Author Topic: SMPS but for a slightly different application  (Read 511 times)

0 Members and 1 Guest are viewing this topic.

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 772
  • Country: gb
  • Life's too short.
SMPS but for a slightly different application
« on: October 12, 2020, 02:18:55 pm »
I have been wandering if this is possible, my idea is to use two high current server power supply's, but use them in place of a linear transformer. I would wire both server power supply's in series for 24 Volts (floating DC negative output, primary earth bonding in tact) Then put that 24 Volts DC as input for a linear regulator with multiple pass transistors high current output at 13.80 Volts. I think it's possible, but not exactly sure how the linear voltage regulator and pass transistors will see the input voltage. In as much as, because it's not linear DC input, would the semiconductors act in a different way, than if it was a linear transformer.
I do have linear transformers, but like the idea of trying, thinking outside the box ideas. I now I could use a buck or boost converter in this configuration, but I'm looking for RF friendly power source.
Thanks for reading, any thoughts appreciated.
 

Offline SpecialK

  • Regular Contributor
  • *
  • Posts: 214
  • Country: ca
Re: SMPS but for a slightly different application
« Reply #1 on: October 12, 2020, 02:54:42 pm »
I don't think it will work.

I believe a switching power supply's output is earth referenced as it doesn't have a big old transformer to do the isolation.  You would be able to test for this with a multimeter.  If there is voltage between the positive output of one and the negative of the other that would indicate the negative are tied to earth.

Secondly, taking 24V and bleeding down to 13.8V through linear regualtors will make a lot of heat if you are draing any kind of load.  That's 10.2 Watts per every amp drawn.

There are instructions on-line how to make server power suplies output 13.8V.  Maybe you should look at them.  Pretty easy to get 1000W that way I am told. Probably could double them up for more current too, but in North America, they would need to be on separate outlets. In UK it would probably work.
 

Offline eblc1388

  • Frequent Contributor
  • **
  • Posts: 400
  • Country: gb
Re: SMPS but for a slightly different application
« Reply #2 on: October 12, 2020, 03:07:54 pm »
To get 13.8V vs linear regulator from two 12V server supplies in series is silly. You will waste nearly half of the power to whatever you're powering.

If all you're after is 13.8V with high amperage, then there are many schemes floating on the net about hacking a server power supply's output from 12V to 13.8V. However, each different supply model would require different hack so there is no one single solution to all the power supplies. Many of the schemes are workable and many people have succeeded in doing so. It involves just two steps, turning up the output voltage and readjust/disable the overvoltage protection shutdown.



 
 

Offline davelectronicTopic starter

  • Frequent Contributor
  • **
  • Posts: 772
  • Country: gb
  • Life's too short.
Re: SMPS but for a slightly different application
« Reply #3 on: October 13, 2020, 01:55:09 am »
Thanks for your replys. I have modded many server power supply's of different models. They are great for the power output, but most are electrically noisy. They just raise the HF Radio noise floor, a few are better than others. I was just looking to use them for an input to a linear circuit. Going back about 5 years now, I did attempt this idea, but it had issues. It gave an output, but there was what seemed unstable voltage, it was constructed for 12 Volts, but that would jump from 11.60 Volts to just up to 12.80 Volts with no load. The pass transistors where two MJ11015 darlington transistors. The regulator was an L7812, I've no idea why the voltage did this. I assumed it was oscillation going on.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf