SYMATTR InstName C5
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use, >multisourced.
piglet
On Thu, 22 Aug 2024 10:36:27 +0100, piglet <erichpwagner@hotmail.com>
wrote:
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use,
multisourced.
piglet
I like the auto-flyback circuit, flyback drive into the center tap of
a DRQ127. Then maybe a voltage doubler to harvest the p-p swing.
And LTC3803 rocks.
https://www.dropbox.com/scl/fi/8jwi35owo41lfmonc8y92/LTC3803.jpg?rlkey=l3thy5h4533sxpd152ve24c26&raw=1
Is there an off the shelf part I can use for the transformer in this?
It's based on a design someone else posted in this group a long time ago (maybe 10 years ago by now).
I can't remember the name of that individual.
Watch out for wrapped lines and character encoding issues for L1, C3, C5
If it doesn't simulate L1 likely has a u symbol issue.
If it doesn't open, line wraps need to be fixed.
On Wed, 21 Aug 2024 12:39:55 -0400, "Edward Rawde"
<invalid@invalid.invalid> wrote:
Is there an off the shelf part I can use for the transformer in this?
It's based on a design someone else posted in this group a long time ago (maybe 10 years ago by now).
I can't remember the name of that individual.
Watch out for wrapped lines and character encoding issues for L1, C3, C5
If it doesn't simulate L1 likely has a u symbol issue.
If it doesn't open, line wraps need to be fixed.
Gets 'singular matrix node error' on Q1e here.
Discrete circuits are fun but unappreciated.
RL
john larkin <jlarkin_highland_tech> wrote:
On Thu, 22 Aug 2024 10:36:27 +0100, piglet <erichpwagner@hotmail.com>
wrote:
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use,
multisourced.
piglet
I like the auto-flyback circuit, flyback drive into the center tap of
a DRQ127. Then maybe a voltage doubler to harvest the p-p swing.
And LTC3803 rocks.
https://www.dropbox.com/scl/fi/8jwi35owo41lfmonc8y92/LTC3803.jpg?rlkey=l3thy5h4533sxpd152ve24c26&raw=1
Very slick part but LT/AD usually too luxurious pricewise for me. The
UC384x is slow and old but tenth the cost. I like the 5V reference output
can often support a low power MCU etc
On Thu, 22 Aug 2024 10:27:43 -0000 (UTC), piglet
<erichpwagner@hotmail.com> wrote:
john larkin <jlarkin_highland_tech> wrote:
On Thu, 22 Aug 2024 10:36:27 +0100, piglet <erichpwagner@hotmail.com>
wrote:
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use,
multisourced.
piglet
I like the auto-flyback circuit, flyback drive into the center tap of
a DRQ127. Then maybe a voltage doubler to harvest the p-p swing.
And LTC3803 rocks.
https://www.dropbox.com/scl/fi/8jwi35owo41lfmonc8y92/LTC3803.jpg?rlkey=l3thy5h4533sxpd152ve24c26&raw=1
Very slick part but LT/AD usually too luxurious pricewise for me. The
UC384x is slow and old but tenth the cost. I like the 5V reference output
can often support a low power MCU etc
True about the price. Our stuff isn't very price sensitive.
TI has some great little switchers lately. We're paying 21 cents for
the TPS562208DDCT spread-spectrum synchronous buck switcher.
I wonder if it would be practical to use one of those to make the
12/200 volt converter. That's just possible. Let's play with that
idea.
Unfortunately, it's a nuisance to plug the TI models into LT Spice.
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a single inductor or a dual coupled inductor like DRQ127
series. if necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use, multisourced.
piglet
Is there an off the shelf part I can use for the transformer in this?
It's based on a design someone else posted in this group a long time ago (maybe 10 years ago by now).
I can't remember the name of that individual.
On Thu, 22 Aug 2024 10:36:27 +0100, piglet <erichpwagner@hotmail.com>
wrote:
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use,
multisourced.
piglet
I like the auto-flyback circuit, flyback drive into the center tap of
a DRQ127. Then maybe a voltage doubler to harvest the p-p swing.
And LTC3803 rocks.
https://www.dropbox.com/scl/fi/8jwi35owo41lfmonc8y92/LTC3803.jpg?rlkey=l3thy5h4533sxpd152ve24c26&raw=1
On 8/22/24 11:44, john larkin wrote:
On Thu, 22 Aug 2024 10:36:27 +0100, piglet <erichpwagner@hotmail.com>
wrote:
On 21/08/2024 17:39, Edward Rawde wrote:
SYMATTR InstName C5
If you want to stick with that way-over-complex design you could look
into versapac 6 windings coupled inductors.
I would not even try that circuit. Better to redesign to use either a
single inductor or a dual coupled inductor like DRQ127 series. if
necessary follow with a cockcroft-walton multiplying rectifier.
UC3843 control chip would be ideal, cheap, reliable, easy to use,
multisourced.
piglet
I like the auto-flyback circuit, flyback drive into the center tap of
a DRQ127. Then maybe a voltage doubler to harvest the p-p swing.
And LTC3803 rocks.
https://www.dropbox.com/scl/fi/8jwi35owo41lfmonc8y92/LTC3803.jpg?rlkey=l3thy5h4533sxpd152ve24c26&raw=1
$0.15 for the first 60V, then the rest of the way with doublers
https://jlcpcb.com/partdetail/LegendSi-LGS6302EP/C5280693
Is there an off the shelf part I can use for the transformer in this?
It's based on a design someone else posted in this group a long time ago (maybe 10 years ago by now).
I can't remember the name of that individual.
On 22/08/2024 2:39 am, Edward Rawde wrote:
Is there an off the shelf part I can use for the transformer in this?
It's based on a design someone else posted in this group a long time ago (maybe 10 years ago by now).
I can't remember the name of that individual.
<snipped .asc file>
There's a more elegant solution to that problem that was discussed here a few yeas ago.
sci-hub.do/10.1109/tpel.2007.909192
Abramovitz, A., & Smedley, K. (2007). A Resonant DC-DC Transformer With Zero Current Ripple. IEEE Transactions on Power
Electronics, 22(6), 2344-2351. doi:10.1109/tpel.2007.909192
It's essentially the Baxandall class-D oscillator with two extra windings on the feed inductor. The paper talks about integrating
all the windings on single - carefully gapped - I-E core, but using two cores gives you an easily designed way of getting to the
same advantage.
At the time I posted an LTSpice simulation that made the point. The output wasn't entirely ripple-free, in that there were
switching transients which would have had to have been filtered out, but that needed a much smaller R,L and C values than you'd
have need to get rid of the switching frequency components.
It won't be attractive if you have a pathological fear of transformers or an aversion to winding them yourself a or finding a
nearby transformer winding shop to wind a few of them for you.
For volume production you'd use printed windings, but thye need a specialist supplier too.
--
Bill Sloman, Sydney
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 415 |
Nodes: | 16 (2 / 14) |
Uptime: | 108:18:01 |
Calls: | 8,692 |
Calls today: | 1 |
Files: | 13,257 |
Messages: | 5,948,381 |