Power supply question? Negative and positive power

Thread Starter

wifiguru10

Joined Oct 16, 2015
3
So I have a stack of unused 1150w power supplies.
They input 110v and output 12V@25A and -52V@16A. That's 300W on the 12V side and 830ish on the -52V side.


So my goal is to power a remote device with 24V @ 40A, using the thinnest cable possible. That's about 900W, only really need 20A but 40 is the goal.

My question is for the genius's out there, and possibly to keep me from killing myself.

Logic tells me that i can get a combined power differential of 64V if i wire + to the 12V+, and - to the + on the 52V side. But if one side provides 300W and the other side provides 900W, will i have an issue if I draw too much current? (read: Will the 12V side "blow" above 300W of current?)

Second question. Would it be better to run two power supplies in serial and run -104V through the cable? or in parallel for more amperage and @ -52V. Assuming on the other side i have a power supply that can handle 19-72VDC OR 72-144VDC (depending on part number). Since it's negative voltage, i'd just swap polarities at the DC-DC converter at the other side.

Also cable thickness, I know that there's voltage drop over distance, so i guess running 104V would be better since it's lower amperage. But as far as using the thinnest cable possible, would i run three in series to get to 160V and let the cable loss bring it down in the 144VDC range.

Cable length would be 100-200'. Looking for as thin as possible, to make this work. What gauge wire would be recommended?

Can someone check my logic here before i do something stupid?

Power supply at the other end is MW SD-1000H-24 or SD-1000L-24 depending on input voltage range.

Thanks in advance!
 

wayneh

Joined Sep 9, 2010
17,496
Your cable should carry the mains supply. That would minimize the required size, if it's possible to do so.

As to your first question - You are limited by the weakest power supply.
 
Last edited:

Thread Starter

wifiguru10

Joined Oct 16, 2015
3
So looking a loss calculator.

So if i wanted to use 24AWG and used 4 PS in series to get to 200VDC, the power drop would be 25% @ 5 amps (1000W). Which at the other side would be 750-ish Watts. Which is 24V @ 30A.

If i can use the 64V as a supply, which is 250V supply @ 5 amps, that would be a 20% drop which would be closer to 800W.

Is my mad-science math right?
 

wayneh

Joined Sep 9, 2010
17,496
Right or wrong, it seems improper to me to design for such large losses in the wire, unless there is some very special reason to make the wire the weak link. Is it not possible to run high voltage AC over your narrow gauge wire and then feed it to the power supplies?
 

Thread Starter

wifiguru10

Joined Oct 16, 2015
3
Well, it's basically a tether for a drone. So weight comes into play, efficiency isn't an huge concern. From what i've seen on forums, the best bet is DC-DC since AC requires a heavy transformer. It's been done (on RC forums) up to 1000' feet but with a light drone.

The issue is the weight of the cable, hence the desire for lighter cable. Commercial applications run in the 2-3kV range DC, with some crazy isolation on their cables. Those solutions run $100K+, as a hobbyist, i don't have that kind of cheddar laying around ;P So looking for a easy/affordable solution.

So am I correct in assuming the voltage drop comes with the increased current? So with no load it would be 200VDC, but at load it'd then drop to 150VDC? Vdrop=IR? So if current is low, voltage drop is minimal. So i'd need to make sure the zero load and high load was within the specifications of the power supply at the other end?

From what I'm gathering (and reading), wattage doesn't matter, just current. That's why power companies run 50kV across their lines. So the higher the voltage the better?
 

tindel

Joined Sep 16, 2012
936
It's all ohms law... The higher the voltage, the less current and vise versa to deliver the same power. That's why there are so many suggestions here to run mains in your tether. But your point is valid that that requires additional weight in your drone which is not desireable. Your power delivered to your drone with the power supplies you mentioned before could be as high as 64V @ 16A or 1kWwith one power supply. You want a power bus that delivers 24V @ 40A or 1.1kW. So that is not enough power in to supply the power out. And physics dictates that you MUST have more power in than power out.

If you run two supplies in series as you suggested then you will have 64V*2 @ 16A or 128V @ 16A = 2kw. You should be about to buck that down on the drone... and it will be smaller than a mains converter, but not by much. Now you're close to mains voltages and power though. so it will be more simplistic to run mains voltages. You have a problem still though - there's no commercially available AC to DC or DC to DC power modules that will convert 120AC or DC to 24V @ 40A. I only checked digikey though so some more searching may turn up a good solution somewhere. A 1.5kW power supply will be pricey too. a few hundred bucks I bet if you find one.

