Prehaps the 'icy' glare of Hg vapor lamps would cool you off?

TTFN
HP

HP it still just drives me crazy that high PE light is called _cold_ and low PE light is called _warm_ ! HP I know you were saying ppl probably call red end of spectrum _warm_ cuz it's reminiscent of firelight and associated with other IR heat sources and all that but it's still totally lame! So they just need to think on totally basic middle school thermodynamics to have concept of bb radiation and color temperature!

 

HP I'm sry for more delays cuz MMK cap just melted

I don't wonder -- Such aplication of polyester film caps is best realized via distribution of capacitance over several units (in, for instance, so called 'MMC' arrangements) - as you seem to apprehend, the issue is not that of dielectric 'breakdown' per se but, rather, heating thereof...

which I say wasn't to do with different setup just polyester dielectric isn't suited for resonance cap so it just happened to fail when it did

Agreed...

Anyhow I'm back to square one with original circuit trying to find right capacitance using MKP and FKP

'Tuning' of the circuit is sufficiently arduous sans 'factoring' of dielectric material into the 'permutations' - please be advised that single-unit MKP 10 capacitors are wholly satisfactory in your application whereas FKP 1 is 'over kill'

finding resonant like you say _sweet spot_ is so hard cuz just close is same performance as way off so there's nothing to go by

HP If you have any suggestions I could use the help!

1) Measure the parallel equivalent inductance of the entire primary winding (all other windings 'open').
2) Start with a capacitance calculated to resonate with the above determined inductance at Ca. 45kHz.
3) Be advised that manic bouts of auto-depilation, apodysophilia, head 'thumping' and/or attempted self-immolation are merely nature's way of telling you it's time for a break

I have tomorrow through weekend free so I'll keep at it

Pleased to hear it!

HP don't you dare try to say I'm just stalling cuz if I was disingenuous puke I could have just made up plausible figures to post!

Where is that coming from? -- Seriously! The thought never crossed my mind! -- When you don't want to do something you refuse outright!

So plz just be patient

Agreed and ditto!

Very best regards
HP

 

HP it still just drives me crazy that high PE light is called _cold_ and low PE light is called _warm_ ! HP I know you were saying ppl probably call red end of spectrum _warm_ cuz it's reminiscent of firelight and associated with other IR heat sources and all that but it's still totally lame! So they just need to think on totally basic middle school thermodynamics to have concept of bb radiation and color temperature!

I hear ya! -- FWIW I've long regarded embodiment of 'nonentities' (e.g. 'cold' and 'dark') to be second only to 'quantification of units' (can we say 'voltage'?) on the annoyance list -- I wonder if we'll survive?

Best regards
HP

 

HP I've spent days trying different caps for resonance and the fact is the best values still give crap for voltage! It reminds me of those _ZVS_ drivers that 4HV fanboys orgasm over cuz they're too ignorant to understand that it's strike distance not drawn distance that is measure of voltage! HP I'm sry if you disagree but I say all royer topologies including mazzilli suck for HV cuz at best they make just 50khz 200mA intermittent duty NSTs! HP if you say that's wrong I dare you to show me repeatable resonant Royer type circuit capable of greater than 15kV! Sure non nonresonant does even 50kv w/o doubler but 300uA is bull$#!%

HP I say IMPWM is only way to go but you don't like pulse topologies so I'm done!

HP here's another thing! Ltspice needs euthanasia! I'm going back to Orcad cuz at least it tells the truth %50 of the time I say the only reason for popularity of ltspice is its _price tag_ and that attitude is more than lame its tragic and scary!

 

HP I've spent days trying different caps for resonance

Aye! That'd explain your loss of patience -- Recall my warning? To wit:

Be advised that manic bouts of auto-depilation, apodysophilia, head 'thumping' and/or attempted self-immolation are merely nature's way of telling you it's time for a break

Thus it seems too much work and no play makes Aleph a dull radical girl!

It reminds me of those _ZVS_ drivers that 4HV fanboys orgasm over cuz they're too ignorant to understand that it's strike distance not drawn distance that is measure of voltage!

