Most basic solution to power Arduino from 100V + supply

Thread Starter

CasualKilla

Joined Dec 12, 2015
21
I am theorizing using an arduino uno for the controller of a 100V to 24V power supply, however I have come to the predicament where I first need to power the arduino from 100V before I can use the arduino to make the 24V supply. I have included this info for background, but it is not so relevent to the problem. My question is what is the simplest way to power the arduino in this case. Efficiency does not need to be high.

I am thinking that perhaps a zener diode and resistor will do the trick? Perhaps hooking up another voltage regulator or capacitor to the diode may help too? One more thing, the power is generally low ripple, but it can vary slowly between 60V to 150V over time.
 

dl324

Joined Mar 30, 2015
9,317
Your zener idea is probably the simplest. For acceptable voltage regulation, you need the zener current to be at least 10X larger than the maximum current the microcontroller will draw. Use the lowest voltage you anticipate and insure maximum zener ratings aren't exceeded for the highest voltage.

What voltage do you want to operate the Uno at?
 

Thread Starter

CasualKilla

Joined Dec 12, 2015
21
Your zener idea is probably the simplest. For acceptable voltage regulation, you need the zener current to be at least 10X larger than the maximum current the microcontroller will draw.

What voltage do you want to operate the Uno at?
I will aim for something between 7V to 10V for the arduino Vin pin, depending on what zeners I can find. Ok cool, thanks for the thumb-rule. I will assume 200mA max for the arduino, as I do not have much details on the arduino circuit design yet.
 

dl324

Joined Mar 30, 2015
9,317
I will aim for something between 7V to 10V for the arduino Vin pin, depending on what zeners I can find. Ok cool, thanks for the thumb-rule. I will assume 200mA max for the arduino, as I do not have much details on the arduino circuit design yet.
200mA is too much current for a zener regulator. You'll need to add a transistor for current gain.
 

Thread Starter

CasualKilla

Joined Dec 12, 2015
21
200mA is too much current for a zener regulator. You'll need to add a transistor for current gain.
Ok yeah mybad, that would equate to 2A @ ~10V through the zener XD. Guess I could shoot for lower consumption, but how does the transistor work? I am assuming I will require a high Vbe capability BJT?
 

dl324

Joined Mar 30, 2015
9,317
Ok yeah mybad, that would equate to 2A @ ~10V through the zener XD. Guess I could shoot for lower consumption, but how does the transistor work? I am assuming I will require a high Vbe capability BJT?
You use the transistor current gain to reduce the load on the zener. But power dissipation is still going to be an issue (somewhere).
 

RichardO

Joined May 4, 2013
2,268
Yep there is a AC - DC conversion in the unit before they step the voltage down so they will work just fine on a DC input.
tcmtech is too quick for me. :)
I will add that the 100 volts DC may be too low for a typical wall wart. For the wall wart to run on 100 volts DC implies that it is designed to run from as low as 70 volts RMS.
 

dl324

Joined Mar 30, 2015
9,317
I will add that the 100 volts DC may be too low for a typical wall wart. For the wall wart to run on 100 volts DC implies that it is designed to run from as low as 70 volts RMS.
It's even worse. Supported input range needs to be 60-150VDC.
 

tcmtech

Joined Nov 4, 2013
2,868
I wouldn't worry about the slightly lower input voltage if the load is well below the units rated capacity.

I have used SMPS based power supplies at input voltages way below their spec values without problems. Most will actually work just fine down to well under 40 VAC at partial loads.
 

Thread Starter

CasualKilla

Joined Dec 12, 2015
21
Wall wort idea is cool and all, but I was hoping to do a design from scratch. I am more interested in learning than making something that just works. Could somebody elaborate on the zener + transistor regulator/supply idea?
 

Veracohr

Joined Jan 3, 2011
694
Like this. The graph is the output voltage with an input going from 60V to 100V. 10V zener, and about 200mA going to the load.

Screen Shot 2016-01-15 at 11.24.11 AM.png

But the transistor power dissipation goes from 10W to 18W.
 

#12

Joined Nov 30, 2010
18,167
I was hoping to do a design from scratch.
I think you were hoping WE could do a design from scratch, then you would build it.

The basic thermodynamics of any analog regulator that changes 100 volts into 10 volts at 0.2 amps is 18 watts of waste heat. Either decide to dissipate 18 watts or abandon the analog methods in favor of switching supplies or Wall Warts.
 

Thread Starter

CasualKilla

Joined Dec 12, 2015
21
I think you were hoping WE could do a design from scratch, then you would build it.

The basic thermodynamics of any analog regulator that changes 100 volts into 10 volts at 0.2 amps is 18 watts of waste heat. Either decide to dissipate 18 watts or abandon the analog methods in favor of switching supplies or Wall Warts.
Hmm, yes I see your point. To tell you the truth my end goal was to build a switching regulator with the arduino as the controller, this is why I am hesitant to use a switching reg to power the Arduino. Also there isn't really any low power switching regs that do 100V + to 5V.
 

#12

Joined Nov 30, 2010
18,167
To tell you the truth my end goal was to build a switching regulator with the arduino as the controller, this is why I am hesitant to use a switching reg to power the Arduino.
That's a false fear. Any correct size switching supply can power the arduino without causing any interference with what the arduino is doing. The worst case scenario means you have to add a capacitor where the power plugs into the processor board.
 

bertus

Joined Apr 5, 2008
20,170
Hello,

What is the input voltage range of the cube?
Can you post a picture of the cube?
Most SMPS are using a rectifier at the input with a capacitor.
Feeding it with DC would not give a problem.

Bertus
 
Top