LinuxMCE Forums
General => Users => Topic started by: Armor Gnome on August 20, 2012, 01:40:32 pm
-
During download and install of 1004 beta I had some time so, I am once again posting about another project of mine...
I have always wanted to make my own cradle/charging station for my favorite LMCE controllers and other always-near items after seeing someone years ago post a photo of a wooden rack they built for the coffee table. The pinout for my orbiter has been on the wiki page since that time and I finally got around to building the 'guts' of my new console. My 10+ years working with fiberglass molding, plastics and aerospace composites should help me produce a pretty nice showpiece but I am having some trouble with the wiring. I was hoping someone either has a OEM dock for this and could give me some info or is good with electronics and can tell me where I am too off the wall.
Proof of Concept Photo (http://i1270.photobucket.com/albums/jj604/jeremy_mcdiarmid1/proof.jpg) In summary I black taped all 18 pins on the bottom of the DT366. I then cut out the tape from the pins I wanted to contact, giving me less chance of hitting the wrong pin. The cigarette box and duck-tape was special just for tschak909. After struggling for too long to balance and line up my exposed wires to exposed pins I realized this probably will not work unless the 12v hot and GRD is also connected. So I never actually got proof of concept...
Moving to Something More Permanent (http://i1270.photobucket.com/albums/jj604/jeremy_mcdiarmid1/under.jpg) Okay besides bad soldering (it was late and heating paperclip sections that softened the vera before solder could flow sucked) This is the underside showing the two connections I am making. A dock to charge my iPhone and an open USB-A to charge whatever. You can also see that a standard PC power cord will power this console when complete.
Homemade Pins (http://i1270.photobucket.com/albums/jj604/jeremy_mcdiarmid1/up.jpg) This view is what will actually face up and the DT will cradle into. Obviously I need to straighten the pins and set the height but that can wait until after this comes out of the mold. The switches are momentary contact switches on the 12v hot from the transformer. They must both be pressed down for the cradle to power on. I added this because I didnt like the idea of 12v + and GRD next to each other in the open. (The tape was only there to make my pins stay in place while I wired the underside tschak909)
Where I could use some feedback is on my idea of a safety switch, lack of any diodes or resistors on these circuits and any other suggestion that would give me power to the USBs even if the DT is not cradled. I thought about a nice flushmount pushbutton "ON" toggle on the face of the cradle but decided against it. If you have an OEM dock for the DT I would really appreciate a photo or component model numbers/colors.
The body is coming along nice but sorry no photos yet. My mold is built with heavy foam atm and I havent picked up clay to coat the curves and fire it to make a reusable mold. Much appreciation to whomever posted the wiring pin-out on the wiki. I thought about chasing down the pinout for the USB-B port but decided 1 open USB is fine so long as I don't have to chase my iPhone cord ever again.
-
Don't have one, but when you mentioned it, I was curious, because I recall that it had the same CPU as the PepperPad 3 (one of which I own). But I found a forum dedicated to the device, here;
http://webdt.org/forum/
There seems to be recent activity. Maybe someone can answer your questions...
-
Maybe a very dumb question but dont the pad have i charger cable that you can just put in the mold?
BR Daniel
-
From what I've seen in pics,... The unit has a standard wall-wort type charger, but can also be charged via it's bottom connector. Plus I think the wall-wort plug input is up near the top of the unit... no convenient. I think Armor Gnome is looking to do a charging while docked thing.
-
You trying to build one of these?
http://lmcecompatible.com/#ecwid:category=280104&mode=product&product=839081
Tim
-
Tim-
No, I am building a custom device with the features of the OEM cradle + my styling and a built in iPhone charger. See my two comments below about your store items. I don't think I am foot stepping at all but message me if you want and we can chat about it.
dont the pad have i charger cable that you can just put in the mold?
BR Daniel
Actually that was a great question and one I tried to incorporate into the dock. I thought about doing something similar to a door latch, rounded on one side and spring extended into the dock area. Sliding the pad down into the dock would push in the power connector away and when it reached bottom the plug would be spring pressed into the jack. The problems with this idea were the power jack (on the side) is level with the front 4 buttons and to expose them I would have to modify the way it sits into its slot. I could do that but it would be almost impossible to match with the iPhone dock below it. The other problem and this is function instead of style, I do not think the bottom pins are live until the unit receives power from the bottom. When hunting for the USB-B pins I didn't get anything on the multimeter anywhere. It's possible the bottom pins are disabled in BIOS but to be honest I am not sure how to get into BIOS on that pad, but short of using jumpers to disable connections I haven't came across old tech that could kill all power to a bus.
