New HF Driver Idea

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

New HF Driver Idea

Post by Liam David »

I watched a video a couple days ago about H-bridges and how they can switch the direction of a motor's rotation. I got the idea that I could build a DC to AC converter, but possibly drive it up to high frequencies using a timer or oscillator. Here is my first circuit schematic.
Image

The resistors in the 555 timer part would be determined by the formula above the schematic. What would be a good target frequency? 10kHz, 100kHz?
Would this work, or would I end up blowing everything out?

-Liam David
User avatar
Werner Engel
Posts: 145
Joined: Sun Aug 18, 2013 11:51 am
Real name: Werner Engel
Location: Vienna, Austria
Contact:

Re: New HF Driver Idea

Post by Werner Engel »

The frequency you should use depends on a lot of parameters.
There is the inductivity, the material of the core, the load, the geometric of the core (air gap), ...

I found out it's quite difficult to build a HV transformer.
Try to vary the frequency and measure the voltage.
Working on it - see the picture.

BR,
Werner
my ferrit core with air gap of about 6mm (thanks to Philipp for helping me winding the wires)
my ferrit core with air gap of about 6mm (thanks to Philipp for helping me winding the wires)
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

The transformer I'm trying to build this for is one I bought from Andrew Robinson.
Image

The core is 2 separate pieces and about 1 inch square. You might be able to see a small bump on the left side of the core where it separates. The gap between the coils and the core is as thick as the plastic case, so about 5mm.
I guess I'd better find some core saturation and induction equations.
User avatar
Werner Engel
Posts: 145
Joined: Sun Aug 18, 2013 11:51 am
Real name: Werner Engel
Location: Vienna, Austria
Contact:

Re: New HF Driver Idea

Post by Werner Engel »

And what is the core made of?
I used the following: http://uk.rs-online.com/web/p/transform ... s/0180834/
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: New HF Driver Idea

Post by Chris Bradley »

You are re-inventing what does not need re-inventing if you do not use FET drivers. Your 'NPN' transistor arrangement looks poor for a variety of reasons that you can find in most good texts on the subjects. Save your efforts and buy high current (for maximum efficiency) non-inverting (to avoid feedback) gate drivers.
User avatar
Bob Reite
Posts: 576
Joined: Sun Aug 25, 2013 9:03 pm
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: New HF Driver Idea

Post by Bob Reite »

You have to be careful that the two sides of the bridge do not turn on at the same time. That's how MOSFETs get blown out. I would duplicate Steve Ward's "High Votage Capacitor Charging PSU" circuit shown at http://www.stevehv.4hv.org/ccps1.htm
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

That's why I (tried) to add a NOT gate for one of the bridge sides, to ensure that I don't get a short. Only one side is on at a time. It's the part of the circuit with the transistor labeled NPN. I'll look into Steve's design. Thanks for the lead.
User avatar
Bob Reite
Posts: 576
Joined: Sun Aug 25, 2013 9:03 pm
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: New HF Driver Idea

Post by Bob Reite »

You need to have a few microseconds delay between the turn off of one side and the turn on of the other.
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

I have seen that there are a couple of microseconds of shoot-through every time the circuit switches, and with 120v, this might be an issue. I'll see how I can build in a delay between the switches.
User avatar
Rich Feldman
Posts: 1470
Joined: Mon Dec 21, 2009 6:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: New HF Driver Idea

Post by Rich Feldman »

Hi Liam.
The core in your transformer picture has rounded corners, a style I'm used to seeing in tape-wound laminated steel cores.
To follow Werner Engel's question, are you sure it's ferrite? The magnetic properties and the efficient frequency range are very different for steel.

Here is one of many reviews of core material categories.
http://www.ti.com/lit/ml/slup124/slup124.pdf

Either way: since you can take the core apart, I suggest that you put it together
without the high voltage windings, while you use it in experiments for learning.
You can make your own experimental windings on folded cardboard bobbins.
The electrical behavior of a primary winding
with an unloaded (open circuit) secondary
is practically the same as that with _no_ secondary.
All models are wrong; some models are useful. -- George Box
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: New HF Driver Idea

Post by Chris Bradley »

Liam David wrote:I have seen that there are a couple of microseconds of shoot-through every time the circuit switches, and with 120v, this might be an issue. I'll see how I can build in a delay between the switches.
Why not buy an H-bridge driver that has a built-in delay, like anyone else would?

