Switched Mode Power Supply

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.
Niels Geerits
Posts: 55
Joined: Tue Jun 13, 2017 10:25 pm
Real name: Niels Geerits

Re: Switched Mode Power Supply

Post by Niels Geerits » Sat Jun 24, 2017 9:33 pm

John. I noticed that with the Vdc/2 too. I left out the capacitors while simulating the circuit. Have almost gotten it to work as I thought it would just got to do some more tweaking.

Niels Geerits
Posts: 55
Joined: Tue Jun 13, 2017 10:25 pm
Real name: Niels Geerits

Re: Switched Mode Power Supply

Post by Niels Geerits » Fri Jun 30, 2017 2:16 pm

So most of my supplies have arrived (except high voltage wiring so I am limited to 1kV tests right now). My arduino is now creating two 300 khz 40% duty cycle out of phase switch signals (I measure the frequency using the arduino too). I have also been doing so reading: I found out that the topology I referred to in my initial post is called the flyback topology this only requires one MOSFET to function. I have also been thinking about the half bridge topology. I am having trouble finding sources on how it actually works. If MOSFETS were actually just switches it would be easy. A problem I can see is that in the drawing I posted here there is no load on the MOSFET drain. So I was wondering if anyone has any sources that I can read to learn more about the half bridge topology.

John Futter
Posts: 1233
Joined: Thu Apr 22, 2004 2:29 am
Real name:
Contact:

Re: Switched Mode Power Supply

Post by John Futter » Fri Jun 30, 2017 9:46 pm

300kHz is way too fast for HV supplies
this causes dIdt problems with the multiplier caps and the diodes
most modern HV supplies use 30 to 100kHz max with HV diodes for 100kHz costing many many dollars each

User avatar
Rich Feldman
Posts: 884
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Switched Mode Power Supply

Post by Rich Feldman » Fri Jun 30, 2017 10:47 pm

>> wondering if anyone has any sources that I can read to learn more about the half bridge topology.

Uh, LMGTFY. Here's the top result from a Google search for half bridge topology. http://www.irf.com/electronics/topology-fundamentals Looks like it would not be a bad place to start. It has some details I could quibble about, but not now.


>> in the drawing I posted here there is no load on the MOSFET drain

Unless you've changed that half-bridge schematic snippet, it shows no MOSFETs and no drains. (hint: IGBT)

Sure, you could build a switch bridge with the load connected to nothing but drains (or collectors), and DC rails connected to nothing but sources (or emitters). Just use complementary devices as the top side switches, and change the gate drive appropriately. In fact, that's the most popular by a factor of a billion or so.
cmos_inverter.gif
cmos_inverter.gif (2.18 KiB) Viewed 468 times
Or put the complementary devices on just the bottom side. Then all switches have drain (collector) on a rail, and source (emitter) on the output. Yet another set of drive waveforms. That option has been around since the dawn of BJT's.
complementary_class_B.gif
complementary_class_B.gif (11.01 KiB) Viewed 468 times
Richard Feldman

Niels Geerits
Posts: 55
Joined: Tue Jun 13, 2017 10:25 pm
Real name: Niels Geerits

Re: Switched Mode Power Supply

Post by Niels Geerits » Tue Jul 04, 2017 1:09 pm

Hey Rich I found that link too I just could not get that much out of it. You hinted at the IGBTs (never heard about these in any of my electronics courses which now I find strange), so I looked them up, got a few and played around with them. I think I get how they work now. It seems though that they are slower than MOSFETS.

When I was talking about drains earlier I assumed you could replace the IGBTs in the drawing with MOSFETS. However using a p-Mosfet and an n-Mosfet makes sense.

I booked some success this weekend. I put together a flyback driver (two mosfets, one low voltage attached to the arduino and one high voltage with the gate connected to the drain of the low voltage mosfet). I used a 100 ohm 50W resistor to limit current and measured the voltage on the secondary of my transformer. I measured using a simple multimeter. At first I saw nothing because the multimeter averages the voltage. So I added an LED (yes. I ended up destroying about 7 of them) because I don't have any spare diodes and then there was light. I put a resistor in series with the LED to measure the voltage drop. It seems I am getting a voltage ratio of 1:1 even though my turns ratio is 1:2. In addition if I increase the voltage over the primary the voltage drop over my 100 ohm resistor decreases. It makes sense for the voltage drop over the resistor to decrease as I increase the switching frequency as the current will not have enough time to rise to its maximum value (coil impedance etc.). I still need to make sense of the decrease in voltage drop when the overall voltage is increased (perhaps it has to do with the core being saturated?). I will do some math today and solve the DE for a simple LR-circuit with a pulsed input. And I will get some diodes and capacitors to double and rectify the output voltage.

