HV power supply for a project I am thinking of doing.
216 views
Skip to first unread message
Phill Scanlon
Jan 12, 2018, 3:21:24 AM1/12/18
to neonixie-l
Hi all,
Looking at open source power supply designs.
The one below has been kindly made open source.
I like because it already has the eagle files, so I am guessing that the layout is (maybe) ok.
Even then I am not sure what is going on with this circuit. Pin 3 (timing capacitor) is connected to a diode and a transistor, so am a bit confused about whether this is even a real working supply.
The power supply is the biggest headache for me (to get small and neat anyway).
I figure that if I use that one in my devices (with a tripler circuit for the dekatrons) I am not going to be stepping on anyone's toes should I decide to sell a few.
The inductor is marked as 100uH but none of the other critical components have values. Even then I would just have to have a stab at the amp rating of the choke.
I am know that the switching diodes and transistors are similar in criticality as all these sorts of circuits re: speed etc. Although I have no idea what speed that thing is running at.
Would I be right in thinking that a good fast enhanced mosfet such as a IRFU220 as in threeneuron's HVPS would be acceptable ?
From the jimdo.com site I found some other useful information regarding a power supply he sells as a kit. It's all through hole stuff.
The trouble is he uses stuff that is not readily available from mouser Australia.
For instance the fast diodes he uses (HER105/SF-16/SF-18) are from taiwan and are "non stocked" from mouser, so I am figuring that the BAV21's that threeneuron's uses would be an acceptable substitute.
To a point, a fast diode is a fast diode (amps aside) more or less, right ?
All I am trying to do is drive a bunch of my new favorite toys/addiction (Dekatrons) and a few nixies in a bit of an art project.
I have seen others .... but Jimdo seems to have nailed the size.
Any guidance would be much appreciated.
Cheers,
Phill
gregebert
Jan 13, 2018, 1:53:06 AM1/13/18
to neonixie-l
Dekatrons require substantially higher voltage than nixies; I'm running my A101 at almost 500V (datasheet says 450v). I use a voltage tripler from the 120Vrms line (US).
Higher voltage is fine for nixies and dekatrons; you just need to have more anode-resistance (or equivalent) to keep the current in the proper range, Though they run at higher voltages, dekatrons require much less current than nixes, around 300-400uA.
The boost converter shown above is not ideal for such high voltage "gains"; a better choice would be it's close-cousin the flyback converter, which uses a transformer instead of an inductor. A transformer is nothing more than 2 or more coupled inductors. I suggest you run a lot of simulations on it before building it, so you can optimize the design.
Phill Scanlon
Jan 13, 2018, 2:47:50 AM1/13/18
to neonixie-l
Yup I am aware that the voltage requirements are high and current a lot lower than standard Nixie type tubes and as you correctly say the circuit above will not do that.
Hence the tippler comment I made in the original post.
Leaning now towards the 2nd circuit now by Jan Rychter .. it is well documented.
The circuit above is kind of strange with that setup on the sense resistor input, makes no sense (pun intended) to me :-)
Re: flyback... no not for me ... pain in the backside all round with the transformer setup needing special windings etc.
Simulation software is one thing .. but from my research it seems that layout is probably more important. That, and the correct selection of inductor and mosfet or transistors.
I have a few different supplies from various sellers and most of them have "voodoo" in them.
Michael Moorrees (threeneurons) have been the best for me as I can get all the components and don't have special custom inductors in them.
He is also a good teacher and has good explanations of what is going on in his circuits.
But .. they are copyright. Which is fair enough.
This quest is because I don't just want to buy a 3rd party "unknown" design and put it in my project because if you do that you are then locked in and have no recourse should the maker stop production.
I also would like the option of selling stuff should I decide too ....
Thus my search for something opensource.
Power supplies I have tried.
Nice little supply with no noisy windings. Voodoo inductor, scratched off critical component values.
Beast power supply ... easy to break (have fried 2 so far.... full of voodoo including a custom chip (which you can get separately) Great though for driving lots of IN-9's (to a point as these suckers really chew up the amps.)
