Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Power Supply Assistance Requested

Status
Not open for further replies.

EvilGenius

Member
Hello
I have opened up the case to a large transformer that converts Input: 120V, 60 HZ, 6A to Output:12V AC, 25A, 300W
I need to convert it to a 20A, 12V DC with 12V minimum requirement.
I measured the output of the transformer with a Multi-Meter and it gave me 12.2VAC with no load
My logic says that it wont work for the following reasons, but I am open to suggestions and clever thinking to make it work.
12.2VAC going thru a bridge rectifier (min Vf of 1.25V) will reduce the voltage to 10.95vDC without a load
With a load (20A x 12v = 240W) I am estimating that my voltage drops to 90% of this value or 10.95 x 0.9 = 9.86v which is not acceptable
The idea is to utilize what I have or scrap it and buy a 12VDC 400W power supply at $28.00
Any suggestions or cheap ideas are very much welcomed....
 
Last edited:
Here is what Duncan says:

dunc.png


Download, put in better values than I did, and re-run. Have fun.
 
12.2VAC going thru a bridge rectifier (min Vf of 1.25V) will reduce the voltage to 10.95vDC without a load

DC voltage from a bridge = 1.414 * 12.2 Vac- 0.6 * 2 ; DC Current is less than rated AC current.
Power out cannot be greater than power in.

EDIT: slight wording change
 
DC voltage from a bridge = 1.414 * 12.2 Vac- 0.6 * 2 ; DC Current is less than rated AC current.
Power out cannot be greater than power in.
Are you saying the voltage after a high power bridge rectifier is 32.8v?
I am flexible on load current as long as I get 12 to 12.2V dc.
I need minimum of 15A but prefer 20A plus.

P.S.: My loads are LED's and resistors, so there is no inductance load.
 
I just read a bit more. I believe what I measured with Multi-meter was Vavg. If so Vrms=1.11 x Vavg = 13.54v Is this correct?
If so, If I put a 0.8 ohm load on it, I will get 11.7v, 14.5A at 170W. Is this correct?
If yes, I will be perfectly happy with that. Nevertheless I will need a very large capacitor bank!
 
Last edited:
A VU meter is for indicating Volume Units, i.e. audio.

Are you using a Digital Multimeter in the VoltsAC mode? If so, they come in two types. One measures true RMS. The other actually measures something else, but is calibrated to display Vac RMS, but only if it is being used to measure a sine wave.

So what did you measure with nothing connected to the transformer secondary?
 
A VU meter is for indicating Volume Units, i.e. audio.

Are you using a Digital Multimeter in the VoltsAC mode? If so, they come in two types. One measures true RMS. The other actually measures something else, but is calibrated to display Vac RMS, but only if it is being used to measure a sine wave.

So what did you measure with nothing connected to the transformer secondary?
The multimeter I used is inexpensive so I will no assume it has sophisticated electronics. With that in mind I put it on 200V AC setting and measured the secondary voltage of transformer without a load.
 
The multimeter I used is inexpensive so I will no assume it has sophisticated electronics. With that in mind I put it on 200V AC setting and measured the secondary voltage of transformer without a load.
Can you use a lower AC range like 20Vac? That will result in a more accurate reading.
 
Can you use a lower AC range like 20Vac? That will result in a more accurate reading.
Unfortunately not. I have 200 and 750 setting.

What's on your mind?
I might just board the darn thing with bridge rec and large cap bank with no load and see what DC output I get. Then we go from there.
 
I reran the Duncan simulator with some better estimates of Transformer impedance and Filter Capacitor ESR. Here is what is predicted:

dunc.png


Peak voltage of ~13.3V; minimum voltage of ~10.8 when the load approaches ~20A. The peak current in the rectifier bridge is >60A, so it will have to be a big bridge. You could reduce the ripple some by going to an even bigger filter capacitor, say 220mF which would keep the minimum voltage at ~12V
 
Peak voltage of ~13.3V; minimum voltage of ~10.8 when the load approaches ~20A. The peak current in the rectifier bridge is >60A, so it will have to be a big bridge. You could reduce the ripple some by going to an even bigger filter capacitor, say 220mF which would keep the minimum voltage at ~12V
I got similar results using another online sim. I did not want to complicate the situation but I have two of these transformers in the big box. So I can divide the load current up with a common ground and get away with 14.5A per transformer. At 0.8 ohms 14.5A I got 11.7v with cap of 0.4F (low ripple). I went on Mouser to price 0.1F cap and they are $15 each x 2 = $30 just for the caps then add cost of BR to it. How do these PSU companies build such large-amp devices so cheap? 400W 12VDC for $28 free shipping!
I think I will give up making what I have work since there are cheaper alternatives. Your thoughts.....
 
-
How do these PSU companies build such large-amp devices so cheap? 400W 12VDC for $28 free shipping!
I think I will give up making what I have work since there are cheaper alternatives. Your thoughts.....

I expect that you're looking at a Switch Mode power supply. They can do it because they do it differently. They don't use a large heavy, expensive 60Hz transformer. They use a much smaller, lighter transformer running at a high frequency that is much cheaper. The bridge rectifier might be a 6 or 8 amp part that will be 1/10th the cost of the one that you are looking at. The front end capacitance is probably less than 500 uF (0.0005F). And, of course, they are buying the parts and building them in much higher quantities that the one or two that you might be building.

