USB- how to get 500ma

Thread Starter

Java

Joined Dec 6, 2010
4
Background- my son wants to make a USB powered beverage cooler/warmer.

I went here. http://www.beyondlogic.org/usbnutshell/usb2.shtml and have soldered a 1.5K resistor to get the USB bus to supply power. The USB supplies power to a fan and a peltier (TEC). I suspect that I am only getting 100ma though.

I understand that if the USB device sends the correct signal the USB bus will supply 5V at 500ma.

How/what do I need/have to do to create the signal that will make the USB bus to send 5v at 500ma?

I did a search and I did not come up with anything that I could identify that would help. This is my first post and my first time soldering anything electronic, so please bear with my ignorance.

Thank you,

Paul
 

mtripoli

Joined Feb 9, 2010
44
Going from memory, the USB device goes through a couple of "back and forth's" with the PC to jump from the 100mA to 500mA. I don't think you can get there passively...

Mike T.
 

t06afre

Joined May 11, 2009
5,934
You have to negotiate with the host to get more power. Google usb power negation and you will find some links. Your estimation is quite correct. 100 mA is the limit without negotiation for more power.
Using a microcontroller with USB controller onboard may do the trick
 

Thread Starter

Java

Joined Dec 6, 2010
4
Wow, you guys are fast!

Thank you, I will google USB negotiation.

At 5v and 100mA the peltier gets hot enough to act as a cup warmer, it is hot to touch. However, it needs more power to cool anything down. It does get cold but at under .5 watts I cannot expect too much. I used a wall wart with 13v at 1A and the peltier worked great.

Off to google, thanks again.

Paul
 

SgtWookie

Joined Jul 17, 2007
22,230
If you have a 1.5k resistor in series with your load (fan, TEC) then you have less than 5v/1500 Ohms = 3.333...mA available.

TECs are very power-hungry and not very efficient.

For heating, you'd be better off using just a simple resistor - but even then, you only have 2.5 Watts available; about 1/3 the heat output of a night light.
 

tom66

Joined May 9, 2009
2,595
Just ignore it. Although USB is specified for 500mA after negotiation, most devices can draw up to 1 amp even without negotiation. This isn't to be accepted as a rule, because some don't, but most do. So if it's a one time thing, try it.
 
Last edited:

thatoneguy

Joined Feb 19, 2009
6,359
Many cheaper USB Powered hubs do not have the current limiting circuitry. Some will allow the full 2A supply out to one port with no communication.

I'd suggest using a "Wall Wart" power supply that gives 5V @ 2A at least. These are sold in many places for powered hubs' power sources.
 

Thread Starter

Java

Joined Dec 6, 2010
4
I realized I made a mistake by adding the resistor. I thought that it was necessary for the USB bus to send power to the device but that is not so. I removed the resistor and power is still sent.
I understand that a wall wart would be better as a power supply but this is my son's project and he is determined to use the power supplied by a USB. We have wired up another USB heater/cooler with both USB and wall wart power supply connections to show the benefit of a higher powered supply.
I understand a simple resistor heating circuit would be better suited but he wanted a beverage cooler and heater. He has wired a rocker switch to reverse the polarity supplied to the peltier making it both a heater and cooling plate.
He is going to lap a stainless steel cup so that it sits flush on the peltier to get the most out of the low wattage. Yes copper would be better but they are not readily available. The cup sides will be insulated as well.

Thank you for your help on this project, I do appreciate it.

Any and all ideas are welcomed as I help him figure out how to best utilize the power supply that he has chosen.

Paul
 

SgtWookie

Joined Jul 17, 2007
22,230
Stainless steel is a very poor conductor of heat; about 1/15 that of copper.

My Dad made a beautiful stainless steel frypan when he was a teenager. They took it camping in upstate NY during the winter. They tried to cook some freshly-caught fish in the new frypan over a roaring campfire. The fish didn't get more than lukewarm; it even froze on the very top.

Aluminum is about 57% as efficient as copper.
Water is 1.4 times better at conducting heat than copper.

It is very stressful on a Peltier device to suddenly reverse the direction of current. The rocker switch should have a center off position, and he should let it sit for a minute in order for the temps to equalize before reversing the direction of current flow.
 

t06afre

Joined May 11, 2009
5,934
I could be an idea to use a hub USB then testing the circuit. Then experimenting things tend to get wrong. And it is better to fry a cheap USB hub, then render a laptop useless. Just a friendly tip :)
 

Thread Starter

Java

Joined Dec 6, 2010
4
Points taken.

I use an old 5V 800mA wall wart for testing and then I go to an old P1 laptop that is not even worth repairing.

I did not know SS was such a poor conductor. I have copper that my son could be cut to fit the peltier . Cut hole in a cup the same size, set with two part epoxy, and seal the copper inside of cup with a lacquer.

The peltier does not get switched from hot to cold until it has been unplugged from the power source. An on-off-on switch would be ideal but the one I had was too big and took up too much of the heat sink.

Great ideas, thank you.

Paul
 

BB00

Joined Jan 5, 2018
1
Aluminum is about 57% as efficient as copper.
Water is 1.4 times better at conducting heat than copper.
I know this thread is old, but I felt like this should be corrected. Water is actually far less thermally conductive. Copper is about 650 times more thermally conductive than water (400W/mK vs 0.59W/mK).

Water cooling is used because water has a high specific heat capacity (about 10x higher than copper), so it can absorb more heat from a hot thing and then travel to the radiator to cool down. If the water wasn't moving, and relied on heat conducting through it to the radiator, the device would overheat very quickly.
 
I realized I made a mistake by adding the resistor. I thought that it was necessary for the USB bus to send power to the device but that is not so. I removed the resistor and power is still sent.
I understand that a wall wart would be better as a power supply but this is my son's project and he is determined to use the power supplied by a USB. We have wired up another USB heater/cooler with both USB and wall wart power supply connections to show the benefit of a higher powered supply.
I understand a simple resistor heating circuit would be better suited but he wanted a beverage cooler and heater. He has wired a rocker switch to reverse the polarity supplied to the peltier making it both a heater and cooling plate.
He is going to lap a stainless steel cup so that it sits flush on the peltier to get the most out of the low wattage. Yes copper would be better but they are not readily available. The cup sides will be insulated as well. Many liquor stores sell copper cups for so-called Moscow mules for about $6

Thank you for your help on this project, I do appreciate it.

Any and all ideas are welcomed as I help him figure out how to best utilize the power supply that he has chosen.

Paul
 
Top