Isolated, regulated, toroidal step down transformer AC power supply design.

[Bob Parker]

I fact the easiest way to get rid of me is just give me a good solution.

A good solution is to fix the problem *you've* created by replacing
perfectly good capacitors and disturbing the circuitry, not to try and
fix the symptoms of the fault *you* have created.
But that concept is far beyond your limited intellect and I think
we're all getting tired of this thread. I certainly am.

 

[Trevor Wilson]

Anyone who is genuinely mystified by my belligerence, either:

a) Didn't read or understand the thread and the chronology of its
development.

**Nope. That ain't it. I've been here since the start. I asked you what the
application was. I also asked you what Voltages you had measured on the
regulator inputs. So far, no answers.

b) Doesn't understand the basic rules and principles of equality, or
understands them but does not apply them in newsgroups and believes that
different criteria or rules should be applied.

**Here's how it works in the real world:

* You ask a seemingly reasonable question (from your, very un-knowledgable
viewpoint).
* You are asked to supply more information, so WE (the knowledgable) can
decide if you request is reasonable or not.
* After we are supplied the information (which includes Voltage
measurements), then a better solution can be supplied.

In which case, it would be helpful if you explained the basis on which you
discriminate
between my belligerence and Phils (or anyone else's.)

**I'm not discriminating between yours and anyone else's belligerence. I am
simply stating fact: You are acting belligerently. And, in several annoying
ways:

* You persist in top posting. Think of it in terms of a normal convesation.
One person makes a comment and the other person responds, AFTER the original
comment. Top posting reverses that convention. A convention which us humans
have dealt with for many thousands of years. Top posters seem unable to
understnad this uttery basic fact of life.
* Despite repeated requests, you STILL refuse to supply basic information
about the device. You tell us what speakers you use, what capacitors you
use, you even tell us about your family life, yet you STILL refuse to supply
the fundamental information required for us to give you a sensible answer.
* Despite repeated suggestions that we have superior solutions, you continue
to return to your own, quite insane one.
* Depsite repeated advice that your regulators are unlikely to be faulty,
you persist in insisting that they MUST be faulty, when it is more likely
that something else, far more prosaic is obviously wrong (that the
regulators simply do not have enough input Voltage).

 

[jasen]

I am not compelled to explain anything more than what I already have and
what I have explained is more than sufficient for my purpose.

Neither are we compelled to answer you civily,

This isn't rocket science. A minimal design AC source can be achieved with
only ONE component but the output is then at the mercy of the vagaries of
the input voltage.
yup.

A better design might involve transformation to a nominal VAC (+/- x%) ,
then rectification and regulation to a nominal DC (+/- y%<x) , filtering and
decoupling and then inversion back to 9 VAC (+/- 1%).

Even Google'sperts should be able to understand that.

yeah, if that's what is needed, usually it isn't,
as a result such supplies aren't common

Bye.
Jasen

 

[jasen]

Good boy, Trevor. At least you grasped that much. An SMPS is not an option
or I would have simply purchased one.

The reason WHY I need 9 VAC is because the application needs it and for
reasons that I have already explained.

I may post a schematic of the applications rectification, regulation and
filtering circuit, but I don't think this is necessary.

Most people can understand a statement like "240 VAC in, 9 VAC out",
regulated to a fine tolerance (Say 1%) and isolated from the toroidal step
down transformer.

How do people with city power run this device?

Have you tried using an online UPS, or active line conditioner, that has
sine-wave output, (this won't be cheap) and then connect the ordinary
powersupply.

Bye.
Jasen

 

[swanny]

Thanks Terry. Now I'm thinking about how to make a meter which shows
musicality in capacitors. Maybe do a frequency sweep to see how it
performs at the bass vs midrange vs treble ends of the spectrum? ;-)

Bob

There seems to be a trend to put clear cases on everything these days... Maybe
if you can shine a light through it you can measure its 'transparency'?

 

[jasen]

I really don't care. I'll design my own solution. I have no doubt it will
not be optimal. I'm not even sure its going to work, but **** this for a
joke.

well, you atleast know the whole story, there's a chance that someone
elee here has encountered a situation like yours and fixed it in a
totally different way. If you want to keep secrets that's your problem.

