Why a 1-port powered hub? I'm happily injecting +5V down the USB lines without issue.

Also, since these adapters are quite expensive anyway (£5 each + £5 international delivery!) - a slight cost hike for added functionality shouldn't be too much of an issue in the long run.



If push came to shove and you wanted to disconnect the main supply from the N810 you could use a 2.5mm TRS socket that disconnects the internal supply from the external.


(I know I said Nokia Style connector earlier - I should've said the above!)


You should be able to source small inline 2.5mm TRS sockets quite easily.



Why I'm pushing for this: piggybacking male-female plugs over and over again to achieve one specific solution at a go results in a heavier Frankensteinesque dongle. An all in one elegant solution like this would be best for all concerned. Few devices work on a <100mA supply anyway [some memory card readers, keyboards, self-powered USB devices, and that's about all that spring to mind] - extra power injection is needed in the long run.



Also - while we're on the subject of possible improvements in manufacturing. Would you consider using a 90 degree Micro-B plug. Again - I know you have the 3D swivel for any degrees of freedom - but with a 90 degree right angle plug you'll decrease the amount the plug sticks out of the size much further. Costs in terms of manufacturing should be limited as the plugs themselves are only a few cents more than the normal ones.



My 2 pence, if you're manufacturing these from scratch then we may as well get all the bells and whistles at once!

I agree with wierdo why not just a short micro male A to female B pigtail adapter? It would be much cheaper to make and serve the same purpose and less fragile.

Thanks for the feedback so far everyone! Yabbas, a few points to discuss:
1) You say that you're injecting 5V onto the USB no problem; if you are simply adding 5V to the USB power lines I'm quite surprised that your devices are working without software hacks/modifications. The reason that you need a powered hub/intelligent device is that the USB standard dictates that when you connect a device, it sends a message to the host requesting a certain amount of power be supplied; the host (in this case the tablet) then responds by either supplying the power, or disabling the device if sufficient power is not available. If you just do a direct injection the tablet still thinks it's only capable of supplying it's limited 125mA, so even though the device technically has sufficient power through your external source, the tablet's kernel SHOULD refuse to enable it. Other members of the forums here (along with myself) have done experiments in the past to confirm this, and that the only way around it is to forcibly disable USB power management in the kernel, which at the end of the day really isn't a good plan. I don't remember the exact thread, but the discussion comes up about once a week whenever somebody asks about "my USB device needs more power!".
Also you're talking about cost... right now you're saying 10 quid is quite expensive; to integrate a USB hub chip into the adapter would add the cost of the components (chip plus a few resistors and a cap or two, not a big deal.. say another few dollars tops) PLUS the cost of PCB production and integration (adds a few more dollars per unit). In addition the size of the adapter would have to increase to accomodate both the components required and the power port (the power port itself while it looks small on the outside actually takes a fair amount of space). So at the end of the day your basically doubling the cost of the adapter to add a power port to it; not only that, but since there's a hub chip in there now you've actually reduced the amount of power available for devices when if you're travelling around and don't have an external power source to plug in available (usb hubs consume approximately 80-100mA according to USB standards... again whether they really do or not the issue, it's how much the kernel says they do, and thus how much it assumes it has left to allocate to other devices).
Don't get me wrong, I'd love to have a solution like you're talking about, but I've been researching it for about 2 months now and at the end of the day it just doesn't make any sense. The power supplied by the tablets currently is sufficient for most "on the go" type of equipment (thumb drives, etc). With the exception of the extra card readers (and I know what you mean) most other USB devices are connected when you're stationary somewhere, so having a second small cable/adapter/your own USB hub whatever for those times is not a huge issue as you're not concerned about ultra mobility. A better bet is to try and convince Nokia to make a bit more power available on the USB bus from the tablet itself in future software/devices... (perhaps improvements in battery technology/power management can make this possible?)

