AC->DC converter question

[James]

Hi,

I have an elementary quesion regarding development of a simple AC to
DC converter. I was hoping that I could get some input on the
components that I've chosen before I go off an start buying
transformers.

The input is US 110-120VAC @60Hz. The output needs to be ~2.8VDC. I
plan to use an LM317T for regulation after the rectifier. The
application will source < 300mA at this voltage. My question is
regarding the transformer and bridge rectifier. Given the overhead
voltage requirements of the linear regulator (~2V), and the
possibility that I may want 3.3V or 5V DC at some point (to drive a
MAX233A, for instance). I was considering this transformer:

http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=221292&pa=221292PS
(sorry, I don't know where to get a datasheet for this one), and a
KBP04M Bridge Rectifier (which I have on hand), the datasheet for
which can be found here:

http://www.ortodoxism.ro/datasheets/GeneralSemiconductor/mXrruqz.pdf

If anyone has time to look over these components or suggest an
alternative to what I've mentioned, I would appreciate the input.
Anything I can buy from Mouser, Digikey, Jameco, etc will work. Sorry
for the newbie question. I didn't think a schematic was necessary, but
I can convey this if it helps.

James

 

[D from BC]

Hi,

I have an elementary quesion regarding development of a simple AC to
DC converter. I was hoping that I could get some input on the
components that I've chosen before I go off an start buying
transformers.

The input is US 110-120VAC @60Hz. The output needs to be ~2.8VDC. I
plan to use an LM317T for regulation after the rectifier. The
application will source < 300mA at this voltage. My question is
regarding the transformer and bridge rectifier. Given the overhead
voltage requirements of the linear regulator (~2V), and the
possibility that I may want 3.3V or 5V DC at some point (to drive a
MAX233A, for instance). I was considering this transformer:

http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=221292&pa=221292PS
(sorry, I don't know where to get a datasheet for this one), and a
KBP04M Bridge Rectifier (which I have on hand), the datasheet for
which can be found here:

http://www.ortodoxism.ro/datasheets/GeneralSemiconductor/mXrruqz.pdf

If anyone has time to look over these components or suggest an
alternative to what I've mentioned, I would appreciate the input.
Anything I can buy from Mouser, Digikey, Jameco, etc will work. Sorry
for the newbie question. I didn't think a schematic was necessary, but
I can convey this if it helps.

James

Vsec too high then LM317 too hot.
Vsec too low then LM317 goes into drop out.

Take care with ripple, transformer loading and rectification drop.


D from BC

 

[John Popelish]

Hi,

I have an elementary quesion regarding development of a simple AC to
DC converter. I was hoping that I could get some input on the
components that I've chosen before I go off an start buying
transformers.

The input is US 110-120VAC @60Hz. The output needs to be ~2.8VDC. I
plan to use an LM317T for regulation after the rectifier. The
application will source < 300mA at this voltage. My question is
regarding the transformer and bridge rectifier. Given the overhead
voltage requirements of the linear regulator (~2V), and the
possibility that I may want 3.3V or 5V DC at some point (to drive a
MAX233A, for instance). I was considering this transformer:

http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=221292&pa=221292PS
(sorry, I don't know where to get a datasheet for this one), and a
KBP04M Bridge Rectifier (which I have on hand), the datasheet for
which can be found here:

http://www.ortodoxism.ro/datasheets/GeneralSemiconductor/mXrruqz.pdf

If anyone has time to look over these components or suggest an
alternative to what I've mentioned, I would appreciate the input.
Anything I can buy from Mouser, Digikey, Jameco, etc will work. Sorry
for the newbie question. I didn't think a schematic was necessary, but
I can convey this if it helps.

I think I would use a transformer with a center tapped
secondary and use two diodes in a half bridge (perhaps
Schottkys like 1N5817 or SR102 to minimize voltage loss).
Then if you decide to raise the voltage to 5 volts, you can
change to a full bridge and double the raw DC voltage. A
6.3 VCT secondary would put out a peak voltage of about 4
volts with the half bridge. This is a bit low, unless you
get a transformer with a very conservative current rating
(say, more than 1 ampere), so you may need to go with a 10
volt CT or dual 5 volt windings that you can series as a
center tapped winding.

 

[Tam/WB2TT]

I think I would use a transformer with a center tapped secondary and use
two diodes in a half bridge (perhaps Schottkys like 1N5817 or SR102 to
minimize voltage loss). Then if you decide to raise the voltage to 5
volts, you can change to a full bridge and double the raw DC voltage. A
6.3 VCT secondary would put out a peak voltage of about 4 volts with the
half bridge. This is a bit low, unless you get a transformer with a very
conservative current rating (say, more than 1 ampere), so you may need to
go with a 10 volt CT or dual 5 volt windings that you can series as a
center tapped winding.

There is also a way of getting two voltages out of a center tapped
transformer without extra parts. Put a bridge rectifier across the whole
winding, with the negative side of the bridge output grounded. There will
now be rectified DC at 1/2 the voltage at the center tap. The voltage at the
center tap will only be one diode drop down. Add the usual capacitor at the
bridge output, and another from the center tap to ground. This works because
there is no AC voltage at the center tap. To get 2.8V from 4V would probably
require an LDO, or a discrete circuit with a PNP pass transistor. I have
done this with a 12.6VCT transformer to get +12 and +5.

Tam

 

[James]

There is also a way of getting two voltages out of a center tapped
transformer without extra parts. Put a bridge rectifier across the whole
winding, with the negative side of the bridge output grounded. There will
now be rectified DC at 1/2 the voltage at the center tap. The voltage at the
center tap will only be one diode drop down. Add the usual capacitor at the
bridge output, and another from the center tap to ground. This works because
there is no AC voltage at the center tap. To get 2.8V from 4V would probably
require an LDO, or a discrete circuit with a PNP pass transistor. I have
done this with a 12.6VCT transformer to get +12 and +5.

