+15v and -15v regulated from 9V battery

[anagrama]

Is it possible to get the sort of +15/-15V regulated power supply required for most synth circuits from a 9V  battery?  From looking at RG Keen's +9V *and* -9V from one battery and One 9V battery gives +18, +25, +33V  circuits I'd imagine it would be, but unfortunately, as a beginner, I don't posses the knowledge necessary to convert them into providing +15V/-15V regulated.  Can anyone tell me how to do this, or point me towards a suitable schematic?  Oh, and what does a "regulated" power supply actually mean?

[ExpAnonColin]

1) Wire up the 33v circuit
2) Wire a voltage divider (10k from the 33v, 10k to ground, 10uf cap to ground with the negative side to ground)
3) Use the output of this as ground in your circuit, the 33v as 16.5v+ and your real ground as -16.5v.  It's close.

-Colin

[anagrama]

thanks, that's a lot simpler than I thought it would be
...but
is it regulated? (whatever that means)

[ExpAnonColin]

Actually, an even better way to do that (on second thought) would be to feed the 16.5 volt output into a 15v regulator, which they do sell... then it would be 15v and regulated.  You'd be using the output from the divider as ground, remember.  Get a positive and a negative regulator, use the divider's output as ground, and then put ground through the input and at the output you'd have a good, regulated 15v out of one and a -15v out of the other.  I warn you I've never done this, can anyone verify?

-Colin

[Samuel]

if it's from a battery...i think so but I'm really not 100% sure.

regulated = delivers its nominal (listed) voltage (or close to it) regardless of how much current is being drawn. unregulated PSs can vary in voltage depending on load...

[niftydog]

There are problems associated with simple voltage divider networks like the one Colin suggested. Namely, the voltage lost through the resistor network (which is a problem in itself) varies with the load current. They can also dissiapte a large amount of energy, requiring highly rated resistors and generating a lot of heat.

Perhaps it would be better to wire up the +18VDC circuit and use a second MAX1044 chips to invert it to create a -18VDC.

Either way, you're looking at a large amount of wasted (dissipated) energy which will drain your battery quite quickly. Perhaps it's better again to use two batterys and two MAX1044s to get what you're after. But then you face the possibility of one battery draining faster than the other!

A regulator attempts to maintain a constant voltage despite variations in current. A typical linear regulator requires a couple of volts above it's rated voltage in order to do it's job. This means that to get a 15VDC regulator to work properly, you need to feed it at least 17VDC (dropout voltage of 2V for most 78XX series regulators)

Adding regulators to a battery powered circuit also increases the wastage of energy from the battery. Also, as the battery voltage drops (and it will quite quickly) the regulator may stop working.

It's possible, the question is "is it worth the effort?"!

[ExpAnonColin]

You have a good point-although if we were to use an adapter, this would be no biggy so long as it could release enough current.

-Colin

[Mark Hammer]

Every single one of those synth circuits you see running off a +/-12v or +/-15v supply operate under the assumption that there will be wads of current available, maybe 50-100ma per side.  Although it may well be possible, in principle, to produce a supply that would magically transform +9vdc into +/-15vdc, the process will invariably result in only a comparatively small amount of current being deliverable to the circuit for not very long.  Unless you have an uncle who can regularly supply you with cases of heavy-duty alkaline 9v batteries, your mission is one that strikes me as ultimately bound to disappoint.

Of course, if the goal is not to operate the circuits with said supply for very long, but merely to have nice hum-free supply so you can fire up a circuit, scope it for 20 seconds to see if it's working, then shut it off, you may have a possible outcome that will make you happy.

[anagrama]

ok thanks

I asked this because I was wondering whether it would be more practical to make a second power supply w/ +15v/-15v for any synth circuits in addition to using a 9v power supply for my normal pedals, or whether it would be better to stick w/ one power supply and an adaptor on each synth circuit (w/ an added bonus of these synth circuits being battery operatable). The battery thing is out then; is it also pointless to make an adpater for converting the 9v from a normal pedal power supply to 15v/+15v?

From what you've said is sounds like it would be less effort to just build a seperate -15v/+15v power supply, but I didn't want to go and do that if there was some simpler solution I didn't know of.  thanks again.

[niftydog]

From what you've said is sounds like it would be less effort to just build a seperate -15v/+15v power supply

well that's debateable. The issue here is practicallity. Like Mark put so eloquently (your mission is one that strikes me as ultimately bound to disappoint.), a power supply will save you loads of hassle and you won't have a need to dispose of a shipping container full of 9V batterys each month.

Also, why limit yourself? Who knows when you'll have the need for a beefy ­±15V PSU?!?!

[Hal]

I'm not sure how those MAX1044 chips work, but if they simply invert, would it be possible to just take a 15v wall-wart, and use that chip?  That seems pretty simple to me, and mouser has wall-warts (probably including 15v) for like $4.

[niftydog]

sure, it would work, but it's subject to the same issues that Mark and I have flagged. Limited current and high power drain.

It's all dependent on the actuall circuit one intends to power from such a supply. Some will work fine, others are far too demanding.

[Hal]

yea, but when working with wall warts, this matters less (not at all?) than when working with batteries.  And it would be really simple - something he wanted.

[niftydog]

power drain is less of an issue, but the 1044 is limited in how much current it can supply before ripple becomes an issue.

Operation is guaranteed from 1.5V to 10V with
no external diode over the full temperature range. They
deliver 10mA with a 0.5V output drop.