It works!! Even when others said it wouldn't.

I posted a circuit awhile back and a number of people said it would never work. Well because so many implied that I was crazy or worse I went ahead and built the circuit. Here is the schematic I originally posted:

Now I don't have a 24 volt source but I do have an 18 volt source. Also I only had 7809 and 7909 regulators not the 7815 and the 7915 in the schematic but the concept is the same. Here is the circuit working.

The gray meter in the back is the supply voltage. The meters in the front shows the output of the 7809 regulator and the output of the 7909 regulator (the 7909 is the negative value). Now I probably should throw some caps at the circuit but this shows it works.

So my point here is don't believe the nay sayers. There seems to be a number of people that want to insult and degrade people. So for those just ignore them. If someone wants to help you, or you want to help, those are the people you need to associate with. Remember you can always build the circuit and see for yourself.

oh wow thats a nice design, this can power op amps, i will be using this, was looking for something like this for a while now.

Do the voltages hold when you start drawing current? That's the real test.

joseph_m:
I posted a circuit awhile back and a number of people said it would never work.

Well actually, going back and looking at the thread, only a single person said it would never work. The rest were people asking questions about what you were trying to do and discussing possible issues with the circuit.

joseph_m:
So my point here is don't believe the nay sayers. There seems to be a number of people that want to insult and degrade people.

My point here is, before you go off and get an attitude towards criticism, realize that a part of designing circuits is listening to what people say--especially their questions. Engineering minded people are very curious by nature and will sometimes point things out by asking questions. Sometimes it is because they are right and sometimes it is because they just want to learn too.

joseph_m:
I posted a circuit awhile back and a number of people said it would never work. Well because so many implied that I was crazy or worse I went ahead and built the circuit.
There seems to be a number of people that want to insult and degrade people. So for those just ignore them. If someone wants to help you, or you want to help, those are the people you need to associate with. Remember you can always build the circuit and see for yourself.

It would have been nice if you had simply updated your subject
http://arduino.cc/forum/index.php/topic,103413.0.html

You are emotionally over-invested in your "design".
Start drawing some current from that, from both "outputs" -- at the same time.

CrossRoads:
Do the voltages hold when you start drawing current? That's the real test.

I didn't want to put any load on it until it had some heat sinks on it. And I wanted to flush out the circuit with some caps. But I am sure it will work.

I didn't update the original thread because I didn't want to rub your nose in your error. But you just did it yourself.

I never said this was my design. Actually I said I found it on line so it's difficult to make the point that I am "over-invested" in it. What I am invested it is something I have seen many times on this forum. The negativity and degrading comments to others. I think that is just plain wrong. Now I am not pointing the finger at you but you didn't have anything to offer except negativity in the original post. The first comment you made was totally unhelpful and very negative. If that's how you want to live your life I will just ignore your posts in the future. But I would hope that you want to be a good person and help others not try and degrade them.

You have not been degraded.
You are making a claim which is not supported by fact.
And nobody is a bad person for disagreeing with you.
I hope you will have as much character as you have chutzpah once you start drawing current from this "circuit" of which you are so incredibly enamored ("yours" or not) and faithfully report the results.

I think it best to take the advice apparently negative or positive as a professional would, critisicm that makes you think a little harder and maybe take an extra effort to test and prove an idea,
mind you its a forum so the topic is open to the publics eye, they can either comment, help, degrade or appreciate what's posted but you have to remember its the internet and alot can and will go
and just to add, I can't imagine that circuit sustaining much load on the negative side of things
and a question, idk if the original designer noted, why the virtual ground from the negative and the ground of the positive regulator?

joseph_m:
Now I don't have a 24 volt source but I do have an 18 volt source. Also I only had 7809 and 7909 regulators not the 7815 and the 7915 in the schematic but the concept is the same. Here is the circuit working.

An issue with your circuit is that you do not have a common ground for the two voltages. Load or no-load, you will never read -9V between 7909 output and ground (as marked in your schematic) and so it is not appropriate if/when a dual supply is required.