Thanks for the link JaseP. I will spend some time hunting around that site. I did see quite a few threads of their users looking for OEM docks, perhaps I could drop a linuxmcecompatible link? (don't know the rules on that one so I will get with the admins and tkmedia about that one). The cast is drying now so I should have some new photos available shortly. Since these are being produced with a mold there is no reason I can't make a few and offer them freely without electronics, sorry but selling a working unit even for donations would be in competition and I won't do that. Tell me what color and pay the shipping, add your own guts.
The wall-wort plug I am housing inside the cradle is other than the PC standard 110-120V line plug, exactly the same as the OEM charger. 12v 3.5A. Something I am wondering now is if the cradle takes a higher Amp power supply?
*Edit, the switches I grabbed to work as safeties for the 12v +, they were normally closed... I have to head to the store to buy the right ones now.
-
@ Armor Gnome;
Sounds like you need a 3D printer...
-
The wall-wort plug I am housing inside the cradle is other than the PC standard 110-120V line plug, exactly the same as the OEM charger. 12v 3.5A. Something I am wondering now is if the cradle takes a higher Amp power supply?
Same power supply, no additional requirements for power. I *believe* the charger charges the pad and the usb ports on the cradle are then powered from the pad, not the charger directly. What portion(s) of a cradle would you like pics of?
J.
-
I *believe* the charger charges the pad and the usb ports on the cradle are then powered from the pad, not the charger directly.
This is exactly what I was thinking, and is why I stopped the 'proof of concept" phase of the build where I was only trying to connect the 4 matched USB pins + shield ground. I will also say I *believe* the usb ports on the cradle are only accessible when 12v power is supplied by the cradle. Thanks for confirming my suspicion, that stops me from reattempting the project in six months to see if I really needed power.
What portion(s) of a cradle would you like pics of?
I am not sure if access to the wiring is a simple screw turn or not but I am curious about wiring from the power adapter jack on the cradle up to pin 9 and 16. My plan is to add in contact switches to prevent power always going to pin 9 but I am curious about how DT Research did it. Is it wired from jack to pin direct or does it have components such as resistors, switched diodes or caps?
The small pads on the orbiter don't leave much room for error when this thing ultimately settles into my homemade socket, so anything I can do to prevent 12v surging into my device because it misaligned and hit another pad would be worth it.
The photo request was for anyone willing but without knowledge of components, if you are comfortable detailing or describing the diagram that would be much easier than uploading a photo.
Thanks for the help!
-
Well. I have the following, based on the diagrams on the wiki page (note that these power pins differ from what is on the wiki):
P1 @ pins 9, 12, 13, 16. These pins are in a square group. No protection from plug to pins. (+ve)
P2 @ pins 1, 4, 15, 18 & 7. First 4 pins are the two outer sets of two pins. (gnd) Connected to all usb grounds and usb housings as well.
I read exactly the same resistance from these pins to the power connector as I read resistance from my leads alone, makes the pins to power connector = 0 ohms.
There is a small cap between +'ve and gnd. A few more caps around the usb connectors.
This is the cradle that came with the WebDT 366LXs that I own.
The model of this board is DT-375 Cradle Board Rev 1.1 with an attached DT-375 Cradle Usb Board Rev 1.1
The cradle is model ACC-008-10, Input DC12V, 3.5A
Four pins for +/gnd is much more realistic to me for the current requirements of an in-use charging webdt with peripherals attached.
J.
-
That was thorough, thank you!
I actually had a pretty long response typed just now but decided to message you about the parts I am unclear on, especially where the +ve to gnd happens in the cradle. I was wiring it to tie into the webpad's gnd and leave that part open otherwise. If nothing else it sounds like there may be room to improve on the wiki.