To be fair, it has long been my observation that the 'core' 4HV membership are highly competent -- But yes! The so called 'ZVS' (a.k.a. Mazilli Royer oscillator) is both widely misunderstood and prey of 'co-option' by abject ignoramuses the web over

HP I'm sry if you disagree

No you're not!

HP I say IMPWM is only way to go but you don't like pulse topologies so

The trouble with 'flyback' (i.e. 'energy storage') topoligies in this application are several -- Principally (in descending order of significance):

1) The duty cycle is such that power at EHT potentials represents a rather small fraction of the output energy -- while such is acceptable in certain applications (e.g. ignition systems and demonstration apparatus wherein the fleeting EHT 'spike' merely 'opens' an ion path for a protracted, relatively high current/low EMF discharge) it is useless in applications requisite of high duty-cycle EHT at currents > 1mA

2) Pulse transient times are such that rectification is complicated by the requirement of external equalization/sharing networks --or-- 'avalanche rated' EHT rectifiers (which being both difficult to source and, in the eyes of many, prohibitively costly)

3) The inherent asymmetry of such schemes negatively impacts versatility of application....

For all that - said scheme isn't entirely without merit -- especially where circumvention of objection #1 is practically realizable...

HP here's another thing! Ltspice needs euthanasia! I'm going back to Orcad cuz at least it tells the truth %50 of the time

It is my experience that Spice (in general) is best regarded as a powerful educational tool only! -- That said - I find it rather curious that LtSpice (being, as it is, a product of Linear Technologies - largely a manufacturer of power supply components) is far less suited to 'power electronics' than are certain 'general' electronics CAD suites

I say the only reason for popularity of ltspice is its _price tag_ and that attitude is more than lame its tragic and scary!

Actually, Spice packages aren't particularly expensive (a few hundred dollars being typical for a single EUL) -- FWIW I feel LtSpice's popularity owes to it's ubiquity - my $.02...

I'm done!

When you've done stamping your feet and holding your breath -- please try this:

Construct the 'plain vanilla' Royer circuit (sans the transformer and resonant capacitor) in a project box. Use 2N3773 BJTs mounted to a heatsink via TO-3 'sockets' (for further convenience you may wish to consider using 'thermal-mat' style insulators/heat-transfer media).

Bring out the base and collector connections via color coded leads to alligator clips/test hooks.

Solder small alligator clips/test hooks to several 330nF MKP10 caps (400VAC/1kVDC)

With the above described arrangement you will be able to rapidly evaluate many combinations of resonant capacitance, winding ratios, etc... sans undue 'agro'

Very best regards
HP

 

When you've done stamping your feet and holding your breath -- please try this:

Construct the 'plain vanilla' Royer circuit (sans the transformer and resonant capacitor) in a project box. Use 2N3773 BJTs mounted to a heatsink via TO-3 'sockets' (for further convenience you may wish to consider using 'thermal-mat' style insulators/heat-transfer media).

Bring out the base and collector connections via color coded leads to alligator clips/test hooks.

Solder small alligator clips/test hooks to several 330nF MKP10 caps (400VAC/1kVDC)

With the above described arrangement you will be able to rapidly evaluate many combinations of resonant capacitance, winding ratios, etc... sans undue 'agro'

HP Sry but NO! Cuz you don't get it! I'm all done with self oscillating circuits cuz they don't work for me! HP you have like charmed touch with Royer and Royer-like HV inverters! Cuz when you power your designs up it's all corona, arcs, sparks, ozone and just every sort of HV commotion! All I can get with pwr oscillator design is weak 20kv or hot 10kv arcs and IT DOESN'T HELP NO MATTER HOW MUCH I EFF AROUND WITH IT!

HP now I'm going to tell you something! Successful application of Royer topology to wide range of transformers is more like art than science! So I say is bad topology for ppl new to HV design cuz even when it works it's not usually repeatable! HP I'm not trying to be a bummer but I say nothing turns prospective students and hobbyists off more than BS of struggling with finicky, touchy circuits no matter how much potential they have! So like I said I'm all done with pwr oscillators! But we can still get symmetrical sinewave output driving with pulses as long as primary circuit is resonant! And BTW be happy cuz I don't give a rat's tail section if it simulates at all! All I care is that it works reliably in real life!

