Maker Pro
Maker Pro

tiny 2.3VA 115VAC 60Hz transformer will get powered from 240VAC 200Hz

P

positivebalance41m

I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).

Since I only need an output of constant 10mA from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.

Unfortunately in trying to find a 220VAC transformer, the cost goes from
US$2.00 to US$7.00 so hence my motivation to be "unconventional" on this.

I know my transfomer losses will be higher at 200Hz compared to 60Hz but I'm
not concerned with efficiency as much as I am concerned with low cost and
reliable functioning.

Can anyone suggest the correct TVS voltage rating? Will the 200Hz pose any
specific problems for the 60Hz designed transformer that could be an
unpleasent surprise? I thought instead of a series TVS, to have a series
capacitor or even a resistor but the 220VAC frequency changes by about 30%
so any R/C ideas to my mind appeared to be too frequency sensitive...but
maybe I miss some clever ideas, component(s), or confirgurations. Perhaps a
shunt TVS and a series capacitor would ensure the transformer primary sees a
more constant 115VAC. I am guessing that on the primary side of the
transformer, there will only be about 1.2 to 1.8mA of ac current. That is
because I will have a fairly constant 12.6VAC at 10mA out of the secondary,
and I assume a few percent will go to core laminations magnetization
current.

Comments anyone?

THANKS :)

~TM
 
I

Ian Stirling

positivebalance41m said:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.
Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).
Since I only need an output of constant 10mA from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.
Unfortunately in trying to find a 220VAC transformer, the cost goes from
US$2.00 to US$7.00 so hence my motivation to be "unconventional" on this.

I think you need to look at your vendor.
The first one I looked at was http://www.rswww.com/

Has 240V primary transformers at $4 or so.
I know my transfomer losses will be higher at 200Hz compared to 60Hz but I'm
not concerned with efficiency as much as I am concerned with low cost and
reliable functioning.
Comments anyone?

I'd get the most detailed spec sheet that the manufacturer gives, and
consider if it might be an idea to run the 110VAC transformer at 220V.
The insulation should be rated considerably above this.

If you double the frequency of the input, to 120Hz, the transformer
won't saturate until double its rated voltage.
The iron losses will be a tad higher, but the copper losses lower due to
the lower volume of copper.

If you only need a watt, it should all work fine.

Do check that half-wave rectification doesn't cause increased current
in the primary.
 
J

John Woodgate

I read in sci.electronics.design that positivebalance41m
com>) about 'tiny 2.3VA 115VAC 60Hz transformer will get powered from
240VAC 200Hz', on Fri, 24 Oct 2003:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

NO, DON'T use half-wave rectification, because the d.c. load current
flows though the transformer secondary winding and may cause saturation
problems. Also, the r.m.s secondary current will be about 2.5 6times the
d.c. load current. Use a bridge rectifier - no residual d.c. in the
transformer and the r.m.s. current is only about 1.6 times the d.c.
output current.
Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).

You don't need anything in series. A transformer suitable for 115 V
input at 60 Hz will accept 115 x 170/60 = 326 V input at 170 Hz. But, of
course, if you put 220 V in, you will get 220/115 times the specified
output voltage. You may be able to get a transformer that is spec'd at
half the output voltage you actually require, i.e. 6.0 or 6.3 V.
Since I only need an output of constant 10mA

I suppose you mean that is the d.c. output current; with a bridge
rectifier the r.m.s. secondary current will be about 16 mA.
from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.

12.6 V x 16 mA = 202 mVA. You are hardly loading the transformer
Unfortunately in trying to find a 220VAC transformer, the cost goes from
US$2.00 to US$7.00 so hence my motivation to be "unconventional" on this.

I know my transfomer losses will be higher at 200Hz compared to 60Hz but I'm
not concerned with efficiency as much as I am concerned with low cost and
reliable functioning.

The core loss would be up by a factor of about (170/60)^1.8 if you put
326 V in, but at 220 V, it won't be increased by much, if anything. See
how hot it gets (compare with another sample running at 115 V 60 Hz with
a comparable secondary load VA).
 