Here's what I'd do. I'd run mains in the tether. I'd guess that as much as 80% of your power capacity is in the startup of your motors and that the 24V is intended to only go to the motors. I'd get a couple supplies to power the motors individually and a separate supply for the control and telemetry portion of your drone. This will save some money for the supplies. It may add some weight to have multiple supplies, but then you can get larger motors and blades to produce more lift if it's needed.

I'd run 14AWG stranded for the mains. I'd also isolate the mains on the ground for safety. This will require a large isolation transformer, but could possibly save your life these can be found used on ebay or craigslist for a $200 or so (last I checked) for a 2kVA isolation transformer. Do not get a medical grade one because they put the earth through - defeating the isolation (which I've never understood). You can sometimes snip the green wire on the medical grade ones though making them perfect for this application. Still remember though that either way you go AC or DC that there is 120V on you drone and that it CAN kill you! Keep one hand in your pocket when working on it while powered, and always assume it's powered unless you know for a fact that it's not (always double check with a multimeter!)

When I was working in aerospace... the ~1kW spacecraft main power bus were using a 28V battery unregulated power bus. As you got into higher power birds >~4kW they shifted to 120VDC regulated power busses to conserve weight on cabling.
 

wayneh

Joined Sep 9, 2010
17,496
Well, it's basically a tether for a drone. So weight comes into play, efficiency isn't an huge concern. From what i've seen on forums, the best bet is DC-DC since AC requires a heavy transformer. It's been done (on RC forums) up to 1000' feet but with a light drone.
Well that explains a LOT! Thanks.

So the challenge is to pass about 1 kW to the drone over the lightest cable possible.

There are only two levers to pull: 1) lighten the cable for a given gauge of copper and 2) get the voltage as high as possible. Your options to lighten the cable are minimal but there must be stranded copper wire available with minimal weight insulation. (You can't really do much about the copper, only the insulation around it.) I don't have experience but would look for what they use in airplanes. I expect they want low weight and resistance to abrasion, and these are both criteria which should matter to you. I think aluminum wire requires more weight for a given amp flow, but you might research this.
So am I correct in assuming the voltage drop comes with the increased current? So with no load it would be 200VDC, but at load it'd then drop to 150VDC? Vdrop=IR? So if current is low, voltage drop is minimal. So i'd need to make sure the zero load and high load was within the specifications of the power supply at the other end?
All true.
From what I'm gathering (and reading), wattage doesn't matter, just current. That's why power companies run 50kV across their lines. So the higher the voltage the better?
I don't see anything useful about thinking that wattage doesn't matter. Once you specify current and voltage, wattage is just the product of the two, so these things are not independent and they all matter. The important thing to know is that for any given wire, the power that will be lost as heat in that wire is proportional to the square of the current passing through it: P=I^2•R where P is the power lost in the wire in watts, I is the current in the wire in amps, and R is the resistance of the length of wire in ohms.

Since the power to the load at the other end is Voltage times Amps, the loss described above is minimized by upping the voltage and lowering the current, while keeping the product of those the same. In your case, an example choice might be 100V • 10A = 1kW or 200V • 5A = 1kW. The power loss in your cable would be 100R or 25R for those options respectively. So you can use a thin, light wire with high R if your voltage is high enough.
 

GS3

Joined Sep 21, 2007
408
Well, it's basically a tether for a drone.
About a year ago I bought a small drone and soon found out two things: (1) I was terrible and needed lots of practice and (2) the battery charge lasted little and took much longer to charge. Given this I immediately thought that if I could power it with cables, even over a very limited distance, I could get some practice for later real flying. But after some calculations of current, wire thickness and weight it became apparent that it just wasn't practical.

This is a very small drone. I suspect that as they get bigger they should be able to lift more cable weight in proportion but I still doubt it is practical for real outdoor flying.
 

tindel

Joined Sep 16, 2012
936
But after some calculations of current, wire thickness and weight it became apparent that it just wasn't practical.
That's pretty much where I was getting in post #6. I was just throwing out a hail mary to have several different power supplies on board.
 

Veracohr

Joined Jan 3, 2011
772
Are you trying to power the drone with this setup or just instruments that are on the drone? Seems to me that your drone wouldn't be able to fly very far on a 200' tether, so maybe a better solution would be to not use a powered drone. If you can do what you want from a single position in the air, maybe you can float the instruments up on a balloon and not have to worry about powering a drone, just the instruments.
 
Top