But you'll get a power supply that will be smaller and lighter than what you were going to do. Plus you'll get a regulated output that will pretty much stay at 12Volts even if the AC input voltage drifts around a bit.
 
It looks like you gentlemen are confusing some details.

First, a non-true-RMS will read very close to the actual EMS value for a sine wave. So a 12VAC reading is as expected for 12 volt output on the transformer.

The peak DC voltage is 1.414 x the RMS voltage - 1.2 volts for the diode drop = 14.1 volts.

No one has mentioned filter capacitors. Without filter caps, the output will vary from 14.1 volts to zero at twice power line frequency. If you use a large enough filter cap as a function of load current, the output voltage will be near the peak DC voltage.
 
It looks like you gentlemen are confusing some details.

First, a non-true-RMS will read very close to the actual EMS value for a sine wave. So a 12VAC reading is as expected for 12 volt output on the transformer.

The peak DC voltage is 1.414 x the RMS voltage - 1.2 volts for the diode drop = 14.1 volts.

No one has mentioned filter capacitors. Without filter caps, the output will vary from 14.1 volts to zero at twice power line frequency. If you use a large enough filter cap as a function of load current, the output voltage will be near the peak DC voltage.
I appreciate your comments. The question was, what did I measure with a multi-meter when I set it on AC volts and got 12.2V and how that translates into DC volts after the bridge rectifier. Then you add large capacitor bank and filtering cap (0.1 uf or so) to it and get some ideal DC output voltage hopefully around 12-12.6V. We abandoned the idea of working with such a large transformer, large load current (I need) at this voltage since it would be very costly to use parts that would make it work. From Mike's sim. he got 200mF (huge, 200000uf) to get rid of most of ripple, and I got 400mf (redunculous, 400000uf) for a very smooth linear voltage. The high amp full wave bridge rectifier is also very expensive. Not to mention that we have not even regulated the output after all of this that will require additional parts and expense.

The question is using the transformer I got, would it be economical to make it work to get the desired output specs with inexpensive parts, or am I better off buying a more modern and sophisticated pre-made power supply unit for $28 that has all the bells and whistles, has more than adequate voltage and current supply for my project. After search for part prices, we came to the conclusion that I am better off buying something off the shelf.

Here is the simulation, hover mouse over load resistor for details: https://lushprojects.com/circuitjs/circuitjs.html?cct=$+1+0.000005+11.086722712598132+30+5+48 v+416+400+416+112+0+1+60+14.1+0+0+0.5 w+416+112+544+112+0 w+544+112+544+176+0 d+544+176+608+240+1+0.8 d+544+304+608+240+1+0.8 d+480+240+544+176+1+0.8 d+480+240+544+304+1+0.8 w+544+304+544+400+0 w+544+400+416+400+0 w+480+240+480+336+0 w+608+240+656+240+0 c+656+240+656+336+0+0.26+12.206016707106794 w+656+240+736+240+1 w+656+336+736+336+0 r+736+240+736+336+0+0.8 w+656+336+608+336+0 w+608+336+480+336+0 o+0+32+0+34+20+204.8+0+-1 o+14+64+0+162+20+12.8+1+-1
 
Last edited:
The Lush simulator is very simplistic. It does not account for transformer source resistance, rectifier series resistance, capacitor effective series resistance. The Duncan simulator accounts for these non-ideal behaviors. The ripple is much worse than Lush shows...

The reality is that large electrolytic capacitors have gotten very expensive. The supply you can buy is a switcher. Its filter capacitors are 1/100 to 1/10,000 the size of what it takes to filter 120Hz.
 
The Lush simulator is very simplistic. It does not account for transformer source resistance, rectifier series resistance, capacitor effective series resistance. The Duncan simulator accounts for these non-ideal behaviors. The ripple is much worse than Lush shows...

The reality is that large electrolytic capacitors have gotten very expensive. The supply you can buy is a switcher. Its filter capacitors are 1/100 to 1/10,000 the size of what it takes to filter 120Hz.
I know. We are in agreement. I was concluding the post by saying why the project was abandoned. There are so many inexpensive modern power supplies out there. sometimes you got to draw the line and say that there are simpler and less expensive alternatives to a design already exist and call it the day. I will utilize the transformers for an AC application such as solenoid switching for synchronized water fountain or something else.
 
If you want a regulated 12 volt DC power supply at 15+ amps, go to your local thrift stores and look for a Microsoft X-Box power supply. These monsters can supply approximately 15 amps depending on the exact model.

These pictures shows show one of these I bought for 3 bucks over the weekend.

20160908_084330-768x1024.jpg


20160908_084339-768x1024.jpg


If you cut the plug off, there are 4 yellow wires and 4 black wires to supply 12 volts. There's a red wire that is a continuous +5v. The blue wire must be connected to the red wire to turn on the 12v output.

20160908_084445-1024x768.jpg
 
If you cut the plug off, there are 4 yellow wires and 4 black wires to supply 12 volts. There's a red wire that is a continuous +5v. The blue wire must be connected to the red wire to turn on the 12v output.
Is that 4 X 4A 12vdc?
I don't have a thrift store close to me but I can look for similar and/or check with Walmart as well. Looks promising and cheap enough. Have you tested it to see if it puts out that much current? It looks rather small!
 
No, the wires are in parallel to supply that much current.

These supplies are breathtakingly expensive new but Goodwill stores often have them hanging around for less than $10.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top