We'll all just have to assume it's to power the irrigation controller
for your attic hydroponics setup, os some monitoring device for a meth
lab in your garage

FWIW the 50W subwoofer amplifier sounds like a good approach if you need
regulated AC, maybe you can get one with a damaged speaker cheap on ebay.

for the signal generator maybe a Wein-bridge oscilator, or DDS.

Bye.
Jasen

 

[Trevor Wilson]

Give up Trevor. Even if he coughed up the info, he wouldnt understand the
suggested fix. Although its hilarious that he cant measure ripple current
but wants to build a 1% AC regulator.

and its more pig-headed than belligerent behaviour - one can be
belligerent and right, which he is not.

**Exellent point.

I recommend a 2kVA true-sinusoidal UPS to fix the problem. but use monster
cables to connect the 9Vac, and ensure the mains plug is gold plated.
Hopefully whilst mark is earning the money to pay for this stuff, he will
stop posting (top or bottom)

some people should not be allowed soldering irons. or keyboards.

**Oh yeah. One of my mates wanted to install silver wire in his amplifier,
CD player or something. I asked if he could solder and he assured me that he
could. After the stuff stopped working, I told him to bring it down. YIKES!
He was using acid flux paste to solder with. The corrosion was extensive.
God knows what kind of iron he had. I gave him one of my old temperature
controlled things and a roll of Multicore solder and an hour's worth of
instruction. I haven't seen any of his gear back on the bench, so I guess
all is now OK.

 

[swanny]

Yes it is.

Because I believed the voltage of the AC domestic supply was straying
outside the operating range of the existing regs (after transformation and
voltage doubling rectification) and because I wanted to replace an existing
wall-wart AC power supply with a regulated AC power supply and because the
application displays errant behaviour during times of over voltage supply.

Once again, your assumptions appear incorrect. The AC mains moves around a bit,
usually to within +/-10%. Manufacturers know this and product is designed to
handle it.

You have a little 9VAC wall wart transformer and these small transformers
usually have poor load regulation so its output when lightly loaded is probably
around 15VAC.

You don't have a voltage doubler, you have a pair of half-wave rectifiers which
give you a bipolar DC supply which is unregulated. The unregulated DC supply is
a good place to put those super caps, so as to smooth out the ripple in the
supply to the regulators.

The regulators will accept a minimum of around +/-17VDC to regulate properly.
With your wall wart, you probably have about 20VDC on their inputs so it works
OK, until you load it down a bit. Then the wall wart AC output will drop and the
inputs to the regulators will drop below 17V. Now your regulators will not be
able to regulate. Your mystery circuit may not like this very much.

If you replace your wall wart with a better 9VAC transformer then your
regulators may have an input below 17VDC most of the time and fail to regulate.
You should probably replace the wall wart with a 12-15VAC transformer with good
load regulation specs. Then your regulators will be happy with their input
voltage and regulate all the time.

However, you must be aware that linear regulators get hot if they have a large
voltage drop across them when under load, so you may need to make sure they are
mounted to a heatsink to dissipate this heat.

All of this has been spelt out to you many times here by many people. Is there
anything here you don't follow?

 

[Poxy]

Once again, your assumptions appear incorrect. The AC mains moves
around a bit, usually to within +/-10%. Manufacturers know this and
product is designed to handle it.

You have a little 9VAC wall wart transformer and these small
transformers usually have poor load regulation so its output when
lightly loaded is probably around 15VAC.

You don't have a voltage doubler, you have a pair of half-wave
rectifiers which give you a bipolar DC supply which is unregulated.
The unregulated DC supply is a good place to put those super caps, so
as to smooth out the ripple in the supply to the regulators.

The regulators will accept a minimum of around +/-17VDC to regulate
properly. With your wall wart, you probably have about 20VDC on their
inputs so it works OK, until you load it down a bit. Then the wall
wart AC output will drop and the inputs to the regulators will drop
below 17V. Now your regulators will not be able to regulate. Your
mystery circuit may not like this very much.

