2
\$\begingroup\$

enter image description here

Imagine that the AC1-2 is 170VAC at 150kHz~.

I simply want to turn the MOSFETs on, and let the AC flow through a inductive-capacitive load.

1) If I can turn the MOSFETs on to saturation by applying a control voltage, is it okay just to leave them in saturation for however long I want to have the load active?

enter image description here

2) If I cannot turn the MOSFETs on to saturation by applying a control voltage, is it okay to use a bootstrap setup switching at the same frequency as the AC flowing through AC1-2 for however long I want to have the load active?

enter image description here

3) If all of this ends up being kind of shitty, what would you recommend I do? I kind of just want to have something that can pass 150kHz 170VAC as if it was a 'wire' i could 'turn on'. It's all just to add capacitance to a high voltage inductor.

enter image description here

At this point, any help would be appreciated, and I hope I've been clear.

\$\endgroup\$
  • \$\begingroup\$ I think I might be asking too many questions at a time or something. \$\endgroup\$ – ARMATAV Mar 2 '15 at 22:46
2
\$\begingroup\$

1) Yes: however, your driver must be isolated as the gate nodes need to be Vgs(on) above the common source node. Photovoltaic isolator ICs or driver-transformer-rectifier combinations are usually used for this task -- either way, it's a switchable, isolated, low-current 10V gate supply. You'll want to work out how much current the isolated gate supply must pull, though (Qg is your friend here), so that you can determine if a PVI IC will suffice, or if the more-parts approach (transformer driver IC, small transformer, and rectifier) is needed.

2) N/A -- 1) is certainly possible

3) A mechanical relay could be used instead; however, you would need to be very careful to snub the relay's contacts correctly lest you get a pile of arcs and sparks every time the relay switches.

\$\endgroup\$
  • \$\begingroup\$ Gotcha. I'll start working on it. I'm kind of confused on something though, when you say 'isolated because it needs to be Vgs(on) above Vs', does that mean that this circuit should not have common ground with the H-bridge that is providing power to it? Or do you mean I should simply have the 10V going through a transformer and still heading to common ground? Because I'm assuming that if the latter is true, the circuit will never actually be 10V above Vgs(on), and it would need to be driven by a bootstrap configuration in order to hit that Vgs(on) + 10V which is 170 + 10V. Like an H-bridge. \$\endgroup\$ – ARMATAV Mar 3 '15 at 0:01
  • \$\begingroup\$ And AFAIK you can't bootstrap 100% duty cycle. \$\endgroup\$ – ARMATAV Mar 3 '15 at 0:02
  • 1
    \$\begingroup\$ Kaboom, imgur.com/Rf39pmx I think I'm done here. I hope the BSR has enough current to open the little transistors that will provide current to the PVIs. \$\endgroup\$ – ARMATAV Mar 4 '15 at 7:04
  • 1
    \$\begingroup\$ @ARMATAV -- it was a figure of speech :) it'd cross-couple from one channel to another, causing everything to quit working. \$\endgroup\$ – ThreePhaseEel Apr 25 '15 at 15:20
  • 1
    \$\begingroup\$ Actually, I think it's super easy and I'm just looking at it wrong. I've totally got this now. I will use your MAX253. I still wish I could just use a second wall wart and opto-coupler plus transistor control though. But whatever, if this works, great! Anyway, thank your for the help. I think everything will go smoothly. Just gotta rectify the output on the secondary of the MAX253. \$\endgroup\$ – ARMATAV Apr 25 '15 at 15:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.