Quick and dirty 5v to 3.3v supply that's stable

Status
Not open for further replies.

Thread Starter

programmer6502

Joined Feb 1, 2014
132
Hi,

I'm working with a MCU and am interfacing with a chip that wants 3.3-4v VCC when my only power source is 5v. Initially I thought I could use a resistor ratio setup to drop 5v to 3.3v and it worked, but as soon as there's a load the voltage drops too low (which I guess should have been obvious). I also tried using a Mosfet setup, as well as PWM with the same results. I know that a regulator would do the trick, but I've already made two parts orders with this particular project and was wondering if there's a way to build a stable 3.3v "regulator" with common components like transistors, caps, resistors, etc. (This 3.3v source would only be required for one chip.)

Yes? No?
 

Thread Starter

programmer6502

Joined Feb 1, 2014
132
How much current does your 3.3 volt circuit draw? What is the minimum and maximum current for it?
I'll measure and find out. The chip was salvaged and I can't find any specific information for it. It's a decoder from an RC radio system that I reverse engineered.
 

Papabravo

Joined Feb 24, 2006
21,158
Where do you think I got the other three attempted methods from? ;)

But thanks a bunch! I'll check that out!
I don't know, but you didn't exactly give a definitive description of a zener regulator with a pass transistor. You did describe a couple of other hair-brained schemes, but what do I know? Where are the schematic drawings?
 

Thread Starter

programmer6502

Joined Feb 1, 2014
132
I don't know, but you didn't exactly give a definitive description of a zener regulator with a pass transistor. You did describe a couple of other hair-brained schemes, but what do I know? Where are the schematic drawings?
True, but it was intended to be more of a general question and I just wanted to give enough details for the question to be answered, which thanks to you, was answered! That's what I was looking for, so I don't have or intend to provide schematics.

I apologize if I was rude, I just research the best I can before coming here. Even if the answer was in plain sight lol.
 

ronv

Joined Nov 12, 2008
3,770
True, but it was intended to be more of a general question and I just wanted to give enough details for the question to be answered, which thanks to you, was answered! That's what I was looking for, so I don't have or intend to provide schematics.

I apologize if I was rude, I just research the best I can before coming here. Even if the answer was in plain sight lol.
Here are 3 ideas:
It is a simulation of 3 circuits. The outputs are shown at the top with a current step from .1 ma to 40 ma.
1- is a transistor - similar to your resistor divider, but powered up by the transistor
2- Some diodes in series with the 5 volts to drop it down.
3- What is called a shunt regulator where the diodes clamps the 5 volts to a lower voltage.

The diodes don't let you set the voltage of course.
The transistor requires your 5 volt supply to be solid. If it varies some of that variation will be seen at the output.
 

Attachments

Thread Starter

programmer6502

Joined Feb 1, 2014
132
Here are 3 ideas:
It is a simulation of 3 circuits. The outputs are shown at the top with a current step from .1 ma to 40 ma.
1- is a transistor - similar to your resistor divider, but powered up by the transistor
2- Some diodes in series with the 5 volts to drop it down.
3- What is called a shunt regulator where the diodes clamps the 5 volts to a lower voltage.

The diodes don't let you set the voltage of course.
The transistor requires your 5 volt supply to be solid. If it varies some of that variation will be seen at the output.
Circuit #1 did the trick with what I had on hand! Thanks a mill!

I think it's about time I start using one of those circuit simulators. It would save a lot of time and I wouldn't risk destroying anything.
 

dl324

Joined Mar 30, 2015
16,839
I think it's about time I start using one of those circuit simulators. It would save a lot of time and I wouldn't risk destroying anything.
You still need to know what you're doing. I ran a simulation and LTspice happily operated a 2N3904 at 1A; 10X the rated current.

I simulated a switching regulator using a linear regulator and the program was very picky about feedback ratios that would work.
 

atferrari

Joined Jan 6, 2004
4,764
Few days ago I implemented a regulator with an LP2950, two caps plus a load resistor. Do not waste time in imprecise solutions and posting about them.

EDIT /
5 to 3V3
/EDIT
 

Thread Starter

programmer6502

Joined Feb 1, 2014
132
Few days ago I implemented a regulator with an LP2950, two caps plus a load resistor. Do not waste time in imprecise solutions and posting about them.

EDIT /
5 to 3V3
/EDIT
You have a point, it isn't practical but it was still a solution and it solved my problem:



The point was to not make a 3rd parts order so I'm happy! Made from scratch using salvaged parts :D
 

atferrari

Joined Jan 6, 2004
4,764
Maybe I was too rough, and sorry for that. The reason is that in my case I was preparing to feed an RN42 which costs dearly here and the associated 8 level translators I need with mi 5V micro.

Yesterday, the second took say, 3 minutes to implement and the output was 3.30 V. Well, my DMM says so.
 
Last edited:

Thread Starter

programmer6502

Joined Feb 1, 2014
132
Maybe I was too rough, and sorry for that. The reason is that in my case I was preparing to feed an RN42 which costs dearly here and the associated 8 level translators I need with mi 5V micro.

Yesterday, the second took say, 3 minutes to implement and the output was 3.30 V. Well, my DMM says sow.
Yeah, I get what you're saying. My first choice would've been to pay 3x the normal cost to get a regulator at my local Radio Shack just to be "done", but I couldn't find anything standalone in the 3v range. They did however, have an adjustable regulator like you mentioned that I would have gotten if I needed provide power for more than one chip. But since I only needed to power one chip, I thought it would have been a little overkill. Not to mention that building that little transistor circuit took roughly the same time (5 minutes) it would've taken for me to implement the external circuitry for the adjustable regulator.

So in my particular case I think it was a just decision, but in most circumstances I would do something like you did, especially if the chip in question was sensitive and of high value.

If anything, I learned something new! :D
 
Status
Not open for further replies.
Top