# Need some guidance making a power supply

#### scl23enn4m3

Joined Nov 4, 2011
9
Hey AAC! I'm looking for help on a small project, but Google doesn't seem to be helping me so I thought I'd try here. I've tried posting this question about three times, but the server keeps timing out on me. Edit: It went through!

As the title says, I'm looking to make a power supply that converts 110vac to 1.5vdc at 15amps. I thought the easiest way to go about this is to buy a power supply that already outputs the right amount of wattage, and then converting it from there. But then I started reading that it may be easier to convert while the current is still AC? That's just the beginning of my confusion. Can someone help me design this or provie me the right terminology to search for? Thanks!

#### wayneh

Joined Sep 9, 2010
17,153
How tightly do you need to reach 1.5V? I mean, a typical transformer-plus-rectifier-plus-filter cap arrangement could get you close but will have some ripple left over. You could add voltage regulation (LM371 for instance) but you're getting close to the bottom end of voltage that can be run that way. And your current need is not trivial.

And yes, you'll want to make the voltage conversion in the AC domain. That's what a transformer or a switch-mode power supply both do. But you need to design for the voltage drop that will occur in going to regulated DC. A conventional bridge rectifier will drop ~1.4v, and a regulator IC such as the LM317 needs about 2V. There are such things as low drop out regulators, but the point is they still need some overhead.

#### scl23enn4m3

Joined Nov 4, 2011
9
Thanks for the reply wayneh! For the 1.5v, I think a <20% tolerance is acceptable. Is that realistic?

EDIT:
So the transformer will drop the AC voltage and increase the amperage, the rectifier will convert AC into DC, and the capacitor will clean it up? I just want to make sure I understood that right.

As for the voltage drop from the rectifier, does that mean I have to get a transformer that ouptuts 3V to compensate? What about the current?

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,221
I'm curious what you're planning on doing with that much current at that low of a voltage?

It would likely help you to have a better outcome of your project if you tried to explain what you are attempting to do.

It frequently happens that a member will have a specific idea on how they want to accomplish something, and only later realize that there was a much more simple/practical way to accomplish the same thing.

#### scl23enn4m3

Joined Nov 4, 2011
9
I'm curious what you're planning on doing with that much current at that low of a voltage?

It would likely help you to have a better outcome of your project if you tried to explain what you are attempting to do.

It frequently happens that a member will have a specific idea on how they want to accomplish something, and only later realize that there was a much more simple/practical way to accomplish the same thing.
I can't really delve into specifics, but it's for a heating element that remains close to your body. I need it to be low voltage due to not wanting to get shocked lol.

#### SgtWookie

Joined Jul 17, 2007
22,221
It would be SO much easier (and cheaper) if you'd use 3.3v or 5v.

You would be able to use much smaller gauge wiring as well. Shock hazard would be virtually nil.

#### thatoneguy

Joined Feb 19, 2009
6,359
It would be SO much easier (and cheaper) if you'd use 3.3v or 5v.

You would be able to use much smaller gauge wiring as well. Shock hazard would be virtually nil.
Old electric blankets used up to 50V and the insulation of cloth to prevent shock.

12V on non-wet contact isn't a shock hazard, it isn't until 24 VAC (Peaks over 30V) that dry skin shocks are dangerous (not unfelt, just not dangerous to most people).

#### scl23enn4m3

Joined Nov 4, 2011
9
It would be SO much easier (and cheaper) if you'd use 3.3v or 5v.

You would be able to use much smaller gauge wiring as well. Shock hazard would be virtually nil.
Well I like easy and cheap, and if 3.3v will work then I'm down with it. Thanks for the suggestion!

#### wayneh

Joined Sep 9, 2010
17,153
...it's for a heating element that remains close to your body.
Consider EMF. I believe some years ago they found that hair dryers and electric blankets were major sources of EMF exposure in the home. High currents close to the body will give a higher exposure to electromagnetic fields than a lower current. You can get the same heat at a higher voltage and lower current, thereby reducing EMF. If you care.

BTW, if you go to 5V, you can just pull the PSU out of an old computer (for free) and get the amperage you need, ready built with an on-off switch, fuse, over-current, over-temp protection, the works.

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,221
You can actually get 3.3v, 5v and 12v. However, 5v is preferred, as you need to put a load on that voltage anyway to get good regulation.

#### scl23enn4m3

Joined Nov 4, 2011
9
I thought about going the computer PSU route (since I have quite a few laying around here), but they are just too bulky. I'm trying to make this as small and contained as possible, hence wanting to make it from scratch.

#### t06afre

Joined May 11, 2009
5,934
It will depend on how you place this thing. What I suspect will be used to create a standardized pain stimuli by creating heat. If you make it floating and place it in good distance from the heart. If the current in case of fault will be localized to a small area. You may use much higher voltage than 1.5 volt. And it should "safe" regarded shock hazards. What I have not taken into account is the risk of burns, if you use 20 watt to heat a small area. The outcome may quickly become severe burns. I think it is best you tell us what you plan. As for now. I consider reporting this thread as unsafe

#### SgtWookie

Joined Jul 17, 2007
22,221
If you want compact, then you might consider looking at laptop aka "desktop" power supplies.

15A @ 1.5v is 22.5 Watts. You should de-rate a power supply by 20% (never run it at maximum output or reliability will suffer). So, you need ~28W.

