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.

Single supply analog - splitting the rail

Status
Not open for further replies.

srobertjames

New Member
When I make an analog circuit using a single supply (say, 5V), I split it (using a voltage divider and a smoothing cap), to provide a reference for op amps and the like. I use that middle voltage to bias coupling capacitors.

Two questions:
1. I generally connect the 0V to the ground plane, and have the 2.5V somewhere in the middle. Would I get better performance if I put the 2.5V on the ground plane, and had a positive supply and a separate negative supply, with a smoothing cap from each one to the "ground"? Or does it make no difference, since either way I have 5V powering, and a reference half way?

2. I find that my voltage divider reference needs either small resistors and large caps, or is very noisy. Is there an IC that will do this for me, and do a better job? Is there a regulator which will give me two voltages: a full V and a half V? (I couldn't find one.) Or should I just use two voltage regulators, one delivering 5V and the other 2.5V?
 

crutschow

Well-Known Member
Most Helpful Member
The ground reference has to have a low enough impedance to handle any ground current at that node. If you have large currents in the split ground then a separate power supply to generate a negative voltage may be preferred. If the ground current is small you could use a rail-to-rail op amp to generate the split ground. For larger currents you can add a NPN-PNP complementary emitter follower driver.
 

dougy83

Well-Known Member
For larger currents you can add a NPN-PNP complementary emitter follower driver.
I can imagine that this could potentially have quite bad response to noise spikes due the crossover region of the buffer and the slew rate of the opamp.
 

crutschow

Well-Known Member
Most Helpful Member
I can imagine that this could potentially have quite bad response to noise spikes due the crossover region of the buffer and the slew rate of the opamp.
The op amp is just to provide a DC reference. Its response time is not important. The noise spikes are absorbed by the required decoupling capacitors (typically something like a 0.1µF ceramic and a 10µF electrolytic) between the split ground and the circuit common.
 

dougy83

Well-Known Member
Its response time is not important.
I didn't mean the response time. Unless both emitter-followers have a quiescent current, I would think they would be presenting a high-impendance/reverse-bias in the case of certain polarity spikes.
 

crutschow

Well-Known Member
Most Helpful Member
I didn't mean the response time. Unless both emitter-followers have a quiescent current, I would think they would be presenting a high-impendance/reverse-bias in the case of certain polarity spikes.
As I said, any spikes should be absorbed by the capacitors until the op amp has time to respond, if the spike causes a long-term change in the average split ground current. The op amp will provide the proper bias to the transistors as required depending upon the direction of the split current.

Exactly what type of "certain polarity spikes" are you concerned about?
 

Conrad Hoffman

New Member
Doing this with a passive divider just isn't very good. Either divide and bias the inputs, AC coupling the amplifier, or buy a TI "rail splitter" IC, or build the same thing using an opamp buffer to buffer a resistive divider.

Best,
CH
 

gechihoney

New Member
Hi, i was given these equipment to design a dual rail dc Power Supply:

2 off diodes (IN4001) 2 off Zener diodes
2 off Resistors (1 kΩ) 2 off Capacitors (1 μF)
1 off Oscilloscope, DVM, AC supply source ( ±12V)

Using an appropriate selection of the components listed above, design and construct a direct voltage supply having two outputs, one positive and one negative.
Im not too sure how to go about it, plssss helppp.. Thanks
 

crutschow

Well-Known Member
Most Helpful Member
Hi, i was given these equipment to design a dual rail dc Power Supply:

2 off diodes (IN4001) 2 off Zener diodes
2 off Resistors (1 kΩ) 2 off Capacitors (1 μF)
1 off Oscilloscope, DVM, AC supply source ( ±12V)

Using an appropriate selection of the components listed above, design and construct a direct voltage supply having two outputs, one positive and one negative.
Im not too sure how to go about it, plssss helppp.. Thanks
Please start a new thread. Don't hijack this one.
 

Artificer

New Member
If you are in a position to use a "recycled" PC power supply, you could use the 5 volt lead as common. The +12 becomes about +7 and the ground becomes -5. Yes, there is a -12 supply in a PC supply, but no current to speak of. That's a "git by" way......

I would look upstream of the input to the +5 regulator circuit, see if it can be split off there. For bias voltage on an Op Amp, it need not mirror the positive rail, merely be clean.

In industrial instrumentation, the usual practice is to use a center tapped transformer, grounding the center tap. Each end is then rectified through a FULL WAVE bridge to produce full output voltage rail to rail. Filtering may be done across full positive and negative or from both to ground. Or both, depending on how sensitive the circuits.

Lastly, use a second supply, connect them in series, ground the middle. Brute force but it'll give you exactly what you want. (ie +12 and -5)
 
Status
Not open for further replies.

Latest threads

Top