Tam- Hide quoted text -

- Show quoted text -

Thanks for the detailed feedback, I appreciate it. Would it be
considered wasteful to regulate 12.6V down to 2.8-5V using an
adjustable linear regulator? I could certainly use something like
this:
http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=210956&pa=210956PS

Also, with respect to using both outputs in parallel, would this
require two rectifier circuits or just one with the primary and
secondary in parallel ahead of the rectifier? I would also like to
figure out how to determine approximate load on the transformer when
the DC load is sourcing (I think that's the way to say it) 200mA
through a linear regulator. I'm sure the datasheet on the LM317T
provides information on losses, etc. Is this easy to sort out?

Once again, thanks for your patience!

 

[ehsjr]

Hi,

I have an elementary quesion regarding development of a simple AC to
DC converter. I was hoping that I could get some input on the
components that I've chosen before I go off an start buying
transformers.

The input is US 110-120VAC @60Hz. The output needs to be ~2.8VDC. I
plan to use an LM317T for regulation after the rectifier. The
application will source < 300mA at this voltage. My question is
regarding the transformer and bridge rectifier. Given the overhead
voltage requirements of the linear regulator (~2V), and the
possibility that I may want 3.3V or 5V DC at some point (to drive a
MAX233A, for instance). I was considering this transformer:

http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=221292&pa=221292PS
(sorry, I don't know where to get a datasheet for this one), and a
KBP04M Bridge Rectifier (which I have on hand), the datasheet for
which can be found here:

http://www.ortodoxism.ro/datasheets/GeneralSemiconductor/mXrruqz.pdf

If anyone has time to look over these components or suggest an
alternative to what I've mentioned, I would appreciate the input.
Anything I can buy from Mouser, Digikey, Jameco, etc will work. Sorry
for the newbie question. I didn't think a schematic was necessary, but
I can convey this if it helps.

James

Here's something better, at about 1/2 the cost:
MPJA stock # 12254 PD is a 9 V DC at 1 amp
wall wart for $3.50. That's enough current for
all three regulators so you can have 2.8, 3.3 and
5 volts simultaneously. Use a 7805 for the 5 volts,
and a 317 for the 2.8 and 3.3 volt supplies. The
Jameco transformer would be marginal for the 5 volt
supply, and would require a rectifier and filter cap,
and power only one regulator at a time, while costing
close to twice the MPJA DC supply.

I'd drop the voltage between the wall wart and the
regulator chip with a 5 watt resistor so that you
have close to 3 volts headroom at the input of the regulator
when the current is 300 mA. This will keep the maximum heat
in the regulator to roughly .9 watts. To figure the value of
the resistor, measure the output from the wall wart with a
300 mA load attached, then subtract 3 volts for headroom, and
then subtract the desired regulator output voltage. Divide the
result by .3 to get the value in ohms, and select the closest
standard resistance value equal to or lower than your figure.

For example, say the wall wart output with a 300 mA load is 15
volts, and you want 2.8 volts: 15-3 = 12, then 12-2.8 = 9.2,
then 9.2/.3 = 30.6666 ohms. So you would use a 30 ohm, 5 watt
resistor. At 300 mA, it would drop the voltage at the input of
the regulator to 6 volts, leaving 3.2 volts to be dropped in the
regulator. At 300 mA, that would be about .96 watts

Shown below is a diagram of all three regulators, based
on an assumed 15 volts output from the wall wart with
a 300 mA load.


------
| Wall |---------+---[22R]-------+---[7805]---+---> +5
| Wart |---+ | | | |
------ | | [.33uF] | [.1uF]
| | | | |
Gnd | +------+-----+
| |
| Gnd
|
+---[27R]--+---[317]-------+---> +3.291
| | | |
| [.1uF] +-[240R]--+
| | |
| | [392R]
| | |
| +-----+
| |
| Gnd
|
+---[30R]--+---[317]-------+---> +2.781
| | |
[.1uF] +-[240R]--+
| |
| [294R]
| |
+-----+
|
Gnd

Since power dissipation in the regulators will be about
1 watt worst case, they need to be installed on heat sinks.
The power resistors - 22 ohms, 27 ohms and 30 ohms need
to be 5 watts each. Those values will need to be re-computed
if the wall wart output, when supplying a 300 mA load, is not
15 volts.

Ed

 

[D from BC]

Thanks for the detailed feedback, I appreciate it. Would it be
considered wasteful to regulate 12.6V down to 2.8-5V using an
adjustable linear regulator? I could certainly use something like
this:
http://www.jameco.com/webapp/wcs/st...&catalogId=10001&productId=210956&pa=210956PS

Also, with respect to using both outputs in parallel, would this
require two rectifier circuits or just one with the primary and
secondary in parallel ahead of the rectifier? I would also like to
figure out how to determine approximate load on the transformer when
the DC load is sourcing (I think that's the way to say it) 200mA
through a linear regulator. I'm sure the datasheet on the LM317T
provides information on losses, etc. Is this easy to sort out?

Once again, thanks for your patience!

Some current is for LM317 control...the rest goes through the load.
For a 200mA Iout, the LM317 ADJ current is too insignificant.

Most often on transformer datasheets Vsec is spec'd at a load current.
With lighter loads, Vsec will be higher. At no load, I think 20% to
25% higher is typical.
On the odd occasion I've seen a V(I) graph on transformer datasheets.


D from BC