Powered by Invision Power Board


Forum Rules Forum Rules (Please read before posting)
  Reply to this topicStart new topicStart Poll

> Opto-coupler At 100khz?, Driving smps MOSFETs at 100KHz?
treez
Posted: January 02, 2010 09:03 pm
Reply to this postQuote Post


Unregistered










Hello,
.
I wish to make an offline 2 Transistor forward converter for UK Mains input.

Switching frequency will be 100KHz.

Will it be OK for me to drive both MOSFETs ON with an opto-coupler?……….being more specific, I will use the opto-coupler to drive a MOSFET driver next to each mosfet…...their will be an isolated supply for the high-side mosfet driver

…..the “slowness” of the opto-coupler will be accounted for by the fact that the MOSFET driver input behaves like a Schmidt trigger and so both the MOSFETs will quickly be driver ON.

Is this a valid method,? –using an opto-coupler like this is OK?
Top
Sch3mat1c
Posted: January 02, 2010 09:29 pm
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 19,361
Member No.: 73
Joined: July 24, 2002




What opto?

I find even 6N136's suck in that sort of duty, and you won't get a damn thing out of a 4N25 at 100kHz.

Transformer is better. You're driving a transformer anyway, so it's not like you need DC coupling.

Alternatives include a bootstrap supply with source follower wiring, so you have a high voltage switch at the bottom (small MOSFET or BJT) pull down the top side gate driver input. It acts like a source follower, except it's a lot faster and saturates as long as the bootstrap holds up. Downsides include large voltage swing and instability with any kind of driver chip (a good old fashioned emitter follower will work better here).

Tim


--------------------
Answering questions is a tricky subject to practice. Not due to the difficulty of formulating or locating answers, but due to the human inability of asking the right questions; a skill that, were one to possess, would put them in the "answering" category.
PMEmail PosterUsers Website
Top
treez
Posted: January 03, 2010 03:47 pm
Reply to this postQuote Post


Unregistered










Top
Sch3mat1c
Posted: January 03, 2010 10:26 pm
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 19,361
Member No.: 73
Joined: July 24, 2002




Yeah, t_on ~ 3us, and that's with a 100 ohm load -- fractional volt output, not saturated. Most optos are on the order of 20us when used as a saturated switch.

The funny part is, with specified values (like, they might give I_f = 20mA, R_L = 4.7k and Vcc = 5 or 10V), rise time is about five times faster than fall time. You can strip out all that storage charge faster by adding a base-emitter resistor: this reduces CTR, but improves switch times greatly.

This thing switches in about 5us:
http://webpages.charter.net/dawill/Images/...tring%20(2).gif
and it's driving a small MOSFET. Turnoff is assisted with a PNP. It runs at 5-10kHz, so the rise time is good enough. It would be possible to add an LC filter if it were faster (>100kHz), which would be nice as this thing definitely puts some trash on the local electric field (it's no help that 5kHz is audible, fortunately my audio cables are shielded!).

As far as optical coupling goes, your only hope is basically a photodiode amplifier tuned for ~0.1ns response time. Faster would be even better. The 6N136 is a start, but its transistor is sensitive to dV/dt, it really needs to be better shielded. Could be faster anyway.

Optically, it's possible to do, but that doesn't mean it's a good idea. There are better ways, direct or transformer coupled.

Tim


--------------------
Answering questions is a tricky subject to practice. Not due to the difficulty of formulating or locating answers, but due to the human inability of asking the right questions; a skill that, were one to possess, would put them in the "answering" category.
PMEmail PosterUsers Website
Top
GPG
Posted: January 04, 2010 01:57 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Spamminator Taskforce
Posts: 6,216
Member No.: 792
Joined: April 25, 2004




PMEmail Poster
Top
Sch3mat1c
Posted: January 04, 2010 04:21 am
Reply to this postQuote Post


Forum Addict ++
Group Icon

Group: Moderators
Posts: 19,361
Member No.: 73
Joined: July 24, 2002




Oooh, I'll have to order some of those. Everything I've been doing in discrete gate drives. Except the desat detector, but hey, that's still only an extra LM393.

Tim


--------------------
Answering questions is a tricky subject to practice. Not due to the difficulty of formulating or locating answers, but due to the human inability of asking the right questions; a skill that, were one to possess, would put them in the "answering" category.
PMEmail PosterUsers Website
Top
johansen
Posted: January 04, 2010 05:14 am
Reply to this postQuote Post


Forum Addict ++
*******

Group: Trusted Members
Posts: 3,087
Member No.: 10,912
Joined: September 06, 2007




I've been cooking up an h bridge today, using some P80NF12 Fets, fets with 7 nf gate capacitance. (measured, datasheet says 5)
Complementary c945 and a733 transistor gate drive is getting me at best .5us rise & fall times, 5 times longer than it needs to be...

Another thought: use two opto isolators, one to turn it on, the other to turn it off.


--------------------
Can the Ethics Committee have a closed door meeting without imploding from cognitive dissonance?
PMEmail PosterUsers Website
Top
0 User(s) are reading this topic (0 Guests and 0 Anonymous Users)
0 Members:

Topic Options Reply to this topicStart new topicStart Poll

 


:: support us ::




ElectronicsSkin by DutchDork & The-Force