(Sorry if I am not getting what the thread is about.)
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

The inside of the core is shiny and metallic, so it's definitely not ferrite. It might be steel. Unfortunately, I can't slide the windings off the core because the core is slightly up-turned where it separates.

Chris,
I'm just trying to build a cheap power supply using mostly parts that I already have laying around or still stuck on circuit boards. Purchasing parts for a ZVS driver capable of driving higher than 24V would cost me over $30, any while that isn't that much money, my limited budget tells me otherwise. What I'm trying to figure out is whether building this circuit and driving my transformer off of it would blow anything up.
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: New HF Driver Idea

Post by John Futter »

Liam

chris has given sage advice

you will end up with a bucket of dead silicon and probably the transformer if you continue on this train of thought. Switching psu design is for experts and even experts do not like high voltage such as you want

the best and cheapest option is to drive the transformer with a big audio amplifier with a sine wave generated from your 555 and filtered to get it sinusoidal
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

Well, in that case I'll drop the circuit and go with something "simpler" and less likely to blow up.
User avatar
Andrew Robinson
Moderator
Posts: 688
Joined: Sun Aug 11, 2013 1:54 am
Real name: Andrew Robinson
Location: Raleigh, North Carolina
Contact:

Re: New HF Driver Idea

Post by Andrew Robinson »

Hey Liam,

Did you get in touch with "One Tesla" like I recommended? They bought an identical core and have lots of HF experience. You might try asking them for some input on how they are driving their core.

PS, don't forget to submerse the core in oil. For some people, it's easy to forget this critical step when doing testing on the bench with drivers. DONT FORGET haha
I can wire anything directly into anything! I'm the professor!
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

Well, I posted on the OneTesla forums about the transformer about a month ago and the only reply I got was one recommending posting on 4hv.org. Did the founders/owners of OneTesla buy it?

I found a website that gives away free samples of CCFL transformers. http://www.coilcraft.com/ccfl.cfm. Since these transformers operate at up to 100kHz, would any of these work for driving a HF x-ray transformer (after sec. voltage is decreased by a lot)?
Just trying to find the cheapest and easiest way to drive the transformer.
Jeroen Vriesman
Posts: 276
Joined: Mon Feb 07, 2011 4:08 pm
Real name: Jeroen Vriesman
Location: Netherlands
Contact:

Re: New HF Driver Idea

Post by Jeroen Vriesman »

You could use a ready-made h-bridge driver to avoid shoot-through:

http://cds.linear.com/docs/en/datasheet/11602fb.pdf (max 60V...)

http://www.st.com/web/en/home/catalog/s ... 1501/SC911 (overview)

for example this one:

http://www.st.com/st-web-ui/static/acti ... 290377.pdf (600V, I-out is about 0.3A)

when driving bipolar transistors, see that I-out * hFe > the current you need.

When using mosfets, look at the gate capacity and see that Cgs * Vthreshold < I-out * ( required switching time) by at least factor 5 or so....very rough indication, the exact calculations are more complex, but LTspice simulations can be helpfull, keep in mind that the required switching time should be much less than the dead time (320 ns for the L6398)
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: New HF Driver Idea

Post by Jerry Biehler »

User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

If the CCFL transformers won't work, then I will use this full bridge gate driver: http://www.mouser.com/Search/ProductDet ... 453DSTRPBF. So, would the CCFL transformers I linked to work?
NicKolas Garrett
Posts: 8
Joined: Thu Jan 08, 2015 7:48 pm
Real name: nicholas garrett

Re: New HF Driver Idea

Post by NicKolas Garrett »

hey guys, I'm super new here! and i didn't expect to be posting so soon :-) but I'm thinking of a solution.
Inside of a crt television you will find just what i think you need. It's called a flyback transformer. The hallmark of the flyback transformer, is it's insane V/turn ratio and turn to turn ratio. You can easily achieve 5-10 volts per turn, and with an estimated 1:100 pri:sec ratio.... Well, lets just say thee lil guys are nice. They're epoxy encapsulated, they are internally rectified and they will happily output an excess of 45kv! On top of that they weigh less then a pound and can output 10's to 100's of miliamps. But the best thing of all! Is they are as common as the transistors/mosfets/igbts you need to drive them!!! you will find at least one suitable mosfet or IGBT very close to the transformer on the PCB inside the television.