I checked out webdt.org btw and when searching for cradle pinout, I found a copy/paste of the linuxmce wiki entry (credit was given). I just thought it was cool that a dedicated device forum found an answer to their questions from our wiki. Following that wiki back to these forums I found that a member did the pin mapping himself and posted it for not only lmce users or even linux users but anyone searching that device to get information. If someone ever wants to make a dock starting with a cigarette pack I guess they know who to contact now. =)
-
You're welcome. :) There are also two guide posts on the docking station that 'guide' the webdt onto the cradle so that the pins only contact the locations they are supposed to. You may try to incorporate something like that to prevent mis-alignment of your webdt/cradle pins. These posts mesh with small holes on either side of the pin locations.
J.
-
There is a small cap between +'ve and gnd. A few more caps around the usb connectors.
The caps will be important, as those are for DC power conditioning and noise suppression, so Armor Gnome will need to match those values in his build. You could always post a pic of the parts if they're surface mount, and we could figure out the values for him.
Armor Gnome,
Make sure your ground pins are longer than your power pins. For 'hot insertion' you always want ground to make first, and break last to prevent damage to your electronics.
HTH!
/Mike
(former telecom R & D lab technician)
-
I have used USB pins on mobos straight to the USB and I have pulled them from cases where the mobo leads went to a board with caps and diodes before going to the USB housing. Both have worked so I wasn't overly concerned with that side of it. On the iPhone USB hardwired dock I need to power up the WebDT with the orbiter running and see if the detection of USB storage causes any problems. I have no reason to suspect it would but its an easy test to see how the iPhone is handled if present at startup, when coming out of suspend and detected while running.
Great tips on the ground pin length and alignment issue. I have planned to address the fitting issue with epoxy around the pins. Using cellophane and clay I can pull a cast of the DT's underside. That would give me an exact mold of the shape to match around my pins (and the alignment pins):
I I I
I I I
cast to
[ I I I ]
[ I I I ]
shaped down to
. . .
[ I I I ]
I will have to consider new methods if I want to extend my ground pins as recommended. The normally open momentary contact switches would resolve this being a 'hot insertion' but only if the throw on the switch is long enough. Everything is soldered rigid onto vera board atm, so adding a floating pin would take some thought. Maybe I should scrap the vera + steel pins, and rebuild the project using coated pins from a non-working floppy drive. If I mount that to PCB I could ammonia etch my own traces and duplicate the board that was used OEM. It's a shame I am not in Arizona anymore as I had access to several etchers including a mini Cupric Chloride model that would allow me much finer and tighter spaced traces. With this much thought into it already should rebuilt it right so that it not only works but is clean and will continue to work for years.
The homemade glue I used on my form is almost set so I can start shaping the final body design with plaster and then clay. The rough shape is pretty huge and could definitely use about half of its volume reduced, If I add a solid wood bottom I can keep the mass while shrinking the volume. Ideally the shape of the cradle should compliment both devices and that is rough. Where the webpad has a durable look the iPhone has the no-edge sleek style. Without laying both devices almost on their backs I am having a hard time getting a dock that shows both screens but isn't unnecessarily tall. Sometimes even the sleekest looking 3D model ends up bigger than desired once cut into foam.
I appreciate the input on this. I will try to add a few more photo links for those who want to see the project as it develops.
-
The caps will be important, as those are for DC power conditioning and noise suppression, so Armor Gnome will need to match those values in his build. You could always post a pic of the parts if they're surface mount, and we could figure out the values for him.
As the same power adapter is used to plug directly into the webdt OR the cradle, the cap is likely not a requirement. It will be smoothing any ripple in the DC voltage and suppressing noise as you suggest. I can read the part numbers if they are required but I would expect they are not entirely necessary in this case.
On the original cradle the four outer pins are raised to 'make' the gnd connection before the +'ve, as you have suggested AG should do.
I have used USB pins on mobos straight to the USB and I have pulled them from cases where the mobo leads went to a board with caps and diodes before going to the USB housing. Both have worked so I wasn't overly concerned with that side of it. On the iPhone USB hardwired dock I need to power up the WebDT with the orbiter running and see if the detection of USB storage causes any problems. I have no reason to suspect it would but its an easy test to see how the iPhone is handled if present at startup, when coming out of suspend and detected while running.
