Author Topic: Configuration of a half bridge.  (Read 957 times)

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Configuration of a half bridge.
« on: July 07, 2020, 09:07:10 PM »
In the search for a desirable driver chip for a ressonant converter, I have come upon a problem which seems to suggest, that I have been missing something very fundamental for all this time.

I have always taken it for granted, that a half bridge converter would be configured like this:





What I focus on here is, that the tank circuit is connected to the midpoint of the power switches in one end, and a capacitively derived midpoint between the power rails in the other end.

However, in the datasheets, I see this circuit:





Here the tank is connected to the power switches midpoint and the power rail ground.
Even on more elaborate application note diagrams, this odd configuration is used, in this case the tank goes from power switch midpoint to power rail high.





What I am missing is this: Instead of reversing the voltage across the tank, as in the first example, the latter two seems to charge the tank in the first half cycle, then discharge it in the second half cycle.
I have a problem seeing these two variations as being equivalent.

Any help is appreciated here.

Cheers, Finn Hammer
« Last Edit: July 07, 2020, 09:08:59 PM by hammertone »

Offline johnf

  • High Voltage Technician
  • ***
  • Posts: 108
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #1 on: July 07, 2020, 09:20:12 PM »
Finn
The circuits are essentially the same the first using two capacitors is less ideal because the resonating caps are usually expensive.
You need two caps in your first example because you are dc isolating the transformer primary.
Its cheaper to DC isolate the half bridge (one capacitor) and earth one end of the primary. For EMI reasons earthing one end of the primary has advantages if the transformer is wound correctly

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #2 on: July 07, 2020, 09:36:53 PM »
John,

Thanks for your quick answer.

I was not thinking of the two capacitors between the rails as ressonating capacitors, but rather as large electrolytics.
In the first example, there will be half buss voltage across tank, leading to a smaller inductance in the transformer primary, whereas the latter two will see full buss voltage, and thus a larger inductance. This may be beneficial with regard to the ressonant tank capacitor size.

But I am courious about your last remark, how do you define an incorrectly wound transformer.

Cheers, Finn hammer

Offline ElectroXa

  • High Voltage Enthusiast
  • *
  • Posts: 27
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #3 on: July 07, 2020, 10:09:22 PM »
But I am courious about your last remark, how do you define an incorrectly wound transformer.


Hi

I would characterize an incorrectly wound transformer as a transformer having either a high leakage inductance (causing ringing with parasitic capacitances for example), a loose coupling between primary and secondary, core saturation, and / or other defects (like too much interwinding capacitance, etc...)  ;)

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #4 on: July 07, 2020, 10:28:37 PM »
ElectroXa,

Core saturation should not be an issue in a high frequency converter >20kHz, where the magnetics are limited by core loss, at a flux around 200 - 300mT.
Instead, the converter being of the ressonant variety, leakage inductance caused by low coupling is much desired.

Cheers, Finn Hammer

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #5 on: July 08, 2020, 08:39:10 AM »
All,

Quote
In the first example, there will be half buss voltage across tank, leading to a smaller inductance in the transformer primary, whereas the latter two will see full buss voltage, and thus a larger inductance. This may be beneficial with regard to the ressonant tank capacitor size.

Edit:

My bad, from answering the reply too quick, before actually understanding the implication of it, like:
from +50% to -50% equals from 100% to 0%. -Still need to try it out in practice to get used to the consept, though.

Cheers, Finn Hammer

Offline ElectroXa

  • High Voltage Enthusiast
  • *
  • Posts: 27
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #6 on: July 08, 2020, 10:03:25 AM »
my bad, I completely forgot the resonant characteristics of these half bridge, where leakage inductance is welcome, focalizing on hard switched half bridge inverters  :(

sorry if my previous message causes misunderstanding

Offline Mads Barnkob

  • Administrator
  • High Voltage Expert
  • *****
  • Posts: 1457
  • Karma: +24/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Configuration of a half bridge.
« Reply #7 on: July 08, 2020, 04:25:53 PM »
Hi Finn

Good to hear from you again :)

We actually discussed this only a few weeks back, or just check the date, here: https://highvoltageforum.net/index.php?topic=1078.msg7723#msg7723

It depends on the load you are driving, which configuration can be used and in regard to the tank resonance, it is never a resonant circuit as the DC blocking capacitor is chosen to work as such and not a resonant LC circuit.

My DRSSTC4 was initially a midpoint to ground half-bridge, but that did not turn out well with a LC load and I had to add 2x 10uF capacitively coupled midpoint for the load. I think a common mistake is to think that the capacitive voltage divider is only needed if you are using a voltage multiplier on the input, which Steve Ward always did in his SSTC designs, but it actually has to do with the driving topology and load characteristics. Resistive loads does not care about having to be discharged like a L, C or LC circuit would, so you commonly see these midpoint to ground half-bridge in DC motor drivers.
http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics
https://www.youtube.com/KaizerPowerElectronicsDk60/join - Please consider supporting the forum, websites and youtube channel!