S

Spehro Pefhany

I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.
<snippage>

As JW (Lin2 Men2 Zhong1) said, I think you can use a transformer like
that one with 220VAC in. I don't think you'll have any trouble with
half-wave rectification either at this power level (just use the CT to
one side and leave the other half of the winding unconnected). Or full
wave rectify it with two diodes and CT and halve the filter cap. The
half wave will give you a bit more voltage margin. You'll be using
about 15mA DC out of the filter cap (maybe 220uF/25V).

Caution that some of Mouser's transformers (including that one) are
not UL (or any other safety agency) approved, so do assure yourself it
won't cause a fire or whatever.

Best regards,
Spehro Pefhany
 
N

N. Thornton

Ian Stirling said:
I'd get the most detailed spec sheet that the manufacturer gives, and
consider if it might be an idea to run the 110VAC transformer at 220V.
The insulation should be rated considerably above this.

If you double the frequency of the input, to 120Hz, the transformer
won't saturate until double its rated voltage.
The iron losses will be a tad higher, but the copper losses lower due to
the lower volume of copper.



Sounds good, just one question. I dont know what regs/approvals you
have to comply with wherever you are, but you might find that the
transformer is approved for 110 use but the approval doesnt cover 220v
200Hz use. So I'd look into that aspect more.


Regards, NT
 
F

Fritz Schlunder

positivebalance41m said:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).

Since I only need an output of constant 10mA from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.

Unfortunately in trying to find a 220VAC transformer, the cost goes from
US$2.00 to US$7.00 so hence my motivation to be "unconventional" on this.

I know my transfomer losses will be higher at 200Hz compared to 60Hz but I'm
not concerned with efficiency as much as I am concerned with low cost and
reliable functioning.

Can anyone suggest the correct TVS voltage rating? Will the 200Hz pose any
specific problems for the 60Hz designed transformer that could be an
unpleasent surprise? I thought instead of a series TVS, to have a series
capacitor or even a resistor but the 220VAC frequency changes by about 30%
so any R/C ideas to my mind appeared to be too frequency sensitive...but
maybe I miss some clever ideas, component(s), or confirgurations. Perhaps a
shunt TVS and a series capacitor would ensure the transformer primary sees a
more constant 115VAC. I am guessing that on the primary side of the
transformer, there will only be about 1.2 to 1.8mA of ac current. That is
because I will have a fairly constant 12.6VAC at 10mA out of the secondary,
and I assume a few percent will go to core laminations magnetization
current.

Comments anyone?

THANKS :)

~TM



There are a number of reasons why I would not recommend what you are
proposing. The first problem is transformer primary insulation. It was
only intended to handle ~160V peak voltage accross the winding (for 115V
operation). Although it will most likely handle double this with little
trouble, at least in a one off case and for a limited test trial, running it
well beyond its voltage rating is asking for trouble when your specific
goals are reliability and large scale manufacturing. The cost of a product
recall/excess failures in the field/fires would be extremely high.

A bidirectional TVS diode isn't the best choice for reducing the voltage
while maintaining good reliability. A 160V TVS would probably be able to
keep the peak winding voltage within specification. However, 1.2-1.8mA of
primary side current is probably much smaller than it really would be. Low
output power rated line frequency transformers usually have very large
magnetizing currents compared to the reflected load current. Mouser
provides specification sheets (primative but it has much of the information
on it) for many of their transformers. I'm guessing you are considering
using some device like the 41PG006 since it is approximately what you are
describing.

http://www.mouser.com/catalog/specsheets/114109.pdf

The device is rated at 12.6V CT @0.06A @60Hz. The reflected primary current
should therefore be something in the vicinity of 8mA at full load. This
number isn't too large, and would be smaller if it wasn't fully loaded.
Notice however in the specification page it indicates a primary exciting
current of 20mA. As you can see the primary exiting current is even larger
than the fully loaded reflected load current by a factor of more than two.