This is not specifically indicicative of you iphone but I was unable to get my WebDT cradle usb ports to charge my Android phone last night. I only tried once and didn't muck with it to determine what the issue may be. My android phone *does* charge if I simply connect gnd and +5v.
J.
-
On the original cradle the four outer pins are raised to 'make' the gnd connection before the +'ve, as you have suggested AG should do.
Perhaps the pin floats on a convex bubble above its' contact, kind of like joystick D-pad or old keyboard? If that is the case then it is achieving gnd first AND normally open functionality like I plan to do on the circuit, all at the pin level. That is an great idea for me to add into my build.
My android phone *does* charge if I simply connect gnd and +5v.
I am no electronics expert by a long shot but I read a really good article on the way power to USB is handled. There is a minimal amount of voltage available on the line (sorry I forget the exacts) until the device requests more power from the bus. So the host does not power the USB port fully unless there is some logic at the connected device level to request it. This is why when stealing power off a USB port you wire two pin feeds together (again forgot the specific colors but I believe its white and green) this gives you a closed connection which tells the bus something is attached and provides level 2 of its available 3 power levels to the 5v line. I am guessing here because I don't have the article infront of me but it was something like: +1.5v constant, +3.5v when something contacts all pins, +5v when the device requests all available power. This understanding was helpful when understanding how USB hubs function, because some devices ask for the power they need and others request full power as soon as attached.
-
Perhaps the pin floats on a convex bubble above its' contact, kind of like joystick D-pad or old keyboard? If that is the case then it is achieving gnd first AND normally open functionality like I plan to do on the circuit, all at the pin level. That is an great idea for me to add into my build.
It's floating on a spring, or similar, and is not N/O, it is connected all the time.
J.
-
Makes sense the more I think about it. If all pins are rigid and hot, with a floating extended gnd, you would need the gnd always available for that pin. It would then be first touched on insertion and last broken on device removal. To add my switch idea I would need to make all pins floating and active, with my +ve rigid. To really complicate things I could have all pins rigid hot, all gnd floating hot, and +ve shorter than the rest of the pins and floating N/O - solder the height difference onto the orbiter pads and I could assure +ve would be first to break on removal and last to connect on insertion. Or all pins rigid, 4 +ve floating high N/O and 1 gnd floating high not switched. Gnd touches first, then all but +ve and then +ve switches closed, on removal +ve breaks first, then all but gnd, and last would be the 1 floating gnd.
How I finally wire it is going to depend on availability of floating pins. If I locate 18 always active floating pins cheap then all I need to do is clip height on all but gnds. If floating pins are hard to come by or expensive to buy each then I use rigid pins for 17 contacts and a floating pin for 1 gnd. It may seem excessive but I still don't like the idea of +ve and gnd available for anything to cross or get bit by. (I doubt many here realize it but I am in the occupational health & safety field) It's only 12v but to me it's not quite right unless its disabled when not in use.
Here is my new current build idea
| | | | side view of the receiving portion of the dock
| | | |
| | [ ][][][ ]-[ ][][][ ]-[ ][][][ ] | | floating perforated board like vera riding on springs - showing the alignment pins non electrical which guide the webpad down
| | i | | showing my one floating always on gnd pin "i"
| | | | | | | i i | | | | showing my rigid pins, the "i" being +ve on N/O contact switches
| | | | | | | | | | | | |
| | contact point for +ve
... I have way too much fun formating! The only thing I couldn't show was the springs which would just guide the perforated board up and down.
-
You ARE having way too much fun! ;-) Let me know if there is anything else I can help with/provide.
J.
PS, keep having fun!
-
I have the webpad sitting in clay attempting to pull a mold for my pin guides and alignment posts. While that sets I decided to scrap the old cradle design and came up with an all new one. After building a basic 3 angle view I was able to take point measurements and count pixels to get my dimensions. Would have been a lot easier if I used something other than GIMP but use what you have.