Offline johnf

  • High Voltage Technician
  • ***
  • Posts: 108
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #8 on: July 08, 2020, 09:00:04 PM »
Mads
The clue here was resonant converter
hence my answer
in reference to winding order I have put the second half of a primary on last on many high frequency transformers with the earthy end outermost so as to shield the noisy internal bits---- gives less for the external flux shorting band to do

Offline Mads Barnkob

  • Administrator
  • High Voltage Expert
  • *****
  • Posts: 1457
  • Karma: +24/-0
  • Denmark
    • View Profile
    • Kaizer Power Electronics
Re: Configuration of a half bridge.
« Reply #9 on: July 08, 2020, 09:43:20 PM »
Mads
The clue here was resonant converter
hence my answer
in reference to winding order I have put the second half of a primary on last on many high frequency transformers with the earthy end outermost so as to shield the noisy internal bits---- gives less for the external flux shorting band to do

Miss one little word and its all a different world :)

Regarding the schematics, they are not completely comparable, are these really operating with the same gate control?

1) A 2-element series resonant half-bridge converter
2) A lower leg 2-element series resonant half-bridge converter (watch out for which lines are VDD and VDC)
3) A upper leg 3-element series resonant half-bridge converter

http://www.kaizerpowerelectronics.dk - Tesla coils, high voltage, pulse power, audio and general electronics
https://www.youtube.com/KaizerPowerElectronicsDk60/join - Please consider supporting the forum, websites and youtube channel!

Offline johnf

  • High Voltage Technician
  • ***
  • Posts: 108
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #10 on: July 09, 2020, 08:37:52 AM »
Mads
I agree
all very general
still a good first post

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #11 on: July 09, 2020, 11:46:10 AM »
Mads, JohnF,
Let me tell what It is that I am trying to do:
I want to build a 1000W powersupply, 50kV, 20mA.
I know this is very ambitious, but then, that is me.
It will be based on a the LLC topology, full SiC switches and clamp diodes,and will employ a 5kV secondary transformer with a number of CW doubler stages.
I have done the first prototype, first based on a full bridge, later shaved down to a half bridge.
To test it, I use this load with a transformer that has 1000V output:

.

 I used the IR2086 controll chip, backed up by UCC21520 gate drivers to get the ultimate of sub50nS delay times, "just because I like it".
To get the high voltage transformer wound, I have a MoReCo winder, which works well with silk covered wire:







But not forgetting the inevitable pitstop of every ambitious transformer coil winder:



Which leads to the necessity of vacuum potting the secondary pies:



Anyway, to load the supply down in high voltage mode, I have this load:



At this point I want to order the second prototype board, and thus I started to search for a more suitable driver chip.
This chip should have various protection modes, preferably non-latching, and soft start when it recovers.
I have already decided on the FAN7387, it looks like a very good controller for this job.

It was in the search for this driver chip, that I notiched the possibility of tying the far end of the load to the rail directly, and this at first looked very odd.
But since everybody says it is how to do it, I will give it a go.
I vaguely remember mention of splitting the ressonant capacitor into two, each going to the other rail, but I have forgotten where I read it. Probably one of the more general texts on LLC converters.

I hope this clears up something.

Cheers, Finn Hammer





Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #12 on: July 09, 2020, 12:16:32 PM »
Sorry for double posting, but I could not make the "edit" button perform its function.

I just found the text, which goes into some detail about splitting the ressonant capacitor in two, and spreading them across both rails, the link leads to a download of the paper AN2644:
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwj859rt9L_qAhUiJcUKHUIQAf8QFjAAegQIAxAB&url=https%3A%2F%2Fwww.st.com%2Fresource%2Fen%2Fapplication_note%2Fcd00174208-an-introduction-to-llc-resonant-halfbridge-converter-stmicroelectronics.pdf&usg=AOvVaw3dAe-EM99J39v6scxv91EF





It is worth a read.

Cheers, Finn Hammer

Offline johnf

  • High Voltage Technician
  • ***
  • Posts: 108
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #13 on: July 09, 2020, 09:23:38 PM »
Finn
Now I see
and yes splitting the cap in two is a good idea
this is not a cheap build so an extra cap to take some strain off it is a good idea

I probably would have done this as a current fed current mode inverter ie overlapping conduction in the bridge.
fewer transients and better control of leakage inductance artifacts and of course little or no need for snubbers on the bridge devices along with better control of current / power during arc over events

Offline davekni

  • High Voltage Engineer
  • ****
  • Posts: 466
  • Karma: +26/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Configuration of a half bridge.
« Reply #14 on: July 10, 2020, 06:57:36 AM »
Yes, I like the split resonant capacitors.  Done that a couple times for exactly that reason - to reduce VBus ripple current.

