C
Chris
Mon, Sep 1, 2014 10:54 PM
Thanks for the inputs everyone,
One of the direct replies got me the data I needed!
Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at 4ma
does not seem to be available!
Cheers,
Corby
Hi Corby,
The way I would approach the problem would be not ask where to get a
given pot core, but how do I generate 3700v @ 4mA, starting from
scratch. I would limit the secondary winding count and use a voltage
multiplier, Cockroft-Walton style to bring the volts up to the required
level. 22w is not an insignificant power level, so I would use a switch
mode regulator chip, driving a pair of small power mosfets, with a
switching frequency 100KHz or higher to keep the magnetics, winding
turns count and the multiplier (Use polypropylene) caps small. Lower
voltage at the secondary also makes it easier in terms of rectifier
diode selection. All the info is in the chip manufacturers application
notes. and suggest Unitrode as a starting point. Have done a few of
these in the past and they are pretty strightforward. Even the magnetics
are covered by the core manufacturers data books. A pair of small E
cores sounds sounds about right and will have matching moulded bobbin
options with tags. Pot cores are a pita in comparison and awkward to
wind / terminate using foil, which you really need when a primary
winding may only have 4 to 8 turns, yet be carrying amps.
The other point is that I would never use any switcher, however good
it's claimed to be, to drive sensitive analog electronics. While many of
the cheap switcher modules are fine with stable line and load, the
transient response ids often dreadful and they take some time
(milliseconds) to recover with step function change at either. Such
instability can damage driven electronics as well. While it's common to
use a switcher for initial conversion, that would always be followed by
monolithic linear regulators, which typically have output noise levels
of mV. For example, say you have an input voltage of 18-32v and a
required output of 12v at 1A, an initial switch mode regulator to
convert to 15v, then even a 7812 or similar to lose the final 3 volts...
Regards,
Chris
> Thanks for the inputs everyone,
>
> One of the direct replies got me the data I needed!
>
> Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at 4ma
> does not seem to be available!
>
> Cheers,
>
> Corby
Hi Corby,
The way I would approach the problem would be not ask where to get a
given pot core, but how do I generate 3700v @ 4mA, starting from
scratch. I would limit the secondary winding count and use a voltage
multiplier, Cockroft-Walton style to bring the volts up to the required
level. 22w is not an insignificant power level, so I would use a switch
mode regulator chip, driving a pair of small power mosfets, with a
switching frequency 100KHz or higher to keep the magnetics, winding
turns count and the multiplier (Use polypropylene) caps small. Lower
voltage at the secondary also makes it easier in terms of rectifier
diode selection. All the info is in the chip manufacturers application
notes. and suggest Unitrode as a starting point. Have done a few of
these in the past and they are pretty strightforward. Even the magnetics
are covered by the core manufacturers data books. A pair of small E
cores sounds sounds about right and will have matching moulded bobbin
options with tags. Pot cores are a pita in comparison and awkward to
wind / terminate using foil, which you really need when a primary
winding may only have 4 to 8 turns, yet be carrying amps.
The other point is that I would never use any switcher, however good
it's claimed to be, to drive sensitive analog electronics. While many of
the cheap switcher modules are fine with stable line and load, the
transient response ids often dreadful and they take some time
(milliseconds) to recover with step function change at either. Such
instability can damage driven electronics as well. While it's common to
use a switcher for initial conversion, that would always be followed by
monolithic linear regulators, which typically have output noise levels
of mV. For example, say you have an input voltage of 18-32v and a
required output of 12v at 1A, an initial switch mode regulator to
convert to 15v, then even a 7812 or similar to lose the final 3 volts...
Regards,
Chris
BC
Bob Camp
Mon, Sep 1, 2014 11:41 PM
Hi
Keep in mind that the supply in question is likely running in the vicinity of an atomic clock. Things like stray magnetic field are pretty important if that’s the case …. It’s also quite possible that the original supply had some “supply dies not the tube” features built into it.
Bob
On Sep 1, 2014, at 6:54 PM, Chris syseng.greenfield@btconnect.com wrote:
Thanks for the inputs everyone,
One of the direct replies got me the data I needed!
Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at 4ma
does not seem to be available!
Cheers,
Corby
Hi Corby,
The way I would approach the problem would be not ask where to get a given pot core, but how do I generate 3700v @ 4mA, starting from scratch. I would limit the secondary winding count and use a voltage multiplier, Cockroft-Walton style to bring the volts up to the required level. 22w is not an insignificant power level, so I would use a switch mode regulator chip, driving a pair of small power mosfets, with a switching frequency 100KHz or higher to keep the magnetics, winding turns count and the multiplier (Use polypropylene) caps small. Lower voltage at the secondary also makes it easier in terms of rectifier diode selection. All the info is in the chip manufacturers application notes. and suggest Unitrode as a starting point. Have done a few of these in the past and they are pretty strightforward. Even the magnetics are covered by the core manufacturers data books. A pair of small E cores sounds sounds about right and will have matching moulded bobbin options with tags. Pot cores are a pita in comparison and awkward to wind / terminate using foil, which you really need when a primary winding may only have 4 to 8 turns, yet be carrying amps.
