Phase shift PWM / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Phase shift PWM

1 2 next

Move Thread

LAN_403

Linas

Sun Sept 11 2011, 05:31PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Hello.
right no i am trying to make reliable 2KW smps for my 120KV x ray tube, and i am searching best controller possible.
after digging in TI and LT i find phase shift controllers for full bridge. but they cost around 20$ just for dip20 chip
is it worth it to give a try for them ? right now just finished design for SG3527 PWM
note: i don't need stabilization

Electra

Wed Sept 14 2011, 07:35PM

Registered Member #816 Joined: Sun Jun 03 2007, 07:29PM
Location:
Posts: 156

Not sure if itâ€™s the SG3875 phase shift modulator ic you were planning on using, but if you want to experiment, I remember seeing an application note where you can make one from a D flip/flop and some Ex-or gates driven from a Pwm ic.
(just found it again, intersil.com/data/an/an9506.pdf )

Since the phase is shifted with respect to the other side of the full bridge the gate drive remains a square wave so this may make the gate drive easier if you are passing it through a transformer.
I think the other advantage is shorter pulse widths can be achieved, as the other device on the opposite side of the bridge is all ready on or off.

Sounds like a reasonable choice for a kw level supply, The only disadvantages I can think of with this method is it may not be easy or possible to implement the cycle by cycle kind of peak current limiting.

When you say you donâ€™t need stabilization, do you mean you just want to control the duty cycle from a pot or something, no feedback.

Linas

Wed Sept 14 2011, 08:30PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

i need 50:50 duty cycle
well, maybe turn off then current go to high, no more...

jpsmith123

Wed Sept 14 2011, 08:43PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

You're working on the circuit to drive your transformer?

The transformer secondary winding is already finished and encapsulated in epoxy?

Linas

Thu Sept 15 2011, 05:27AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Yes, i need H bridge controller and topology to drive as reliable as possible .
From bridge side, i was told is good idea add 10nF on mosfet DS to reduce dv/dt, it may save mosfet
well it was, but now i dissemble it, because i need two additional turns. now will have 7 turns for primary ( around 20V/t )

Shrad

Thu Sept 15 2011, 07:37AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

could you please tell me if your SG3527 is the same pinout/characteristics as the SG3524/3525 ? I'm looking for clear application examples for this model and can't seem to find what I need (your example would be nice to have, for instance, as I'm looking to build a negative DC 2kV 200mA power supply with current control for some helium-cadmium laser heads)

Dr. ISOTOP

Thu Sept 15 2011, 08:59AM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

If you don't want feedback, a TL494 is a very reasonable way to drive a bridge.
If you do, then things get harder, since PWM'ing a soft-switched inverter is hard.

Linas

Thu Sept 15 2011, 10:33AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

SG3525=inverted output SG3527
because i have lot of TC4421 ( they are inverting) so i need inverting output to drive GDT .
Well, i don't know, it would be nice to have current stabilization, because transformer can draw as many current as my fet's can handle
but as far as i know, stabilization will cause a lot of problems

I think maybe i should add 1-2s shutdown if current go to high

Shrad

Thu Sept 15 2011, 09:08PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

my current feedback will be derived from a measurement resistor of say 1K 5W, and the whole HV part has to be positive ground, so for me to have a common ground with the LV part, I'll have to invert everything past the transformer

I'm still hesitating between a hall current sensor or a measurement resistor with an optocoupler powered by a small tertiary winding

jpsmith123

Fri Sept 16 2011, 12:07PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I'm leaning toward using a SG3525 to drive my transformer.

I'm wondering if I'll have to add a buffer stage to drive four IRFP450s in a full-bridge configuration (in this case through two Coilcraft SD250-3L GDTs)? I suppose I should breadboard it and do some experimenting.

For overcurrent protection I will use a Triad CST306-1T current sense transformer.

Does anyone have any good snubber design info for a full-bridge configuration?

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob