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 Design & Zeners

Not open for further replies.


Ummm.... this turned into a long post, I'll give cliff notes:

1. To find the value of a current limiting resistor for a zener diode, do you look in the diode spec sheet for a breakdown current, and then use (VCC - Zener V) / current?
2. We are running some devices off of a normally closed channel of a relay in a car. I'm worried that since the relay control is driven by an opamp/NFET combo that the relay's ability to maintain a connection on its normally closed channel. Is this concern warranted?

Ignore everything below if you read the cliff notes :) its mostly a rant anyway.....

I'm curious as to how some people would view a design feature of a power supply. My senior design group is working on a device for a car and all the parts operate off of 5 volts. Matt did the power supply design for the most part. I gave him a 7805 with a PNP pass transistor, he said he didn't understand it, put it on a breadboard with a PFET (after I told him to use a blasted BJT multiple times) and it didn't work. The output was too high, so he replaced the BJT with a PFET and an opamp.

Here's the basic design...

A 7805 if used to get a 5 v reference, the output is sent to an opamp.
The opamp is used on the gate of a PFET.
The output of the PFET is intended to be 5v, and also goes back to the opamp for comparison.
The opamp adjusts the gate until the PFET drain reaches 5v

I was fine with the design up to here though I would have been more comfortable with a PNP pass transistor. What I have a question about is the following....

The 5v output of the PFET goes to a relay
The electronic devices are connected to the normally closed channels of the relay.
An opamp is used with a 5.1v zener diode & current limiting resistor and also a voltage divider to drive an NFET.
The opamp compares the zener voltage to the divider voltage to drive the NFET.
The NFET drives the relay....

If the voltage on the divider gets too high that means the 5v source is too high, the relay is switched and it connects VDD to GND through a 1 ohm 10 watt resistor, this blows the main fuse, and while all of this is going on it disconnects the digital electronics. The resistor is WAY under rated for this. Its one of those wire wound sand resistor, and the duration is very short. We'll test it to make sure it can last...

Is this actually a sound design? I don't like it, its going in a car and I don't know how much vibration the relay can take without the devices losing power. Also, he's running the divider off of the 5v line (which I agree with) and the 5.1 zener and current limiting resistor off of the 5v line (which I disagree with.)

One of us doesn't understand diodes I guess.... He says that current doesn't matter, if its a 5.1v zener, it'll maintain a 5.1v no matter the size of the resistor. Basically, he has a 1 M ohm resistor then the zener. Yeah, SURE its going to maintain 5v.... but only because its on a 5v source! If he hooked it up to 12v like I said it should have been, with a resistor that large it would be at 12v or just a little below right? It takes a minimum amount of current/power to maintain breakdown if my head is straight....

5.1 zener means 6.9v across the limiting resistor. Look on zener data sheet, find current 'I' in breakdown, and resistor should be around 6.9v / I right?

These circuit boards cost us $80 for two of them, I think we're going to end up needing to order more... and we've already spent over $1000 on this project.

Matt has a problem (in my opinion), and that is he has to do things in a "clever" way though it usually comes out way overdone.... he turned a 7805, resistor, and pnp transitor into a 7805, four resistors, a relay, a zener, a PFET, and an NFET. I also think the opamp PFET may be wired wrong.... its different that what he has on his breadboard, and he said his breadboard works. I could also strangle him, his fuse holder is no where near being able to fit on this board, and he knew it didn't fit before we ordered the board.


  • power_supply.gif
    23.3 KB · Views: 590
Well it turns out that the op amp was wired wrong as I suspected, ran some jumper wires and the power supply works.... and I guess if testing shows that the relay will not be reliable in a car then we'll have to go with something else there. I don't want to be forced into ordering more boards. We're already in over $1000 and the ONLY income we may see from this is if we get hired on the basis of our senior design :p
If you want to post a circuit, post a schematic. Very few of us have the patience to translate a layout to a schematic, and even if we did, component values would be missing.
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips