USB connection -- one more time

mynet43

I'm designing a Propeller board with a USB interface. I've done this several times before.

When I power the board from the USB port, I have no problem. When I power it from a battery or power supply, the USB sometimes causes problems.

In the past, I've read several of the threads as to how to solve this. There seems to be some controversy over how to best solve this.

What I'm looking for is the easiest way to overcome the problem, without adding active components. At one time, I believed that the problem would go away if you removed the LED's on the Tx and Rx lines. Does anyone know for sure if this will work?

I haven't followed this recently, but does anyone know if there is one agreed-upon easy solution for this problem?

I've attached a schematic of an old circuit that works fine, but it's powered by the USB port.

Thank you for your help.

Jim

Rayman

I guess you mean the problem where the TX or RX lines can power up the FTDI chip and then reset the Prop...

I think your choices (without using active components) are to either make sure nobody puts out VDD on the TX or RX lines, install 1k or bigger resistors between the Prop and the FTDI rx&tx, or install a jumper on the reset line between the Prop and the FTDI chip.

mynet43

Rayman,

Thanks for the feedback. If I install a jumper like you suggested, will I still be able to program the Prop over the USB port?

It looks like removing the LED's would remove the path to the 5V on the port. Do you think this would do it? (see my schematic)

Thank you for the help.

Jim

Rayman

I'm not clear on how your LEDs are powered in your new circuit...

Anyway, I think adding a jumper to the reset line is a simple and effective solution. Just install jumper to allow reprogramming, remove jumper to prevent unwanted resets...

mynet43

Rayman,

Now I understand about the jumper. Thanks!

As far as the LED's. I was proposing to remove them completely from the circuit. They're inside a box, so it doesn't matter anyway.

It looks to me like this would remove the feedback path to the FTDI Vcc. Does that seem right?

Thank you for your help.

Jim

Bobb Fwed

Why no active components? We use SN74LVC2G07 on our [production] product to prevent the reset problem when the USB is not plugged in, it's about as simple and low maintainable as an active device can get. It will fix your problem with one component, and it won't require you to manual manipulate the board (adding/removing a jumper) every time you reprogram it. This technique has been used by many of us on the forums.

mynet43 wrote: »

Rayman,

Now I understand about the jumper. Thanks!

As far as the LED's. I was proposing to remove them completely from the circuit. They're inside a box, so it doesn't matter anyway.

It looks to me like this would remove the feedback path to the FTDI Vcc. Does that seem right?

Thank you for your help.

Jim

mynet43

Bobb,

I really don't have an objection to using the SN74LVC2G07 on my board. I'm just looking for a simple proven solution.

Can you give me a sample schematic to show how it's used. There was so much chatter on the earlier threads that I was really confused as to the best approach.

This sounds like it may well be the simplest solution, which is what I'm looking for.

Thank you for your help.

Jim

Bobb Fwed

Here you go. I just noticed you don't have pull-ups on your RX and TX lines to the propeller. I didn't think the communication would work without that. Either way, this circuit definitely needs them.