dual voltage power suply

[ikalogic]

hi.
741 op-amp need dual voltage power supply, i did something that worked very fine but i am not sure if it is really ok...

it is a simple voltage divider circuit:

(+)---/\/\/\/\/----(GND)------\/\/\/\/\/\---(-)

the 2 resistance have to be very high (1 M-ohm), and voila, i've got the +ve, -ve and ground, and it works fine with the op-amp

but there is something strange, if the value of the 2 resistances are very near (~equal to each others), the op-amp will takew a very long time to respond to input change...which makes me ask, is what i have done to get a dual voltage wrong from the begining??? if how can i convert from single polarity to dual voltage??

thx

 

[panic mode]

You don't need high resistance, it is not incommon to use lower
value such as 10k or less. What you didn't show is capacitor
accross one of the resistors. It is used to provide low impedance
path for high frequency signals. Because your divider uses such
high values, it takes time to charge the capacitor and reach
the 1/2 point.

 

[mozikluv]

voltage divider

hi,

your approach using the voltage divider is alright, also take the advise of panic mode, however you also mentioned having a single supply but would like to use it as a dual supply. well it can be done but you did not mention at what voltage you are using now. if that is a 12v supply you can have a dual supply but the voltage will be half (6v-0v-6v). the same principle like the voltage divider is used. the jucntion between the two resistor will be your center tap.

 

[ikalogic]

hi, thanks very much.

however, i've still got something that i really dont't understand...

Even after a pair of 5K resistances the response to changes in input is very slow and it fades out like if there was a capacitor, but there is not.

then i have tried geting REAL dual voltage from 2 9V batteries inseries :

(+v)--[+9V][battery][-9v]--(GND)--[+9v][battery][-9v]--(-v)

and the problem is still there, does that mean that the problem of slow response comes from the circuit it self??

the problem is however solved if i use appreximately 2k-ohm and 8k-ohm resistances for the voltage divider.... op-amps are so misterious!


thx a lot

 

[Ravi]

I think why the op-amps are behaving mysteriously is due to their input offset voltage.This is the voltage which,when applied at the input, provides an out put voltage of exactly zero. Offset may be minimised by applying large amounts of (-) feedback or by using offset null facility provided by the op-amp @ pin 1 & 5.

Usually op-amps needs not only a dual supply but also a well balanced supply. Meaning both supplies magnitudes should be same.

I also think that the slow response, it is due the slow slew rate of the op-amps?

 

[Nigel Goodwin]

I think you need to post the complete circuit, so we can see what you are doing. But as already mentioned, you don't want high value resistors for your potential divider - they must pass enough current to swamp any uneven supply demands from the circuit. Presumably you have a decoupling capacitor across at least one of the resisitors?.

It's common practice to use another op-amp to buffer the mid point, this makes things a lot better.

However, this slow change sounds like you are doing something wrong somewhere.

 

[ikalogic]

here it is all of it!

here is all the project
well as you can read i am a newbie! so no laughing ok?!

it is simple light detector which will light a led (D1) when the light beam between the red led (D2) and sensor (Q1) is cutten.

NOTE: that Q1 is not a transistor i just did not find anything else in the drawing software [actually it is the sensor found in all old computer mouse, I apply a voltage on 2 leads and i get an output voltage if there is an infrared light striking it. Red led simply does the job of emmiting infra red light very fine!]

So when voltage divider R1 is at center position, all my output will be delayed by about 0.5 a second, from the time an input change was applied.

BUT: when i adjust R1 so that (+V = 0.2 v) and (-V= 9.8 v) the whole thing works very fine with no delays at all, and in both 2 position:
- D1 glow when light does not strike the sendor
- or D1 glow when strike the sensor
which i adjust by inversing D1 and connecting it to -V

so .. does anybody understand ??

 

[Nigel Goodwin]

Well I'm rather baffled by the circuit, but one obvious mistake jumps out and grabs me! - you don't have a series resistor with the LED - that's certainly likely to upset your dual-rail supply. Also I'm bemused by the way the inverting input of the opamp is connected?.

 

[ikalogic]

ok, sorry for the non-prefessional design of the circuit!