Interesting idea of a current-fed bridge or half-bridge.  By current-fed, is that inductor-fed like the ZVS  (Royer) oscillator circuits?  Would make sense, and definitely something I need to think about more for future designs.
David Knierim

Offline johnf

  • High Voltage Technician
  • ***
  • Posts: 108
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #15 on: July 10, 2020, 08:48:33 AM »
Dave
Yes
so the input stage to the bridge is fed from the first stage inductor output
real clever people also make the current feed stage the PFC as well
outside my experience adding the PFC

Offline hammertone

  • High Voltage Enthusiast
  • *
  • Posts: 16
  • Karma: +1/-0
    • View Profile
Re: Configuration of a half bridge.
« Reply #16 on: July 10, 2020, 08:55:30 AM »
Davekni, Johnf,

The buck current fed bridge was my first choice too, and I have the UC3827 driver chips to do it with. I made one smaller version a few years ago.
However, I got put off by the noise at the buck switching node, and decided to try the ZVS operation of the LLC converter.
Now there is no way back.
I have had the plant working at 100kHz @1kW, mosfets running lukewarm without heatsinking, and absolutely clean voltage transitions, it is a marvel to behold.

With regard to the split capacitor, I have read further in AN2644, under "Converter startup", where my initial concern is acknowledged (the initial double voltage across tank), but there is also an explanation of the following build up of a DC level on the cap, which I had not deduced myself, so all well.

-----------------------------------------------------------
This phenomenon is well-known in push-pull and half-bridge topologies and is sometimes
referred to as "flux doubling" because the transformer's magnetic flux excursion, which
normally swings by 2Bpk from -Bpk to +Bpk under steady state operation, in the first cycle
goes from 0 to 2Bpk. Fortunately, if the integrated magnetics approach is used, there will be
no risk of saturation. Flux doubling will concern only the resonant inductor Ls, which is
mostly associated to the transformer's leakage inductance that, by definition, cannot
saturate because the relevant flux is developed in air.
Although not inherently hazardous, (capacitive mode and ZVS loss will typically occur in the
first two-three cycles, then the associated stress level is practically negligible) this
phenomenon is not nice to see.

-----------------------------------------------
Anyway, the split cap it will be, and with Diode Clamps too, which will clamp cap viltage excurtions beyond the rail voltage, a situation occurring during output shorting. The following explanatory schematic appears to have one ressonant  capacitor too many, though



I greatly appreciate all input in this discussion, it has helped me in opening my eyes to aspects of the LLC whitch were previously hidden to me.

Cheers, Finn Hammer
« Last Edit: July 10, 2020, 09:02:41 AM by hammertone »

Offline davekni

  • High Voltage Engineer
  • ****
  • Posts: 466
  • Karma: +26/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Configuration of a half bridge.
« Reply #17 on: July 28, 2020, 04:16:20 AM »
Just happened to be reading that same application note AN2644 that you quote, so recognized the excerpt.  It's not correct.  Leakage inductance usually does include the ferrite core.  The saturation current for leakage inductance is higher than for the primary parallel inductance, but not infinite.  Consider the typical configuration as you have, U-cores with primary winding on one leg and secondary winding on other leg.  Even with the secondary shorted, primary flux is passing through the ferrite inside the winding, then from the exposed remainder of the core through air to the other side.  The air path has much larger cross-sectional area than the core.  Thus inductance will drop quite significantly as the part of the core inside the primary winding saturates.

The primary saturation current with secondary shorted is much higher than with the secondary open, roughly by the ratio of primary inductance for those two cases.  If leakage inductance is 20%, saturation current will be a bit over 5x.  Saturation is unlikely, but not impossible.

This entire post is probably not relevant to anything practical.  I just couldn't get that application note's misstatement off my mind.
David Knierim

High Voltage Forum

Re: Configuration of a half bridge.
« Reply #17 on: July 28, 2020, 04:16:20 AM »

 


