I am trying to step up my voltage from 5 volts to 36 volts.

The only way that I am seeing this is to have a DC DC Converter. I am planning to use a DC converter which steps up 5 volts to 12 volts each and by using three of them and then later on connecting them in series assuming the voltage would add up.

Would this be possible?

DC/DC converter, semi reg, 3W 12V specifications:

  • DC / DC Converter Type: Isolated POL
  • DC / DC Converter Output Type: Fixed
  • Input Voltage DC Max: 5V
  • No. of Outputs: 1
  • Output Voltage Nom.: 12V
  • Output Current Max: 250mA
  • Output Power Max: 3W
  • Depth: 19.5mm
  • Width: 7.2mm
  • Height: 10mm
  • Isolation Voltage: 1kV
  • Output Current: 250mA
  • Output Voltage: 12V
  • Power Rating: 3W
  • SVHC: No SVHC (19-Dec-2012)
  • Supply Voltage: 5V
  • 1
    \$\begingroup\$ I dont see why this would not be possible with a boost converter? Boost converters can easily multiply the voltage practically & safely around times 10(in this case 36V should be achievable using that). \$\endgroup\$ – Sherby Jan 23 '14 at 2:20

This is possible only if the DC/DC converters are of the isolated kind. In an isolated converter there is no ohmic connection (no d.c. current path) from the output to the input. If you try this with non-isolated converters you will probably destroy them.

I can't say whether your plan is the best way to get 36V from 5V. I suspect you would get better efficiency from a single converter but I don't have a particular part to recommend.

  • \$\begingroup\$ From the OP's "Type: Isolated", I would think that's a safe bet. ;-) \$\endgroup\$ – Spehro Pefhany Feb 22 '14 at 3:15
  • \$\begingroup\$ @SpehroPefhany The original question did not contain that specification, it was added after I posted my answer. \$\endgroup\$ – Joe Hass Feb 22 '14 at 10:15

Here, I found 21 boost regulators that will do what you need.

One that looks pretty good is the LT3759 (the output can go much higher than the 12V shown in this example)



I've done this on a product before... although that was adding multiple isolated 30V supplies to get 60V. The datasheet of the modules also explicitly shows it to be safe to do so.

You can't get 36V more directly?

I'm assuming you're talking about powering the supplies all in parallel, and tying the isolated outputs together in series.

I would at a minimum do appropriately sized diodes across the outputs of each supply to make sure that if one supply is off or slower to come up than the others then you don't get -24V across it's outputs. A unipolar filter cap wouldn't like that too much. Even with diodes there will be a reverse leakage current that I'm a bit nervous about (although it probably doesn't matter... ?).

I'd recommend trying to find a more direct solution, and if that fails to look for isolated supplies that explicitly allow for adding the outputs in series (probably with said diodes).

  • \$\begingroup\$ Thanks darron. Tried resourcing for different alternatives but its not possible though :) i shall look at placing diodes :) \$\endgroup\$ – muren Nov 24 '13 at 10:49

Yes its possible and others have covered that .People ask why would you do this ? on the face of it you wouldnt because the boost ratio is well within the realms of available SMPS technology as sbell has shown.There is in fact a gross EMC advantage of doing what you propose because the peak switching volts inside each convertor is lower ,this means that you will get less RF voltage between input and output terminals .In your example it could mean almost 10dB on a radiated plot .Now do you really need to have a galvanicly isolated output .Isolated DCDC convertors are more expensive and less efficient and larger than thier non isolated counterparts.


Look at the datasheet's Absolute Maximum Ratings. I guess it should have input voltage like Vin= VCC+ n% ( which n may be a small number, in your example Vin Max may be around 12 volts). When you convert with first one you are feeding 5v into the converter. for the second one , input is 12v, and for the third one is 24 volts.

My guess is the second one gets hot and the third one burns! By all, it was better to mention your part number to check all of these.

  • \$\begingroup\$ Hmmm. When the OP said that the converters would be connected in series I assumed that the voltage outputs would be connected in series, not that the output of one converter would be fed to the input of another. Perhaps we can get a clarification. \$\endgroup\$ – Joe Hass Nov 23 '13 at 13:28
  • \$\begingroup\$ @JoeHass yes the question is rather ambiguous \$\endgroup\$ – Aug Nov 23 '13 at 13:30
  • \$\begingroup\$ Im sorry for not being clear.the converters would be in parallel but later on intended to be connected in series.is that possible? as then i would be having three output of 12 V and would connecting three of them give me 36 V? \$\endgroup\$ – muren Nov 23 '13 at 14:01
  • \$\begingroup\$ 3 x parallel 12 volts will give you 12v and not 36v. \$\endgroup\$ – Aug Nov 23 '13 at 14:13
  • \$\begingroup\$ Read in a portal that :If the required output voltage is higher than the rated voltage of the chosen given converter, then the outputs of two converters can be connected in series in order to achieve the desired output voltage. It is recommended that two converters with the same model are used for this application. \$\endgroup\$ – muren Nov 23 '13 at 14:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.