I have added the resistances in series with the led,
and have exchaged led (D2) with a 'real' infra red led and thing go much better, i mean the response is faster, but still the response is even better if i adjust R1 so that approximatly: (+V = 0.2 v) and (-V= 9.8 v).

does the feedback reistance has anything to do with the deleyed response??

thx a lot.

 

[Sebi]

Some thoughts:
definitely need two decoupling caps (10uF) and connecting the pot wiper to end. A resistor need from inverting input (-) to GND, because the feedback not working...
The canibalized part from mouse a dual photodiode, show me Your connection in circuit.
And what is the function of circuit? What is the end-application?

 

[ikalogic]

wow... thanks very much.. i really appreciate giving this precious effort.. really thank you...i was allways wondering what is the sensor from the mouse.. now i know it is a dual photodiode... gr8!

well here is what you asked, my connexions in the circuit... i know it looks strange but this exactly how it is connected in my circuit...

Actually there is no specific end application, it is many done just to learn .. i am an absolute beginner... but i think it can be used in alarm systems by connecting the output to the trigger of a monostable timer that outputs a voltage to a buzzer ..for example!

 

[ikalogic]

hey hey hey..i thought about something.. tell me if am wrong.. can't the slow response of the output be caused by the slow response of the led which shows me the output ???

 

[Sebi]

O.K. You need only a half section and a bias pot.

 

[B-o-b]

741 and a single supply

Greetings!

Hmmm... I'm not sure if I understand how yer applying this circuit to the opamp. However, if it works, then I think yer on the right track.

There is a way to "trick" the critter into thinkin' it has a dual supply, when it's just a single supply. Yer resistors are correct, but again, I don't know how yer connecting them in conjunction with yer 741.

I'd like to recommend the Engineer's Mini-Notebook (Radio Shack
part # 62-5011) and on page-16, it discusses the fool-proof way to use a single supply with a 741.

Since I can't see yer circuit, I can't give U a proper suggestion right now. But if U'd like, U can email me the schemo (bob4analog@yahoo.com) and I can look at it further.

Till then, I think U'll really enjoy the Engineer's Mini-Notebook I suggested. I've found it to be very helpful.

Cheerz,

Bob

 

[Roff]

Don't fart around with adjusting resistors. You need an op amp to create a stable artificial ground that doesn't "move around" as you play with the circuit. The only limitation is that the difference in the currents between the "new" positive and negative supplies can't exceed the output current capabilities of the "ground" op amp. You might want to add capacitors between the two supplies and ground. Don't put capacitors across either of the 100k resistors - it will make things worse.

 

[plot]

put a capacitor in parallel with each resistor as suggested, should work. it's what i do whenever i'm powering an opamp from a single supply.. never had a problem with it.

 

[ikalogic]

thx

 

[ikalogic]

hi
why does and op-amp for a dual voltage poer supply is better than just a simple voltage divider?
what makes it so special ?

i have heard b4 about using op-amps for dual voltage supply but never understood why ..Thx

 

[Nigel Goodwin]

hi
why does and op-amp for a dual voltage poer supply is better than just a simple voltage divider?
what makes it so special ?

i have heard b4 about using op-amps for dual voltage supply but never understood why ..Thx

A potential divider is fine, as long as the current drawn from both the positive and negative rails is the same (or very similar). Using an opamp to generate the split supply allows them to be different (within reason), the opamp corrects for the difference.

 

[Roff]

hi
why does and op-amp for a dual voltage poer supply is better than just a simple voltage divider?
what makes it so special ?

i have heard b4 about using op-amps for dual voltage supply but never understood why ..Thx

Expanding on what Nigel said:
The problem with using the voltage divider without the op amp is that if the loads on the supplies vary, the supply voltages will fluctuate relative to the "ground" created by the divider. This may not be a big problem for other op amps in the circuit, because they generally have good supply rejection, but it can wreak havoc on other circuits using the same supplies. For example, a common-emitter transistor amplifier generally has very poor power supply rejection. Fluctuations of the supply voltage will show up on the output. Circuits such as these are one of the reasons we use voltage regulators.
I have re-posted the schematic of the circuit with the buffered ground below.