Niels Geerits
Posts: 55
Joined: Tue Jun 13, 2017 10:25 pm
Real name: Niels Geerits

Re: Switched Mode Power Supply

Post by Niels Geerits » Sat Jul 08, 2017 8:02 pm

Another update: I bought some cheap 1kV diodes and capacitors and built a tripler for my test setup. Using this I have reached a new personal record voltage of 3kV (and I can still double the voltage on my variac). After making some arcs I got back to measuring the output voltage. I noticed that when using my MOSFET (IRFP250N) the system saturated somewhere at around 2kV (by saturated I mean the output voltage no longer increased linearly when I increased the Input voltage). Then I put in my IGBT (G4PC50FD, which isn't made for my frequency btw) and easily made it to 3kV. I could have gone further, like I said my variac was only half open. In both cases I was driving at around 200 kHz.

I am wondering why I was getting higher output voltages with the IGBT. Perhaps once you get close to the limits of your transistor it no longer performs as it should (MOSFET max voltage is 200V. I went pretty much right up there and the MOSFET was getting very hot), as a result the MOSFET may not have actually been switching at 200kHz.

User avatar
Richard Hull
Site Admin
Posts: 10625
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Switched Mode Power Supply

Post by Richard Hull » Sun Jul 09, 2017 7:56 am

Yes, tone that frequency down. 25-50khz is a good bet with the lower frequencies working better in the kilowatt handling range.
The magnetics are also less of an issue.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Bob Reite
Posts: 251
Joined: Mon Aug 26, 2013 1:03 am
Real name: Bob Reite
Location: Wilkes Barre/Scranton area

Re: Switched Mode Power Supply

Post by Bob Reite » Mon Jul 10, 2017 3:03 am

My supply seems to work best at 30 KHz, due to the transformer characteristics.
The more reactive the materials, the more spectacular the failures.
The testing isn't over until the prototype is destroyed.

User avatar
Rich Feldman
Posts: 884
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Switched Mode Power Supply

Post by Rich Feldman » Mon Jul 10, 2017 2:24 pm

Niels, you can get better answers here if you show us a schematic. Whole circuit, including voltage tripler, and your attenuator for HV measurement. I don't know how the voltage triplers are configured in CRT anode power circuits.

>> why I was getting higher output voltages with the IGBT?

The actual breakdown voltage of your IGBT specimen might be higher than that of your MOSFET.

In normal operation of a flyback converter, the secondary current needs to have a DC component (time average of instantaneous current). Its magnitude is the same as the DC current in primary winding divided by the turns ratio, N. In every cycle: When the primary switching device turns off, the secondary voltage is that at which the load conducts. Or N times the voltage at which the primary switch breaks down, if that happens first. Somebody else made that comment recently -- perhaps on another forum I read.

As others have said here: 200 kHz or even 100 kHz is a stupidly high switching frequency, especially for IGBTs, and especially for beginners, even beginners with oscilloscopes. I was just working on a power issue in a board with about a dozen buck converters running at 500 kHz, and could post some salient details if you want.
Richard Feldman

Niels Geerits
Posts: 55
Joined: Tue Jun 13, 2017 10:25 pm
Real name: Niels Geerits

Re: Switched Mode Power Supply

Post by Niels Geerits » Tue Jul 11, 2017 11:18 am

Hey all, I followed your advice: first I lowered the frequency to 100 kHz and noticed I was getting a higher V_out:V_in, however the V_out maxed out for lower V_in. I then lowered the frequency further to 50 kHz, which had the same effect, higher V_out:V_in and lower max V_in. I then increased the number of turns on the primary side which did very little for my max V_in. Then I added a 15nF cap and 100 ohm resistor across the primary to act as a filter against Deltas (primary inductance estimated at 1-5mH). This decreased my V_out:V_in (not in resonance), however it did increase V_in max. So I am thinking about adding an inductor with a known inductance and using that as my filter inductor instead of my unknown primary inductor.

Another sidenote I might upgrade my circuit to a push pull topology, however I see no need for that yet.

As for my circuit diagram:

Image

and

Image

Note I haven't drawn in the filter cap and resistor yet, but it is simple enough to imagine I think. I use a multimeter to measure the current through the 2M\Omega resistor. I have also used a 2M ohm followed by a 100k ohm resistor and used a multimeter to measure the voltage across the 100k ohm resistor. I measured a max voltage of around 3kV.

Post Reply