If you replace your wall wart with a better 9VAC transformer then your
regulators may have an input below 17VDC most of the time and fail to
regulate. You should probably replace the wall wart with a 12-15VAC
transformer with good load regulation specs. Then your regulators
will be happy with their input voltage and regulate all the time.

However, you must be aware that linear regulators get hot if they
have a large voltage drop across them when under load, so you may
need to make sure they are mounted to a heatsink to dissipate this
heat.

All of this has been spelt out to you many times here by many people.
Is there anything here you don't follow?

I think you'll find that you've offended Mark in many, many ways. What you
think is logical, sensible advice based on the operating principles of the
components involved is in fact a bitter, unjustified attack on him.

He's TOLD you what you need to know to help him, so stop crapping on about
irrelvant thigs like "voltages" and "drop outs" and "specifications" and
so-called "facts" and start agreeing that he's clearly suffering from a bad
pair (!) of regulators and the one dual-rail power supply design in the
world that only functions correctly with an insanely complex thing that
generates regulated AC.

 

[Poxy]

well, you atleast know the whole story, there's a chance that someone
elee here has encountered a situation like yours and fixed it in a
totally different way. If you want to keep secrets that's your
problem.

We'll all just have to assume it's to power the irrigation controller
for your attic hydroponics setup, os some monitoring device for a meth
lab in your garage

I don't quite see why an irrigation controller would need dual rails, but
some kind of sensor that required an instrumentation amplifier might need
it, and depending on how serious Mark is with his hydro setup, he might be
sensing a range of parameters such as temperatures - both ambient and
root-zone, CO2, TDS, PH and other nutrient attributes etc.

 

[jasen]

The way I understand it, the regulation that already occurs within the
application can only operate within a tolerance of the regulating IC's
nominal input voltage. The application is using 15 Volt DC regulators
(output), after the bridge rectifier, and can only operate (+/- x%) of 15
volts DC input.

you probably only need larger heatsinks on the regulators, higher voltager
reservoir capacitors and a slightly higher AC voltage input

If the domestic supply comes in at 264 VAC (+10%), the existing linear
regulator now supplies 9.9 VAC to the application, the bridge rectifier
doubles that to 19.8 VAC.

it does not.

I monitor the domestic supply and regularly see voltages coming into our
house outside the range of +/- 10%.

One solution, of course, would be to redesign the power supply within the
application's case, including upgrading the regulating IC's input voltage
tolerance to at least 40% of its output, if such an IC can be found!

here's another way.

.-||-+------------+------||--.
| | | |
| | o---+---o |
F1 | | _-~ ~-_ |
.-oxo-------+----|--+-~ : o---+---o : ~--+-----
| | | : | : |
--+-oxo-. ||| _____| | |~| | |~| |
F2 3|||(_ === |\| | |\| === out
"240V" 3|||(_ 36V | |_| | |_| | 240V +/- 9%
+/-24% 3|||(________|_________|__________|
--+-----' ||| ry1 ry2
| TR1  "+15%" "-15%"
`---------------------------------------------

activating relay 1 will boost the output voltage by 15% use it when the
input goes past 8% low relay 2 will reduce the output by 15%, use it when the
input goes past 8% high, the circuit to tetect over or undervoltage and
switch on the apropriate relay is not shown.

the relays should be rated to handle switching 48VAC and have atleast 1000V
isolation, the fuses ( --oxo-- in the diagram ) sdhould be rated
apropriately for thr transformer, the capacitors should be non-polarised
about 1uF seems a good size.

a circuit not completely unlike the above is used at the power sub-station
to regulate your power, but having it closer to the appliance means fewer
compromises need to be made.

 

[jasen]

Three things:

1) "7815" gives about 4 million hits on Goggle.

Try again. Hint: he's not calling you "datasheet"

2) I asked for the help a little over 48 hours ago, and since that time I
have had to suffer about 30 attacks, but NO HELP what-so-ever.

You aren't cooperating with those who attempt to help.