The other point is that I would never use any switcher, however good it's claimed to be, to drive sensitive analog electronics. While many of the cheap switcher modules are fine with stable line and load, the transient response ids often dreadful and they take some time (milliseconds) to recover with step function change at either. Such instability can damage driven electronics as well. While it's common to use a switcher for initial conversion, that would always be followed by monolithic linear regulators, which typically have output noise levels of mV. For example, say you have an input voltage of 18-32v and a required output of 12v at 1A, an initial switch mode regulator to convert to 15v, then even a 7812 or similar to lose the final 3 volts...
Regards,
Chris
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi
Keep in mind that the supply in question is likely running in the vicinity of an atomic clock. Things like stray magnetic field are pretty important if that’s the case …. It’s also quite possible that the original supply had some “supply dies not the tube” features built into it.
Bob
On Sep 1, 2014, at 6:54 PM, Chris <syseng.greenfield@btconnect.com> wrote:
>> Thanks for the inputs everyone,
>>
>> One of the direct replies got me the data I needed!
>>
>> Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at 4ma
>> does not seem to be available!
>>
>> Cheers,
>>
>> Corby
>
> Hi Corby,
>
> The way I would approach the problem would be not ask where to get a given pot core, but how do I generate 3700v @ 4mA, starting from scratch. I would limit the secondary winding count and use a voltage multiplier, Cockroft-Walton style to bring the volts up to the required level. 22w is not an insignificant power level, so I would use a switch mode regulator chip, driving a pair of small power mosfets, with a switching frequency 100KHz or higher to keep the magnetics, winding turns count and the multiplier (Use polypropylene) caps small. Lower voltage at the secondary also makes it easier in terms of rectifier diode selection. All the info is in the chip manufacturers application notes. and suggest Unitrode as a starting point. Have done a few of these in the past and they are pretty strightforward. Even the magnetics are covered by the core manufacturers data books. A pair of small E cores sounds sounds about right and will have matching moulded bobbin options with tags. Pot cores are a pita in comparison and awkward to wind / terminate using foil, which you really need when a primary winding may only have 4 to 8 turns, yet be carrying amps.
>
> The other point is that I would never use any switcher, however good it's claimed to be, to drive sensitive analog electronics. While many of the cheap switcher modules are fine with stable line and load, the transient response ids often dreadful and they take some time (milliseconds) to recover with step function change at either. Such instability can damage driven electronics as well. While it's common to use a switcher for initial conversion, that would always be followed by monolithic linear regulators, which typically have output noise levels of mV. For example, say you have an input voltage of 18-32v and a required output of 12v at 1A, an initial switch mode regulator to convert to 15v, then even a 7812 or similar to lose the final 3 volts...
>
> Regards,
>
> Chris
> _______________________________________________
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
AP
Alexander Pummer
Tue, Sep 2, 2014 3:42 AM
forward converter [not fly -back]
look for pot-cores material 3C80, 3C81, 3C85 or equivalent, you will
need a cca 21 to 28mm dia pot-core if you run it at 80 to 120kHz,
driver ic TL495 , which is a PWM push pull controller
estimate the efficiency for 60% for the input power, look for very good
and large 10Megohm resistors you wil nee four of them 3700V/01mA = 37
Meg ohm, the must be temperature stable and must be able to carry 1000V
each, you need a high quality resistor for the bottom part of the
feedback divider same sort as the 10Megohms value cca 57k use 56k + an
1k pot.
Philips has fast high voltage rectifiers you need some with very short
recovery time
how to design a transformer:
http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCMQFjAA&url=http%3A%2F%2Fwww.mag-inc.com%2FFile%2520Library%2FProduct%2520Literature%2FFerrite%2520Literature%2FMagneticsFerritePowerDesign2013.pdf&ei=RTUFVOKzIMn3igLAoYGgBQ&usg=AFQjCNE63ziQj-MYzYmYzNfLrHThFJu1lg&bvm=bv.74115972,d.cGE
a more elegant way is a resonant converter, less noise, less
complicated high voltage transformer, and the recovery time is not so
critical, but that is for more experienced DIY people
73
Alex
On 9/1/2014 3:54 PM, Chris wrote:
Thanks for the inputs everyone,
One of the direct replies got me the data I needed!
Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at
4ma
does not seem to be available!