On the 90 degree point... I think it's quite mute really; the adatper still has to be the width of the female A connector on the bottom anyways in order to fit the "360 swively hinge thing", and the male micro is actually quite short (it looks larger in the picture). In it's normal configuration it would stick out about 3/4", with a right angle it would be about 1/32" smaller, and you'd loose some of the benefit of the 360 as the hinge would be butted up against the tablet itself on one side.

I'm all for doing it right the first time (especially in this sort of quantity!) but it has to be viable. (Also if you're concerned over shipping costs... group buys or finding somebody/setting up your own local distribution for them would make shipping more afordable as I can send probably about 15-25 for that $11, and higher quantities aren't a lot more... it's just that's the minimum to ship anything overseas and have it actually get there).

Thanks again for all the ideas and feedback!
-Rob

Welcome aboard, Antikx

It's an odd one then as the two tests I've done show otherwise.

I've successfully injected +5V on two devices that clearly wouldn't work (and didn't work) on the N810's 100mA supply by disconnecting the internal +5V and Ground and running it through my own battery supply. This includes a larger SD Card reader that wouldn't power with the N810 supply, and a Huwaei HSDPA Modem that was detected fine (dmesg logs said so, but had no drivers installed) - without power injection the modem wasn't detected at all.


Strictly speaking you're right about power negotiation (part of Host Negotiation Protocol) - but I assumed that the host would try to initialize the device when in low power mode, retry negotiation despite there being insufficient power "just in case" there is enough residual power to kick things off - warn you that power is insufficient (draws a lot of current) but maintain the device connection if it can. Kind-of like a fallback.


It may well be down to USB2 sporting the: Battery Charging Specification 1.0: Released in March 2007.

Adds support for dedicated chargers (power supplies with USB connectors), host chargers (USB hosts that can act as chargers) and the No Dead Battery provision which allows devices to temporarily draw 100 mA current after they have been attached. If a USB device is connected to dedicated charger or host charger, maximum current drawn by the device may be as high as 1.5 A. (Note that this document is not distributed with USB 2.0 specification package.)



A lot of these devices "power up" when connected to dedicated PSUs -- my mobile phone for example, the Huwaei modem (blinks and negotiates a 3G connection as LED turns blue), a typical card reader lights green, a portable hard drive spins up. Perhaps when powered up with higher current (as is the case when using Power Injection or your own USB charger) their "power negotiation" protocols in HNP are skipped?


The Application note here: http://www.maxim-ic.com/appnotes.cfm...e_number/1822/ briefly covers OTG and has some interesting info on the type of circuitry involved.


*shrugs* All this is moot now anyway since you're saying there are conflicting reports on what works and what doesn't.





I'd probably go with a solution that a few others have mentioned - a pigtail. But add the right angle option so you could stick it to the back of the N810 and forget it's there.

Doesn't the pigtail design have these 3 advantages over the right angle option?

1 - Places less stress on the N810 microplug connector, therefore less likely to break. I have twice had failures at the connection point with other devices because of stress breaking connection on the circuit board.

2 - Simpler, therefore probably more reliable

3 - Simpler, therefore possibly cheaper

Misunderstood me when I say right angle option. I mean one where the side profile is minimized.

See: http://www.hosiden.co.jp/web/english...se/hpc1520.htm




One of those on the Nokia end, going straight to a USB-A socket...

Put me down for one for one of those hinged adapters (as per 1st post).

Cheers

I'll take one as well.

-kenroy

I'll take three (or maybe more) of the hinged, though I would like a 360 version for my n800 as well.

One other thing that would be useful would be a short (maybe retractable) cable with micro-A on one end and a mini-B on the other. The camera, GPS, powered hub, and most other things I plug into the n810 use this connector.

On the powered hub matter, injecting 5v might make something work, but then the device is lying if it says it only needs 100mA. The N8x0 will reject anything that tells it that it requires more than 100mA over USB (hence the 2G iPod works because it recharges/powers over the Firewire slot, but later ones won't work because they want to be recharged via USB even though all have batteries). I just got a few small portable hubs with external power connectors and they solve the problem.