So the question is, how much power must your TVS diode dissipate? Making
this calculation is not trivial. You would need to calculate the
transformer's magnetizing inductance (2*pi*f*L=Xinduct=PriVoltage/Excitation
Current). From this information the easiest route to go would be to wip out
some SPICE (not drugs, the computer software package) and setup a 220V 170Hz
AC source connected up to a 160V TVS or whatnot in series with an inductor
of the same value you found from the above calculation. Run the simulation
and experiment with it to determine your power dissipation. From the
results you would also need to add on the dissipation contribution from the
reflected load current. I suspect when you get all done with everything you
will find your power dissipation is something perhaps in the vicinity of or
greater than 1W, but much higher for larger transformers.

By far and away the best course of action for high reliability on a mass
production scale would likely be to use a 220V rated transformer to begin
with. Find a cheaper transformer. The Mouser 41FW300 might be a better
choice.

http://www.mouser.com/catalog/specsheets/114136.pdf

It has a whole bunch of taps on it and is quite versatile while still being
cheap ($2.80 single, $1.95 @ 100+). Ironically Mouser's price for the exact
same part is higher than my local electronics parts shop charges (Circuit
Specialists aka Web-Tronics). You can purchase it online from their store
instead of Mouser:

http://www.web-tronics.com/passive-electronic-components-transformers-multi-
trapped-power-transformers.html

I'm holding one of these in my hand right now. It is kind of a rats nest
with all the wires hanging off of it, but I guess that is the price you pay
for a one size fits all part.
 
W

Walter Harley

positivebalance41m said:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.


Why don't you just use a resistor and a zener diode off the 340VDC supply?
Is there some sort of isolation requirement you're not telling us about?

Use a 22k series resistor. The resistor will be dropping 330V at (say)
15mA, so it would need to dissipate 5W, but that doesn't take a heatsink.
The zener is only dissipating a max of 15mA * 10v, which is nothing.

If you need really good regulation, use a 15v zener and then follow it with
an 78L10. Total parts cost, including bypass capacitors, somewhere under
$5.
 
F

Fritz Schlunder

Walter Harley said:
isolated


Why don't you just use a resistor and a zener diode off the 340VDC supply?
Is there some sort of isolation requirement you're not telling us about?

Erhrmm... Cough... (clearing throat sounds) Uhh... Perhaps I might
suggest reading the very first line of the OP's post again.
 
R

Rich Grise

I think you can accomplish the same thing with a couple of
capacitors. You'd use a Zener or other shunt regulator, so
that the cap/diode would be called on for a constant
current.

Cheers!
Rich
 
T

Tony Williams

John Woodgate said:
You don't need anything in series. A transformer suitable for 115
V input at 60 Hz will accept 115 x 170/60 = 326 V input at 170
Hz. But, of course, if you put 220 V in, you will get 220/115
times the specified output voltage. You may be able to get a
transformer that is spec'd at half the output voltage you
actually require, i.e. 6.0 or 6.3 V.

I've run 50Hz transformers at 400Hz. With lamination
transformers, leakage inductance is the thing to watch
out for, especially when using a bridge rectifier into
a cap.
 
J

John Woodgate

I read in sci.electronics.design that Tony Williams
) about 'tiny 2.3VA 115VAC 60Hz transformer will get powered from
240VAC 200Hz', on Sat, 25 Oct 2003:
I've run 50Hz transformers at 400Hz. With lamination
transformers, leakage inductance is the thing to watch
out for, especially when using a bridge rectifier into
a cap.

That could indeed be a factor. But could you please enlarge on 'watch
out for', because it isn't entirely obvious what the implications are?
Are we into impaired regulation or smoke?
 
F

Fred Bloggs

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

Since when is 0.1A @ 12.6V 0.25VA? That transformer is a cheap PoS- will
not begin to hack an automotive/mobile environment- if it doesn't shake
apart then it will burn up and/or corrode. The stuff about
mass-production is pure bull-in your dreams.
 