 

HP Sry but NO! Cuz you don't get it! I'm all done with self oscillating circuits cuz they don't work for me!

@Aleph(0) You surprise me! -- Temperament doesn't become you! Perhaps you should step back for a few days then re-approach it with a fresh perspective? -- I assure you that power oscillator topologies represent practical, tried and true design - no mysticism required! - That said we'll explore flyback, SBC, CMC/PCMC, etc... in due time!

With regards
HP

 

@Aleph(0) -- When you've done with your little cry please advise me of same -- I can send the 'retro-breadboarding' supplies as soon as you let me know...

Best regards
HP

 

You surprise me! -- Temperament doesn't become you! Perhaps you should step back for a few days then re-approach it with a fresh perspective? -- I assure you that power oscillator topologies represent practical, tried and true design

HP I know it but it just drives me crazy when everything I try to improve it from non resonant setup just lowers output voltage and raises output current I'm also tired of arguing with total idiots who don't understand that voltage determines strike distance! By their faulty logic 2kv MOT is over like 100kv just cuz arc can be drawn to like 15 cm

So it's not really _temperament_ it just galls me when nothing I do works and I can't analyze circuit properly cuz like you say HV and modern test gear don't mix

I can send the 'retro-breadboarding' supplies as soon as you let me know...

HP tnx plz just send whatever you think I need! Also plz post images cuz I can't get it in my mind? I say to post on forum cuz it's something other interested ppl will want to see too!

 

I know it but it just drives me crazy when everything I try to improve it from non resonant setup just lowers output voltage and raises output current

That's where test 'rigs and jigs' are invaluable for their reduction/mitigation of 'research fatigue'

I'm also tired of arguing with total idiots who don't understand that voltage determines strike distance!

Then I suggest you don't argue with them! -- Merely advise them of the facts, provide them with one or more references and move on!

I can't analyze circuit properly cuz like you say HV and modern test gear don't mix

But there's no need for 'modern gear' in that pursuit! -- Merely use a Tektronix 'boat anchor' (e.g. A 545B equipped with a 'CA' PIU) with appropriate probes, etc - and all should be peachy!

Also plz post images cuz I can't get it in my mind? I say to post on forum cuz it's something other interested ppl will want to see too!

Will do -- it may be as late as tomorrow evening/early Thursday AM though...

Very best regards
HP

 

@Aleph(0) -- Please see my post on the 'EHT design...' thread

Best regards
HP

 

HP tnx plz just send whatever you think I need! Also plz post images cuz I can't get it in my mind? I say to post on forum cuz it's something other interested ppl will want to see too!

Here ya go!
Note that 'guarded' and 'open' blocks are shown on the left and right respectively...

For the benefit of interested parties; The board may be fashioned of any 'non-aromatic' hardwood (Hickory or Oak preferred) -- Please do not use pine, redwood, mahogany, ebony, walnut/black walnut, cedar nor any manner of chemically treated/laminated wood, etc! -- As a 'rule of thumb'; 'If the lumber smells - it stinks for this application' Prior to initial use please saturate the board with hot (180°C /356°F) petroleum 'wax' via immersion or flooding - then, following cool-down, scrape the surface clean of non-absorbed 'wax' -- Finally; fit the board with (at least) four 'feet' (Small ceramic insulators or PCV pipe 'caps' serve well in this regard)...



Best regards
HP

 

HP how does that work? Plz show closeups of what you're calling _blocks_ HP why is some wood better to use than others? You say to seal board with wax but why? Also how do connection _blocks_ fasten to board? Even if mine is all assembled I say ppl are wondering abt it so cuz I say HV breadboard should be recommend as required materials plz upload pics to blog with detailed description so everyone can benefit!

Sry cuz I don't have time right now! Back next week!

 

Hi,

Wood can be partly conductive and at high voltage that can be a big problem.
Of course it also depends how you are using it.