* Recent Topics and Posts

post Re: making HV bobbins for ferrite transformers
[Transformer (Ferrite Core)]
davekni
Today at 06:20:04 AM
post Re: Welcome new members, come say hello and tell a little about yourself :)
[General Chat]
RoadReaper
Today at 05:40:19 AM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
Max
Today at 02:07:26 AM
post making HV bobbins for ferrite transformers
[Transformer (Ferrite Core)]
Teravolt
Today at 01:27:22 AM
post Re: A Tesla coil and an antenna
[General Chat]
johnf
August 09, 2020, 09:23:30 PM
post A Tesla coil and an antenna
[General Chat]
TheFreq
August 09, 2020, 09:02:32 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Steve Ward
August 08, 2020, 10:22:04 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
T3sl4co1l
August 08, 2020, 03:36:40 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
acmq
August 08, 2020, 02:07:52 AM
post Re: LSI Nitrogen Laser tear-down and Homemade TEA Nitrogen and Dye Lasers
[Light, Lasers and Optics]
LesWright
August 06, 2020, 10:20:12 PM
post Re: LSI Nitrogen Laser tear-down and Homemade TEA Nitrogen and Dye Lasers
[Light, Lasers and Optics]
Weston
August 06, 2020, 08:45:59 PM
post Re: Cathodes in Compact Fluorescent Lamps
[Light, Lasers and Optics]
klugesmith
August 06, 2020, 05:41:04 PM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
Jun
August 06, 2020, 03:53:58 PM
post LSI Nitrogen Laser tear-down and Homemade TEA Nitrogen and Dye Lasers
[Light, Lasers and Optics]
LesWright
August 06, 2020, 03:28:14 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
AstRii
August 06, 2020, 12:59:03 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
Mads Barnkob
August 06, 2020, 10:51:43 AM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
UB_Serpent
August 06, 2020, 08:40:57 AM
post Re: compact midi interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
TMaxElectronics
August 06, 2020, 03:22:38 AM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
davekni
August 05, 2020, 11:48:55 PM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
davekni
August 05, 2020, 11:37:13 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
AstRii
August 05, 2020, 08:01:11 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Lightning On Demand
August 05, 2020, 06:34:08 PM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
Uspring
August 05, 2020, 12:56:32 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Mads Barnkob
August 05, 2020, 12:55:52 PM
post Re: Bipolar high voltage multiplier project - more than 70cm hot arc discharge
[Voltage Multipliers]
Teravolt
August 04, 2020, 09:21:09 PM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
johnf
August 04, 2020, 09:16:55 PM
post Re: Does anyone know how this site works.
[General Chat]
thedoc298
August 04, 2020, 05:53:03 PM
post Re: Does anyone know how this site works.
[General Chat]
thedoc298
August 04, 2020, 05:51:03 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Lightning On Demand
August 04, 2020, 05:17:26 PM
post Re: Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Max
August 04, 2020, 02:14:56 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
Max
August 04, 2020, 02:08:07 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
futurist
August 04, 2020, 10:48:35 AM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
UB_Serpent
August 04, 2020, 10:02:39 AM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
Jun
August 04, 2020, 08:47:07 AM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
Jun
August 04, 2020, 08:43:00 AM
post Re: Does anyone know how this site works.
[General Chat]
Mads Barnkob
August 04, 2020, 08:33:47 AM
post Re: Does anyone know how this site works.
[General Chat]
MRMILSTAR
August 04, 2020, 05:09:26 AM
post Does anyone know how this site works.
[General Chat]
thedoc298
August 04, 2020, 01:54:49 AM
post Fun with wireless power
[Solid State Tesla Coils (SSTC)]
Lightning On Demand
August 03, 2020, 11:55:41 PM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
Teravolt
August 03, 2020, 09:17:40 PM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
MRMILSTAR
August 03, 2020, 08:30:38 PM
post Re: Budget USB cameras
[High Speed Filming]
missoum0514
August 03, 2020, 06:56:08 PM
post Re: Dual MOT tesla coil
[Spark Gap Tesla Coils (SGTC)]
Jun
August 03, 2020, 05:23:15 PM
post Re: Is my xray transformer a ferrite core and how would I drive it?
[Transformer (Ferrite Core)]
nielsquake
August 03, 2020, 05:02:26 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
Max
August 03, 2020, 04:31:04 PM
post Re: Easy to build Polyphonic MIDI Interrupter
[Computers, Microcontrollers, Programmable Logic, Interfaces and Displays]
futurist
August 03, 2020, 10:05:59 AM
post Re: Is my xray transformer a ferrite core and how would I drive it?
[Transformer (Ferrite Core)]
Mads Barnkob
August 03, 2020, 08:21:26 AM
post Re: Is my xray transformer a ferrite core and how would I drive it?
[Transformer (Ferrite Core)]
nielsquake
August 02, 2020, 11:45:38 PM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
davekni
August 02, 2020, 08:10:47 PM
post Re: How much is the voltage used in Corona Discharge tubes for ozone Generation?
[Transformer (Ferrite Core)]
Uspring
August 02, 2020, 07:27:33 PM

Sitemap 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 
SimplePortal 2.3.6 © 2008-2014, SimplePortal