MPJA has a couple of desktop supplies that would work for you, both ~$10: http://www.mpja.com/12V-3A-Desktop-Supply/productinfo/18794+PS/ http://www.mpja.com/12V-333A-Hipro-Desktop-Supply/productinfo/18951+PS/ They both need power cords, but they sell those too. You wouldn't be able to build a supply that cheaply. Thread Starter #### scl23enn4m3 Joined Nov 4, 2011 9 It will depend on how you place this thing. What I suspect will be used to create a standardized pain stimuli by creating heat. If you make it floating and place it in good distance from the heart. If the current in case of fault will be localized to a small area. You may use much higher voltage than 1.5 volt. And it should "safe" regarded shock hazards. What I have not taken into account is the risk of burns, if you use 20 watt to heat a small area. The outcome may quickly become severe burns. I think it is best you tell us what you plan. As for now. I consider reporting this thread as unsafe It's not at all related to pain stimuli or for the body what so ever. I'm not into that kind of stuff. The reason I mentioned it is close to the body is because the heating element is exposed (as in not contained in anything), and to remain safe I don't want an accidental long exposure to hurt someone. If you are going to report this for being unsafe, that's fine. Though I'd like to state my inability to share details shouldn't automatically lead to the assumption of foul play. On a sidenote, and by pure coincidence (as in this project has nothing to do with my profession), I work in the healthcare industry and have extensively studied and used electrocardiography. I'm quite familiar with the conduction of the human body. Thread Starter #### scl23enn4m3 Joined Nov 4, 2011 9 If you want compact, then you might consider looking at laptop aka "desktop" power supplies. 15A @ 1.5v is 22.5 Watts. You should de-rate a power supply by 20% (never run it at maximum output or reliability will suffer). So, you need ~28W. MPJA has a couple of desktop supplies that would work for you, both ~$10:
http://www.mpja.com/12V-3A-Desktop-Supply/productinfo/18794+PS/
http://www.mpja.com/12V-333A-Hipro-Desktop-Supply/productinfo/18951+PS/

They both need power cords, but they sell those too. You wouldn't be able to build a supply that cheaply.
Thank you for the links! So those PSUs provide ~36watts. They state their output, but how would I be able to draw the low wattage and high current instead of the high wattage low current they are designed for? Are is that automatically done based on power requirements?

#### t06afre

Joined May 11, 2009
5,934
You should have known how many questions we have in this forum. Regarding unsafe devices like converting bug zappers to human zappers. And as I self work with medical instrumentation. And know the safety rules quite well. Your request created a raised eyebrows. If you can argue for that you have done some risk management. I can not see any problems

#### SgtWookie

Joined Jul 17, 2007
22,221
Thank you for the links!
You're welcome.
So those PSUs provide ~36watts.
Up to 36W for the 1st one, and ~40W for the 2nd, yes. ~29W and under should be handled nicely by either of them.

They state their output, but how would I be able to draw the low wattage and high current instead of the high wattage low current they are designed for?
I think you meant low voltage, high current?

If you place your load in series instead of parallel, you wind up with the same power dissipation.

For example, if you had 16 loads that drew 1A each at 1.5v when wired in parallel (each load was connected directly across the 1.5v supply), you could operate them wired in series of 8 in a string, as 12v/1.5v=8. So, two strings of 8 loads in parallel gives you the same power requirement; just operated at a different voltage.

Are is that automatically done based on power requirements?
A resistive load will only allow as much current flow as Ohm's Law allows: I=E/R, or current = voltage/resistance.

#### scl23enn4m3

Joined Nov 4, 2011
9
You're welcome.
...
Up to 36W for the 1st one, and ~40W for the 2nd, yes. ~29W and under should be handled nicely by either of them.
...
I think you meant low voltage, high current?
...
If you place your load in series instead of parallel, you wind up with the same power dissipation.
...
For example, if you had 16 loads that drew 1A each at 1.5v when wired in parallel (each load was connected directly across the 1.5v supply), you could operate them wired in series of 8 in a string, as 12v/1.5v=8. So, two strings of 8 loads in parallel gives you the same power requirement; just operated at a different voltage.
...
A resistive load will only allow as much current flow as Ohm's Law allows: I=E/R, or current = voltage/resistance.
Boy that post was full of grammatical mistakes. That's what I get for attempting to post from my tablet. But yes, watt = volts, and Are = or.

Thanks for clarifying that! So the resistance of the entire circuit would allow me to control the voltage and amperage? Would a pot affectively allow me to control it on the fly then? What guage wire would you recommend I could safely use for this circuit? On hand I have both stranded and solid in various sizes from 12awg to 30awg. Thanks again! Not only for the knowledge but for taking the time out of your day for such detailed replies!

EDIT:
One more question:

What if I just had one load? Wouldn't that take the full 12v?

Or are you suggestion the PSU has a 1.5v rail on it?

Last edited:

#### scl23enn4m3

Joined Nov 4, 2011
9
You should have known how many questions we have in this forum. Regarding unsafe devices like converting bug zappers to human zappers. And as I self work with medical instrumentation. And know the safety rules quite well. Your request created a raised eyebrows. If you can argue for that you have done some risk management. I can not see any problems
That's pretty disturbing. I've accidently shocked myself enough times to confidently say I don't enjoy it.

#### wayneh

Joined Sep 9, 2010
17,153
I've accidently shocked myself enough times to confidently say I don't enjoy it.
If you don't enjoy it, you're not doing it right!

No really, there's a whole cottage industry (as well as legitimate industry) of applying low levels of electrical stimulation to the body. Some devices have received FDA approval as safe and effective. The techniques can really make a difference for folks with the "right", treatable ailments. The technology helps a lot of people. But it's off-topic and probably a forbidden topic here.