Very nice supply with a custom inductor. No noise.
Another by the same guy but very noisy (sounds) at higher loads. Also a custom inductor.
People want to protect their work... which is understandable.
Paul Andrews
Jan 13, 2018, 7:46:51 AM1/13/18
to neonixie-l
I open sourced my flyback converter. If you use it for commercial purposes I’d be grateful if you gave me a credit. http://www.nixies.us/projects/nixie-tube-power-supply/
You would have to modify it to get the voltages you want. That might just be modifying the feedback. It is all documented in ridiculous detail, including the whole decision process, LTspice models, KiCAD files, BOM and experimental results.
You would have to modify it to get the voltages you want. That might just be modifying the feedback. It is all documented in ridiculous detail, including the whole decision process, LTspice models, KiCAD files, BOM and experimental results.
Phill S
Jan 14, 2018, 12:45:22 AM1/14/18
to neonixie-l
Hi Paul,
It looks like you have put a lot of time into it.
Re: Temperatures.
John Taylors supplies get as hot as hades when running at high loads also.
It will take me some time to digest this.
The only real issue I have with flyback designs is, once you have a design worked out you are sort of locked into that particular transformer.
And as I am sure you know transformers come and transformers go.
All the issues I ever had back in ye olde days when I was fixing TV's was bloody flyback power supply systems. Since all TV's had them back then it was a familiar problem.
Probably 80% of faults were in these circuits so I am a bit biased. They were a little beefier than yours of course but still.
Thank you for sharing your project.
Much appreciated.
gregebert
Jan 14, 2018, 1:47:06 AM1/14/18
to neonixie-l
TV flybacks of yesteryear are an entirely different animal.
First, the flybacks in TVs ran at much higher voltages, up to 25kV for color sets.
Second, the earlier versions were wax-impregnated. So, when you have a TV set with a lot of tubes you have a lot of heat, and after a few years, just enough wax has crept away to allow insulation breakdown. Then all hell breaks loose. What a noise, and what a smell when the flyback went out......No wonder they were always entombed inside a metal cage....
In the early 1970's, silicone-based insulation was used in flybacks and I dont recall any failures in my TVs thereafter. And they became cage-free.
None of those problems will happen with a nixie supply.
Just about every electronic device manufactured today that is line-powered has a switch-mode power supply, and they operate with the same underlying principles as a flyback converter. Very safe,efficient, and reliable when properly designed.
Phill S
Jan 14, 2018, 2:26:44 AM1/14/18
to neonixie-l
You are right of course.
Much higher voltages in the old CRT tv's.
Still, I have changed out more flyback transformers than I can count, having said that I am guessing it was mostly the triplers inside failing, and as they were all potted there was not much you could do.
This was over 35 years ago (early 80's). I never encountered any metal cage ones though. All the ones I saw were an all in one potted type design with the focus and brightness (from memory) knobs on them.
Every tv manufacturer seemed to have their own take on them even between models (for obvious reasons) and you needed to find the correct one for the correct tv. All round pain in the backside.
I like the simpler inductor only designs, which are not dependent on an esoteric transformer.
It's just my personal bias of course.
GastonP
Jan 14, 2018, 12:21:13 PM1/14/18
to neonixie-l
IIRC the metal cage had more to do with X-Ray protection from the (tube) diode rectifier than the noise, to which the tube electronics was quite impervious.
It also dealt well with the corona effect on all the parts where as you say the isolation had gone away or cracked.
I still remember the smell of ozone of the aged TV sets...
It also dealt well with the corona effect on all the parts where as you say the isolation had gone away or cracked.
I still remember the smell of ozone of the aged TV sets...
On Sunday, January 14, 2018 at 3:47:06 AM UTC-3, gregebert wrote:
TV flybacks of yesteryear are an entirely different animal.
.No wonder they were always entombed inside a metal cage....
Reply all
Reply to author
Forward
0 new messages