NesPi Needs More Power

Question

So i am doing the nespi project and have found that i need power to a usb hub and power to the pi. I only have installed one usb 3.0 port for power to come into the nes case. If i get this power supply --> https://www.amazon.com/Adapter-Portable ... wer+supply Can i take two usb cables from ^^^this power supply from a usb splitter into a single usb cable then split it back out into two usb connections. Then have one end go to the pi and the other end go to the usb hub to power both devices. Another possibility i have thought of is trying to bring more amps to the usb hub but i do not know it it can handle more power in and if that would allow it to output more power per port. I would also greatly appreciate any other ideas on how to get power to both devices with only using one usb port to get power into the case. Attached is the setup before the powered usb hub was added in to power the 4 controllers separate from the pi's power. And the back port is the single usb 3.0 port i have to bring power into the case. Hopefully later i can get a pic of that to help with any questions on the setup. 
 Attachments:

Former Member · Accepted Answer

No worry's i appreciate the offer. I do have a good update. I have found that i am getting enough power to the pi with the single usb 3.0 port and cable with the 5v 3a power supply. I found the issue was related to the dual port usb i had running controller 3 and 4. Adding a splitter to the dual usb port with the controllers caused the pi not to boot. Running the pi off the usb hub caused low power to the pi. I then was still having issues with controllers 3 and 4 going in to the dual port usb. So i decided to bypass them and boom everything was working with no issues. It seems that the dual port usb some how scrambles the power and data between the ports which messed up the controllers powering on/working correctly. On a side note the wires that i used for carrying th power in are both 20awg usb 3.0 cables so shouldn't have any issues with those. The two red squares is where my issues occurred from once eliminated everything seems to be working ok again.

Robert Peter Oakes · Answer

you may learn a bit more about this by reading my artical here Raspberry PI:- USB power cables, crashing and other problems

or here Raspberry PI:- USB, Ethernet, crashing and other problems | The Breadboard

Peter

rew · Answer

RW is me? Call me "REW" on forums... :-) No, I'm in the Netherlands.... Western Europe. Some people don't dare buy stuff off ebay/aliexpress/alibaba because they got burned once. Fine. But with some care, you can get good stuff at great prices from there. Sometimes you get something you didn't expect. So somehow I got good probes for my multimeter. So I wanted them again for my other multimeters. I order them, and I get "scale models". They look exactly the same, just at 80% scale. In that case, I was lucky: the seller had mentioned the size of the product in the offer, and that corresponded to what I already had. Got an "oops sorry, here are the correct ones". Another thing people fall for is that they buy stuff that has specs that most people can't verify easily. Like batteries or powerbanks. There you are likely to get 10 times less capacity than claimed! But the leds I ordered yesterday are pretty good. How much current do most people push through a led? 20mA? They still glow visibly at 60nA. That's 300 thousand times less current than allowed. You need to look carefully to see that, but at 20uA they function great as indicator lights.

rew · Answer

When measuring a voltage, the measurement device will always draw a small current, which hopefully is negligible to the source being measured. The input impedance of the meter needs to be as high as possible for this When measuring a current the measurement device will always introduce a small voltage in the current-path being measured, which hopefully is negligible to the system being measured.... The input impedance of the meter needs to be as low as possible for this... That voltage is called the "burden voltage". Now if we accept a burden voltage of 0.2V, and use a 10M shunt resistor, the meter will show 200mV full-scale for 20nA full scale current...... That 10M shunt resistor is builtin to most multimeters. (beware of the flukes... Those may have 17M input impedance, so the measurements are more difficult to evaluate on sight...). As 20nA was too little to see, I had to accept a burden voltage of 600mV for the measurement of 60nA. As long as you're aware of that, this is not much of a problem....