The basic shape is cut from stacked foam blocks and hand carved to my drawings. Keep in mind everything is undersized in these photos so that when I go over the foam with clay I can add everywhere and get the geometry exact. The design I am using is pretty basic but if the lines are not straight and angles don't match then I wasted my time and would have been better off sticking with the cigarette box. :)
The next set of photos for the cradle body should be clay, fired clay, mold, and then first pass attempt. On the electronics side I am stalled for parts so it's likely I will have a color molded cradle with all my finishing touches waiting on guts. The design is actually pretty universal and would work with slight modifications for ipad/iphone, nokia 800/900 and phone, etc. Once the mold is made it really is no trouble to make small changes in opening sizes to fit other devices. I just dremel away glass for bigger phones and add glass for smaller phones. I love tinkering so as stated earlier no charge. Linuxmcecompatible has this exact unit or message me about customizing something for a different device. My own builds are simple based on broken-wallet-ness but I would love to get some of my ideas out of my head and into someone's living room.
Foam only - Devices resting comfortably (http://i1270.photobucket.com/albums/jj604/jeremy_mcdiarmid1/Foam.png)
Fairly happy with the shape of the foam, keep in mind the final product is in places going to be 3/8 thicker. The angle of the devices is almost perfect though and I was impressed by how well foam was able to keep a line. Total measurements currently = 10.5" tall, 11.5" wide, 7" deep. The orbiter sticks up about 0.5 inches above the backrest so its viewable area starts approximately 5.5" from the base and because of its lean back extends to about 10.5" from the base. Configuration was set for desktop use next to a 22-25" monitor. Depending on how you sit it should fall just below eye-level and just about perfect placement for button presses.
Empty Foam Template (http://i1270.photobucket.com/albums/jj604/jeremy_mcdiarmid1/FoamEmpty.png)
This shot shows the angles of my design a little better. You can also see the area I left myself for adding electronics, which will be critical when customizing these. The relief cut into the backrest is for the external battery of the wedpad, it can get warm sometimes so I didn't want to box it in completely. Even in foam the cradle is very stable, I have a slight lean back to my devices but they sit low enough to not cause an issue. The base of the unit that houses the opening for a USB port will actually support an up to 3/4" wood bottom plate. If it so much as wobbles after that I would shocked.
Easy configuration options from here I may add to other pulls on this mold. Adding an LCD alarm clock instead of a phone, adding a powered USB hub internally to make this a 4-port charging station for multiple users, adding an easy open back panel for hiding things such as chargers for rechargeable batteries, I would LOVE to get my hands on Squeezebox dimensions to put audio playback just below a full orbiter UI. Of course I have my old crappy parts and weird projects that will find their way into these... integrated wireless AP, old receiver/amplifier parts to power audio from the core, add a chumby-sized orbiter below and a standard small LCD screen above from an embedded MD (no messing with touchscreen support, and keep full screen media with orbiter functions below, extend the base depth wise and add a compatible networked phone...
Finally, I haven't uploaded any of my finished model renderings because I have some little extras in mind that I would rather "Taaa Da!" than try to describe.
-
Awesomesauce.
You make me want to make things again.
-
He makes me want to spend money that I really shouldn't (and probably can't) on a Makerbot or a Solidoodle...
And L3ME... You DO make things!!! ... Just sayin'
-
... Well I hope it can be saved. If not I have more foam and my dimensional drawings.
Typically for something like I this I would finish off my wooden/foam/wire-mesh/whatever with DuraBond 90 in a few thin fast coats and be done with it. I have worked with clay before but only for small parts where I shaped the part out of a solid block. Using clay as a covering was an entirely different process that took me a while to get the hang of. Instead of nice thin coats shaping sharp edges this wet clay I used actually worked best when applied an inch think and then dug back down to the shape I wanted. I never did get the hang of a nice sharp line so deviated from my plans and added some slow curves. - an hour later and it resembled a vase, an hour after that a standard fair clay pot. Once I caught on to the idea of carving into excess clay for straight edges I was running out of steam. Tomorrow I will decide if I like what I see, any chance anybody wants an ashtray/pot/vase?
l3mce - My classes are all on-line so combined with a LMCE hobby its pretty typical for me to have a string of bed-computerchair-bed days. The day spent building something was a nice break that will let me get back to some template creation and C++ practice exorcises while I work on the wiring here at my desk/test bench.
I am aware that just because the device I am placing on top of this thing is supported under LMCE, this thread is not forum relevant. Thank you for allowing me to post about it. Because of feedback received I am going in a completely different direction electronically that without a doubt will make the difference between a working cradle and a pile of fiberglass and solder.