A switch would be easier to design but more cumbersome for the user. I'm thinking of it doing it automatically.... assuming i can figure out a way.

Originally Posted by Bxsteez

A switch would be easier to design but more cumbersome for the user. I'm thinking of it doing it automatically.... assuming i can figure out a way.

Why not just another port? You can put a charging port, usb host port and usb data port.

Originally Posted by used2hvatreo

Why not just another port? You can put a charging port, usb host port and usb data port.

That is a good idea but i believe there are is only one serial pair for the Tab. Then you would only be able to use one at a time but I might make two ports that are different. That would help with the confusion. Maybe a USB B for syncing and USB A for USB Host. Hmmmmmm... I have to think about it a bit. I have to make this adapter as cheap as possible too so i have to be a little relaxed on the circuit components.

It sucks not being large enough to buy 10000 FETS or resistors. They are expensive when you only buy like 10 or 100. So far the parts are only 2 or 3 bucks, of course that is excluding the PWB and whatever housing i use to package this thing.

I was looking over my schematic this afternoon with a coworker and it looks like i have my PMOSes oriented incorrectly and on top of that the ground from the wall is likely to be floating so the circuit won't work as it is now. I have to go back to the drawing board and correct it. I will update......

Still trying to figure out why my simulations worked.

Originally Posted by Bxsteez

I was looking over my schematic this afternoon with a coworker and it looks like i have my PMOSes oriented incorrectly and on top of that the ground from the wall is likely to be floating so the circuit won't work as it is now. I have to go back to the drawing board and correct it. I will update......

Still trying to figure out why my simulations worked.

Your simulations worked because you have used the same ground everywhere. If you are using the NI Circuit Designer then you definitely have the same ground. When you switch over to design your board, you will have to run a trace for the ground as well, but in the circuit design phase it lets you simulate by just putting the ground symbol there.

As far as the ports go, I was thinking the same thing. mini-USB for charging, micro-USB for data. As far as host, I'm not sure what can and can't be done at the same time.

Also, I was just looking around, and maybe the version 1 answer for HDMI is to simply map the MHL lines to another micro-USB port, and then allow the use of other cheaper usb to hdmi adapters, like the one I found for 11.99 on amazon

Originally Posted by used2hvatreo

Your simulations worked because you have used the same ground everywhere. If you are using the NI Circuit Designer then you definitely have the same ground. When you switch over to design your board, you will have to run a trace for the ground as well, but in the circuit design phase it lets you simulate by just putting the ground symbol there.

As far as the ports go, I was thinking the same thing. mini-USB for charging, micro-USB for data. As far as host, I'm not sure what can and can't be done at the same time.

Also, I was just looking around, and maybe the version 1 answer for HDMI is to simply map the MHL lines to another micro-USB port, and then allow the use of other cheaper usb to hdmi adapters, like the one I found for 11.99 on amazon

Oh i know why the grounding the p-channel gate worked and why it won't work in real life but it was the fact that the P-channel is backwards and the output still worked.

I think i figured it out. I think current is flowing because the voltage is biasing the body diode and the current flows enough to raise the source voltage (load voltage) enough to breach the threshold point and turn on the FET. Its all kinds of wrong with the right answer lol. It was a silly error on my part. I have another idea but its going to force me to add a third FET which I was trying to avoid.

On the connectors. I don't think the ternminals for Micro USB can carry enough current to support chargin the tab at 2A. I could be wrong. I think cellphone charges use 1A. I have to check on the specs.

Originally Posted by Bxsteez

Oh i know why the grounding the p-channel gate worked and why it won't work in real life but it was the fact that the P-channel is backwards and the output still worked.

I think i figured it out. I think current is flowing because the voltage is biasing the body diode and the current flows enough to raise the source voltage (load voltage) enough to breach the threshold point and turn on the FET. Its all kinds of wrong with the right answer lol. It was a silly error on my part. I have another idea but its going to force me to add a third FET which I was trying to avoid.

On the connectors. I don't think the ternminals for Micro USB can carry enough current to support chargin the tab at 2A. I could be wrong. I think cellphone charges use 1A. I have to check on the specs.

I was definitely going more for the idea of multiple diff ports so it would be straight forward as to which port did what, but yeah, def check it out, would hate to be responisble for burning up something because it couldn't handle the amps

Are there any updates on this? I would hope this is still getting done-

No updates yet. I've been kinda swamped with work so I haven't had a chance to update the schematic or start on the layout. Its coming though

Originally Posted by used2hvatreo

Also, I was just looking around, and maybe the version 1 answer for HDMI is to simply map the MHL lines to another micro-USB port, and then allow the use of other cheaper usb to hdmi adapters, like the one I found for 11.99 on amazon

Did you made this? I'm really interested in building this as an adapter. I have a Galaxy Nexus and a TAB. With shomething like this I could use the same MHL to HDMI adapter with both devices.