 

Plz show closeups of what you're calling _blocks_

plz upload pics to blog with detailed description so everyone can benefit!

Will do -- Give me a couple of days...

HP why is some wood better to use than others?

Soft and aromatic woods are quite flammable, exhibit greater electrical conductivity, and tend to warp with age...

You say to seal board with wax but why?

To reduce surface conductivity and seal out moisture/contaminants - additionally, certain hardwoods (especially hickory) are degraded by atmospheric O2 (to say nothing of the levels of O3 exposure inherent to the intended application!)

Also how do connection _blocks_ fasten to board?

Via - would you believe... wood screws -- Such will be clearer when you view the 'detail' images

I say HV breadboard should be recommend as required materials

Agreed!

Wood can be partly conductive and at high voltage that can be a big problem.

Absolutely! Hence my 'strategy' of regarding the board as a 'soft' ground...

Of course it also depends how you are using it.

Although I don't 'count' on the board's electrical characteristics - I take measures to reduce conductivity as much as possible (via, for example; wood species selection for high density/low moisture, 'waxing' and avoidance of treated/laminated lumber) -- Still, I feel it's best viewed as a 'weak ground'

Very best regards
HP

 

@Hypatia's Protege I had time to do more experimenting with drive circuits so I explored cross coupled topology and I say it's promising but maximum gate-source voltage limitation is a bummer cuz it limits drain voltage to like 20v with that circuit! So now I'm thinking that's rationale for mazilli circuit which I say is just cc circuit modified so drain voltage can be higher than gate voltage. HP now I've got to say I'm just not impressed with mazilli circuit at all! Like I said b4 it's strong on current and totally lame on voltage! HP plz just tell me how to overcome that cuz yours is 50kv ac right off the secondary with white arcs! Best I can do is abt 10kv no matter what!

HP here's another question plz just tell me if I'm correct to say that desirability of resonant design is cuz of _softer_ switching? Also what determines frequency of oscillator when secondary is totally shorted?

HP just so you know breadboard kit has arrived at ups center so I can get it tomorrow! So tnx! BTW I still think you should post detailed pics of terminal strips for benefit of others interested in breadboarding hv circuits

 

@Aleph(0) -- Backatca latter on!

Best regards
HP

 

@Aleph(0) -- Backatca latter on!

Best regards
HP

HP So I picked up parcel with breadboard kit! TNX! Cuz I can tell it's high quality components! But I say a few more pics showing wired example circuit will be helpful so plz just breadboard simple circuit on your setup and post pics

 

But I say a few more pics showing wired example circuit will be helpful so plz just breadboard simple circuit on your setup and post pics

I'll get to it (i.e. uping terminal-strip detail images) when I'm good and ready! -- And NO! I'm not going to design/build your circuit for you! (Nice try!) ---@Aleph(0) now you've all the supplies required! -- I expect progress!

With regards
HP

 

Hey @Aleph(0) -- I apologize that I (inadvertently) neglected your inquiries (addressed below) --- BTW please advise me as to whether you yet require the breadboard-related images? -- If not I'll 'up' them when we get back to the tutorials...

Re: cross-coupled switch topology:

So now I'm thinking that's rationale for mazilli circuit which I say is just cc circuit modified so drain voltage can be higher than gate voltage.

Essentially correct...

HP now I've got to say I'm just not impressed with mazilli circuit at all! Like I said b4 it's strong on current and totally lame on voltage!

HP plz just tell me how to overcome that cuz yours is 50kv ac right off the secondary with white arcs! Best I can do is abt 10kv no matter what!

Generally speaking; Well conceived Mazzilli topology/design applied to EHT 'production' will feature greater primary impedance than strictly necessarily with 'straight' resonant Royer implementations (all else being equal)...

HP here's another question plz just tell me if I'm correct to say that desirability of resonant design is cuz of _softer_ switching?

That and sinusoidal output...

Also what determines frequency of oscillator when secondary is totally shorted?

Would you believe... the leakage inductance and 'tank' capacitance!?!? -- Kindly dare to think, Aleph!

Best regards
HP