Using 9 volt batteries for +/- 9v op-amp power

[Category 5]

I am going to be matching some transistors for Fabio's op-amp boards aqnd was wondering If I can do this.

I have ordered some 741 op-amps to make the transistor test circuit. It needs +/- 10v to work. Since I don;t have a spare power supply I can use I was thinking of using 9 volt batteries.

Can I wire one correctly (tie - to ground) and the other one backwards (just tie positive to ground and use the - as my - voltage source)?

In my mind this sounds like it will work. Most of the 9 volt batteries I have tested give off about 9.4 volts which seems like it should be enough to get the matching circuit to work properly. I would hate to have to build a power supply just for matching transistors.


I figure if all goes well I can get a hundred or so transistors. I'll match them into pairs, and I can surely share pre-matched pairs with other in the group looking to build op-amps as well.

Shane

 

[AMZ-FX]

Can I wire one correctly (tie - to ground) and the other one backwards (just tie positive to ground and use the - as my - voltage source)?

Yes, that will do it.

regards, Jack

 

[clintrubber]

Which test-circuit are you using ? Don't know the '741 DC-specs, but could imagine the offset-performance is not very good and could influence your matching results ?

Bye,

Peter

 

[lanxe]

if i understand your question correctly, you can do it like this:

Ryan

 

[NewYorkDave]

Funny, I was just thinking about this the other day. I'm in the preliminary design phase--in other words, "thinking about it"--of a circuit that will most likely run off two 9V batteries.

Wiring up the batteries as shown above will work fine most of the time. The only problem I can see is that if one battery is weaker than the other for whatever reason, you'd get offsets in your circuit. One approach that might be worthwhile is to wire the two batteries in series and, rather than grounding the junction of the two batteries, use a simple "rail splitter" circuit to ensure that the supplies are always symmetrical with regard to ground. Such a circuit is shown below in "skeleton" form (with optional decoupling caps and reverse-polarity protection diodes omitted for clarity):

 

[lanxe]

i just put this circuit together last night so that i could run +/-9Vdc from a wallwart and it seems to work fine for what i am using it for..........check it out here:

http://www.geofex.com/circuits/+9_to_-9.htm

Ryan

 

[StephenGiles]

Well...strap me to a tree and call me Brenda, that's the exact circuit I used for a dual What compressor I built for my nephew a couple of years ago. It runs off an 18v AC adaptor.
Stephen

 

[lanxe]

[quote author="lanxe"]i just put this circuit together last night so that i could run +/-9Vdc from a wallwart and it seems to work fine for what i am using it for..........check it out here:

http://www.geofex.com/circuits/+9_to_-9.htm

[/quote]

i used this circuit to power the synth in my avatar...........BUT there is some minor variation in the oscillators that i cant figure out if it is the power supply or if it is something that i messed up..........but it is very minor.

 

[Bo Hansen]

Here is a nice chip from Texas
TLE-2426, The"Rail Splitter" Precision Virtual Ground.

http://focus.ti.com/docs/prod/folders/print/tle2426.html
http://focus.ti.com/lit/ds/symlink/tle2426.pdf

--Bo

 

[ebartlet]

[quote author="NewYorkDave"]Funny, I was just thinking about this the other day. I'm in the preliminary design phase--in other words, "thinking about it"--of a circuit that will most likely run off two 9V batteries.

... use a simple "rail splitter" circuit to ensure that the supplies are always symmetrical with regard to ground. [/quote]

I did this the other night for a project. I used 1k resistors (which obviously will waste more power than 100k's), which was just a guess, but how do you go about selecting a resistor value for this? How much does it matter?

 

[Category 5]

[quote author="clintrubber"]Which test-circuit are you using ? Don't know the '741 DC-specs, but could imagine the offset-performance is not very good and could influence your matching results ?

Bye,

Peter[/quote]

I lost the link, but I will try to post the .gif of the circuit later tonight. I suppose I could "match" 9-volt batteries for the circuit too, if that's necessary. Could I just use another op amp with better offset specs?

Shane

 

[PRR]

> if one battery is weaker than the other for whatever reason, you'd get offsets in your circuit.

> I suppose I could "match" 9-volt batteries for the circuit too, if that's necessary. Could I just use another op amp with better offset specs?

No, no no. If an "op-amp" is ANY good, power rail unbalance has very MINOR effect. And in fact even on the dumb old 741, you would scarcely be able to measure the offset drift caused by changing from +/-9V to +3V/-25V.

If offset really matters, note that a 741 can have offset of 5mV. And we used to think that was good. (I just saw a high-speed low-noise amp with rated offset under a microvolt.)

And if all you are doing is matching, all in one sitting (by far the best way without precision repeatable test gear), you don't care if all your results are 1mV or 50mV off, so long as they are all off the same way.

My first IC project was a 741 powered by 2 9V batts, just like this. The batts ran many-many hours, the many-mV offset was not a problem feeding the grid of a tube amp.

I did have a moment of panic: I powered-up the first time, then re-read the datasheet. There was something about running with just one power rail, and the implication was that I could blow a very expensive 741 chip (yes, this was the very dawn of chips, and I paid 50 times what a 741 costs today). And you can't disconnect both 9V Batts at the exact same instant. I think I mis-understood the warning, or else a 9V battery does not have the strength to zap a chip that easily: I plugged and unplugged batts in either order every day and never killed my precious 741.

 

[Category 5]

I like the max1044 trick in the above post. Seems that would be useful to have around whenever bipolar power is required since you can easily just put any voltage in via a wal-wart or battery.

I ordered some max1044cpa from digi-key. I will report back. In the mean time if anybody has any tips as far as getting the closest match for the transistors I'd appreciate it.

Shane

 

[Category 5]

Well I built the little max1044 thing and I think that is such an awesome little trick. Perfect +/- power from a single source, and pretty much no offset.

I know this will be fine for my transistor tester circuit, but do you think this method is robust enough for a preamp?

It becomes concievable that (aside from phantom) this could be used to get +/- 18v from 2 nine volts, with plenty of current for a single channel of N*ve or A-P-I. My only concern is the ripple. Apparrently you can raise it to around 60k by tying pin 1 to pin 5 on the 1044. This would put it out of audible range, but would there be other problems, say with phase and oscillation?

Shane

 

[CJ]

Yes the dual 9 volt thing has been around since Jesus was wearing shorts.

Some opamps will run down to 2.5 volts (OP90) so you could actually usee a couple of AA batts. I have a guitar splitter that runs off of 4 AA's and the batts have lasted an eternity.

I once paid 5 bucks for a 741 in 1970.
What did you pay Paul?