Now the best results come from driving them with a (primary resonant or ZVS)fullbridge, And the frequency doesn't really mean much(at least 10khz). The important factor is the duty cycle you drive them at. The most simple and common circuit used for them is the mazzilli ciruit *see attached*. Which operates very close to 50% duty cycle.
This circuit is so simple you can build and have a functional forward converter in less then 1 hour.
Once well optimized these flyback circuit drivers can probably output around 500 watts, so be careful they can be lethal.

If your interested, here is some more information on some of the various ways to drive a flyback. (or any ferrite cored transformer)
http://uzzors2k.4hv.org/index.php?page= ... merdrivers

i hope this can help.
Attachments
Mazzilli ZVS Flyback Driver.png
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

Yeah, I looked over that ZVS circuit but settled for a high powered H-bridge because I can pump much more voltage into it. I'll hopefully put it together this weekend. I'm driving this transformer with the H-bridge.
Image
NicKolas Garrett
Posts: 8
Joined: Thu Jan 08, 2015 7:48 pm
Real name: nicholas garrett

Re: New HF Driver Idea

Post by NicKolas Garrett »

Liam David wrote:The inside of the core is shiny and metallic, so it's definitely not ferrite. It might be steel. Unfortunately, I can't slide the windings off the core because the core is slightly up-turned where it separates.
If this is so, and it is an iron lamination core.... you arent going to get much performance by switching it at anything above 400hz. iron's magnetic impedance is much too low for high frequency signals to permeate.
Rich Feldman wrote:Hi Liam.
The core in your transformer picture has rounded corners, a style I'm used to seeing in tape-wound laminated steel cores.
To follow Werner Engel's question, are you sure it's ferrite? The magnetic properties and the efficient frequency range are very different for steel.

Here is one of many reviews of core material categories.
http://www.ti.com/lit/ml/slup124/slup124.pdf

Either way: since you can take the core apart, I suggest that you put it together
without the high voltage windings, while you use it in experiments for learning.
You can make your own experimental windings on folded cardboard bobbins.
The electrical behavior of a primary winding
with an unloaded (open circuit) secondary
is practically the same as that with _no_ secondary.
^^ bump


You can also apply 5-15VAC(50-60hz) to the secondary and measure the primary's voltage with an oscilloscope. if it looks nice and sinusoidal you can assume the core is iron. And in this case, you SHOULD NOT connect it to an h-bridge. if its jagged and messy, its more likely powdered iron or ferrite. (high frequency) this is safe to do as the secondary's high impedance to the power source you are applying will not allow sufficient current to burn any windings, and will maintain safety because you're stepping down. However I would not suggest holding or touching the bare wire while making the connections since this will cause a voltage spike no mater what kind of transformer is it due to the inductive kick effect (widely used in flyback transformers)
User avatar
Andrew Robinson
Moderator
Posts: 688
Joined: Sun Aug 11, 2013 1:54 am
Real name: Andrew Robinson
Location: Raleigh, North Carolina
Contact:

Re: New HF Driver Idea

Post by Andrew Robinson »

Maybe I can shed a little more light on this subject. This transformer was indeed originally driven by an H Bridge. Unfortunately I donated it long ago to an NC State student for a research project.
12.jpg
12.jpg (28.41 KiB) Viewed 8302 times
9.jpg
I can wire anything directly into anything! I'm the professor!
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: New HF Driver Idea

Post by Liam David »

Well, since I do not have an oscilloscope and am assuming the core is iron, what should I drive this with?
User avatar
Nick Peskosky
Posts: 135
Joined: Wed Aug 08, 2012 9:02 pm
Real name: Nicholas Peskosky
Location: San Diego, CA

Re: New HF Driver Idea

Post by Nick Peskosky »

I would recommend obtaining even an old CRT Tektronix scope (<$100) if you are going to attempt HF driver design or work with PWM type circuits. Without one, you will be woefully unprepared to analyze circuit parameters even at the most basic of levels needed to optimize power matching and core saturation.
Nick Peskosky
NPeskosky@gmail.com

"The whole of science is nothing more than the refinement of everyday thinking." - Albert Einstein
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”