you have an engineering degree, a soldering iron, a few capacitors and a willingness to void your warranty :).
What do you think about the chances of HP releasing a more spacious battery? It's still unclear whether the device could handle more battery power anyhow, and it seems unlikely to me that HP would make it so a mid-level device could take a bigger battery (that's more of a 55xx enterprise thing). But still, that might work...