N

N. Thornton

Fritz Schlunder said:
There are a number of reasons why I would not recommend what you are
proposing. The first problem is transformer primary insulation. It was
only intended to handle ~160V peak voltage accross the winding (for 115V
operation). Although it will most likely handle double this with little
trouble, at least in a one off case and for a limited test trial, running it
well beyond its voltage rating is asking for trouble when your specific
goals are reliability and large scale manufacturing. The cost of a product
recall/excess failures in the field/fires would be extremely high.

I'm sure the TF's insulation rating will be way above what its
designed to run at. The qusetion is whether its high enough to meet
the OP's required spec.


Regards, NT
 
M

Mike Harrison

I'm sure the TF's insulation rating will be way above what its
designed to run at. The qusetion is whether its high enough to meet
the OP's required spec.
Insulation is not the problem - core saturation is. A 115v tranny at 240 will get very very hot
 
S

Spehro Pefhany

Insulation is not the problem - core saturation is. A 115v tranny at 240 will get very very hot

Not at two or three times the frequency it won't.

Best regards,
Spehro Pefhany
 
J

Jeff

positivebalance41m said:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).

Since I only need an output of constant 10mA from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.

Unfortunately in trying to find a 220VAC transformer, the cost goes from
US$2.00 to US$7.00 so hence my motivation to be "unconventional" on this.

I know my transfomer losses will be higher at 200Hz compared to 60Hz but I'm
not concerned with efficiency as much as I am concerned with low cost and
reliable functioning.

Can anyone suggest the correct TVS voltage rating? Will the 200Hz pose any
specific problems for the 60Hz designed transformer that could be an
unpleasent surprise? I thought instead of a series TVS, to have a series
capacitor or even a resistor but the 220VAC frequency changes by about 30%
so any R/C ideas to my mind appeared to be too frequency sensitive...but
maybe I miss some clever ideas, component(s), or confirgurations. Perhaps a
shunt TVS and a series capacitor would ensure the transformer primary sees a
more constant 115VAC. I am guessing that on the primary side of the
transformer, there will only be about 1.2 to 1.8mA of ac current. That is
because I will have a fairly constant 12.6VAC at 10mA out of the secondary,
and I assume a few percent will go to core laminations magnetization
current.

Comments anyone?

Since your power supply draws a fairly constant current, a series resistance
equal to the transformer's reflected resistance would work great! The
resistor will also have very low power dissipation. A zener could be added
to short the supply if the voltage gets too high, thus keeping the current
at a near constant.
 
R

R.Legg

positivebalance41m said:
I have a unique situation where I require regulated 10VDC @ 10mA isolated
supply to run some control circuitry and the only available power sources
are 340VDC or 3 phase 220VAC (which can vary between 170Hz to 280Hz. This
has to be extremely low cost, reliable, easy for large scale manufacture,
and not require heatsinking of any components.

I am thinking to use a tiny 0.25VA 115VAC 60Hz transformer that I can get
from Mouser for USD$2.00. Simple 1/2 half rectify and filter. Then use an
NJR78L10 voltage regulator.

Any thoughts on an idea I am considering to use the very cheap 115VAC
transformer on 220VAC +/- 20% 200Hz, by placing a series Bilateral TVS
(some people know these by the name of Transorb, or back to back zener
essentially).

Since I only need an output of constant 10mA from the transformer secondary
(12.6VAC CT) and I don't care the output AC waveform is distored, I thought
this might work.

At 240mW, capacitively-coupling the transformer primary to one of the
input phases, and shunt-regulating the transformer secondary, is a
practical option. The capacitor is used to limit current, the shunt
regulator (~zener) to define the output voltage (and by reflection -
the voltage across the primary).

The problem is that the lower voltage transformer will not be
considered safe as an isolator, when operated in circuitry outside of
it's class. If it is not used as a safety isolator, then it's
functional isolation will likely suffice.

RL
 
Top