You could achieve the equivalent of the above with a single 7809 regulator and just flip the terminals when a negative voltage is required.

You haven't yet proven that it's a working circuit as just reading the voltages without load current being drawn is a very incomplete 'test'. I would suspect you might find problems when there are unequal load currents being drawn from the two different voltage outputs. So be sure to test by measuring the two voltage output value with one voltage rail supplying say 200ma and the other 100ma and see if the voltages change or not.

I have seen many DC voltage 'rail splitter' designs in the past, some very cleaver, but most with some limitation or other. So perhaps you have indeed come across the simplest voltage rail splitter design to date, but I would suspect you would need to show more evidence to convince some of us. The lack of a true balanced ground connection is very suspicious to me.

In you schematic you have a 24vdc voltage source driving a negative and positive voltage regulators. Now do you mean to tell us that if you hook your volt meter to the +15 and -15 terminals that you will get a 30vdc reading? ( a true dual polarity DC power supply will read 30vdc directly across the two output terminals) How is that possible with using a 24vdc source? Remember that any circuit connect must use just the single ground terminal output pin and one or the other or both output voltage terminals.

Good luck and keep us posted on your results.

Lefty

That circuit as posted is a nonsense. It shows you getting +/- 15V out of a 24 volt battery. So without any form of switching you get a total of 30V from a 24V battery - I think not.
So far what you have demonstrated is a voltage splitting circuit that gets a maximum of +/- half the input voltage.

Note that the numbers you actually came up with -- +9V, and -9V, from an 18V supply -- are theoretically possible with perfect linear regulators. Imperfect regulators will end up causing voltage drops when you start pulling load, but I wouldn't be surprised if the circuit that you're measuring would stay at, say, +7V/-7V with some amount of load on it. That's still totally congruent with an input of 18V.

I could write an even simpler version of that 18V to +9V/-9V circuit right here. Check the attachment for the implementation. Get a couple of 1A 9V Zeners and you're in business!

The numbers in the diagram -- +15V and -15V, coming out of a single 24V supply -- are not going to work with linear regulators. You need non-linear converters, such as switching/PWM regulators, to create a greater total potential than you get from the single voltage source. If you look at what Grumpy Mike said, it's exactly that. So, it sounds like you're hearing something differently from what people are actually saying, and then deciding to take offense at perfectly valid engineering advise. In the end, he who loses out is you, because that's not a way to be successful in life, but if that's really up to you to determine.

18V rail splitter.png

Criticism of your brainchild is not criticism of you. And while it's natural to be protective, that can be counter-productive. It's important to realize that criticism means someone is listening. What's more, they apparently have some grasp about how your idea functions. These are big steps. Don't tune them out. By illuminating the weak points you have the opportunity to strengthen your design. Many shortcomings that are obvious to outsiders are invisible to the creator. -- Gerard Fonte elektor magazine 05/2012

joseph_m:
The gray meter in the back is the supply voltage. The meters in the front shows the output of the 7809 regulator and the output of the 7909 regulator (the 7909 is the negative value). Now I probably should throw some caps at the circuit but this shows it works.

I suggest you haven't connected the black leads of both front meters to your GND. Until you do that and get +/-9V I am sceptical to the max.

Just think if this circuit did work it would be the ultimate in low drop out regulators. That is you can regulate right down to zero volts drop across the regulators.
I wonder why no one has done that before ... hum ..... :~

Less attitude and more facts would be appreciated. If it works we will all be enlightened with no need to gloat. If not there will be less embarrassment.

My experience tells me that you are seeing leakage current through the regulators, but a few milli-amps of load will cause the negative voltage to disappear, however, if I am wrong I would really like to know. I will have to go back to the old reference books and review the internal design of linear regulators.

I must admit that it does appear to work - but only because the regulators happen to be half voltage of the 18V supply. Well, apart from the fact that both outputs surely must be +ve

The 7809 is regulating correctly, producing 9V - the remaining 9v are dropped across the (incorrectly configured) 7909.

If the supply voltage was for instance 24V, the outputs would be 9V and 15V.

John

Did you buy 2 extra meters just for this experiment?