Im confused about power supply for op-amps

[DarkZero]

Hey all !

ok i was looking at a schematic that someone tried to build namely PSU

http://www.electronics-lab.com/wp-content/uploads/2015/03/schem1.gif

now im not interested in building one but i have question about power supply of the op-amp in the schematic from schematic it should be +33V/-5.6 but what i see in datasheet of the TL081 that VCC should be +-18V so how it is working ?

[suicidaleggroll]

Voltages are relative.  What does "+/- 18V" even mean?  What's the difference between that and 0-36V, or 36-72V?  The only difference is which point in the circuit you arbitrarily choose to call "zero".  The op-amp never sees "zero", all it sees is a potential difference between V+ and V-, which as long as that difference isn't greater than 36V, it doesn't care.  You could call those rails 10000V and 10036V for all the op-amp cares, it doesn't matter.

[IanB]

Well, to be fair, the absolute maximum supply voltage for the TL081 is given as +/- 18 V, and yet 33 + 5.6 = 38.6 V, which exceeds 36 V. So it is reasonable to question this.

The datasheet also says that input voltages should not exceed +/- 15 V around the mid-point of the supply voltage, which implies that the amplifier should not be used in an asymmetric configuration.

[DarkZero]

this is what i tought correct me if im wrong if in datasheet says supply for op-amp is +/- 18 then i can put total of 36v (potential difference) max across its supply pins, so in circuit i linked either it is poor design or it will barely work with some unexpected results since ppl tried to make that circuit to work with no result.

now is there any more modern op amp to mc1558 since i need some that can go to atleast 40v and i cant source mc1558

[danadak]

Go to the Digikey web site, select the filter parameters of what you are looking for -


http://www.digikey.com/product-search/en/integrated-circuits-ics/linear-amplifiers-instrumentation-op-amps-buffer-amps/2556125?FV=fff40027%2Cfff800dd&mnonly=0&newproducts=0&ColumnSort=-659&page=1&stock=0&pbfree=0&rohs=0&k=opamp&quantity=&ptm=0&fid=0&pageSize=25


Regards, Dana.

[basinstreetdesign]

Check out the LM741.  It can stand 44V across its power pins.

[Audioguru]

That power supply project at Electronics-Lab has many overloaded parts, a few errors and cannot produce 30V at 3A. Hero and I fixed it and our fixes are in a few very long threads in the forum over there.
We used TL2141 opamps (also MC34071 opamps were used but are not made in a DIL package anymore) that have a 44V maximum supply voltage and inputs and outputs that work at the negative supply voltage which can be 0V The TL081 and 741A inputs and output need a negative supply voltage then their total voltage must be too high for them. With the new opamps then the output opamp does not need a negative supply so its supply is the 38VDC (or 40VDC with no load) from the 28V transformer (higher voltage than the original 24V one). Here are our schematic and parts list:

[Cerebus]

this is what i tought correct me if im wrong if in datasheet says supply for op-amp is +/- 18 then i can put total of 36v (potential difference) max across its supply pins, so in circuit i linked either it is poor design or it will barely work with some unexpected results since ppl tried to make that circuit to work with no result.

now is there any more modern op amp to mc1558 since i need some that can go to atleast 40v and i cant source mc1558

I happen to have a handy spreadsheet precisely of 44V+ op amps, here's an excerpted version of it. (The excerpt is because although the original has full performance data in it, it's in units etc. to suit me and I know what's what. Someone else reading it might easily misunderstand stuff I've left in shorthand so it's safer to just cut it down to the minimum. The alternative would be a few hours tidying it up before I could upload it.) This only lists op amps from Analog, Linear and Texas but they are all at least 44V supply:

[Cerebus]

Well, to be fair, the absolute maximum supply voltage for the TL081 is given as +/- 18 V, and yet 33 + 5.6 = 38.6 V, which exceeds 36 V. So it is reasonable to question this.

The datasheet also says that input voltages should not exceed +/- 15 V around the mid-point of the supply voltage, which implies that the amplifier should not be used in an asymmetric configuration.

How would a TL081 'know' that the voltages were asymmetric, it doesn't have a 'ground' pin or any concept of where ground is? I think you're getting confused with its input voltage limit which is specified as +/-15 from the mid-point of the supply voltage; but that doesn't prevent you having a 'ground' that is several volts away from the supply mid-point.

[Audioguru]

Speaking about the input voltages of the TL081 opamp U2 in the original circuit, did you notice transistor Q1 that shorts its output to ground when the power is turned off? It is because the TL081 has a problem called "Opamp Phase Inversion" where its output suddenly goes as high as it can when an input voltage becomes within about 4V from its negative supply voltage. When the power is turned off then the negative supply collapses first and transistor Q1 detects it before it drops to less than 4V and Q1 turns on. The new TLE2141 opamp does not have this problem so Q1 is not needed.   

[grahamed]

Hi

I don't know whether thnis has been said before but the circuit was published in Practical Electronics Oct 1978.  I built two and they have been working fine for 30 years and more.  No TL opas back then......

[Audioguru]

I don't know whether thnis has been said before but the circuit was published in Practical Electronics Oct 1978.  I built two and they have been working fine for 30 years and more.  No TL opas back then......

I first saw this project at Electronics-Lab.com in 2004 where in the forum there were many complaints of its parts burning out and it not making 30V at 3A. Soon later I saw the kit for sale in a local store.

[Zero999]

Hi

I don't know whether thnis has been said before but the circuit was published in Practical Electronics Oct 1978.  I built two and they have been working fine for 30 years and more.  No TL opas back then......

It doesn't mean it's ideal though.

I've used components outside their ratings/specifications before and they've worked fine but I wouldn't want to publish the circuit in a magazine, make a kit or put it into production. At most I may post it here, along with an explanation of what's wrong with it.