Underfed regulators

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Sounds like the answer is in the Subject line of your post: "underfed".

If you are using nominal 12V either side of center trafos they are at best marginal if they have very poor regulation and thus substantially more than 12VAC coming out at low load. If they were really spot-on 12V the peak voltage would only be about 17V and thus well below the necessary 20 or so to make the regulators happy.
 
OK....

You say the transfos you have are rated at 12VAC (and I'm assuming you are running a full wave bridge), and open circuit, they measure 16VAC.

Going back to the "spec" 12VAC, the the Best Case "raw" supply will be 12 x 1.414, or 16.969 VDC, whic is WELL below the drop out V of an 18 V regulator.

Assuming 16 VAC *under load*, then you have just a bit above 22.6 VDC with an infinitely large "filter cap" hanging off the recttifier.

A 7818 or 7918 requires at least 3V for "dropout" regulation, as I recall.

I also assume you don't have a 1000000000000000000000000 uFd cap hanging off the output of the rectifier.

I would guess you are "starving" the regulators.

In my experience, and as a rule of thumb, the tranny's VAC rating needs to equal the desired VDC output (after regulator) to accomodate for diode voltage drops and ripple on the filter cap.

Bri
 
[quote author="Brian Roth"]OK....

[snip]

In my experience, and as a rule of thumb, the tranny's VAC rating needs to equal the desired VDC output (after regulator) to accomodate for diode voltage drops and ripple on the filter cap.

Bri[/quote]

I agree that's a pretty good rule-of-thumb for voltages in this range. At higher voltages it can be relaxed a bit, and at lower ones it cuts it a little too close (for example a ~6V filament supply), since the diode drops and regulator differential take out a bigger fraction of the voltage.

For Fairchild's 7818 they rate the output under load to 1A with a 3V drop (21V in), although the Dropout Voltage as an input-output differential is given as 2.0V typical. For the first number with 21V in, the output is allowed to sag to as little as 17.1V, and can be as high as 18.9V.

[quote author="buttachunk"]Sorry I was unclear,, the traffos were rated 12VAC, dual secondary,,, they ended up around 16-0-16 with no load... I thought that there was a margin of error of like 3-4 volts under with regs...

Ended up designing and building a little charge-pump, which boosted up to +-20.5-ish volts (per rail). I could have fed the regs with this, but ended up bypassing them and went through 10r resistors instead. The rails are now at +18.3VDC and -18.4VDC under load, measured at the opamp pins,, running far quieter than the reg'd supplies....[/quote]

Linear regs can only reduce their input voltages in order to give one a regulated output voltage---all they are doing, after all, is putting the equivalent of a variable resistance in series with the raw supply. There are some varieties of switching regulators that can both boost and cut as required, but typically even for them it's just one or the other.

Most op amps are fairly tolerant of power supply fluctuations at least at low frequencies, although they can fall apart badly at higher. The reason to use regulators is often more one of providing very low ripple without a lot of passive filtering, rather than tight control of the d.c. output per se.

But in systems where d.c. levels, and not just their ratios, are important, regulators are essential. Many times better system design can make the power supply sensitivity a lot less, but decent global regulation may still be the sensible way to go.
 
> charge-pump, which boosted up to +-20.5-ish volts (per rail). I could have fed the regs with this

No. You have neglected to say what input caps or how much current, but in general you will NOT get clean 18V from raw "20.5-ish V". There will be a volt or more of ripple. Your DC voltmeter shows the average, what you need is the depth of the ripple. This may be 19.5V. 19.5V-18V= 1.5V. The 78xx series will not reliably regulate with just 1.5V to spare, they are "waste" regulators and need plenty of excess voltage to waste. What you really get is 18V dropping to 17V 12 times a second, which is really nasty sounding.

Brian's Rule "VAC=VDC" is good in this range. You want 18VAC. bcarso's footnotes to Brian's Rule are valid things to ponder: you can't get solid clean 6VDC from 6VAC without heroic windings, rects, caps, but 18VDC from 18VAC is well above this range. Given a 12VAC winding, you should have regulated to 12VDC at most. Observing 16VAC no-load, you might stretch this to 15VDC, but your loaded readings suggests that your 16VAC sags toward 12VAC very fast under load.