3) I have a family and one particular dependant who is bed ridden.

That's not a licence to be an asshole.

What did you expect from me?

maybe you could cal;m down a little and look at it from our perspective

Bye.
Jasen

 

[jasen]

Either way, the cheapest solution is sounding like I should exchange the
7815 /7915 pair, for a pair that DOES have that 30V spec.

7815s are good to 35V, do you want something worse?

and in so doing, eliminate the possible cause of the observed clipping
during voltage overloads.

How did you observe this? What do you mean by "clipping"?
A regulated DC power 'rail' is supposed to be a flat line on the scope
all the time...

Bye.
Jasen

 

[jasen]

Duh!

I've NEVER claimed to have any electronics expertise, what-so-ever. Never.

So you aren't the mark who attemped to mock Phil over some misunderstandibng
of your voltage doubler?

you must be an imposter then.

**** off.

 

[jasen]

so just where does one place the magnetic shunt in a toroidal core for a
CVT?

(thinks....)

Between the primary and secondary windings - same as an EI

(I'm guessing that'd mean a layer of GOSS between the primary and secondary
windings )

So a cross-section parallel to the windings would go, radially,
GOSS core,
insulation,
primary,
insulation,
GOSS shunt
insulation,
secondary

besides, most of the people who used to design these have died.

that could be a problem.

Does an R core really regulate much better than a toroidial, they still have
the primary wound on top of the secondary?

Bye.
Jasen

 

[Bob Parker]

How do people with city power run this device?

Have you tried using an online UPS, or active line conditioner, that has
sine-wave output, (this won't be cheap) and then connect the ordinary
powersupply.

Bye.
Jasen

He wants everyone to keep playing his infantile guessing game so he
can remain the centre of attention.
How many commercially manufactured mains-powered electronic devices
are critically sensitive to normal mains voltage variations, requiring
precise regulation of the input voltage? About none that I know of. Even
UPSs have to see a pretty big variation before they cut in.
He's had this problem since he went through this audiophile thing,
ripped out most of the electrolytic caps and replaced them with
audiophile ones. Remember his comment about us being too dull to ask him
if he'd made any bad solder joints?
He's made a pig's breakfast of what was a normally functioning CD
player or whatever it is, and now wants some loony solution to mask a
fault *he* caused.
And no Mark, we're not going to keep trying to guess what it is. My
suspicion is that the reason you don't want to tell is because it's some
audiophile thing you know Phil will laugh his head off about, and you're
scared of Phil from your experience in the aus.hi-fi newsgroup.
Here endeth today's sermon....

 

[Trevor Wilson]

(thinks....)

Between the primary and secondary windings - same as an EI

(I'm guessing that'd mean a layer of GOSS between the primary and
secondary
windings )

So a cross-section parallel to the windings would go, radially,
GOSS core,
insulation,
primary,
insulation,
GOSS shunt
insulation,
secondary


that could be a problem.

Does an R core really regulate much better than a toroidial,

**Nope.

they still have

the primary wound on top of the secondary?

**Nope. That is what makes them excellent for low level audio applications.

 

[Franc Zabkar]

I am no expert, Franc, but I don't think so.

Doesn't look like the one I am looking at (but yours is very difficult to
read).

ASCII art needs to be viewed with a fixed width font, eg Courier. Is
that your problem?

This was the original format:
http://groups.google.com/group/aus.electronics/msg/8b6aca757c02f9aa?dmode=source&hl=en

Anyway, I know that the one I am looking at takes in 9 VAC and outputs +/-
15VDC (after regulation) and looks like two half wave rectifiers with diodes
pointing in opposite directions.

h

- Franc Zabkar

 

[Phil Allison]

"Trevor Wilson"
"jasen"

**Nope. That is what makes them excellent for low level audio
applications.

** Many of then DO have the secondary wound directly over the primary.

Only way to achieve the very low external mag fields.

Not so good for safety, though.



........ Phil

 

[Trevor Wilson]

"Trevor Wilson"
"jasen"





** Many of then DO have the secondary wound directly over the primary.

**I stand corrected. Some are so configured.