# Single Supply

Discussion in 'Electronics Resources' started by shredability, Feb 27, 2014.

1. ### shredability Thread Starter Member

Apr 1, 2012
31
0
I'm designing something with split supply opamps and I only have a single supply power supply. I know about single supply opamps.

I found this:

Which makes sense, the opamp holds it output at BATT/2 because of the voltage divider seen at the non-inverting input. But I'm thinking the max current the can be sourced/sunk from the GND node is a function of the opamp. (typically in the mA range)

So I did this up:

It's based off an on brushed motor driver circuit. I would just like some opinions on if this will fix my current problem.

Cheers

-Walker

2. ### crutschow Expert

Mar 14, 2008
13,496
3,373
Yes, the transistor will increase the amount of unbalanced ground current you can have. But note that the transistors have to dissipate the power from this unbalanced current which equals the current times 1/2 the supply voltage. So you may have to heat sink the transistors depending upon the amount of imbalance.

3. ### t06afre AAC Fanatic!

May 11, 2009
5,939
1,222
You know you can connect two single supply units in order to get a dual supply power. The solution you have suggest. Is in general something I not would recommend at all. As the solution is like asking for problems in most cases. In the picture I have shown how to do it with two 9 volt batteries. But the method is general

KLillie likes this.
4. ### crutschow Expert

Mar 14, 2008
13,496
3,373
Why do you say that? The pseudo-ground circuit is commonly used when a split supply is desired but only a single supply is available (as the op stated).

5. ### ScottWang Moderator

Aug 23, 2012
4,930
777
If you using the second circuit should be very careful, when the R1,R2 lost the balance then the transistor could be burn out.

6. ### crutschow Expert

Mar 14, 2008
13,496
3,373
What do you mean "lose the balance"? How can that occur?

Apr 5, 2008
15,800
2,386
8. ### ScottWang Moderator

Aug 23, 2012
4,930
777
Such as:
Resistor made with wrong color, any reason cause any one of them shorted, 741 damaged, R1≠R2, etc...

The output of this circuit is one kind of balance of terror, because the Totem Pole output was designed to get the max amplitude, it will make Vce as shorted circuit when it was a amplifier, so on the innate, they was a high switch or low switch, but if want to be a power, whatever the switch function of Q3 or Q2 must be inhibited, and if any one loss the function then the transistor could damaged, it depends on how to choose the Vceo and Ic of bjt.

The circuit for amplifier has a load is speaker, but this circuit there is no load, some circuit as this one will in series two resistors with E to protecting the bjts, if using 10 Ω then it will limiting the current to flows to the bjt about 1.2 Amax, when any one of bjt was on(sat) then another one will be off, the Vce will be about 24V, so the Vceo better has 40V or more.

9. ### crutschow Expert

Mar 14, 2008
13,496
3,373
Of course if you have a bad or shorted component a circuit will fail. That's true for any circuit, not just this one.

I don't understand your ramblings about the "balance of terror". This is a simple circuit with the NPN carrying current if the minus side of the load is drawing more current then the plus side, and the PNP carrying current for the opposite condition. It's a perfectly straight forward circuit. This circuit has little to do with one used for audio output.

10. ### shredability Thread Starter Member

Apr 1, 2012
31
0
That's a very good point! sometimes I forget to analyse my circuits from a power point of view!

Apr 1, 2012
31
0
Yeah I know

12. ### shredability Thread Starter Member

Apr 1, 2012
31
0
This seems to be what I have, it just uses a power opamp and a whack more components to make the system more robust, deal w/ noise etc... But I didn't even think to use a power opamp!!

Thanks!

13. ### ErnieM AAC Fanatic!

Apr 24, 2011
7,436
1,626
Any discussion of a virtual ground absolutely requires a priori knowledge of what exactly is being grounded.

Until that initial question is asked no comment nor solution may be offered.

14. ### shredability Thread Starter Member

Apr 1, 2012
31
0
edit: I assume you mean prior. Technically I'm not grounding anything, sorry for poor terminology. This will be a circuit common or "virtual ground"

15. ### ScottWang Moderator

Aug 23, 2012
4,930
777
If any reason cause R1 shorted or make P.3(+) close +24V, Q3 will be On and the Vce will be about 0.9V, and the Vce of Q2 has 24V-0.9V = 23.1V, that's why I mentioned have to choose the Vceo and Ic of bjt, otherwise the Q2 could be damaged, the same reason as R2, if any reason cause R2 shorted or make P.2(-) close GND, Q2 will be On and if E of Q2 has voltage that it will be about 0.9V, and the Vce of Q3 has 24V-0.9V = 23.1V, that's why I mentioned have to choose the Vceo and Ic of bjt, otherwise the Q3 could be damaged.

So the P.3(+) should be keeping at a virtual zero voltage, when it lost the balance, something bad could be happen as the above described, I already met some netizen used that circuit to do the virtual ground and almost damaged the transistors, when the input power as 36V or 48V then the TIP41C and TIP42C should be change to some other better transistors, many users don't know how to make a choice for the Q2, Q3, so it may easy to burn out the bjts.