The regulation on 78xx is good enough that you could have wired the 12-0-12VAC as 24VAC and fed a 2-diode centertapped voltage doubler. This would give +/-36VDC under load, which is a bit high, but ensures the 7818 always has plenty of headroom to do its thing. The observed 16VAC could give +/-45VDC no-load, which is well over the 7818's input rating: I would expect nine out of ten 7818s to work long enough to order the 18-0-18VAC transformer you need.

> ended up bypassing them and went through 10r resistors instead. .., running far quieter than the reg'd supplies....

Which is what I keep saying: chip opamps do NOT need regulated supplies. A well laid out C-R-C pi filter with modern low-cost compact electrolytics will be clean enough for audio. 1000uFd-100Ω-1000uFd is a starting point; considering that 2200 and 4700uFd caps (or a 10-bag of 1000uFd) are not very big or expensive, you can go overboard without bulging the budget.
 
> My DMM is nimble enough to show slower warbles (from ripple) on wildy rails,,, the regulated power honestly looked pretty steady on the DMM (+-.1V)...

Ripple is 120Hz, not a wandering. DVMs won't show it; are often tuned to ignore it.

Try reading the AC of the DC, but many meters get confused when you try to do that. A 0.1uFd blocking cap may help.
 
[quote author="buttachunk"] [snip]
It was odd in this case that the regs were giving me +-18v with the 12VAC-rated transformers. My DMM is nimble enough to show slower warbles (from ripple) on wildy rails,,, the regulated power honestly looked pretty steady on the DMM (+-.1V)... That false sense of security could be very detrimental to these other DIY builders with noisy gear and stable-looking voltage rails.
[snip]

I'm beginning to see the darkside of regulators in audio applications--- is this example basically a simulation of brown-outs and line sag ?[/quote]

It's not odd. DMM's update typically at a few-per-second and are often attempting to suppress mains-frequency components for precisely the reason of making the display more stable. Under these conditions the best tool is a 'scope, or failing that switching to the a.c. range on the meter to see what that says. (ahh I see PRR has jumped in on this as well)

Power transformers, like everything else, are designed to a spec and a cost. Most of them have much higher unloaded/lightly-loaded voltages than their stated rating values. Sometimes this saves you in a rectifier/cap input filter situation if the loading is low compared to the rating.

A number of factors conspire to get you in real-world apps. One of the most pernicious, again best revealed by a 'scope, is when everybody else's power supplies are sucking down the peaks of the a.c. line charging their own capacitors---either directly, with a bridge rectifier directly connected to the line and feeding a big cap, or indirectly through their own power transformers feeding a similar circuit. The result is a flattening of the tops of the a.c. out of the outlet. Since your cap input supplies are counting on something near the canonical peak line voltage (sq root of 2, times nominal r.m.s.) this will cut into your margin for regulation even if everything else is in order.

PRR's point about passive filters is well-taken. The additional R between the two caps throws in much more attentuation of higher frequencies. These frequencies are there because of the nature of the cap charging waveform, as well as junk getting through from the power line. Contrast this to a regulator periodically dropping out on dips in the input voltage---some parts may handle it well and just pass the reduced input to the output as well as they can. But others may have fits as their feedback fights to stabilize and then has to recover when the input-output differential restores. A bigger output cap than the specified minimum on the regulator output will help a bit.

Semiconductor makers, mindful of the desire to reduce regulator losses and knowing that many people are marginal on their cost-driven designs, have pushed "low-dropout regulators". Typically the lion's share of the current goes through a PNP or P-channel FET for a positive regulator, since the drive signal needs only go negative with respect to the input voltage, and the part can operate with a much lower input-output drop compared to a more typical N device. However---the feedback stability margin issues are often a big problem. National actually shipped a part that would oscillate with both too "good" a capacitor on the output as well as too foul a one!
 

Latest posts

Back
Top