Cheers,
Corby
Hi Corby,
The way I would approach the problem would be not ask where to get a
given pot core, but how do I generate 3700v @ 4mA, starting from
scratch. I would limit the secondary winding count and use a voltage
multiplier, Cockroft-Walton style to bring the volts up to the
required level. 22w is not an insignificant power level, so I would
use a switch mode regulator chip, driving a pair of small power
mosfets, with a switching frequency 100KHz or higher to keep the
magnetics, winding turns count and the multiplier (Use polypropylene)
caps small. Lower voltage at the secondary also makes it easier in
terms of rectifier diode selection. All the info is in the chip
manufacturers application notes. and suggest Unitrode as a starting
point. Have done a few of these in the past and they are pretty
strightforward. Even the magnetics are covered by the core
manufacturers data books. A pair of small E cores sounds sounds about
right and will have matching moulded bobbin options with tags. Pot
cores are a pita in comparison and awkward to wind / terminate using
foil, which you really need when a primary winding may only have 4 to
8 turns, yet be carrying amps.
The other point is that I would never use any switcher, however good
it's claimed to be, to drive sensitive analog electronics. While many
of the cheap switcher modules are fine with stable line and load, the
transient response ids often dreadful and they take some time
(milliseconds) to recover with step function change at either. Such
instability can damage driven electronics as well. While it's common
to use a switcher for initial conversion, that would always be
followed by monolithic linear regulators, which typically have output
noise levels of mV. For example, say you have an input voltage of
18-32v and a required output of 12v at 1A, an initial switch mode
regulator to convert to 15v, then even a 7812 or similar to lose the
final 3 volts...
Regards,
Chris
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
forward converter [not fly -back]
look for pot-cores material 3C80, 3C81, 3C85 or equivalent, you will
need a cca 21 to 28mm dia pot-core if you run it at 80 to 120kHz,
driver ic TL495 , which is a PWM push pull controller
estimate the efficiency for 60% for the input power, look for very good
and large 10Megohm resistors you wil nee four of them 3700V/01mA = 37
Meg ohm, the must be temperature stable and must be able to carry 1000V
each, you need a high quality resistor for the bottom part of the
feedback divider same sort as the 10Megohms value cca 57k use 56k + an
1k pot.
Philips has fast high voltage rectifiers you need some with very short
recovery time
how to design a transformer:
http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCMQFjAA&url=http%3A%2F%2Fwww.mag-inc.com%2FFile%2520Library%2FProduct%2520Literature%2FFerrite%2520Literature%2FMagneticsFerritePowerDesign2013.pdf&ei=RTUFVOKzIMn3igLAoYGgBQ&usg=AFQjCNE63ziQj-MYzYmYzNfLrHThFJu1lg&bvm=bv.74115972,d.cGE
a more elegant way is a resonant converter, less noise, less
complicated high voltage transformer, and the recovery time is not so
critical, but that is for more experienced DIY people
73
Alex
On 9/1/2014 3:54 PM, Chris wrote:
>> Thanks for the inputs everyone,
>>
>> One of the direct replies got me the data I needed!
>>
>> Alex, I'd like to by it that way, but A 24VDC input 3700VDC output at
>> 4ma
>> does not seem to be available!
>>
>> Cheers,
>>
>> Corby
>
> Hi Corby,
>
> The way I would approach the problem would be not ask where to get a
> given pot core, but how do I generate 3700v @ 4mA, starting from
> scratch. I would limit the secondary winding count and use a voltage
> multiplier, Cockroft-Walton style to bring the volts up to the
> required level. 22w is not an insignificant power level, so I would
> use a switch mode regulator chip, driving a pair of small power
> mosfets, with a switching frequency 100KHz or higher to keep the
> magnetics, winding turns count and the multiplier (Use polypropylene)
> caps small. Lower voltage at the secondary also makes it easier in
> terms of rectifier diode selection. All the info is in the chip
> manufacturers application notes. and suggest Unitrode as a starting
> point. Have done a few of these in the past and they are pretty
> strightforward. Even the magnetics are covered by the core
> manufacturers data books. A pair of small E cores sounds sounds about
> right and will have matching moulded bobbin options with tags. Pot
> cores are a pita in comparison and awkward to wind / terminate using
> foil, which you really need when a primary winding may only have 4 to
> 8 turns, yet be carrying amps.
>
> The other point is that I would never use any switcher, however good
> it's claimed to be, to drive sensitive analog electronics. While many
> of the cheap switcher modules are fine with stable line and load, the
> transient response ids often dreadful and they take some time
> (milliseconds) to recover with step function change at either. Such
> instability can damage driven electronics as well. While it's common
> to use a switcher for initial conversion, that would always be
> followed by monolithic linear regulators, which typically have output
> noise levels of mV. For example, say you have an input voltage of
> 18-32v and a required output of 12v at 1A, an initial switch mode
> regulator to convert to 15v, then even a 7812 or similar to lose the
> final 3 volts...
>
> Regards,
>
> Chris
> _______________________________________________
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
