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 converter for 3.8V from 60 - 80 V DC

Status
Not open for further replies.

Diver300

Well-Known Member
Most Helpful Member
I need to get 3.8 V at about 2 A from an input of 60 - 80 V DC.

The voltage is too high for most of the simple buck regulators, so I've been looking at using a transformer, as in most mains voltage input SMPSs.

I'm thinking of using something like XP POWER | ECL10US05-T | Switch Mode | Electrical & Circuit Protection | Farnell UK

I am fairly sure that the output voltage is set by two fixed resistors and that I can tweek the output voltage by changing one of them.

What I am less confident about is the input voltage. 60V is only half of the bottom of the DC input voltage range. Will I need to reduce the number of turns on the primary of the transformer? I guess that the primary will be wound on first so that will basically need rebuilding the transformer.

I have found some SMPS controller ICs. Their example circuits seem to be much the same as one of the power supplies. Has anyone used one and has any hints or suggestions in this case?
 
Yes, I assume that reducing the number of turns on the primary side would work but be aware that the current will double so you need to increase the crossectional area by doubling and the switching transistor might not be able to take it.

Also, be aware tha it is has an active power factor correction circuit it will need to be bypassed.
 
driving an LED? if so you need a current source not a voltage regulator.

Where did you get the idea that buck regulators would not handle it? I could direct you to ones that would but since I think you are actually looking to light a power LED I will direct you to the HV9910.

It is a buck mode current source for LEDs that uses an external power FET and runs of DC from 8-500V
 
Thank you for the replies.

I appreciate the fact that the current will double, but the duty cycle is very small so heating will be insignificant.

I am also concerned that there may be too much current for whatever switching device is fitted to the power supply.

There is no active PFC circuit. The power supply is quoted as working from DC as is.

The load is not an LED. I need 3.8V and the load can vary from nothing to nearly 2A. The HV9910 wouldn't really be suitable.

It could be easier for me to use a buck regulator with an external MOSFET. I was concerned that the duty cycle would be very small, and at low load currents the MOSFET would only be on for an extreemly short time, and therefore difficult to control.
 
actually in a buck the duty cycle is related to the voltage ratio with in reason. at low load it will break into intermittent operation and at the high load it will be slightly higher until you saturate the inductor or blow out the switch, which will happen anyhow if you saturate the inductor.

Either way the duty cycle will be between 20% and 30% or a minimum of 4uS on at 50KHz frequency. While not the simplest circuit I would recommend an LM3475 with a level shifted gate drive to you. The reason is that hysteretic control is inherently stable, making it easy for a hobbyist to get running correctly unlike any other control technique.
 
Last edited:
I can't use the LM3475 because I need to run from a 60 - 80 V supply. The LM3475 has a maximum input voltage of 10 V.

I appreciate what you say about the duty cycle not changing much as the load changes. However most of the time I will be using about 10 - 100 mA so it will be running in discontinuous mode.

Also with an input voltage of 70 V and output at 3.8 V the duty cycle will be about 5.5%

At around 70 V the inductor current rises very fast, so the on-time has to be very small or the inductor has to be quite large. That was why I was thinking of using a transformer.
 
That is why I said you would need level shift for the gate drive. You would obviously need a small regulator of some sort to supply its power as well...even a little zener should do the trick.

With an external FET the primary limitation is what the chip will survive... so you need to limit it's supply voltage to a safe level and level shift the output to the 80V rail that the FET is actually on.
 
That'll work, although you'll need to tweak it a bit to accept 80V and output 2A. I don't think it'll be as efficient as the origional though.
 
that is the nice thing about hysteretic control - inherently stable, pick the appropriate inductor and capacitor and the duty cycle and frequency follow.

There is no frequency to worry about, just minimum on time of 180nS and how fast you can get the FET to switch. This inductor looks like it would work:

Digi-Key - M8771-ND (JW Miller A Bourns Company - PM2110-271K-RC)

unless your FET drive is slow in which case you would need a higher value inductor, the circuit will simply respond by running at a lower frequency.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top