First of all thanks to all of you helping and sharing your ideas on this one. Second; Bill this is to work in line with a USB so the supply voltage is 5V DC and the load can vary from 500mA to a theoretical max of 1A. Is it feasible to get the same circuit to work?

Again much appreciated guys!!!

Many thanks

 

I need to know the parts (mainly your timing capacitors), so I can finish drawing it.

I don't think 5 V will be an issue. Lower voltage than I like to go, and you will not be able to pull 500ma from a USB. What is on the other side of the relay?

If it AC read up on SSRs. They are bit more expensive, but they are excellent parts.

 

I want to strip down a usb cable and plug on to the computer the one side of the cable, then the circuit we are discussing and then on the other side a usb modem. so the other side will need 5VDC 1A max but I assume that I can use a Y shaped usb cable to get 1A max out of two usb ports.

 

OK, I'll assume the capacitors, and if you can come up with something closer than the assumption I'll redraw it.

I'll say it again, you are not going to get 1A out of a USB port. You'll probably blow a fuse in the motherboard, if you're lucky it is a self resetting type breaker instead of a fuse. I haven't been that lucky myself, I wound up buying my neice a powered USB hub and a new USB card to allow her to use her IPod.

USB 2.0 is rated for 0.1A, USB 3.0 is rated for 0.15A per port. The total draw for USB 2.0 (with all the devices sucking power out of several ports) is 500ma, and for USB 3.0 it is 900ma. USB 3.0 is brand new, it came out in 2010. This is not out of one port, it is meant to be the total draw.

http://en.wikipedia.org/wiki/Usb

 

I am sorry you are right! I assume max 200mA which I might get from two usb ports. Timing is 23:57 hours on and for 3 minutes off...if that makes any sense...

Thanks again !!!!!

 

You won't get 23:57 accuracy with an RC circuit. You will need a crystal source as I've stated before.
I would recommend a small microcontroller with a 32,768Hz crystal.

What are you trying to control? Is this a school project?

 

We've basically discussed the accuracy issue in the first two pages. It brings it down to two chips (vs. four). The end use isn't known. Me, I like these kind of projects just for the fun of designing.

 

OK, here is the finished schematic, hopefully minus brain farts. I may try building one myself, it is an interesting project. The relay is activated for 1 minute, if you want multiple minutes multiply R6 by how many minutes you want.

 

Thanks a lot for that Bill!!!I'll give it a try and I'll tell you the result!!
Thanks Again for your time!!

 

You can draw 1A from USB. Most ports do not implement the automatic current limit because of the prevalance of "USB toys" which have no USB interface in them with which to negotiate power.

Anyway, I think you need to specify exactly what you are trying to do. You might be trying to solve something in a way which is too complicated.

For accuracy you might want to skip the crystal entirely and use 50 Hz mains - that's how most alarm clocks work. A timer with dividers of 50/60/60/24 would work.

 

....For accuracy you might want to skip the crystal entirely and use 50 Hz mains - that's how most alarm clocks work. A timer with dividers of 50/60/60/24 would work.

If the OP could use the power line then see post #9.

Ken

 

The OP pretty well specified what he wanted, and got it. It remains to be seen if what he thought he wanted is what he really wanted. In any case, if accuracy were needed I would have gone the crystal oscillator route, it would still be much simpler than using AC mains frequency. There is actually a design out there by me already, it is a done deed, and a bit simpler in chip count to boot. I might redraw it differently sometime for 24 hours, instead of multiple days.